Tradition, Transmission, and Transformation from Second Temple

Tradition, Transmission, and Transformation from Second Temple Literature through Judaism and Christianity in Late Antiquity Studies on the Texts of the Desert of Judah

Edited by

George J. Brooke

Associate Editors

Eibert J. C. Tigchelaar Jonathan Ben-Dov Alison Schofield

volume 113

The titles published in this series are listed at brill.com/stdj Tradition, Transmission, and Transformation from Second Temple Literature through Judaism and Christianity in Late Antiquity

Proceedings of the Thirteenth International Symposium of the Orion Center for the Study of the Dead Sea Scrolls and Associated Literature

Jointly Sponsored by the Hebrew University Center for the Study of Christianity, 22–24 February, 2011

Edited by

Menahem Kister, Hillel I. Newman, Michael Segal, and Ruth A. Clements

LEIDEN | BOSTON Library of Congress Cataloging-in-Publication Data

Orion Center for the Study of the Dead Sea Scrolls and Associated Literature. International Symposium (13th : 2011) Tradition, transmission, and transformation from Second Temple literature through Judaism and Christianity in late antiquity : proceedings of the Thirteenth International Symposium of the Orion Center for the Study of the Dead Sea Scrolls and Associated Literature, jointly sponsored by the Hebrew University Center for the Study of Christianity, 22–24 February, 2011 / by Menahem Kister, Hillel Newman, Michael Segal, and Ruth A. Clements. pages cm. — (Studies on the texts of the desert of Judah, ISSN 0169-9962) Includes indexes. ISBN 978-90-04-27408-2 (hardback : acid-free paper) — ISBN 978-90-04-29913-9 (e-book) 1. Judaism— History—Post-exilic period, 586 BC–210 AD—Congresses. 2. Judaism—History—Talmudic period, 10–425—Congresses. 3. Church history—Primitive and early church, ca. 30–600—Congresses. 4. Civilization, Greco-Roman—Congresses. 5. Apocryphal books—Criticism, interpretation, etc.— Congresses. 6. Rabbinical literature—History and criticism—Congresses. 7. Dead Sea scrolls—Congresses. I. Kister, Menahem, editor. II. Newman, Hillel, editor. III. Segal, Michael , editor. IV. Clements, Ruth, editor. V. Title.

BM176.O75 2015 296.1—dc23

2015016654

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 0169-9962 isbn 978-90-04-27408-2 (hardback) isbn 978-90-04-29913-9 (e-book)

Copyright 2015 by Koninklijke Brill nv, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi and Hotei Publishing. 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 xiii

Parabiblical Traditions and Their Use in the Palaea Historica 1 William Adler

Outsider Impurity: Trajectories of Second Temple Separation Traditions in Tannaitic Literature 40 Yair Furstenberg

No Angels before the World? A Preexistence Tradition and Its Transformations from Second Temple Literature to Early Piyyuṭ 69 Yehoshua Granat

Pious Long-Sleepers in Greek, Jewish, and Christian Antiquity 93 Pieter W. van der Horst

Remnants of a Pharisaic Apologetic Source in Josephus and in the Babylonian Talmud 112 Tal Ilan and Vered Noam

Windy and Fiery Angels: Prerabbinic and Rabbinic Interpretations of Psalm 104:4 134 Yaakov Kaduri

Hellenistic Jewish Writers and Palestinian Traditions: Early and Late 150 Menahem Kister

The Severus Scroll Variant List in Light of the Dead Sea Scrolls 179 Armin Lange

Where is the Lost Ark of the Covenant? The True History (of the Ancient Traditions) 208 Chaim Milikowsky vi contents

Satan’s Refusal to Worship Adam: A Jewish Motif and Its Reception in Syriac Christian Tradition 230 Sergey Minov

Stars of the Messiah 272 Hillel I. Newman

Retelling Biblical Retellings: Epiphanius, the Pseudo-Clementines, and the Reception-History of Jubilees 304 Annette Yoshiko Reed

Why is “A” Placed Next to “B”? Juxtaposition in the Bible and Beyond 322 Avigdor Shinan and Yair Zakovitch

The Reception and Reworking of Abraham Traditions in Armenian 343 Michael E. Stone

Index of Ancient Texts 361 Index of Modern Authors 387 Preface

The Second Temple period was crucial to the formation of both rabbinic Judaism and early Christianity, which share common roots in that semi- nal era. The origins of many literary genres, modes and traditions of biblical exegesis, halakhic rulings, and theological notions found in later Jewish and Christian writings can be traced back to that formative period. The Thirteenth International Orion Symposium (February 22–24, 2011) was convened in order to explore the dynamics by which some of these earlier texts and traditions were transmitted to later generations and taken up in other contexts, where they were both preserved and adapted to new cultural and religious settings. The fruits of that symposium are presented in this volume. These explorations span the period from biblical and Second Temple times to the Middle Ages. The papers address compositions ranging from Second Temple Apocrypha and Pseudepigrapha, the Dead Sea Scrolls, and Hellenistic Jewish writings, to rabbinic midrash, halakhic texts, piyyuṭ, and patristic literature; some deal as well with medieval works of Jews, Christians, and Moslems. In many senses these rich and diverse worlds must be seen as parts of a larger continuum, but at the same time we must not blur distinctions or fail to acknowledge the unique features of each culture and corpus of material. Many works that originated in the Second Temple period (such as the books of the Apocrypha and the so-called Pseudepigrapha, the oeuvres of Philo and Josephus, and fragments of Hellenistic Jewish authors) continued to be transmitted in Christian circles and cited by Christian authorities; however, these works were generally unknown to Jewish literature of late antiquity and the early Middle Ages. With few exceptions (notably Ben Sira, the Aramaic Levi Document and the Damascus Document, medieval fragments of which have been discovered in the Cairo Genizah), direct transmission of texts among Jews of works from the Second Temple period is extremely rare. On the other hand, classical rabbinic literature (and to a lesser extent medieval rabbinic literature) often contains traditions that originated in the Second Temple period. Such traditions can likewise be found in early Christianity. Therefore, in order to study the afterlife of Jewish literature of the Second Temple period, one must confront two distinct phenomena: the transmission of texts on the one hand; and the transmission and transformation of traditions on the other. These are not discrete processes, but complementary ones. Scrutiny of both phenomena is crucial for exploring the roots of rabbinic Judaism and early Christianity: rabbinic traditions often provide the key to understanding obscure passages in viii preface works of the Second Temple period, just as the earlier texts often demonstrate the pre-Christian origins of those traditions.

The studies included here analyze different aspects of numerous compositions from diverse periods and contexts; nevertheless, they may be broadly catego- rized according to their content and methodology. Many focus on traditions and their transformation in subsequent contexts, particularly within rabbinic literature. Some of these traditions emerged from the exegesis of particular biblical verses, both narrative and halakhic; others are historical or quasi-historical in nature, though even these latter may be colored by biblical exegesis. Several of the papers highlight the role of shared or contrasting exegetical principles and methods. Taken together, these studies demonstrate that the rabbinic corpus, together with Christian and extrarabbinic Jewish sources, frequently attests to much earlier traditions. That is, many of the traditions found in the later texts were neither created by the rabbis and their contemporaries, nor did they necessarily emerge from the milieux of the specific sages to whom they are attributed or necessarily originate in the place and period of the redaction of the individual compilations. Of course, these traditions did continue to evolve in the course of their transmission; new meanings continued to accrue even for old traditions when put into new cultural and literary contexts. Avigdor Shinan and Yair Zakovitch contend that biblical interpretation must be seen as a continuum extending from passages within the Bible itself to the midrash, and that the implicit methods of inner-biblical exegesis can be illuminated by the more explicit techniques of later material. They investigate a specific exegetical practice: the attribution of significance to literary juxtaposition with the biblical text. Beginning with the explicit usage of this technique in rabbinic literature, they trace it back to implicit occurrences in the writings of the Second Temple period; they ultimately find the roots of this practice in the literary structure of the biblical texts themselves. Yaakov Kaduri (James L. Kugel) analyzes both explicit and implicit exegesis of a biblical verse (Ps 104:4) found in ancient Jewish sources. He demonstrates the continuity of the perception of the nature of angels which is derived from the interpretation of this verse, from the Pseudepigrapha and Qumran through rabbinic literature. Menahem Kister offers an “integrative study” of Hellenistic Jewish texts dealing with creation (that is, texts written in Greek and preserved in citations by the Church Fathers) and parallels in Hebrew and Aramaic from both Jewish (rabbinic midrash, piyyuṭ) and Samaritan sources. He seeks to demonstrate preface ix that common traditions and motifs underlie the divergent sources, and that, in spite of the essential differences and chronological disparities between the corpora, they are mutually illuminating. Yehoshua Granat, in a paper that is thematically connected to the contributions of Kaduri and Kister, examines traditions about the creation of the angels that are found in several Palestinian piyyuṭim and may be traced back to Second Temple literature. These traditions do not accord with mainstream rabbinic interpretation, with which they are contemporary. Chaim Milikowsky explores the history of motifs surrounding the disappearance of the Ark of the Covenant in apocryphal, pseudepigraphic, and rabbinic sources. He notes that although the problem of the disappearance of the Ark is itself exegetically derived, and although all of the traditions surrounding this issue share the same concerns, rabbinic sources offer solutions that differ from those of earlier, nonrabbinic texts. He suggests that this stems from fundamentally different approaches to biblical interpretation. Tal Ilan and Vered Noam discuss the origins of and relationships between parallel historical traditions found in Josephus and the Babylonian Talmud. They contend that two aggadot in the Babylonian Talmud depicting historical events in the time of Alexander Jannaeus are rooted in traditions of the Second Temple period, and are in fact passages of Pharisaic origin (possibly from a common Pharisaic source). In this case, rabbinic traditions in later works may shed light on the earlier material Hillel Newman discusses the eschatological motif of stars as omens of salvation, examining the origins of this motif in biblical exegesis, its development in apocalyptic tradition, and its manifestation in various historical circumstances. The sources considered range from Qumran literature and the New Testament to medieval midrash and Christian works of late antiquity. Special attention is devoted to the messianic significance of the star in the Bar Kokhba revolt. In addition to literary sources, the paper deals with numismatic evidence from the Bar Kokhba period and with later iconographic material. Turning to halakhic traditions and conceptions, Yair Furstenberg identifies an early stratum of rabbinic halakhah concerning purity which is similar to the laws of separation from impurity in Qumran legal texts. The paper raises the question of continuity and discontinuity between halakhic views of various groups in the Second Temple period and rabbinic literature. Armin Lange looks at a unique medieval rabbinic document: a brief list of textual variants to the Torah, attributed in its introduction to a Torah scroll originating in pre-Destruction Jerusalem. In an attempt to assess the textual type of this scroll and its relationship to the Masoretic text, he compares this list to readings found in the Dead Sea documents and in other biblical versions. x preface

Five papers in this volume deal in large measure with the transmission and reception of Second Temple texts and traditions, directly or indirectly, in Christian literature. Two of these papers deal with the transmission of specific traditions over long periods of time and through changing cultural contexts: Pieter van der Horst explores the motif of “pious long-sleepers” from the Hellenistic period to the medieval Legenda aurea. The focus of his paper is the Paralipomena Ieremiou—apparently a marginally reworked Christian version of a Jewish text—and its parallels in the rabbinic stories of Ḥoni the Circle- Maker. He compares these stories to the Christian legends of the Seven Sleepers of Ephesus and argues that the Christian legend is derived from the earlier Jewish tradition. The paper deals also with the relation of Graeco-Roman tales of “long-sleepers” to Jewish and Christian literature. Sergey Minov, in a paper thematically related to Kister’s contribution, describes the transmission and development in Syriac Christian compositions of motifs that echo Jewish traditions of the Second Temple period connected with the fall of Satan and his refusal to venerate Adam. Minov describes the transformation of these motifs from late antiquity through the medieval period and demonstrates how they are used for inner-Christian and Christian anti-Islamic polemics throughout the Middle Ages. Minov’s paper addresses some of the same issues also raised in the other three papers devoted specifically to Christian literature. These articles are valuable for their substantive discussions of the avenues by which Second Temple traditions sometimes became part of the fabric of late antique and medieval Christian cultures. Articulating the Christian reception histories of these ancient Jewish traditions contributes to a more nuanced understanding of the relationships between the ancient materials and their subsequent manifestations in late antiquity and the Middle Ages. Although, as stated at the outset, Jewish and Christian writings adapted material from the Second Temple period through distinctive pathways, both corpora display—each in its own way—a similar interplay of continuity, creativity, and transformation. Annette Yoshiko Reed investigates the appropriation of Jubilees by fourth-century Christian authors, particularly Epiphanius and the Pseudo- Clementines. She argues that the reception and explicit citation of Jubilees traditions at that time should be understood within the context of contemporary discussions of canon and heresy—questions that deeply concerned Christian writers of the fourth century. William Adler analyzes parabiblical traditions in a Byzantine composition of the ninth or tenth century, the Palaea Historica. He demonstrates that this Greek Christian retelling of the Bible should be seen not as a repository of lost Jewish documents with merely a Christian “overlay,” as has been suggested by preface xi some, but as a work incorporating elements from a variety of extant sources and traditions over the course of its composition and transmission, occasionally adding very early traditions at late stages of its development. The most appropriate perspective from which to assess this composition is within its medieval Christian context. Michael Stone surveys narratives concerning Abraham in Armenian Christian literature prior to the tenth century CE and assesses the character of these legends. As he demonstrates, the Armenian Abraham saga shows some resemblance to Syriac and Greek sources and as a whole is a Christian composition, some elements of which reflect traditions that had emerged already in the Second Temple period. The author’s stance is that the interrelationships among these medieval texts is best explained by treating them as “textual clusters,” without imposing rigid paradigms of “conventional stemmatic analysis.”

Although several of the papers are linked to one another by the discussion of common themes or genres, they are best characterized by their diversity. In view of this very diversity, we have chosen to arrange the individual contributions alphabetically by author, rather than to impose topical divisions that might obscure their complexity. The focus on different texts, genres, eras, and cultures has perforce involved the employment of a variety of methodologies and led to a wide array of conclusions concerning the trajectories of transformation and development taken by earlier traditions. But this confluence of different approaches and results, evident at the symposium itself and in this volume, yields—we hope—a richer understanding of the interrelationships among these vast bodies of literature and their attendant cultural contexts. The implications of these studies are far-reaching. For example, for the historian of Judaism, the question of continuity and change after 70 CE is of considerable importance; in some of the papers in this volume the reader will encounter examples of both. In some cases, affinity may be found even across sectarian or cultural lines. Elements found in early rabbinic literature are on occasion anticipated in texts from Qumran or in Hellenistic Jewish works. Such phenomena demonstrate broad lines of common heritage, and at the same time alert us to the unique features of the various groups. Needless to say, this is also true for defining the place of nascent Christianity within Second Temple Judaism and among its successors. The particular persistence of the textual heritage of Second Temple Judaism among Christians in later centuries also raises profound questions concerning the avenues by which Second Temple literature continued to exert its influence on Jews. As stated above, rabbinic literature betrays little direct knowledge of such works. This does not mean that Jews, including rabbinic sages, xii preface could not have possessed any of these books and known them first hand. Yet the disparity between the proportions of documentation among Jews and among Christians is so striking that it suggests that Christians played a considerable role in “reintroducing” Jews to Second Temple literature, if only in translation. This is explicitly the case in the Middle Ages, but the earlier stages of the process remain shrouded in mystery. While each article in this book may be read as a self-contained study, when taken together, the papers reveal a rich tapestry of various facets of the afterlife and continuity of traditions and texts of the Second Temple period. They demonstrate the benefits of such wide-ranging studies, both for the investigation of earlier compositions and traditions and for a more nuanced understanding of subsequent Jewish and Christian literature. At the same time, they also illustrate the methodological considerations and constraints necessary to avoid the pitfalls of facile comparisons when analyzing this changing continuum. The dynamic processes described in these studies will hopefully point the way towards potential avenues for further research of both Second Temple literature and subsequent Jewish and Christian works.

We are grateful to the Hebrew University’s Center for the Study of Christianity and its successive directors, David Satran and Brouria Bitton-Ashkelony, for their cooperation in the planning and realization of this symposium. As always, we extend our heartfelt appreciation to the Orion Foundation, the Sir Zelman Cowen Universities Fund, and the Hebrew University of Jerusalem for their support of the Orion symposia, and to Ms. Ariella Amir and the Orion staff for ensuring the efficient management of the symposium program. Finally, we wish to thank George Brooke, editor of the STDJ series, and the editorial staff of Brill Academic Press, especially Mattie Kuiper and Maaike Langerak, for their assistance in guiding the volume through production to completion.

Menahem Kister, Hillel Newman, Michael Segal, and Ruth Clements Jerusalem, February 2015 / Adar 5775 Abbreviations

AB Anchor Bible AGJU Arbeiten zur Geschichte des antiken Judentums und des Urchristentums AJEC Ancient Judaism and Early Christianity AJS Review Association for Jewish Studies Review AJSL American Journal of Semitic Languages and Literature AnBib Analecta biblica AnBoll Analecta Bollandiana ANF Ante-Nicene Fathers. Ed. A. Roberts, J. Donaldson; rev. A. C. Coxe. 10 vols. Buffalo, N.Y.: Christian Literature Publishing, 1885–96. Repr. Peabody, Mass.: Hendrickson, 1994 ANRW Aufstieg und Niedergang der römischen Welt: Geschichte und Kultur Roms im Spiegel der neueren Forschung. Ed. H. Temporini and W. Haase. Berlin: de Gruyter, 1972– APOT The Apocrypha and Pseudepigrapha of the Old Testament. Ed. R. H. Charles. 2 vols. Oxford: Clarendon, 1913 BAR Biblical Archaeology Review BETL Bibliotheca ephemeridum theologicarum lovaniensium BHM Bet Ha-Midrasch. Ed. A. Jellinek. 6 vols. 2d edition. Jerusalem: Wahrmann, 1967 (In Hebrew) BJRL Bulletin of the John Rylands University Library of Manchester BJS Brown Judaic Studies BKAT Biblischer Kommentar, Altes Testament BSOAS Bulletin of the School of Oriental and African Studies ByzArch Byzantinisches Archiv BZAW Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft CBET Contributions to Biblical Exegesis and Theology CBQ Catholic Biblical Quarterly CCSA Corpus Christianorum: Series apocryphorum CCSG Corpus Christianorum: Series graeca. Turnhout: Brepols, 1977– CCSL Corpus Christianorum: Series latina. Turnhout: Brepols, 1953– CEJL Commentaries on Early Jewish Literature CELAMA Cultural Encounters in Late Antiquity and the Middle Ages CFHB Corpus fontium historiae byzantinae CSCO Corpus scriptorum christianorum orientalium CSEL Corpus scriptorum ecclesiasticorum latinorum CSHB Corpus scriptorium historiae byzantinae xiv abbreviations

DACL Dictionnaire d’archéologie chrétienne et de liturgie. Ed. F. Cabrol. 15 vols. Paris: Letouzey et Ané, 1907–1953 DJD Discoveries in the Judaean Desert DSD Dead Sea Discoveries EDSS Encyclopedia of the Dead Sea Scrolls. Ed. L. H. Schiffman and J. C. VanderKam. 2 vols. Oxford and New York: Oxford University Press, 2000 EncJud Encyclopaedia Judaica. 16 vols. Jerusalem: Encyclopaedia Judaica, 1972 EPRO Etudes préliminaries aux religions orientales dans l’empire romain FAT Forschungen zum Alten Testament FC Fathers of the Church. Washington, D.C.: Catholic University of America Press, 1947– GCS Die griechische christliche Schriftsteller der ersten Jahrhunderte HAR Hebrew Annual Review HCMR History of Christian–Muslim Relations HSCP Harvard Studies in Classical Philology HSS Harvard Semitic Studies HTR Harvard Theological Review HTS Harvard Theological Studies HUCA Hebrew Union College Annual HUCM Hebrew Union College Monographs ICC International Critical Commentary IEJ Israel Exploration Journal IOS Israel Oriental Studies JAAR Journal of the American Academy of Religion JACSup Supplements to Jahrbuch für Antike und Christentum JAJSup Journal of Ancient Judaism Supplements JAOS Journal of the American Oriental Society JBL Journal of Biblical Literature JCP Jewish and Christian Perspectives JECS Journal of Early Christian Studies JJS Journal of Jewish Studies JQR Jewish Quarterly Review JSHRZ Jüdische Schriften aus hellenistisch-römischer Zeit JSJ Journal for the Study of Judaism in the Persian, Hellenistic and Roman Periods JSJSup Journal for the Study of Judaism: Supplement Series JSOTSup Journal for the Study of the Old Testament: Supplement Series JSP Journal for the Study of the Pseudepigrapha JSPSup Journal for the Study of the Pseudepigrapha: Supplement Series JSQ Jewish Studies Quarterly abbreviations xv

JSRC Jerusalem Studies in Religion and Culture JSS Journal of Semitic Studies JTS Journal of Theological Studies LCL Loeb Classical Library MGWJ Monatschrift für Geschichte und Wissenschaft des Judentums MH Medieval Armenian Literature (Matenagirkʿ Hayocʿ) (cited by century and page). Antelias: Armenian Catholicosate Press NHS Nag Hammadi Studies NTS New Testament Studies NovT Novum Testamentum NovTSup Supplements to Novum Testamentum OBO Orbis biblicus et orientalis OCD Oxford Classical Dictionary. Ed. S. Hornblower and A. Spawforth. 3d ed. Oxford: Oxford University Press, 1996 OLA Orientalia lovaniensia analecta OrChr Oriens christianus OrChrAn Orientalia christiana analecta OTL Old Testament Library OTP Old Testament Pseudepigrapha. Ed. J. H. Charlesworth. 2 vols. Garden City, N.Y.: Doubleday, 1983–1985 OtSt Oudtestamentische Studiën PACS Philo of Alexandria Commentary Series PG Patrologia graeca [= Patrologiae cursus completus: Series graeca]. Ed. J.-P. Migne. 162 vols. Paris, 1857–1886 PL Patrologia latina. Ed. J.-P. Migne. 217 vols. Paris, 1844–1864 PO Patrologia orientalis PTS Patristische Texte und Studien PVTG Pseudepigrapha Veteris Testamenti Graece RAC Reallexikon für Antike und Christentum. Ed. T. Kluser et al. Stuttgart: Hiersemann, 1950– RB Revue biblique REJ Revue des études juives RevQ Revue de Qumrân RGG Religion in Geschichte und Gegenwart. Ed. H. D. Betz, D. S. Browning, B. Janowski and E. Jüngel. 8 vols. 4th ed. Tübingen: Mohr Siebeck, 2004 RGRW Religions in the Graeco-Roman World SBLDS Society of Biblical Literature Dissertation Series SBLEJL Society of Biblical Literature Early Judaism and its Literature SBLMasS Society of Biblical Literature Masoretic Studies SBLMS Society of Biblical Literature Monograph Series xvi abbreviations

SBLRBS Society of Biblical Literature Resources for Biblical Study SBLSCS Society of Biblical Literature Septuagint and Cognate Studies Series SBLTCS Society of Biblical Literature Text-Critical Studies SBLTT Society of Biblical Literature Texts and Translations SBLTTPS Society of Biblical Literature Texts and Translations: Pseudepigrapha Series SBLWGRW Society of Biblical Literature Writings from the Greco-Roman World SC Sources chrétiennes. Paris: Cerf, 1943– ScrHier Scripta Hierosolymitana SHJ Studies in Hellenistic Judaism SHR Studies in the History of Religions (supplement to Numen) SJ Studia judaica SJLA Studies in Judaism in Late Antiquity SNTSMS Society for New Testament Studies Monograph Series SPhM Studia Philonica Monographs STAC Studien und Texte zu Antike und Christentum STDJ Studies on the Texts of the Desert of Judah StPB Studia post-biblica SVTP Studia in Veteris Testamenti pseudepigraphica TAPA Transactions of the American Philological Association TBN Themes in Biblical Narrative TDNT Theological Dictionary of the New Testament. Ed. G. Kittel and G. Friedrich. Trans. G. W. Bromiley. 10 vols. Grand Rapids: Eerdmans, 1964–1976 TS Theological Studies TSAJ Texte und Studien zum antiken Judentum / Texts and Studies in Ancient Judaism TUGAL Texte und Untersuchungen zur Geschichte der altchristlichen Literatur VC Vigiliae christianae VCSup Supplements to Vigiliae christianae VT Vetus Testamentum VTSup Vetus Testamentum Supplements WBC Word Biblical Commentary WUNT Wissenschaftliche Untersuchungen zum Neuen Testament ZAC Zeitschrift für Antikes Christentum/Journal of Ancient Christianity ZAW Zeitschrift für die alttestamentliche Wissenschaft ZDMG Zeitschrift der Deutschen Morgenländischen Gesellschaft Parabiblical Traditions and Their Use in the Palaea Historica

William Adler

For students of Second Temple Judaism, the Palaea Historica (“Old Testament History”), an anonymous work composed sometime in the late ninth to early tenth centuries, is probably most noteworthy for its copious assortment of parabiblical legends about biblical personalities.1 The conventional title, Palaea Historica, is meant to distinguish this work from another document, preserved only in Slavonic, known as the Palaea Interpretata (Tolkovaja Paleja [“Explanatory Palaea”]).2 While the two works contain overlapping material, they differ in both character and purpose. The Palaea Interpretata, a more theological and polemical work, interprets scripture from an avowedly Christian perspective. Although the Palaea Historica (hereafter, Palaea) features a

1 For the Greek text of the Palaea cited in this essay, see A. Vassiliev, Anecdota Graeco– Byzantina (Moscow: Universitas Caesarea, 1893), 1:188–292; for the English translation, see W. Adler, Palaea Historica (“The Old Testament History”), in Old Testament Pseudepigrapha: More Noncanonical Scriptures (ed. R. Bauckham, J. R. Davila, and A. Panayotov; Grand Rapids: Eerdmans, 2013), 1:585–672; the introduction to this translation touches more briefly upon some of the topics which are here discussed at greater length. References to the Palaea given here provide chapter and verse number from my English translation, with the page number from Vassiliev in parenthesis. For a discussion of Jewish sources and traditions preserved in the Palaea, see principally D. Flusser, “Palaea Historica: An Unknown Source of Biblical Legends,” in Studies in Aggadah and Folk-Literature (ed. J. Heinemann and D. Noy; ScrHier 22; Jerusalem: The Hebrew University Magnes Press, 1971), 48–79; see also Vassiliev, Anecdota Graeco–Byzantina, xlv–xlvii; S. Lieberman, “Zenihin,” Tarbiz 42 (1972–1973): 42–54 (in Hebrew). For a bibliography of recent studies of the Palaea, see J. Reinhart, “Die älteste Bezeugung der Historischen Paläa in slavischer Übersetzung (Cod. slav. Vindob. Nr. 158),” Прилози за књижевност, језик, историју и фолклор 73 (2007): 61–65. 2 Among the titles in the Greek witnesses: 1) Cod. Ottob. gr. 205: Ἀπὸ τοῦ παλαιοῦ (“From the Old Testament”); 2) Paris Bib. Nat. gr. 37: Ἐκ τῶν παλαιῶν ἱστοριῶν μερικὴ διήγησις (“Partial Narrative from the Old Testament Histories”; 3) Venice, Bibl. Naz. Marc. gr. 501: Παλαία γνῶσις. Ἱστορίαι ἀπὸ τῶν παλαιῶν (“Old Testament Knowledge: Stories from the Ancient Books (= Old Testament?)”; 4) Panaghia gr. 68: Ἱστορίαι τινὲς ἐκ τοῦ παλαιοῦ (“Some Stories from the Old Testament”); 5) Vindob. hist. gr. 119: Ἱστορία παλαιοῦ περιέχων (sic) ἀπὸ τοῦ Ἀδάμ (“History of the Old Testament Extending from Adam”).

© koninklijke brill nv, leiden, ��5 | doi ��.��63/9789004299139_002 2 Adler scattering of biblical figures and events read as types of Christ, its narrative is comparatively free of overtly christological interpretation.3 The rather abrupt ending of the work raises at least the possibility that the author was not able to see his project to a satisfactory conclusion. In the absence of a complete critical edition of the work, sweeping judgments about its style and contents must remain provisional. The overall impression created by Vassiliev’s text is that, while often ungrammatical, the Greek style is not atypical of popular prose literature of the middle Byzantine age. Questionable Syriac etymologies do not reveal very much about the author’s acquaintance with the Syriac language.4 A creedal statement at the beginning of the work (1–2 [188]) and citations from well-known figures of the Christian East establish the Palaea firmly within the orthodox Eastern tradition. Overt religious polemic in the Palaea is rare, confined mainly to the very beginning and end of the work. The author is sharply critical of opposing views concerning the fall of Lucifer (3.8 [189]) and the nature of the tree of the life (5.1–2 [190]). A strongly worded condemnation of the claim that Cain was born of intercourse between Eve and Satan identifies the group propagating this interpretation as the “Phundaitae” (7.5 [191]), a dualist heresy with concentrations of adherents in Bulgaria, the Balkans, and Western Asia Minor, often associated with the Bogomils.5 In the broader context of biblical interpretation in

3 For christological readings, see, e.g.: 5.5 (190): the Tree of Life as “Christ and the Holy Spirit”; 39.1 (210): Abraham’s tithe to Melchizedek as prefiguring Christ’s offering of a pair of doves; 46.7 (213): Melchizedek’s priesthood as a type of Christ’s priesthood; 67.11 (226): Jacob’s blessing of Joseph’s sons as a prefigurement of the cross; 103.8–11 (246): the bronze serpent in the wilderness as a prefigurement of the cross; 115.1–4 (254): Balaam’s prediction of the coming of Christ; 119.4 (256): Moses’s striking of the rock as a figure of Christ; 122.2 (258); 141.15 (272): Jael’s tent peg as a prefigurement of the cross. For typological interpretation, see also 91 (238), where the author interprets the twelve springs and seventy palm trees at Elim (Exod 15:27) as prefigurations of Jesus’s twelve apostles and seventy disciples (cf. Luke 10:1). 4 For Syriac etymologies, see 27.66 [203]: “Haran in the language of the Syrians means ‘excre- ment’ ”; 30.2 [205]: “Mabri (Mamre)” as “cold water” in Syriac. At 58.19 (221), the interpretation of “sabek” as “forgiveness” (and thus a prefiguration of the cross) assumes a derivation of the word from the Aramaic/Syriac root ŠBQ (“loose” or “forgive”). This familiar etymology would not have required an active knowledge of Syriac; for other sources attesting the same etymology, see, for example, Melito, Fragmenta 12 (ed. O. Perler; SC 123; Paris: Cerf, 1966); and ps.-Athanasius, Quaestiones in scripturam sacram (PG 28.740BC). For arguments in favor of the author’s familiarity with Syriac, see Flusser, “Palaea Historica,” 54–55, 78; for an opposing view, cf. Lieberman, “Zenihin,” 52–54. 5 For general background on the Phundaitae, also known as the “Phundagiagitae,” see D. Obolensky, The Bogomils: A Study in Balkan Neo-Manichaeism (Cambridge: Cambridge University Press, 1948), 177–83; G. Ficker, Die Phundagiagiten (Leipzig: Barth, 1908). On the Parabiblical Traditions and their Use in the Palaea Historica 3

Byzantium, there were plenty of opportunities for anti-Jewish invective. To the author’s credit, however, he mostly resists the temptation. The most conspicuous exception is the mildly confrontational language tacked on somewhat awkwardly to the end of the work (169.2–8 [291–92]).6 While colorful and often entertaining, the representations of biblical personalities and the assorted parabiblical legends preserved in the work are sometimes highly idiosyncratic. The unnamed servant whom Abraham sends to find a wife for Isaac bears the improbable name “Andrew” (59.1 [222]; cf. Gen 24:2). After Joab refuses his request to conduct a census (2 Samuel 24), King David assigns the task to “Achria,” a figure unknown in 2 Samuel’s narrative of the event (161.4 [286]). In its narrative of the triumphal career of warrior and judge Deborah, the Palaea recalls how she rescues the city of Jerusalem from destruction first by intoxicating the Persian king Artasyris with her beauty and strong drink and then beheading him (145–46 [275–76]). While this story of Deborah’s exploits against the Persians has no parallel in the Judges narrative about her (cf. Judges 4–5), an attentive reader would have little difficulty recognizing the similarities between her triumph and Judith’s victory over Holophernes, the commander of Nebuchadnezzar’s army (cf. Judith 11–13). A second Jewish leader named “Endor” assassinates Got, another oppressive Persian king, during the period of the elders after Joshua. After offering gifts to the Persian king, he arranges a private meeting with the king, stabs him to death, plunders the Persians’ wealth and returns home in triumph (128–129 [262–63]).7 The Melchizedek of the Palaea is the real oddity, far removed from the exalted, even heavenly figure of older Jewish and Christian tradition.8 The orphaned son of a pagan king and devotee of the god Cronus, he lives alone for forty years in a cave on Mount Tabor. When Abraham meets him, the first thing he does, in obedience to divine instruction, is to shave Melchizedek’s beard and cut his unkempt hair and talon-like nails (32–38 [206–10]).9 One searches in vain in Jewish scriptures or parabiblical literature of the Second Temple period for the origins of the legend about Melchizedek the

birth of Cain from Satan’s (Satanael’s) seduction of Eve, as taught in dualist movements of the Middle Ages, see M. Loos, Dualist Heresy in the Middle Ages (Prague: Akademia, 1974), 86, 93 n. 24, 137. 6 See also 67.11 [226], where the author describes the Jews as “a people who worship the law (νομολάτρης).” See further below, pp. 28–29. 7 Cf. Cod. Scorialensis Ψ.11.20, where he is named “Gog” (perhaps the original reading). 8 For Melchizedek in early Jewish and Christian tradition, see F. L. Horton, The Melchizedek Tradition: A Critical Examination of the Sources to the Fifth Century A.D. and in the Epistle to the Hebrews (SNTSMS 30; Cambridge: Cambridge University Press, 1976). 9 For discussion, see below, pp. 16–19. 4 Adler wild man, Endor the Jewish hero, or the Persian kings Got and Artasyris. We are on a surer footing, however, in tracking down the prehistory of many other traditions found in the work. The Palaea’s narrative of the early years of Abraham rehearses the Jewish legend, made famous by the Book of Jubilees and recycled in Byzantine chronicles, about the patriarch’s estrangement from the pagan religious practices of his father. In its retelling of Abraham’s hospitality to the three angelic visitors at Mamre, the Palaea describes how the calf that Abraham had sacrificed and offered to them as a meal miraculously revived and returned to its mother (48–49 [215]). The story, whose purpose was self-evidently to reassure readers that angels had not physically consumed animal flesh, recalls a similar episode recounted in the Testament of Abraham (6), a work thought to date to the first or second century CE.10 In the third century, Origen had already suggested that the contest pitting Moses against the Egyptian magicians Jannes and Jambres described in 2 Tim 3:8, and the struggle between the archangel Michael and Satan over the body of Moses recounted in Jude 1:9 originated, respectively, in the Apocryphon of Jannes and Jambres and the Assumption of Moses.11 As has often been suggested, the Palaea’s pres- ervation of variant versions of these same two legends (75.10 [232]); (121.6–8 [259]) raises the possibility that its author learned about them directly, and not through the witness of the New Testament.12 There is, in principle, nothing wrong in sifting through the Palaea for vestiges of Jubilees, the Testament of Abraham, and other early Jewish and Christian

10 See D. C. Allison, Testament of Abraham (Berlin: de Gruyter, 2003), 164–67; J. Kugel, Traditions of the Bible: A Guide to the Bible as It Was at the Start of the Common Era (Cambridge, Mass.: Harvard University Press, 1998), 344; Flusser, “Palaea Historica,” 60. The apparent consumption of food by the angels was a problem noted and discussed by other ancient authors as well. See, for example, Theodoret, Quaestiones in Octateuchum 70 (ed. N. Fernández Marcos and A. Sáenz-Badillos; Textos y Estudios “Cardenal Cisneros” 17; Madrid: Poliglota Matritense, 1979), 65.23–66.8; Theodoret also rejects the belief that the angels physically consumed food, on the grounds that “they did not have bodies” (οὔτε γὰρ εἶχον σώματα). 11 See Origen, Commentary on Matthew (on Matt 27:3) (ed. E. Klostermann; 3 vols., in 4; GCS 38, 40, 41; Leipzig: Hinrich, 1933–1955), 1:250.6–12,where he claims that 2 Timothy learned the names of Jannes and Jambres from the Book of Jannes and Jambres. For Origen’s ascription of Jude’s report to a “little treatise” (libellus) known as the Ascensio Moysi, see his De Principiis 3.2.1 (ed. P. Koetschau; GCS 22: Leipzig: Hinrich, 1913), 244. 16–19. This part of De Principiis is preserved only in Rufinus’s Latin translation. 12 On the Palaea as a witness to the Book of Jannes and Jambres, see A. Pietersma, The Apocryphon of Jannes and Jambres the Magicians (RGRW 119; Leiden: Brill, 1994), 160–61, 195. For the Palaea as a witness to the Assumption of Moses or some related work, see further below, pp. 26–37. Parabiblical Traditions and their Use in the Palaea Historica 5 sources. As Flusser has amply illustrated, the project can yield interesting, albeit disputable, results. But without first situating the work in the context of middle Byzantine culture, the whole approach is bound to devolve into little more than ingenious speculation, especially when it comes to identifying otherwise “lost” older traditions and documents. Specifically, how much of this material originated in written works, how much in nonliterary sources (local legend, liturgy, and religious art), and how much was simply a creation of the author’s own imagination?

1 Composition and Sources

1.1 The Palaea and Parabiblical Tradition in Byzantium In the rare instances in which the author of the Palaea injects himself into the narrative, it is to explain and defend his selection or exclusion of subject matter. In each case, the author cites the requirements of brevity. He can only offer his readers an abbreviated account of the Mosaic law (101.1 [245]), because a “hurried account” (προδρομή) such as his must confine itself to matters suitable to the work’s larger purposes. He says much the same thing at the conclusion of his narrative of David’s reign. Although many other things could be told about the king, “in a summary account” (ἐν παραδρομῇ), these stories must be put to the side (161.17 [287]). Modern attempts to identify western counterparts to the Palaea have lik- ened it to the twelfth-century Historia Scholastica of Peter Comestor13 and to the Biblia Pauperum, an illustrated book of biblical stories popular in the late Middle Ages.14 More recent characterizations of the work as “a rewritten Bible,” “chronicle,” or “compendium” do not adequately represent the composition and contents of the work.15 The author was far from a mechanical

13 See M. R. James, Apocrypha Anecdota Second Series (Cambridge: Cambridge University Press, 1897), 157. 14 See Vassiliev, Anecdota Graeco-Byzantina, xlvi; F. H. Marshall, Old Testament Legends from a Greek Poem on Genesis and Exodus by Georgios Chumnos (Cambridge: Cambridge University Press, 1925), xxiii. 15 See, for example, J. R. Davila, The Provenance of the Pseudepigrapha: Jewish, Christian, or Other (JSJSup 105; Leiden: Brill, 2005), 151: “compendium of scriptural stories”; A. R. Michalak, Angels as Warriors in Late Second Temple Jewish Literature (WUNT 2/330; Tübingen: Mohr Siebeck, 2012), 249: “collection of biblical legends”; G. J. M. van Loon, “The Meeting of Abraham and Melchizedek and the Communion of the Apostles,” in Coptic Studies on the Threshold of a New Millennium (ed. M. Immerzeel and J. van der Vliet; OLA 133; Leuven: Peeters, 2004), 1376: “retelling of the Old Testament”; T. Rajak, “Moses in Ethiopia: Legend 6 Adler compiler of miscellaneous and unconnected stories.16 And for a work held to be a paraphrase, the overall treatment of biblical history is remarkably uneven. While virtually silent about the post-Davidic monarchy, the author discourses at length about relatively marginal, or even unknown, figures in the biblical narrative. The single biblical book that the author comes close to paraphrasing in full is Tobit, a book whose namesake the author consistently and unaccountably identifies as “Bit” (163–167 [289–90]). Nor can the Palaea be understood as a “commentary,” at least in the generally understood sense of that word. The author must have known the text of Genesis 1–11 very well, substantial portions of which he quotes almost verbatim. But these sections, which are never identified as “citations,” blend imperceptibly into the author’s own additions and elaborations. Because so much scriptural material is embedded into the narrative of the Palaea without attribution, a reader unfamiliar with the biblical text would be generally unable to disaggregate biblical content from the author’s elaborations. There is only one scriptural book that the author actually comments on with any consistency, and that is the book of Psalms.17 With the exception of Psalms and the Great Canon of Andrew of Crete, direct and attributed citations from both biblical and nonbiblical sources are surprisingly rare.18 In addition to Andrew, the few Jewish and Christian authors that are cited were familiar names in Byzantium: Josephus, Gregory of Nazianzus (7.3–4 [191]), John Chrysostom (53.10 [218]), and Theodore the Studite, abbot of the Stoudios monastery in Constantinople (53.9 [218]).19 But unverifiable citations from even these well-known authorities in the Byzantine church hardly inspire confidence in the author’s direct knowledge of or access to older and more

and Literature,” in eadem, The Jewish Dialogue with Greece and Rome: Studies in Cultural and Social Interaction (Leiden: Brill, 2001), 258: “a Byzantine chronicle”; D. Flusser, “Palaea Historica,” 78–79: “rewritten Bible . . . a mixture of creative fabulation and queer fantasy.” 16 See below, pp. 24–32. 17 See below, n. 63. 18 The handful of biblical quotations would not have required close attention to the biblical text. At 144.17–18 [275], the author quotes Jer 31:15; the reading given is actually the text of Jeremiah cited in Matt 2:18: Ῥαχιὴλ κλαίουσα τὰ τέκνα αὐτῆς. The author also quotes directly from Deut 32:13 (119.8 [257]) and 1 Sam 2:1 (138. 6 [270]). However, these latter two passages would have been known to the author through Byzantine hymnody; see below pp. 19–20. 19 The Palaea’s single reference to Theodore the Studite (759–826; see following note) also provides a terminus post quem for the work. Parabiblical Traditions and their Use in the Palaea Historica 7

recherché Jewish sources, some of which pre-dated him by over a millennium. 20 An examination of the transmission and survival of parabiblical material in Byzantium of the ninth century will only heighten our doubts. Middle Byzantine sources roughly contemporary with the Palaea—most notably universal chronicles and kindred works—represent a rich and largely untapped repository of legends and traditions originating in parabiblical literature of Second Temple Judaism. With but few exceptions, however, none of the authors who cited these sources had direct access to them, depending instead on epitomes, paraphrases, and collections of excerpts, with long and often tortuous prior histories of transmission. A revealing statement by George Syncellus, a chronicler of the early ninth century, shows how this winnowing out process helped to mainstream works that had long before been proscribed as apocryphal. Because of their adulteration by Jews and heretics, he writes, readers should be satisfied only with preselected and approved excerpts.21 His warning could be received in two ways: on the one hand, as a deterrent to readers interested in learning more about the fuller contents of these works; and on the other, as validation of those excerpts found to conform with and promote orthodox doctrine. The stages in this process of “normalization” are probably most visible in the case of the Book of Jubilees, a work whose use in Christian historiography probably extends back to the chronicle of Julius Africanus (ca. 221).22 With

20 According to the Palaea (7.3–4 [191]), Gregory the Theologian stated that “Jesus was baptized at age thirty because of the sin of Adam at thirty years.” Our author identifies his source as “Yesterday on the Illustrious Day of the Holy Lights”—apparently in reference to first line of Gregory of Nazianzus’s In sanctum baptisma (Orat. 40). Although this work does describes Jesus’s baptism at age thirty (PG 36:352A, 400C), it does not mention Adam’s age when he first had intercourse with Eve. At 53.10 [218], the Palaea states that in his work “Concerning the consumption of wine,” John Chrysostom describes how “Lot became a husband to his daughters because of his drunkenness. And the man was both father and grandfather.” The quotation probably originates in ps.-Basil of Caesarea, Sermo 11 (sermo asceticus et exhortatio de renuntiatione mundi): “Lot became a husband to his daughters, his own son-in-law and father-in-law, a father and grandfather” (PG 31:640CD); cf. Vassiliev, Anecdota Graeco–Byzantina, xlvii. The Palaea also ascribes to Theodore the Studite the words: “No one who drinks water acts foolishly, but Noah appeared naked, after trying wine. And Lot produced offspring of evil as a result” (53.9 [218]). The quotation could not be verified. 21 George Syncellus, Ecloga Chronographica (ed. A. A. Mosshammer; Leipzig: Teubner, 1983), 27.11–12. 22 See W. Adler, “The Chronographiae of Julius Africanus and its Jewish Antecedents,” ZAC 14 (2011): 510–13; H. Gelzer, Sextus Julius Africanus und die byzantinische Chronographie (2 vols. in 1; Leipzig: Teubner, 1898; repr. New York: Burt Franklin, 1968), 2:293–94. For a 8 Adler the single exception of Josephus, Jubilees was the most widely used historical source from Second Temple Judaism. In the proem to his chronicle, and without qualification, George Cedrenus (eleventh century) names it as one of his chief sources, along with ecclesiastical historians and “other books” (ἀπὸ τῆς Λεπτῆς Γενέσεως οὐκ ὀλίγα συλλέξαντες καὶ ἀπὸ ἐκκλησιαστικῶν ἱστοριῶν καὶ ἀφ’ ἑτέρων βιβλίων).23 Over the course of its long transmission, excerpts from the work experienced the expected adaptation and deformation. Copyists were not always attentive in differentiating Genesis from Jubilees, the latter work known to Byzantine authors as the Little Genesis (ἡ Λεπτὴ Γένεσις), Details of Genesis, or the Apocalypse of Moses. When a chronicler or a commentator introduced a passage from Jubilees with the words “Moses says,” a reader could mistakenly assume that the author was speaking of the biblical Genesis.24 At some stage in its transmission, a substantial body of material originating in Jubilees began to circulate, suspiciously, under the less controversial name of Josephus. Whether the fusion was deliberate or not, the result was a net plus for Jubilees’ standing; on the basis of the parallels they discerned between the two works, Byzantine chroniclers quite naturally assumed that Josephus borrowed material from Jubilees.25 Even chroniclers who were wary about the standing of apocryphal books found it difficult to disentangle Jubilees traditions from parallel material in Genesis and Josephus.26 Chroniclers who recycled tradi-

discussion of the appropriation of Jubilees by Epiphanius in relation to this earlier trajectory of usage, see the paper of A. Y. Reed in this volume. 23 George Cedrenus, Compendium historiarum (ed. I. Bekker; CSHB; Bonn: Weber, 1838), 1.6.1–4. 24 See, for example, the Chronicon of Symeon Logothete (ed. S. Wahlgren; CFHB 44.1; Berlin: de Gruyter, 2006), 25.11–12 (24.2): Κάϊν, ὡς λέγει Μωϋσῆς, τῆς οἰκίας πεσούσης ἐπ’ αὐτὸν ἐτελεύτησεν (= Jub. 4:31). 25 So see Syncellus, Chronography, 4.20–22, who speaks of “authors of Jewish Antiquities and Christian histories who have written on the basis of the Little Genesis and the Life of Adam” (ὅσα καὶ ἄλλοις ἱστορικοῖς Ἰουδαϊκὰς ἀρχαιολογίας ἢ καὶ Χριστιανικὰς ἱστορίας γράψασι περὶ τούτου εἴρηται ἐκ τῆς λεπτῆς Γενέσεως καὶ τοῦ λεγομένου βίου Ἀδάμ). Translations of Syncellus are taken from W. Adler and P. Tuffin, trans., The Chronography of George Synkellos (Oxford: Oxford University Press, 2002). For discussion of material from Jubilees attributed to Josephus, see further ibid., liv; and Gelzer, Sextus Julius Africanus, 2:278–80. 26 For a parade example, see Michael Glycas, Annales (ed. I. Bekker; 2 vols.; CSHB 26, 27; Bonn: Weber, 1836), 263.20–264.6. Jub. 37–38:3 describes how, during a war between the sons of Jacob and the sons of Esau, Jacob killed Esau with an arrow to the chest. For the author, this fulfilled Isaac’s prophecy to Esau: “And it shall come to pass when you become great, and shake his yoke from off your neck, you shall complete a complete sin unto death. And you seed shall be rooted out from under heaven” ( Jub. 26:34); cf. Isaac’s statement in Gen 27:10, which lacks the words after “off your neck.” Byzantine historians frequently cited Parabiblical Traditions and their Use in the Palaea Historica 9 tions and stories from Jubilees often did so with no apparent awareness of the actual origins of these materials. Over time and repeated reuse, they assumed a life of their own, sometimes far removed from their original setting and purpose. Because of the popularity and wide dissemination of some of the more accessible of these chronicles (most notably the ninth-century chronicle of George the Monk), stories and traditions from Jubilees became part of a received tradition of “biblical history,” familiar to a wide readership. In its own way, the Palaea’s recycling of the same stories was part of the same main- streaming process. The most cited passages from Jubilees were also subject to the most reworking and analysis. Byzantine authors might not have agreed with the specifics of Jubilees’ enumeration of the acts of creation; few were willing to accept, for example, that the angels were one of the acts of creation ( Jub. 2:2). But they were impressed by the uncanny symmetry between the twenty-two acts of creation, the number of patriarchs up to the time of Jacob, the letters of the Hebrew alphabet and the books of the Hebrew Bible ( Jub. 2:1–23).27 In the story of Noah’s division of the earth among his three sons and Canaan’s subsequent violation of their oath, Christian historians found a plausible jus- tification for the Israelites’ reacquisition of Canaan ( Jub. 8:10–18). But in place of Jubilees’ archaic mappa mundi, they substituted a system more in line with the prevailing understanding of the inhabited world.28

the Jubilees passage, sometimes naming Josephus as its source and multiplying the confusion by attributing Isaac’s entire prediction to Genesis, including the additional words: “You shall complete a complete sin unto death” (cf. Syncellus, Chronography, 123.22–25). By the time the passage had found its way to Glycas, and despite his best efforts, it was no longer possible for him to set the record straight. Glycas was deeply hostile to Jubilees, which he later dismisses as a joke (206.20–22). But like other chroniclers, he treats the story of the war between Esau and Jacob as an authentic passage from Josephus. He does know, however, that the additional words of the prediction πλημμέλειαν πλημμελήσεις θανάτου were not to be found in Genesis. He thus tells his readers that these words are a gloss by “Josephus” and not part of the text of Genesis. 27 Syncellus, Chronography, 3.4–18. See also George Cedrenus, Compendium historiarum, 1.9.10–13; Symeon Logothete, Chronicon, 10.14–17 (8.4). Symeon tellingly ascribes the tradition only to Moses (ὡς λέγει Μωϋσῆς). Because the Ethiopic version of Jubilees does not make mention of the twenty-two letters of the Hebrew alphabet, there is uncertainty about its originality; see J. C. VanderKam, trans., The Book of Jubilees (CSCO 511; Scriptores Aethiopici 88; Louvain: Peeters, 1989), 13 (n. ad loc.). 28 See W. Adler, “The Origins of the Proto-Heresies: Fragments from a Chronicle in the First Book of Epiphanius’ Panarion,” JTS 41 (1990): 494–98. 10 Adler

Perhaps the most widely known legend from Jubilees was its dramatic account of Abraham’s estrangement from the pagan religious practices of his father, his destruction of the idol temple in Ur, and his flight from the city ( Jubilees 12). The episode was retold endlessly. If Syncellus can be trusted as a witness, by the ninth century the story was simply accepted as a factual ampli- fication of the more sketchy version of events recounted in Genesis. In recounting the episode of Abraham’s burning of the idol temple in Ur and Haran’s death during the conflagration, he says only that it “was reported often” (ὡς πολλαχοῦ ἱστορεῖται).29 Like other commentators, Syncellus found in it a helpful way to resolve a wide array of questions about the sequence of events preceding Abraham’s flight to Haran: What was Genesis referring to when it spoke of Haran’s death “before” his father (Gen 11:28)? Why did Terah, not Abraham, lead the family from the city (Gen 11:31)? Was Terah still alive when Abraham journeyed on to Canaan. And if so, did he behave disrespectfully in leaving his father behind in Haran (Gen 12:1)?30 The same traditions appear in the Palaea, here again with the expected variations. The author of the Palaea cites the Jubilees-based equation of the twenty-two acts of creation with the twenty-two letters of the Hebrew alphabet (3.13 [189]). But he does not identify the source, and his enumeration of the individual works is at odds with Jubilees. The Palaea’s account of Abraham’s early life in Ur reproduces the essential details of Jubilees: his repudiation of the ancestral religion of his father; his burning of the temple in Ur that housed idols built by his father; the death of his brother when he rushes into the temple to retrieve them; and Terah’s blessing of Abraham before his migration from Haran to Canaan (26.7–9 [202–3]; cf. Jub. 12:1–5). But while the Palaea retains vestiges of the older interpretive history of the Abraham episode, it has been recast with a very different constellation of narrative details.31 Nor is there any

29 Syncellus, Chronography, 107.22. 30 Syncellus, Chronography, 105.6–108.11. For a discussion, see W. Adler, “Jacob of Edessa and the Jewish Pseudepigrapha in Syriac Chronography,” in Tracing the Threads: Studies in the Vitality of Jewish Pseudepigrapha (ed. J. C. Reeves; SBLEJL 6; Atlanta: Scholars Press, 1994), 143–71. 31 Although the Palaea does not identify Abraham’s father by name, it unaccountably identifies Terah as Abraham’s brother and the victim of the fire (26.8 [202]). Abraham’s understanding of the God of the universe from contemplation of the heavens occurs in Ur, not Haran, which the author places in Arabia Felix (26.2 [201]); cf. Jub. 12:16–17. The author of the Palaea makes an additional effort to absolve Abraham of any disrespect for his father. Unlike Jub. 12:1–8, according to which Abraham openly confronts his father about his idolatry, the Palaea says that Abraham asked these questions only to himself (20.5 [202]). In other differences from Jubilees (and the biblical account), Abraham receives Parabiblical Traditions and their Use in the Palaea Historica 11 suggestion that the author had the first clue about the source from which these traditions ultimately originated. The story is simply assimilated, without attribution or qualification, into a continuous narrative of biblical history. In the single case in which he does identify a source, the author vaguely, and incorrectly, attributes to a “book of Josephus” (ἐν τῇ βίβλῳ Ἰωσήπου) the Jubilees-based story about Noah’s division of the earth among his sons after the Flood and Canaan’s violation of the oath (23.14 [200]). The Palaea (16.1 [196]) also gives Josephus unearned credit for the odd claim that Adam’s offspring included “sixty children, who also became thirty androgynes” (Ἐγέννησε δὲ Ἀδὰμ υἱούς, ὡς γράφει Ἰώσηπος, παίδας ἐξήκοντα οἵτινες καὶ ἐγένοντο ἀνδρόγυνα τριάκοντα). As elsewhere in Byzantine literature, the well-regarded Josephus had become a peg on which to hang a stray, or perhaps suspect, tradition. Curiously, the two passages in the Palaea bearing the closest resemblances to Josephus are unascribed. Both of them involve the exploits of the young Moses. In his Antiquities, Josephus credits the young Moses with devising an ingenious stratagem to defeat the Ethiopians (Ant. 2.243–248). His advancing army carries with them ibises in wicker baskets; the birds destroy the venom- ous snakes that block their entrance into Ethiopia. While the basic story line in the Palaea is much the same, the details are quite different (70 [228–29]). Here the birds are storks, not ibises; and the country Moses invades is India, not Ethiopia.32 The Palaea’s other episode about the young Moses barely disguises its numerous secondary layers of tradition. According to Josephus (Ant. 2.233– 236), Moses, while still an infant, taxes the Pharaoh’s patience by hurling his crown to the ground after the Egyptian king had set it on his head. When an Egyptian scribe, interpreting this as a bad omen, advises the Pharaoh to kill him, his daughter Thermuthis rescues Moses before her father is able to act on the scribe’s advice. Later folklore softened the story by treating Moses’s actions as those of a playful child, not a dark omen. In one adaptation, Moses grabs at Pharaoh’s beard. In the other more elaborate renditions, a court sage

God’s command in Ur, following the incident in the temple; Terah blesses him there in Ur and travels with him as far as Haran. Abraham then goes on to Canaan only after his father’s death; cf. Jub. 12:28–31, according to which Terah is still alive when Abraham sets out to Canaan. 32 On this Josephus legend and later variants of it, see Rajak, “Moses in Ethiopia: Legend and Literature.” The Palaea’s statement that Moses invaded India, not Ethiopia, may have arisen from the belief, common in Byzantine sources, that the Axumite kingdom of Ethiopia was “inner India”; see P. Mayerson, “A Confusion of Indias: Asian India and African India in the Byzantine Sources,” JAOS 113 (1993): 169–74. For the identification of the birds as storks, see also Divrei ha-Yamim shel Moshe Rabbeinu, BHM 2:6–7. 12 Adler becomes the hero of the story by performing an experiment meant to establish the child’s innocence. He has Moses choose between the crown and a sword. By reaching for the sword, Moses shows that he is merely fascinated by shiny objects. The other variant of the story has Moses choosing between the crown and burning wax. When Moses grabs the torch and put it in his mouth, his behavior proves that he has no designs on the throne.33 The author of the Palaea was not altogether successful in fusing this jumble of overlapping traditions about the infant Moses together into a single and coherent narrative (69 [227–28]). In other witnesses to the legend about the infant Moses, the story of the burning wax was meant to explain the origins of Moses’s speech impediment. That important explanatory detail has been lost in the Palaea’s retelling.34 Nor has the author’s blending of the different tests produced a fully satisfactory outcome. The sage establishes Moses’s innocence by demonstrating that he was attracted to the crown because it was a shiny object. This argument does not explain, however, why he cast it to the ground. And how would preferring a sword to Pharaoh’s crown have demonstrated that Moses meant nothing sinister by tugging on his beard?35 In the treatment of Byzantine writings thought to preserve much older Jewish traditions, scholars typically engage in a kind of textual archeology, sifting through these works for traces of older source material. But without a clear understanding of these authors’ overall purpose and literary habits, the whole approach is bound to yield misleading and often inflated results. The same observation applies to the Palaea. While vestiges of Jubilees, Josephus’s Antiquities, and other more exotic sources are certainly visible, the likelihood that the author had direct access to any of them is negligible. In most cases, the author has dipped into a fluid and disparate collection of popularly known traditions to create, sometimes maladroitly, a continuous narrative of biblical history.

33 For discussion of the various later elaborations of this story, see Flusser, “Palaea Historica,” 63–67; J. Kugel, Traditions of the Bible, 510–11; L. Ginzberg, The Legends of the Jews (trans. H. Szold; 7 vols.; Philadelphia: Jewish Publication Society of America, 1909–1938), 5:402 n. 65. 34 For the link between this episode and Moses’s speech impediment, see, for example, Exod. Rab. 1:26. Note, however, that the Slavonic translation of the Palaea does make the connection (on which, see Lieberman, “Zenihin,” 49). 35 For further discussion, see Flusser, “Palaea Historica,” 63–67. Parabiblical Traditions and their Use in the Palaea Historica 13

1.2 The Palaea’s Non-Literary Sources In an environment in which sources were at least as much visual and auditory as they were written, it is hardly surprising that Byzantine art and oral tradition provide some of the closest parallels to stories preserved in the Palaea. Of particular interest are the miniatures in Greek and Slavic Psalters. These images depict, down to specific details, many of the same episodes recounted in the Palaea; e.g., Lamech, the blind archer, mistaking Cain for a wild animal; the calf restored to life after Abraham sacrifices it as a meal for the three angels at Mamre; Saul’s quest for the lost sheep and his anointing by Samuel (1 Samuel 9–10); and the angel threatening a penitent David with his sword (2 Sam 24:16–17).36 While some of these images, especially those preserved in Psalters of a later date, may have drawn upon the Palaea, there is no reason to assume that it, or in fact any written source, was the direct inspiration for all of them. Given the overall tendencies of the work, we might just as readily imagine that the author of the Palaea sought to provide a historical explanation for motifs already recognizable from Byzantine art. This inference is further confirmed by comparison of the Palaea with art and pilgrimage narratives from medieval Palestine. For a work generally lacking in historical and chronological information, the Palaea shows a deep interest in the geography and terrain of the Holy Land, and biblical sites located there. When the giant Nimrod measured the earth, writes our author, he discovered that Palestine was at dead center (25.2 [200]). At one point, the author provides the exact distance between the location of Lot’s tree and the Jordan: twenty-four miles (55.3 [219]).37 He also speaks authoritatively about the sources of the Jordan. “This river,” he says, “originated half from the sea of Tiberias and half from the lake of Gennesaret; the [one part] is called Jor, and the other Danes. When the two tributaries flow together downstream, it is called the Jordan” (123.6–7 [259]). Before writing off the author as misin- formed, we should consider that Jerome reports the same odd detail about the tributaries of the Jordan.38 It is one of those pieces of travel lore with an amazing longevity. As late as the eighteenth century, western travelers to Palestine continued to repeat it.39

36 See R. Stichel, “Außerkanonische Elemente in byzantinischen Illustrationen des Alten Testaments,” Römische Quartalschrift 69 (1974): 163–81. 37 On Lot’s tree, see below, pp. 15–16. 38 See Jerome, Comm. in Matt. 16.13 (ed. D. Hurst and M. Adriaen; CCSL 77; Turnhout: Brepols, 1969), 139, 8–11. The same notice appears in Jerome’s Liber de situ et nominibus locorum Hebraicorum 195 (PL 23:890C). 39 See further Flusser, “Palaea Historica,” 74 n. 99. 14 Adler

Another tradition of Palestinian provenance reproduced in the Palaea concerns the prophet Habakkuk. In Bel and the Dragon, one of the Greek additions to the book of Daniel, an angel grabs a reluctant Habakkuk by the hair and transports him to Daniel’s den. After feeding Daniel, he is returned within the hour to “his own place” via the same angel (1:33–36). The depiction of Habakkuk being seized by the hair is a popular motif in medieval art, which typically identifies the angel as Michael (not Gabriel). The actual location of Habakkuk’s “own place” was also known to medieval Christian travelers. According to Abbot Daniel, a twelfth-century Russian pilgrim to the Holy Land, a chapel to the south of Bethlehem was built on the very site of the field where the prophet Habakkuk once fed his reapers. His report also supplies a fine point lacking in Bel: namely, that Habakkuk returned to Judea in time to feed his laborers.40 These same details appear in the Palaea’s narrative of the life of Daniel. When Habakkuk goes out to feed his reapers, the archangel Michael seizes Habakkuk by the crown of his head and carries him by a gust of the wind from Judea to Babylon. After Habakkuk feeds Daniel with the pottage he had prepared for his workers, Michael returns him to his home in time to feed his workers, with provisions in even greater abundance (168 [291]). What we see in this story is the process by which a piece of local lore, tied to a particular place, marked by a religious shrine, widely known to travelers, and depicted in Christian iconography, has now found a home in a larger biblical history. The same process can be observed in the longer, more developed stories in the Palaea. One of these relates Lot’s expedition to Paradise to recover branches from the Tree of Life. Among the many ancient and medieval Jewish and Christian stories about the quest for some saving artifact from Paradise, probably the best known feature Adam’s third son Seth. In one version of Seth’s foiled quest for oil from the Tree of Life, the angel guarding Paradise, while denying his request for the oil, does give him seeds or twigs from three trees: the cedar, cypress and pine. Wood that sprouts forth from them is said to have found its way to Solomon’s Temple and ultimately to the cross.41

40 For Abbot Daniel’s report, see J. Wilkinson, Jerusalem Pilgrimage, 1099–1185 (London: Hakluyt Society, 1988), 149. 41 For Seth’s quest for oil from the Tree of Life, see, for example, the Life of Adam and Eve, 36:1–44:3. In the Slavonic version of this work, Seth returns with twigs from a pine, a cypress, and a cedar tree. For recent discussion of the various witnesses, see B. Baert, A Heritage of Holy Wood (Cultures, Beliefs, and Traditions 22; Leiden: Brill, 2005), 317–22; and, more recently, N. Fallon, “The Cross as Tree: The Wood-of-the-Cross Legends in. Middle English and Latin Texts in Medieval England” (Ph.D. diss.; Univ. of Toronto, 2009), esp. 27–29; 34 (on the Palaea story); see also E. Quinn, The Quest of Seth for the Oil of Life (Chicago: University of Chicago Press, 1962), 2–12, 55, 71. Parabiblical Traditions and their Use in the Palaea Historica 15

The Palaea also has an episode about a journey to Paradise for three pieces of wood (54–55 [218–19]). In this case, however, Abraham commissions Lot to retrieve three fire-brands from the vicinity of Paradise. Against all expectations, Lot, with divine protection, does succeed, returning to Abraham with wood from the cypress, pine, and cedar trees.42 After he and Abraham plant the wood in the shape of a triangle, each piece a cubit from the others, Lot regularly travels the twenty-four miles to the Jordan to fetch water for the wood. The three pieces of wood soon sprout leaves and grow together into a single trunk. The Palaea’s narrative breaks off abruptly here, stating only that the tree remained standing until the time of Solomon, and promising to explain the meaning of the tree at a future time (55.7 [219]). But a tree with a single trunk and the roots of three different trees—a barely concealed symbol of the Trinity—leaves the reader in little doubt about the future course of events. Abraham drives home the point by telling Lot that the tree will be the “aboli- tion of sin” (55.5 [219]). Another version of the story, found in one of the manuscript witnesses to the chronicle of Michael Glycas, provides a more satisfying resolution of the story. From it we learn that Lot’s tree was cut down for use in Solomon’s Temple, but was never actually used for this purpose. Only later did the tree realize its true purpose: as the wood for the cross.43 The sequence of events associated with Lot’s journey to Paradise, including Lot’s watering of the three-rooted tree, is probably most famous from a series of paintings found at the Monastery of the Cross near Jerusalem (built in the eleventh century). A site behind the altar of the main church is said to be the place where Lot planted the wood that sprouted into the tree of the cross.44 A painted panel lining the chapel of the Monastery of the Cross also depicts Satan attempting to thwart Lot’s efforts. Thanks to religious tourists to Palestine, this piece of local folklore became widely known. One pilgrim reports having heard a story about how Satan, disguised as a depleted Russian pilgrim, tried to impede Lot’s mission by drinking the water that he was carrying from the Jordan.45 The Palaea has all of these details, including a somewhat

42 Cf. Isa 60:13 (LXX), which speaks of the cypress, pine, and cedar trees that will “glorify my holy place.” Christian interpreters in the East understood this as a reference to the wood both of Solomon’s Temple and of the cross. The three pieces of wood were also taken as symbols of the Trinity; see further Baert, A Heritage of Holy Wood, 302, 318, 423. 43 Glycas, Annales, 254.38–255.33 (lectiones variae). 44 See V. Tzaferis, “The Monastery of the Cross: Where Heaven and Earth Meet,” BAR 27/6 (Nov/Dec 2001): 35–39. 45 See J. E. Hanauer, Folklore of the Holy Land (London: Sheldon, 1907), 34–36. 16 Adler oblique reference to Satan’s obstructions.46 All of this—together with the fact that our text gives the precise distance from the tree to the Jordan river—raises the possibility that the Palaea’s story of Lot’s quest originated in a tradition of Palestinian provenance coexisting and competing with rival medieval traditions about the origins of the wood of the cross.47 It is thus hardly necessary to trace the origins of the story to a conjectural Apocryphon of Lot.48 Plenty of nonliterary sources existed upon which the author could draw. The same can be said about the most elaborate episode of the Palaea: Abraham’s historic meeting with the hermit Melchizedek on Mount Tabor (32–38 [206–10]). The author displays great ingenuity in explaining how Melchizedek, the son of royalty, chose for himself the life of an ascetic wild man and hermit, with hair and beard extending down to his feet, and nails a cubit in length (36.7 [209]). Even before his encounter with Abraham, the Palaea links together the lives of the two men by patterning the narrative of Melchizedek’s early life and conversion according to its previous account of Abraham’s early life in Chaldaea and his estrangement from the pagan practices of his father. Like Abraham, Melchi (later Melchizedek) was the son of a pagan king Josedek, devotee of the god Cronus. On the eve of a planned sacrifice to Cronus, Melchi stays up late observing the heavens. As he contemplates the orderly motion of the stars, he recognizes, as had Abraham before him, that a single god must be the author of all of this. But when he informs his father about his discovery, his father, outraged and fearful that his son has angered the gods, decides that the only way to right the wrong is for all the inhabitants of the city to sacrifice their male offspring. Melchi flees from the city and takes up the life of a hermit on Mount Tabor. The circumstances of his flight from his ancestral home thus explain how the epistle to the Hebrews could describe Melchizedek as “without genealogy, without father and mother” (Heb 7:3). After God destroys the city of his birth and its inhabitants in an earthquake,

46 At 55.4 (219), the Palaea writes of “Lot’s struggles” (πυκτεύων: lit. “boxing”) while journey- ing to and from the site of the tree. The word that the author uses to describe the conflict (πυκτεύων) appears elsewhere in connection with the contest of believers with Satan; cf. C. H. W. Lampe, A Patristic Greek Lexicon (Oxford: Clarendon, 2003), s.v. For examples of this usage, see John Chrysostom, Adversus oppugnatores vitae monasticae 3.17 (PG 47:378); idem, In sanctum Romanum (homilia 1) 2.4 (PG 50:612). 47 See further Baert, A Heitage of Holy Wood, 319 n. 121; A. Giannouli, “Apocryphon Lot (CAVT, Nr. 93): Zur Entstehung und Entwicklung einer Legende,” in Byzantinische Sprachkunst: Studien zur byzantinischen Literatur gewidmet Wolfram Horandner zum 65. Geburtstag (ed. M. Hinterberger and E. Schiffer; ByzArch 20; Berlin: de Gruyter, 2007), 88–103. 48 Pace A.-M. Denis, Introduction à la littérature religieuse judéo-hellénistique: Pseudépigraphes de l’Ancien Testament (2 vols.; Turnhout: Brepols, 2000), 1:215. Parabiblical Traditions and their Use in the Palaea Historica 17

Melchi is left an orphan, living alone on Mount Tabor and surviving solely on water and wild plants (36.4). For the author of the Palaea, the meeting between Abraham and Melchizedek was no chance encounter, but rather the outworking of a divine plan. Earlier in the narrative, the author describes a terrifying nighttime encounter between Ephron the Hittite king and an angel who threatens him with a sword. To propi- tiate God, Ephron pays Abraham a tithe, which Abraham, only recently arrived from Ur, offers up as a sacrifice (28–29 [203–4]). The Palaea then describes how Abraham subsequently receives a revelation from an angel of God telling him to take costly raiment, bread, wine, and a razor, ascend Mount Tabor, and shout three times, “Man of God!” (31.7–8 [206]). In compliance with the divine decree, Abraham trims Melchizedek’s nails and shaves his head and beard, after which the two make an offering to God and partake of the meal of bread and wine that Abraham has brought with him (37.6–9 [209]; cf. Gen 14:18). Just as Abraham had received a tithe from Ephron, so he then pays a tithe to Melchizedek—thus explaining Heb. 7:9: “Receiving tithes, he paid tithes” (38.7 [210]). The Palaea’s own version of events is detailed and rather well narrated, displaying, as Flusser has observed, elements of the “epic art.”49 Whether the author deserves credit for its composition is another question, thrown into further doubt by the subsequent appearance of a second, more abbreviated account of Abraham’s meeting with Melchizedek after his defeat of King Chedorlaomer (46 [213–14]; cf. Gen 14:1–17). The likelihood that the author’s narrative of Abraham’s meeting with Melchizedek on Mount Tabor did draw upon a literary source is also suggested by various versions of the story that survive in Greek, Syriac, Coptic, and Arabic.50 But given the wide variations among the different written witnesses, we need not assume the existence of a single apocryphal Urtext.51 As with other stories in the Palaea, the account of Abraham’s meeting with Melchizedek is likewise rooted in Holy Land

49 Flusser, “Palaea Historica,” 58. 50 The closest literary parallel with the Palaea’s account is found in ps.-Athanasius’s History of Melchizedek (PG 28:525–29). For an English translation of the ps.-Athanasius text, see S. E. Robinson, “The Apocryphal Story of Melchizedek,” JSJ 18 (1987): 26–39. For discussion of the several witnesses to the legend, see also J. Dochhorn, “Die Historia de Melchisedech (Hist Melch): Einführung, editorischer Vorbericht und Editiones praeliminares,” Le Muséon 117/1–2 (2004): 7–48. 51 Dochhorn, “Die Historia de Melchisedech,” 40–41, has a thorough discussion of the differences between the Palaea and ps.-Athanasius; cf. Flusser, “Palaea Historica” (56–57), who, while acknowledging these differences, argues for the existence of a Melchizedek apocryphon. 18 Adler folklore; for the wider dissemination of which, religious tourists own some responsibility. According to Gen 14:18–20, Abraham meets Melchizedek in Salem, the city over which the latter is king. In the Palaea’s report of the meeting, Salem, the city of Melchizedek’s birth, was founded by the idolater Nimrod and ruled by Melchizedek’s father (32 [206–7]). The actual meeting between Melchizedek and Abraham occurs on Mount Tabor in the Lower Galilee, after Melchizedek has fled the city. That detail, found in almost all the versions of the story, helps to identify the provenance of the story. According to the ecclesiastical historian Nicephorus Callistus (fourteenth century), Helen, the mother of Constantine, erected a church on Mount Tabor near the cave where Melchizedek was said to have lived.52 As with many other stories retold in the Palaea, medieval travelers helped in the wider dissemination of this piece of local folklore. In the early twelfth century, Abbot Daniel describes visiting the same cave: “In this small cave lived the holy Melchizedek, and Abraham came to him here and called him thrice, saying ‘Man of God.’ Melchizedek came out and brought bread and wine, and having made a sacrificial altar in the cave, offered up the bread and wine in sacrifice, and this sacrifice was immediately taken up to God in heaven; and here Melchizedek blessed Abraham and Abraham cut his hair and nails, for Melchizedek was hairy.”53 By Daniel’s time, Melchizedek’s cave, teeming with anchorites, had become a mandatory site for pilgrims. Later in the twelfth century, Joannes Phocas, a monk of Crete, called the grotto of Melchizedek “well worth seeing” (ἀξιοθέατον). Here one could also see “cells serving as habi- tations for ascetics, wherein many of the greatest saints have passed their ascetic lives.”54 Visual representations of the scene had their own role to play. The Palaea’s account of Abraham’s meeting with the hirsute Melchizedek on Mount Tabor turns up in Coptic art and wall paintings in Palestinian monaster- ies, and in illustrated Byzantine manuscripts.55 In a mural scene from the monastery of St. Anthony in eastern Egypt, a skeletal Melchizedek, naked from the

52 Nicephorus Callistus, Historia Ecclesiastica, ch. 30 (PG 146:113C). 53 See Wilkinson, Jerusalem Pilgrimage, 162. 54 Joannes Phocas, Descriptio Terrae Sanctae, 11 (PG 133:937BC). English translation by A. Stewart, The Pilgrimage of Joannes Phocas in the Holy Land: In the Year 1185 A.D. (London: Palestine Pilgrims’ Text Society, 1889), 14. According to Phocas, Melchizedek’s cave was located on the northern side of the mountain. The same geographical detail in the Palaea (31.6 [206]) again points to the author’s familiarity with pilgrimage narrative. 55 For depictions of the scene in Byzantine biblical manuscripts, see A. Böck, “Melchisedek,” Reallexikon zur byzantinischen Kunst (Stuttgart: Hiersemann, 1966), 6:118–19; and P. van Moorsel, “A Different Melchizedek? Some Iconographic Remarks,” in Themelia: Spätantike und koptologische Studien: Peter Grossmann zum 65. Geburtstag (ed. M. Krause and Parabiblical Traditions and their Use in the Palaea Historica 19 waist up, stands in front of a cave, holding out a chalice and Eucharistic spoon to Abraham. Lying before them are the scissors and the knife that Abraham will use to shave Melchizedek.56 While the image of Melchizedek extending to Abraham a chalice and eucharistic spoon is obviously imbued with religious and ritual significance, the meaning of the whole scene, in such striking contrast with the older more exalted representations of Melchizedek, remains obscure. Perhaps, as has been suggested elsewhere, the barbering of Melchizedek had developed an older understanding of Melchizedek as a Nazirite into a biblical prefiguration of the ritual tonsuring of the anchorite at the time of his initiation into more institutional forms of monastic life.57 The fact that the various witnesses configure the power relationship between Abraham and Melchizedek differently may also suggest unresolved tension in the traditions about the relative standing of both figures.58 But although the process by which Melchizedek was transformed in Christian legend from priest/king into an orphaned hermit remains obscure, the story of Melchizedek’s meeting with Abraham on Mount Tabor was in wide circulation by the later ninth century. What the author of the Palaea has done is to provide for his readers the historical backstory for an image already well known to them.

1.3 The Palaea and Byzantine Liturgy While generally lax in citing literary sources, the Palaea does show a deep familiarity with Byzantine hymnody and liturgical texts; notably, the Psalms; the

S. Schaten; Sprachen und Kulturen des Christlichen Orients 3; Wiesbaden: Reichert, 1998), 331 n. 6. 56 See E. S. Bolman, Monastic Visions (New Haven: Yale University Press, 2002), 68–69. See further G. J. M. van Loon, “Priester van God de Allerhoogste: Iconografische en iconologische aspecten van de Ontmoeting van Abraham en Melchisedek en de Apostelcommunie in koptisch Egypte,” Periodical of the Institute of Eastern Christian Studies at Nijmegen 53 (2001): 5–29; idem, “The Meeting of Abraham and Melchizedek,” 1373–92; P. van Moorsel, “A Different Melchizedek?” 329–36. 57 See Robinson, “The Apocryphal Story of Melchizedek,” 35–36, who also notes the description of Melchizedek as a Nazirite in the Syriac Cave of Treasures. 58 Robinson, “The Apocryphal Story of Melchizedek,” 36–37. An odd detail preserved in the Palaea may also suggest some connection with an older Jewish tradition. According to the Palaea, Abraham circumcised every male in his household (including Ishmael), because God had told him to “circumcise yourself in Melchizedek” (περίτεμε σεαυτόν ἐν τῷ Μελχισεδέκ). As evidence against the Jews that circumcision was not an eternal covenant, Christian writers typically claimed that Melchizedek was uncircumcised; see, for example, Justin Martyr, Dial. 19.4; 33.2. See further Flusser, “Palaea Historica,” 59. 20 Adler divine liturgy of Saint Basil [23.2–9 [199–200]); the nine canticles of Byzantine hymnody; and the Great Canon of Andrew of Crete (c. 650–c. 712)—bishop, theologian, and hymnographer. Most notably, there are places in the work where the author takes on the role of guide to and expositor of elements of the liturgy. In Byzantine hymnody, for example, the Song of the Sea (Exod 15:1–19), the Song of Moses (Deut 32:1–43), and the Prayer of Hannah (1 Sam 2:1–10) make up the first three of the nine canticles chanted at Matins. The Palaea inserts verses from all three biblical odes into the appropriate points in its narrative.59 After the citation from the “Song of Moses” comes the following gloss, either by the author himself or by a later scribe: “This is also where this second song was set forth” [ἐν ᾧ καὶ ἐτέθη ταύτῃ ᾠδὴ β´]—the enumeration is an obvious reference to its order in the sequence of the nine canticles.60 The author then goes on to assign to Moses the following admonition: “And let them be taught and know this song. If someone does not know this song through recitation and does not have it on the tip of his tongue, he will be cut off from the people” (καὶ εἴ τις οὐ μανθάνει τὴν ᾠδὴν ταύτην ἐκ στόματος καὶ κέκτηται ἐν ἄκροις χείλεσιν αὐτὴν ἐξολοθρευθήσεται ἐκ τοῦ λαοῦ) (120.1–4 [257]). The meaning of this admonition would not have been lost on an attentive reader. Acting on the assumption that his readers would be familiar with the full texts of these canticles through hearing, reenactment, and recitation, the author provides only a few words in each citation; but this was certainly enough for readers to understand the historical context of the odes that they sang every day. The author follows the same practice in citing from the Psalms and Andrew of Crete. Sung in its full form during the forty days of the Great Lent, Andrew’s Great Canon is the most well-known penitential hymn of the Byzantine liturgy.61 Quite frequently in the Palaea Andrew is identified only as the “wise man” or “author of hymns.”62 The author of the Palaea must have

59 119.8 [257]): the Song of Moses (Deut 32:13); 125.10 [261]: the Song of the Sea (Exod 15:15); 138.6 [270]): the Prayer of Hannah (1 Sam 2:1). 60 This is the reading of Cod. Ottob. gr. 205. Cf. Par. Bib. Nat. gr. 37b: καὶ ἐτέθη ᾠδὴ δευτέρα; Venice, Bibl. Naz. Marc. gr. 501: ἐτέθη ταύτῃ ᾠδὴ δευτέρα. 61 On Andrew’s Canon, see D. Krueger, “The Great Kanon of Andrew of Crete, the Penitential Bible, and the Liturgical Formation of the Self in the Byzantine Dark Age,” in Between Personal and Institutional Religion: Self, Doctrine, and Practice in Late Antique Eastern Christianity (ed. B. Bitton-Ashkelony and L. Perrone; CELAMA 15; Turnhout: Brepols, 2013), 57–98. On the Palaea as a framing narrative for the Great Canon, see below, pp. 25–26. 62 See 8.2 (192); 9.3 (192); 14.4 (195); 16.12 (196); 144.18 (275) (“Andrew” or “Andrew of Crete”); 138.9 (270); 139.5 (270) (the “Cretan”); 127.10, 12 (262); 130.22 (265); 133.9 (267); 136.8 (269); 140.21 (271); 141.14 (272); 144.17 (275); 148.9 (277); 160.10 (286) (the “wise man”). Several of Parabiblical Traditions and their Use in the Palaea Historica 21 considered the words of Andrew and the Psalms in minute detail. Verses from these two works frame many of the episodes described. In several cases, a single word or phrase from one of them is the literary hook for the author’s retelling of the biblical narrative. Only in the case of the Psalms and the Great Canon does he self-consciously and consistently play the role of an expositor, citing a lemma text and providing a historical explanation of its meaning.63 If, then, the author of the Palaea can rightly be called a “commentator,” it is not as a commentator on the Bible tout court, but rather on the Psalms and the Great Canon; that is, the retelling of the biblical episodes provides commentary to these two frame texts. Yet, despite the contribution of Byzantine liturgical texts to the Palaea’s representation of biblical events and characters, the subject, to my knowledge, remains almost completely unexplored. In one of the more creative insertions of elements of Byzantine liturgy into the Palaea’s narrative, biblical characters are actually made to reenact a part of the liturgy, in this case a prayer from the divine liturgy of St. Basil.64 The climac- tic moment of this liturgy, which in the orthodox churches today is normally celebrated ten times during the year, is the anaphora (or eucharistic prayer) of the priest; he raises the bread and wine and utters the words: “We offer you these gifts from your own gifts through all, for all, and in all” (τὰ σὰ ἐκ τῶν σῶν δώρων σοι προσφέρομεν, κατὰ πάντα, καὶ διὰ πάντα, καὶ ἐν πᾶσιν).65 At dramatic moments in sacred history, Byzantine authors were known to put the same words into the mouths of biblical figures. In the apocryphal Apocalypse of John, for example, the eucharistic prayer of the priest and the congregation’s response are woven into the narrative of Christ’s harrowing of Hell.66

the references in the Palaea to a “wise man” or “author of hymns” cannot be confirmed in the text of the Great Canon printed in the Patrologia Graeca. See, for example, 21.10 (199); 22.11 (199); 97.8 (242); 103.11 (247); 114.16 (254); 115.5 (254); 119.5 (257); 161.16 (287) (“wise man”); and 5.3 (190); 52.8 (217); 67.11 (226); 99.10 (243) (“author of hymns”). Because “wise man” and “author of hymns” are the usual ways in which the author cites Andrew, some of these unconfirmed citations may be of value for the textual history of the Great Canon. 63 For quotations from the Psalms, see 16.14 (196), 45.6 (213), 46.5 (214), 100.10 (245), 103.6 (246), 105.15 (248), 106.3,5 (248, 249), 116.10–11 (255), 117.5 (255), 119.7 (257), 125.8 (261), 130.4 (264), 156.14 (283). For the Palaea’s citations from the Great Canon, see preceding note. 64 The influence of the prayer on the Palaea’s account of Noah’s sacrifice was already noted by Vassiliev, Anecdota Graeco–Byzantina, xlvii. 65 Basil of Caesarea, Liturgia (recensio brevior vetusta), 679 (PG 31:1637D). 66 Apocalypsis apocrypha Joannis (versio altera), 37: Τὅτε εἶπεν πρὸς τὰ πνεύματα· Τὰ σὰ ἐκ τῶν σῶν σοὶ προσφέρονται, κατὰ πάντα καὶ διὰ πάντα. Καὶ ἀπεκρίθησαν οἱ ἄγγελοι, καὶ εἶπον· Σὲ ὑμνούμεν. For the text, see F. Nau, “Une deuxième apocalypse apocryphe grecque de saint Jean,” RB 23 (1914): 215–21. 22 Adler

The very same words appear in the Palaea; only in this case, Noah and his sons reenact the liturgy. According to Genesis, Noah’s first act after disembark- ing from the ark was to build an altar and make an offering to God. Appeased by the sacrifice, God vows never to destroy the world again (Gen 8:20–22). In the Palaea’s own retelling of this event, Noah is made to utter the words of the oblation as he raises the bread and wine: “The offerings are not ours, but rather that part of your possessions that was saved by your command we offer to you, O Lord.” His sons recite the antiphon of the congregation: “We sing of you, we bless you, we give thanks to you, Lord, and we pray to you, our God” (σὲ ὑμνοῦμεν, σὲ εὐλογοῦμεν, σοὶ εὐχαριστοῦμεν, Κύριε, καὶ δεόμεθα σου, ὁ θεὸς ἡμῶν) (23.3–5). It is a highly effective narrative technique. The words of their entreaty would have been immediately recognized by readers familiar with the divine liturgy. Putting the words of the eucharistic prayer into the mouths of the “priest” Noah and his sons allowed his readers to participate in events occurring in the distant past. By inserting a few explanatory glosses of his own, the author also seized upon Noah’s offering to explain this most sol- emn event of the Byzantine liturgical year.67 A second class of examples illustrates cases in which the Palaea creates entirely new characters out of something the author heard or imagined he had heard in the liturgy. According to Andrew’s penitential hymn, “Hannah’s child, the great Samuel, was reckoned among the judges, and Arimathea raised him in the House of the Lord (ὃν ἐθρέψατο ἡ Ἀρμαθαίμ ἐν οἴκῳ κυρίου).”68 When

67 “And after making a sacrifice from the clean cattle, he set it before God, entreating him for his favor and saying, ‘We offer to you what is yours from that which is yours in all and for all.’ That is to say (τουτέστιν), ‘The offerings are not ours, but rather that part of your possessions that was saved by your command we offer to you, O Lord.’ [And his sons replied,] ‘We sing of you, we bless you, we give thanks to you, Lord, and we pray to you, our God.’ This means (λέγων δὲ οὔτως), ‘We sing of you who formed us and brought us from nonexistence into existence’” (23. 3–6 [199–200]). For discussion of this passage in relationship to the Divine Liturgy, see further the illuminating comments in Philon d’Alexandrie: Questions sur la Genèse II, 1–7. Texte grec, version arménienne, parallèles latins (ed. J. Paramelle and E. Lucchesi; Cahiers d’Orientalisme 3; Geneva: Cramer, 1984), 29. Also note the relatively early artistic representation of Noah’s sacrifice found in the Ashburnham Pentateuch (perhaps as early as the fifth century), where the stone altar is set with the three chalices of the early Roman and Syrian baptismal Eucharist. See the discussion of this image in R. Clements, “A Shelter amid the Flood: Noah’s Ark in Early Jewish and Christian Art,” in Noah and His Book(s) (ed. M. E. Stone, A. Amihay, and V. Hillel; SBLEJL 28; Atlanta: Society of Biblical Literature, 2010), 277–99 (291–94). 68 Andrew of Crete, Canon (PG 97:1368A). On Arimathea (Gr. Αρμαθαιμ) as Samuel’s place of birth, see 1 Sam 1:1 (LXX). Parabiblical Traditions and their Use in the Palaea Historica 23 he heard this, the author of the Palaea apparently assumed from the word ἐθρέψατο that Arimathea was not a place, but a person. And thus we are told that “a woman by the name of Armathem was the one who raised Samuel” in the Temple (139.2 [270]).69 Elsewhere in the Palaea, the author refers to someone named “Zan” whom God punished for violating the sanctity of the ark of the covenant (107.3, 7, 9, 11 [249]). Although the author sets the event during the time of Moses, the context makes it clear that “Zan” is a corruption of the biblical name Uzzah, the person from the time of David whom God punished for reaching out his hand to steady the wagon carrying the ark (2 Sam 6:2–8). Andrew’s hymn recalls how, “when the ark was being carried on a wagon, and when one of the oxen slipped, Ozan (= Uzzah) only touched it and experienced the wrath of God.”70 But as is clear from the quotation from Andrew in the Palaea, the author must have confused the omicron in the name “Ozan” with the masculine definite article: ὁ Ζάν. And thus the repeated reference to Zan, minus the definite article.71 One of the most remarkable examples of a creative misunderstanding of the liturgy is the Palaea’s extended account of a Jewish hero during the period of the judges, named “Endor” (Ἀενδώρ). According to the Palaea (128–29 [263]), Endor was appointed leader of the Jews when they were being oppressed by the Persians. After offering gifts to the Persian king “Got,” Endor arranges a private meeting with the king, in the course of which he stabs the king to death, plunders his wealth and makes a dramatic escape. But who is Endor? Although Judges knows no such figure, we need not look for an apocryphal Book of Endor to unearth the source of this legend. It comes from the author’s own understanding of Ps 82:11 (LXX; 83:11 MT), which reads: “they were utterly destroyed at (en) Endor” (ἐξωλεθρεύθησαν ἐν Ἀενδώρ). The author of the Palaea, as he does elsewhere, understood ἐν to mean “by,” thereby conjuring out of his imagination a fictional character named “Endor,” responsible for the destruction of the Persians.

69 139.1–5 [270]: καὶ ἦν ἡ γυνὴ Ἀρμαθέμ λεγομένη ἡ ἀναθρέψασ(α) τὸν Σαμουήλ . . . Περὶ τούτου λέγει ὁ Κρήτης· ἐν τοῖς κριταῖς ὁ Ἄννης ἐκλήθη γόνος, ὁ μέγας Σαμουήλ, ὃν ἐθρέψατο ἡ Ἀρμαθὲμ ἐν οἴκῳ Κυρίου, καὶ τὰ ἑξῆς. 70 Andrew of Crete, Canon (PG 97:1375B). 71 This will not have been the only time in the Palaea when the author was thrown off by a definite article. Repeated references to Tobit as “Bit” may originate in the author’s confusion of the first syllable in Tobit’s name (Τωβιτ) with the definite article; see above, p. 6. Why the Palaea predates “Zan’s” unfortunate intervention to the time of Moses is unclear. In the Great Canon, the episode is correctly dated during the reign of David. 24 Adler

The examples cited above are only a sampling of the ways in which either the Greek psalter or Byzantine liturgy and hymnography have influenced the Palaea’s narrative of biblical history.72 While the author’s misunderstandings of the Psalter and Andrew’s Canon do not speak well of his learning or sophis- tication, they do point to a central and recurring feature of the work. Much of the author’s creative inspiration originated not in comparatively obscure written sources, but rather in a store of shared biblical lore familiar to him and his readers through religious imagery, pilgrimage, and liturgy. Because he quotes from memory, the “texts” that he knew the best and quoted most frequently were those that were either sung or performed.

2 Ideological Influences and Recurring Themes

For anyone interested in extracting older source material from the work, the characterization of the Palaea as a “compendium” or “compilation” of biblical stories is an appealing one; thinking of the Palaea as a mere collection relieves the modern interpreter of having to taking into account the author’s own editorial role in selecting and presenting the material. While it is undeniably true that the author was not always adept in incorporating into his narrative stray traditions and legends whose meaning and significance he may not have fully grasped, he was far from a mechanical and unreflective collector. Although the author rarely inserts overtly partisan opinions into the narrative, certain recur-

72 Other types of interpretive traditions have also left their mark. Note in addition the following: 1) In several places, the enumeration of the ten plagues of Egypt (76–84 [232–34]) follows, against Exodus, the sequence of Ps 104 LXX (MT 105):29–36. For discussion, see H. Jacobson, “The Egyptian plagues in the Palaea Historica,” Byzantion 47 (1977): 347; see also A. Pietersma, The Apocryphon of Jannes and Jambres the Magicians, 160–62. 2) In the Palaea’s rendering of the story of Absalom, the author describes how Absalom listened to the “counsel of Hushai and his son Emene” (Ἀκούσας δὲ Ἀβεσ(σ)αλὼμ τοὺς λόγους Χουσὴ καὶ υἱοῦ Ἐμενῆ) (159.5 [295]). Emene, a figure unknown to 2 Samuel, may have originated from a peculiar understanding of the Greek text of Ps 7:1. The heading of the psalm in Greek reads: a “psalm of David which he sang to the Lord because of the words of Hushai, son of Iemeni” (Ψαλμὸς τῷ Δαυιδ, ὃν ᾖσεν τῷ κυρίῳ ὑπὲρ τῶν λόγων Χουσι υἱοῦ Ιεμενi). Cush“ כּוׁש ּבֶ ן־יְמִינִֽי .Presumably, the author understood the words Χουσι υἱοῦ Ιεμενi (= Heb the Benjaminite”) to mean “Iemeni son of Hushai.” 3) The Palaea’s account of Balaam states, contra Num 22:23–35, that the angel encountered Balaam after he had reached Balak (114.5–16 [253–54]); here the author follows an unknown “wise man”: “the fearsome angel appeared to an ass and accused Balaam the diviner on his return of disobeying the ineffable and divine decrees of God made long ago.” Parabiblical Traditions and their Use in the Palaea Historica 25 ring themes determine his choice and shaping of his subject material, including his use of parabiblical traditions. In the discussion that follows, I want to examine two of these organizing themes: 1) biblical history as a record of transgression, penitence and forgiveness; and 2) the inviolability of the priesthood.

2.1 Penitence and Forgiveness Of all of the liturgical texts known to the author of the Palaea, none seems to have exerted a greater formative influence on his treatment of biblical history than Andrew of Crete’s great penitential hymn. Taking the form of a chronological survey of the character and moral failings of biblical exemplars, the Great Canon invited Byzantine penitents to compare their own sinful souls with scriptural models and recognize their need for self-abasement and repentance. The author of the Palaea found in the Canon’s exploration of human frailty a highly useful mechanism for identifying suitable themes and subject matter in biblical history. As we have seen, a misunderstood phrase might serve as the inspiration for a fictional character like Endor. A word or verse from the Canon could also be the point of departure for amplifying Andrew’s admonitions about the consequences of moral weakness. One instructive example concerns the Palaea’s retelling of the savage assault on the wife (no longer the concubine) of the Levite by the residents of Gibeah. According to the account in Judges 19, the Levite had come to Bethlehem to recover his concubine, who had fled to her parents. Under constant pressure from the girl’s father to prolong his stay, the Levite delayed his departure for several days. When he finally did leave, it was almost evening. The Palaea con- structs the chain of events differently (143–45 [273–76]). The reason why the Levite is late in leaving is entirely of his own doing. Laziness and inattentiveness to the time cause him to delay his departure. For that reason, he has no choice but to spend the night in the hostile city of the Benjaminites—a fatal decision leading to the brutal assault on his wife. The citation from Andrew of Crete at the end of the episode (144.17 [275]) explains why the Palaea made the Levite responsible for what happened to his wife: “The Levite among the judges, by negligence (ἀμελείας) divided his wife among the twelve tribes, my soul, in order to proclaim the lawless outrage of Benjamin.”73 For the author of the Palaea, biblical history is one long cautionary tale about the moral excesses decried in Andrew’s Canon, and the means by which contrite transgressors were reconciled to God. The Palaea assigns all manner of

73 Andrew of Crete, Canon (PG 97:1365D). The text in the Palaea (διὰ μελίαν [“by ashwood”]) should accordingly be emended to διὰ ἀμέλιαν (“by negligence”). 26 Adler transgressions to biblical figures, some of them not immediately evident from the biblical text itself. In all, Cain is said to have committed seven sins, a number hinted at in Gen 4:15 (11.1 [193]).74 Scripture, the author says, “everywhere” (πολλαχοῦ) reproaches Lot’s carelessness in becoming drunk and committing an unthinkable act with his daughters (53.11 [218]). The reason why Abraham sends him on what he thinks will be a suicidal and fruitless mission to Paradise is because of his certainty that Lot’s death in transit would be the only way to earn divine forgiveness for the otherwise unforgivable transgression with his daughters. It might have been so, had not God had other, loftier plans for him. Had Esau controlled his hunger, he might never have lost his birthright (62.9 [224]). Laxity and drunkenness led to the downfall of the Nazirite Samson (Judges 16)—moral failings, the author writes, mourned by the “wise man” himself (Andrew of Crete) (136.4–7 [269]).75 The Palaea’s extended account of David’s reign offers a most instructive illustration of the interplay between Byzantine imagery, penitential discipline, and the author’s interpretation of biblical history (151–61 [279–87]). The narrative is organized around two penitential texts: a) David’s abject appeal for God’s forgiveness in Psalm 50 LXX (MT 51), the most commonly recited penitential psalm of the Byzantine liturgy; and b) a citation from an unidentified “wise man” (in this case, apparently not Andrew of Crete), describing how the king’s penitence averted the sword of the angel threatening both him and his kingdom. In amplifying on the meaning of these two passages, the Palaea recounts how, after David had arranged for the murder of Uriah, God instructed a reluctant Nathan to rebuke the king (2 Samuel 11–12). Nathan need not fear retribution from the king; if he resisted Nathan’s admonitions, the angel would slay him with his sword. While the king’s repentance and appeal for God’s forgiveness recorded in Psalm 50 staved off the avenging sword, the angel pronounced judgment on his kingdom with the words: a “sword shall not leave your house.” The angel and his sword reappear in the Palaea’s account of David’s illegal census, when the angel turns away his wrath only after David shows penitence. All of these events, the author writes, confirm the “wise man’s” summary of David’s reign: “To David your prophet, how you stopped your angel smiting the people with the sword” (161.16 [287]).76 More visually-oriented readers would also

74 For a list of Cain’s seven sins virtually identical to the Palaea’s, see ps.-Athanasius, Quaestiones (PG 28:737BC). 75 “In lamenting this (Samson’s humiliations), the wise man (Andrew of Crete) writes as follows: ‘Emulating the laxity of Samson’” (136.8 [269]). 76 The inspiration behind the image of the sword was probably generated by Nathan’s statement to David in 2 Sam 12:10 that, because of his murder and adultery, “the sword shall Parabiblical Traditions and their Use in the Palaea Historica 27 find in the narrative an explanation of a common image in Byzantine Psalter manuscript miniatures, from as early as the ninth century: their depiction of Psalm 50 (LXX) features an angel brandishing a sword and hovering over a penitent David, lying prostrate before Nathan.77 It is hardy unexpected that the author should press parabiblical traditions about biblical patriarchs into the service of the theme of sin and repentance. One such tradition concerns the figure of Lamech, the descendant of Cain. Lamech’s mysterious words in Gen 4:23—“I have killed a man for my wound- ing, and a young man to my own hurt”—made a rich target for legendary elaboration. A well-known story, familiar to both Jewish and Christians commentators, described how Lamech, an archer blind from birth, inadvertently killed Cain with an arrow, and then in a fit of anger and frustration slew the young man who was guiding him.78 With this act, Syncellus found in the figure of Lamech a type of the deicidal Jews.79 But the Palaea extracts from the same episode a less disparaging meaning. Guided by the words of Andrew of Crete, our author casts Lamech as a type of the repentant sinner. Lamech’s statement in Gen 4:23 represents the words of a man confessing to a double homicide. “This Lamech,” the author writes, “was the first to become a type of confession and received forgiveness from God, because he, of his free will, passed judgment on himself” (14.5 [195]).80

never depart from your house.” The sword imagery also supplied the link to the census story; cf. 1 Chr 21:26–27, where, following David’s offerings to avert the plague, the angel of God returned his sword to its sheath. 77 L. Brubaker, Vision and Meaning in Ninth-Century Byzantium (Cambridge: Cambridge University Press, 1999), 352–56. On the theme of David the penitent king in Byzantine art, see also M. Kuyumdzhieva, “David Rex Penitent: Some Notes on the Interpretation of King David in Byzantine and Post-Byzantine Art,” in The Biblical Models of Power and Law (ed. I. Bilarsky and R. G. Pāun; Rechtsthistorische Reihe 366; Frankfurt am Main: Lang, 2008), 133–51. For Psalter miniatures depicting scenes from the reigns of Saul and David, as described in the Palaea, see further Stichel, “Außerkanonische Elemente in byzantinischen Illustrationen,” 173–75. On the connections between the Palaea’s narrative of David’s reign and illuminated Psalter manuscripts, see now the detailed discussion in S. H. Wander, “The Paris Psalter (Paris, Bibliothèque Nationale, cod. gr. 139) and the Antiquitates Judaicae of Flavius Josephus,” Word & Image 30/2 (2014): 90–103, esp. 94–101. 78 See J. Kugel, “Why was Lamech Blind?” HAR 12 (1990): 91–103; V. Aptowitzer, Kain und Abel in der Agada (Vienna: Lowit, 1922), 59–68. 79 Syncellus, Chronography, 9.6. 80 On Lamech’s confession as the source of God’s forgiveness, see also ps.-Athanasius, Quaestiones (PG 28:740A): “He escaped punishment through the confession of the transgression, and by imposing the judgment on himself averted the judgment of God.” 28 Adler

The author bends to the same end another well-known parabiblical motif: the notion of antediluvian monuments of stone and wood. In the Hellenistic Near East, stories of pre-Flood monuments preserved for later generations were a common means by which historians could explain the survival of the collective learning of earliest mankind. One such story, first attested in Josephus’s Antiquities, described how Seth and his descendants, after receiving a warning from Adam about an imminent catastrophe of either fire or water, erected two sets of monuments, one of brick (in case of a fiery cataclysm), the other of stone (in the case of a flood of water). On them were inscribed their learning in the celestial sciences.81 Because this story formed a vital link in reconstructing the chain of the transmission and dissemination of world culture, Byzantine historians recycled the legend in varying versions. The Palaea’s own rendition of events bears only a distant connection to Josephus’s report. It is Enoch, not Seth and his offspring, who now erects the two monuments. And what he records has nothing to do with the transmission of higher learning. Rather the author exploits another, competing tradition about Enoch—not as the sage and culture hero, but as the prophet of repentance.82 On the eve of the Flood, Enoch exhorts the sinning giants to repent and glorify God, warning them that the world would be destroyed either by water or fire. In preparation for the calamity, Enoch does nothing other than record the mighty acts of God on monuments of brick and marble, presumably as a warning for later generations (20.4–6 [198]). As elsewhere, the author was far from graceful in integrating the story into his own narrative. Why would Enoch need to warn the sinning giants, when Noah was already doing the same thing? In any case, his own take on the legend of the two pre-Flood monuments, far better suited to the morally edifying aims of the work, reveals the wide cultural divide separating Josephus, the Hellenistic Jewish historian, from the pious Byzantine author of the Palaea.83 The centrality of the theme of sin and repentance may also explain the abrupt and uncharacteristic criticism of Judaism affixed to the end of the work. Because this critique is rather muted, and because the work is otherwise devoid of overt invective against Judaism, Flusser viewed the Palaea’s

81 Josephus, Ant. 1.68–71. 82 Cf. Sir 44:16: Ἐνὼχ εὐηρέστησεν κυρίῳ καὶ μετετέθη ὑπόδειγμα μετανοίας ταῖς γενεαῖς. On Enoch’s witness to the sinning pre-Flood generations, see also 1 Enoch 12–16; Jub. 4:22 (to the Watchers). 83 For a discussion of this passage in the Palaea, see A. A. Orlov, “Overshadowed by Enoch’s Greatness: ‘Two Tablets’ Traditions from the Book of Giants to Palaea Historica,” JSJ 32 (2001): 149–51. Parabiblical Traditions and their Use in the Palaea Historica 29 conclusion as more or less a pro forma exercise, an obligatory ending to a “book dedicated to the biblical history of the Jews.”84 But that judgment may not be entirely warranted. Now that the means for atonement provided for in the Bible of the Jews are no longer available, the author of the Palaea asks, “whence, O Jew, do you have hope for the remission of transgressions? Where do you have a scapegoat?” (169.9–10 [291]). His challenge to the Jews is in fact very much of a piece with the author’s interpretation of Israelite history as a story of sin and repentance.

2.2 The Sanctity of the Priesthood In its account of the golden calf incident, the Palaea fuses themes of moral indolence and ascetic discipline with another motif in the work: the inviolability of the office of the Aaronid high priesthood. For the Palaea, the episode of the golden calf was only secondarily a story about idolatry. Its primary message was to warn readers about the dangers of breaking a fast (92.4–93.5 [239]). Before Moses withdraws to Mount Sinai, he tells the Israelites to keep a fast for forty days. But when their lust for food gets the better of them, they implore Aaron to build for them an image of God. For they know that they may not sit down to feast until they first make an offering to God; and to that end, they need a visible idol. When Aaron asks them to surrender their jewelry to be smelted into an idol, he assumes that they will be reluctant to part with the goods that they had taken from Egypt. That is a miscalculation; the gluttony of some of them exceeds even their attachment to gold and silver. Any Byzantine Christian familiar with the rigors of the pre-Paschal forty-day Great Fast would also have had no difficulty understanding the underlying point of the Palaea’s retelling of the incident of the golden calf: do not break the fast, no matter how hungry you are. From the perspective of the Palaea, Aaron, the first high priest, shares no complicity in the making of the golden calf. If he might be held guilty of anything at all, it is in failing to appreciate the depths of the Israelites’ gluttony (93.2–5 [239]). Absolving Aaron of any culpability has the ring of partisan- ship—a suspicion borne out by the ensuing narrative. There, the author takes care to record the dire penalties imposed on those found guilty of flouting ritual law and priestly prerogatives. Even violations done innocently are subject to extreme penalties. When “Zan” attempts to steady the ox pulling the cart bearing the ark, God causes his hand to wither (cf. 2 Sam 6:7; 1 Chr 13:10); this was because “there was no provision in the law for the unholy to make contact with the ark” (107.6 [249]). God later kills the five sons of the priest

84 Flusser, “Palaea Historica,” 77. 30 Adler

Eli for partaking of the sacrificial offerings before the priest has blessed them (140 [270–71]). As with the figure of Aaron, the author has an exalted view of his descendants. According to Num 14:30, Caleb and Joshua were the only two adult Israelites from among those who had left Egypt who would be allowed to enter Canaan; this was because they were the only spies who assured Moses that God would deliver the land to them. By contrast, the Palaea replaces Caleb with Phinehas, the grandson of Aaron and high priest in the wilderness; because they praised the land, they alone of the older generation are allowed to enter Canaan (104.4–12; 105.13 [247–48]). In another example, Numbers 16 relates that God killed Korah, a Levite, and Dathan and Abiram, from the tribe of Reuben, for instigating a revolt against Moses. The Palaea reports the chain of events differently (100 [244–45]), making only Dathan and Abiram conspira- tors. And their transgression is not that of conspiring against Moses, but rather that of demanding a right reserved exclusively for the line of Aaron: “Why don’t we also offer incense to God, rather than Aaron alone? Did God appear only to Aaron? [Haven’t all of us also seen him on Mount Sinai?] Aren’t we also Jacob’s offspring, and don’t we also make up the twelve tribes?” (100.2 [244]). When they seize the censers, spread incense and make smoke, the ground divides and swallowed up the entire assembly of Dathan. It is only because Moses makes intercession to God that the assemblies of both men are not destroyed. Protection of the sanctity of the priesthood steers the author into a highly contested subject in Byzantine political ideology: the relationship between priesthood and kingship, and the division of the two realms. In his narrative of the origins of the two offices, the author does not disguise his own views on their relative standing. When God gives Moses instructions about the building of the Tabernacle, he tells him that Aaron will “make atonement for . . . the air and the storms and kings and rulers and the people and the whole world” (98.8 [243]). Soon thereafter, the other tribes are seized with jealousy, demanding to know “why Moses has not given us, too, a share of the priesthood, and not just his brother Aaron?” The leaves that subsequently sprout from Aaron’s rod after the other Israelite tribes complain about being barred from the Tabernacle ratify the divine origins of his high priesthood (99.3–8 [244]). That lofty view of the divinely ordained and privileged office of the Aaronid high priesthood contrasts sharply with the Palaea’s report of Saul’s ascent to the throne. According to 1 Sam 9:1–10:8, Kish, a man of great wealth, dispatched his son Saul and a servant to recover lost asses. When they arrived at Samuel’s city, young maidens drawing water instructed him and his servant to find Samuel. Because God had previously told Samuel about meeting a man who would save the people of Israel, he immediately recognized Saul as the future king of Israel and treated him accordingly. Parabiblical Traditions and their Use in the Palaea Historica 31

The sequence of events recorded in the Palaea is decidedly different. In Andrew of Crete’s Canon, Saul’s coronation is called a πάρεργον, that is, an inci- dental event: “When Saul once lost his father’s asses, my Soul, he incidentally found the kingdom” (Σαούλ ποτε, ὡς ἀπώλεσεν τοῦ πατρὸς αὐτοῦ, ψυχή, τὰς ὄνους πάρεργον τὸ βασίλειον εὗρεν).85 Proceeding from this single word, the author of the Palaea recounts in elaborate detail the rise to power of this accidental king, the first king of Israel (147–48 [276–77]). In response to the demands of the people, God, with seeming indifference, tells Samuel to anoint as king whomever he lights upon at the gate of the city. By chance, it turns out to be Saul, a poor manual laborer who had fallen asleep at the city gates following a hapless and disorganized attempt to recover the sheep that he had lost. When Samuel finds him, he unceremoni- ously anoints him and presents him to the people as king. The ensuing account of Saul’s reign goes on to describe him as a one-eyed tyrant who, among his other brutalities, gouges out the eyes of his subjects (148.6–7 [277]). While nothing like this is found in 1 Samuel’s record of his reign, a reader would have little difficulty in seeing Saul’s reign as foreshadowing that of Byzantine tyrants, for whom enucleation was a form of punishment documented from as early as the reign of Justinian II.86 In the highly contentious eighth-century controversy over the priestly prerogatives of royalty, both sides combed sacred scriptures for biblical prototypes to justify their own conflicting ideologies. As Dagron has amply demonstrated, the center of the discussion was Melchizedek, the priest/king. For emperors seeking to legitimate the fusion of priest and king into a single office, Melchizedek was the biblical exemplar of the king whose priesthood actually exceeded that of the hereditary priesthood.87 Advocates of this understanding of Melchizedek’s office would find little solace in the Palaea. Genesis 14:18 calls Melchizedek priest and king; the Palaea describes him only as “priest of God, the Most High” (36.9 [209]). That, of course, is self-evident from the narrative. Melchizedek, the solitary hermit on Mount Tabor, orphaned son of a murderous pagan king, was anything but a suitable model for the Byzantine emperor and priest. It is equally telling that, in presenting the history of the post-David monarchy, the Palaea mentions only one king, the Judahite king Uzziah.

85 Andrew of Crete, Canon (PG 97:1368CD). 86 See J. Lascaratos and S. Marketos, “The Penalty of Blinding during Byzantine Times,” Documenta Ophthalmologica 81 (1992): 133–44; cf. Flusser, “Palaea Historica,” 76 n. 104. 87 G. Dagron, Emperor and Priest: The Imperial Office in Byzantium (Cambridge: Cambridge University Press, 2003), 173–91. 32 Adler

(162 [287–88]). Opponents of iconoclast emperors saw God’s punishment of Uzziah for usurping priestly functions (2 Chr 26:19) as an admonition against overreaching emperors. In a fictionalized epistolary exchange between the iconoclast Emperor Leo III the Isaurian (c. 685–751) and pope Gregory II, Leo is famously reported to have written: “I am emperor and priest (βασιλεὺς καὶ ἱερεύς εἰμι).” Although ps.-Gregory’s response is a historical muddle, he did see in this claim a parallel between Leo and his impious biblical prototype: “In truth, Uzziah was also your brother, and he had your arrogance; and he tyrannized the priests of his time as you do today.”88 This is also the Palaea’s message. Like Leo III, Uzziah insists that his office grants him the authority to perform a priestly function, in this case that of burning incense in the Temple (162.5–6 [287]): “Am I not a priest? [Do I not wear the purple?] (οὐχὶ ἱερεύς εἰμι [καὶ ἀλουργίδα περιβέβλημαι])? “You are indeed king,” the priest replies, “but you are not allowed to burn incense” (ἀληθῶς βασιλεὺς εἶ σύ, ἀλλ᾿ οὐκ ἔξεστί σοι θυμιᾶσαι). As punishment for the king’s insistence upon burning incense anyway, God afflicts him with leprosy. When, out of respect for Uzziah’s office, the people compound the wrong by failing to remove the leprous king from the city, God refuses to communicate with them until Uzziah has died. The significance of the altercation between King Uzziah and the high priest over the right to burn incense could hardly have been lost on readers familiar with the long-standing controversy over the sacerdotal prerogatives of kingship.

3 Influence and Later Transmission

While Flusser’s characterization of the work as “unknown” may be an apt description in some circles, the Palaea was hardly an obscure or inconsequen- tial work in the centuries after its initial publication. The translation of the Palaea into Slavonic in the twelfth century, and subsequent extracts and con- densations of the work in that language, assured its standing in Russia and south Slavic countries; it continued to be read and used there well after the Bible first became available in a complete Slavonic translation in the late fifteenth century.89 An illustrated Greek poem on Genesis and Exodus, of some

88 J. Gouillard, “Aux origines de l’iconoclasme: Le témoinage de Grégoire II,” Travaux et Mémoires 3 (1968): 287.140–141. See further Dagron, Emperor and Priest, 164–66. Ps.-Gregory oddly confused king Uzziah’s impiety with Hezekiah’s removal of the brazen serpent. 89 See E. Turdeanu, “La ‘Palaea’ byzantine chez les Slaves du Sud et chez les Roumains,” Revue des Études slaves 40 (1964): 195–206 (reprinted in idem, Apocryphes slaves et rou- Parabiblical Traditions and their Use in the Palaea Historica 33

2800 verses, composed by the Cretan poet Georgios Chumnos around the year 1500, further attests to the work’s enduring influence. In both the contents and the arrangement of extrabiblical material, the poet was deeply indebted to the Palaea.90 As is to be expected in a work meant for popular edification, the transmission of the Palaea was fluid. In one manuscript of the Palaea (Paris Bib. Nat. gr. 37b), the scribe, upon reaching the end of the Palaea’s narrative of the reign of David, affixed to it the widely known chronicle of George the Monk (ninth century), thereby continuing the narrative down to the reign of the emperor Heraclius. Because the Palaea is neither a continuous nor a dated chronicle, the transition is hardly elegant.91 But in fusing the two works, the scribe recon- figured the Palaea into a kind of prehistory of the Byzantine empire, thereby reinforcing the connection between Byzantine emperors and their Israelite royal prototypes. Recent studies of the use of the Palaea in Slavic versions point in the same direction. Here, too, the Palaea’s account of the ancient Israelite monarchy seems to have functioned as a kind of charter for the creation of a political identity in newly created Slavic states.92 In the later transmission of the work, excerptors and copyists also ampli- fied the use of the Palaea as an expository guide to Byzantine liturgical texts. In a recently published article, J. Reinhart reported the existence of a manuscript featuring a section of the Palaea interleaved with “ein serbisch-kirchen slavischer Psalter mit liturgischen Supplementen,” dating to the year 1385.

mains de L’Ancien Testament [SVTP 5; Leiden: Brill, 1981], 392–403); Reinhart, “Die älteste Bezeugung der Historischen Paläa in slavischer Übersetzung,” 45–75. 90 For the text, see G. A. Mégas, ed., Hē Kosmogennēsis: Anekdoton stichourgēma tou 15. aiōnos emmetros paraphrasis tēs Geneseōs kai Exodou tēs Palaias Diathēkēs (Athens: Athenian Academy, 1975); for the Palaea as a source for the poem, see Mégas’s introduction, pp. 31–35. For an English translation, see Marshall, Old Testament Legends. For a discussion of the influence of the Palaea in the Greek-speaking world, see most recently A. Giannouli, “Apocryphon Lot,” 95–96. 91 In the Paris manuscript, the fusing of the two works occurs after the words: Δαυὶδ προφήτᾳ τῷ σῷ ὡϛ ἔπαυσαϛ ποτὲ λαῷ τὸν λαὸν πατάσσοντα ῥομφαίαν σου ἄγγελον (f.179v). See also Ven. Bibl. Naz. Marc. gr. 501, which, following the text of the Palaea, includes excerpts from the chronicle of George the Monk. For versions of the Palaea circulating together with George the Monk and other translated Greek chronicles, see I. Sorlin, “La diffusion et la transmission de la littérature Byzantine en Russie prémongole du XIe au XIIIe siècle,” Travaux et Mémoires 5 (1973): 385–408. Sorlin’s article deals with the transmission of the Hronografičeskaja Paleja, a work more allied with the Palaea Interpretata. 92 See I. Biliarsky, “Old Testament Models and the State in Early Medieval Bulgaria,” in The Old Testament in Byzantium (ed. R. Nelson and P. Magdalino; Washington, D.C.: Dumbarton Oaks Research Library and Collection; 2010), 255–76. 34 Adler

As Reinhart discovered, ff. 193r to 202v of the manuscript (Cod. slav. Vindob. Nr. 158), which the catalogue description called an “apokryphe Erzählung über David und Saul,” originated in the Palaea’s extended account of their reigns.93 Some of the later scribal additions also reflect the work’s use in a liturgical setting. Section headings in the copy of the Palaea preserved in Cod. Scorialensis Ψ.11.20 (gr. 455), a manuscript dating from the thirteenth century, seem to supply the day and the week when the passage was meant to be read publicly.94 In the middle of the narrative of the tower of Babel, the partial text of the Palaea preserved in Cod. Vatopedinus includes, rather abruptly, a lengthy collection of excerpts from the Psalms, along with running commentary. As Paramelle and Lucchesi have observed, the scribe, probably inspired by the recitation of the divine liturgy by Noah and his sons (along with explanatory glosses) appearing only a few lines before, extracted this material from a widely circulating liturgical commentary.95 Nor was this learned copyist reluctant to smuggle into the narrative of the Palaea parabiblical expansions originating in older Jewish sources. As Jubilees had done earlier ( Jub. 2:23), the scribe points out the symmetry of the twenty- two acts of creation with the twenty-two letters of the Hebrew alphabet, the twenty-two patriarchs from Adam to Jacob, and the twenty-two books of the Hebrew Bible as well. For extra measure, he even supplies the Hebrew names of each of the twenty-two biblical books (in Greek transliteration). Rehearsing the same explanation found earlier in the Sibylline Oracles and 2 En. 30:9–15, the copyist also inserts an exposition of the symbolic significance of the four Greek letters comprising the name Adam.96

93 J. Reinhart, “Die älteste Bezeugung der Historischen Paläa in slavischer Übersetzung,” 61–75. 94 See, for example f. 2v.: τῇ εʹ τῆς βʹ ἑβδομάδος περὶ τοῦ Ἀδάμ (“on the fifth day of the second week [of the Great Fast?)], concerning Adam”; f.3v: τῇ [πα]ρασκευῇ τῆς . . . ἑβδομάδος . . . Ἐνώχ (“on the Parasceve [i.e., sixth day] of the . . . week . . . Enoch”). Unfortunately, these headings (presumably in red ink) were not completely legible in the black and white reproduction of the manuscript. 95 Paramelle and Lucchesi, Philon d’Alexandrie: Questions sur la Genèse II, 1–7, 29–31. The authors demonstrate that the interpolation was derived from a commentary on the Divine Liturgy known as the Historia Ecclesiastica, and dubiously attributed to Germanus I, patriarch of Constantinople (715–730). 96 The copyist interpreted Adam’s name as an acrostic for the four directions of the world: “The forefather, formed by the hand of God, has the name Adam, consisting of four elements: ‘alpha,’ the first letter, meaning ‘east’ (ἀνατολήν); the second letter meaning ‘west’ (δύσιν); ‘alpha,’ the third letter, meaning ‘north’ (ἄρκτον); ‘mu’ the fourth letter meaning ‘south’ (μεσημβρίαν).” For a preliminary discussion of this manuscript as a witness to Parabiblical Traditions and their Use in the Palaea Historica 35

Other scribes inserted their own narrative expansions, many either drawing upon or paralleling Jewish traditions. According to the text found in Scorialenis (f. 33v), Jesse and his wife, following the birth of their first six sons, agree by oath to forego future marital relations. When Jesse decides to have another son, however, he approaches his wife’s maidservant, who informs her mistress about her husband’s intentions. Jesse’s wife then disguises herself as the maidservant. From this union, carried out in violation of their mutual oath, David was conceived. The story, which is meant here to explain David’s lament in Ps 50 LXX (MT 51):5 about “being conceived in sin,” is strikingly similar to a midrash quoted in the Yalqut ha-Makhiri of Makhir ben Abba Mari, a compilation probably originating in southern France and dating to the late fourteenth or fifteenth century.97 In another manuscript (Paris Bib. Nat. supp. gr. 928, ff. 7r–v), the copyist takes exception to the Palaea’s account of the death of Cain at the hands of the blind Lamech. After explaining his reasons for rejecting this widely known legend as not credible, he replaces it with a rival tradition, also first attested in Jubilees, that Cain died when his house collapsed around him ( Jub. 4:31). Because these traditions about the death of Cain and the symbolic meaning of the number twenty-two and Adam’s name were already known to other Byzantine writers, there is no need to assume that the scribe had unmediated access to Jubilees, 2 Enoch, or the Sibylline Oracles. The scribe responsible for the text of the Paris codex says in fact that he learned about the circumstances of Cain’s death from ps.-Dionysius the Areopagite.98 But what these examples do illustrate is the ongoing process of integrating traditions from Jewish parabiblical literature into the Byzantine construction of “biblical history.”

the text of the Palaea, see Paramelle and Lucchesi, Philon d’Alexandrie: Questions sur la Genèse II, 1–7, 22–24. 97 For an English translation of and commentary on this passage, see P. L. Culbertson, A Word Fitly Spoken: Context, Transmission, and Adoption of the Parables of Jesus (Albany: State University of New York Press, 1995), 250–52. For a discussion of the medieval witnesses to the same tradition, see also Ginzberg, Legends of the Jews, 6:246 n. 11. 98 For the same interpretation of the letters of Adam’s name, see Glycas, Annales 43.8–18. For the Jubilees-based tradition about Cain’s death when his house collapsed, see Syncellus, Chronography, 11.4–6, and above, n. 24. For the symmetry of the twenty-two acts of creation with the twenty-two letters of the Hebrew alphabet, the twenty-two patriarchs and the twenty-two books of the Hebrew Bible, see above, p. 9 and n. 27. 36 Adler

4 Prospects for Future Research

In its account of the death of Moses (121.1–8 [258]), the Palaea recalls how Moses, with Joshua at his side, ascends the mountain, and after surveying the land stretched out before him, instructs Joshua to return to the people and inform them of his death. Upon his death, Satan, identified here as “Samouel” (a corruption of “Sammael”), attempts to bring his body down the mountain to his people, so that they might worship him as a god. In the ensuing struggle over Moses’s body, the archangel Michael chastises Satan with the words, “the Lord rebukes you, Devil” (ἐπιτιμᾷ σε κύριος, διάβολε). He then routs him, and, at the behest of God, removes the body of Moses to an unknown place.99 Among New Testament scholars, the particular interest in this story arises from an allusion in the epistle of Jude to the same tradition: “But when the archangel Michael, contending with the devil, disputed about the body of Moses, he did not presume to pronounce a reviling judgment upon him, but said, ‘The Lord rebuke you’ (Ἐπιτιμήσαι σοι κύριος).’” (v. 9). Because the wording of Michael’s rebuke in the Palaea approximates the language of Jude, scholars have generally treated the Palaea as an independent and more developed witness to the same parabiblical work known to Jude, or at least one allied to it. Theories about the identity of this source abound, among them the Testament of Moses, the Assumption of Moses, or some later secondary conflation and expansion of both works.100 Speculation like this imposes a heavy burden on a passage in the Palaea numbering a scant fourteen lines of printed text. Like most of the other parabiblical material preserved in the work, the story is folded, without attribution, into the narrative. In formal terms, there is little in the Palaea’s notice of Moses’s death suggestive of either a “testament” or an “assumption.”

99 Although the variants in the Greek witnesses to this passage are too numerous to enumer- ate here, one reading in Vassiliev’s edition is almost certainly a corruption. The Vienna manuscript on which he based this reading states that Michael removed the body of Moses to a place “where he was commanded by Christ our God” (προσετάχθη παρὰ θεοῦ τοῦ Χριστοῦ ἡμῶν). Lacking in every other manuscript, the words τοῦ Χριστοῦ ἡμῶν were added by the copyist. 100 For a sampling of opinions, see R. H. Charles, The Assumption of Moses (London: Black, 1897), xlix–l; Flusser, “Palaea Historica,” 72–74; K. Berger, “Der Streit des guten und des bösen Engels um die Seele,” JSJ 4 (1973): 13–14; R. Bauckham, Jude and the Relatives of Jesus (Edinburgh: T&T Clark, 1990), 249–52; J. Tromp, The Assumption of Moses (Leiden: Brill, 1993), 281–82; Davila, The Provenance of the Pseudepigrapha, 151; and most recently F. Grierson, “The Testament of Moses,” JSP 17 (2008): 265–80. Parabiblical Traditions and their Use in the Palaea Historica 37

The purpose of the Palaea story is not to describe Moses’s parting words to Joshua, or his assumption into heaven. It is rather to explain why the location of the burial place of Moses remained unknown (cf. Deut 34:6). One common explanation was that angels had removed the body in order to prevent his corpse from becoming an object of cult. The Palaea’s narrative stands in the same tradition: by removing the corpse to an unknown place, Michael thwarted Satan’s plans to turn the Israelites to idolatry. Because Sammael’s plot is unattested in any of the other putative witnesses to the Testament or Assumption of Moses, scholars have tended to treat it as a later accretion.101 But this supposed “secondary development” is also the whole point of the narrative, and one incidentally well attested in both Christian and Jewish sources of late antiquity and the Middle Ages.102 I have cited this example in order to illustrate what in my view is at the heart of the analysis of the sources of parabiblical content in the Palaea. For students of Second Temple Judaism, the conventional approach is either to peel away the Christian overlay superimposed on a conjectural older stratum of Jewish tradition, or to identify remnants of some lost or fragmentarily preserved document. The improbable result of this approach is the proliferation of written sources to which the author, a monk of at best middling learning and literacy, allegedly had access: e.g., the Apocryphon of Lot, the History of Melchizedek, the Book of Jannes and Jambres, the Testament of Abraham, the Testament of Moses, and/or the Assumption of Moses. A tendency to cast the inquiry in purely textual terms also defines the understanding of the author’s relationship with the “Old Testament.” Explicitly or not, the default position assumes that, in writing the Palaea, the author sat down with a biblical codex, which he then under- took to paraphrase, rewrite, and supplement with parabiblical traditions and source material. There is no reason to assume, however, that the author viewed the “Old Testament” as a self-contained object of study and explanation, or even that he maintained a clear distinction in his own mind between “biblical” and “parabiblical.”

101 See Bauckham, Jude and the Relatives of Jesus, 252. 102 See, for examples, Origen, Selecta in Numeros (PG 12:577b); Aphrahat, Demonstrations, 8.9; Theodoret, Quaestiones in Octateuchum, 262.2; and the Armenian History of Moses. For an English translation of this last-named work, see M. E. Stone, “Three Armenian Accounts of the Death of Moses,” in Studies on the Testament of Moses (ed. G. W. E. Nickelsburg; SBLSCS 4; Cambridge, 1973), 120. For discussion, see further Bauckham, Jude and the Relatives of Jesus, 252 n. 51; B. Bitton-Ashkelony, “The Life of Peter the Iberian,” in Christian Gaza in Late Antiquity (ed. B. Bitton-Ashkelony and A. Kofsky; JSRC 3; Leiden: Brill, 2004), 119–23. 38 Adler

It is thus true, but materially irrelevant, to say that when, as is frequently the case, the author of the Palaea strayed from the biblical narrative, he showed “lack of attention to the text of the Bible,” “wrote without looking in his Bible,” or “neglected the biblical text.”103 Byzantine Christians were not Protestants. According to some estimates, the majority of Byzantine Christians would have heard no more than fifteen percent of the Bible in the course of the liturgical year.104 Knowledge of biblical history was available to them through many channels, not all of them literary. Given this, it would be more useful to think of the author not as a biblical commentator, but rather as an expositor of a shared “biblical history,” familiar to a wider audience through popularly known chronicles, religious images, pilgrimage narratives, hymns, lectionaries, and homilies. A piece of local folklore or the image of Abraham cutting Mechizedek’s hair, of Lot watering a tree, or of an angel’s sword dangling ominously over the head of David, would have stirred the author’s creative impulses in the same way that a troparion or even a word or phrase in a Byzantine penitential hymn could. The exploration of parabiblical material in the Palaea would thus be better served by extending the study of the work beyond the search for lost documents.105 The copious parallels between the Palaea and parabiblical themes

103 Flusser, “Palaea Historica,” 70, 75, 79. 104 See Krueger, “The Great Kanon,” 15. 105 It is lamentable that modern scholarship on the Greek text of the Palaea continues to rely on an obsolete 1898 edition, for which the editor consulted only two manuscripts. Apart from resolving difficulties in the Greek text of Vassiliev’s printed edition, the other unedited complete or partial Greek witnesses (numbering about 11) have much to contribute to our understanding of the transmission and reception history of the work. For a few examples of the textual value of these unpublished witnesses, see, for example, 32.5 (207): “They thus remained in the palace built by Nimrod, and continued their rule up to the reign of Archisedek, king of Babylon” (καταμένοντες ἐβασίλευον ἀρτὶ τῆς βασιλείας Ἀ̓ρχισεδὲκ τῆς Βαβυλῶνος). Flusser (“Palaea Historica,” 57) rightly took note of the “strange- ness” of the name. The preferred reading is to be found in Ven. Bibl. Naz. Marc. gr. 501, f. 13r– v: καταμένοντες ἐβασίλευον ἀρτὶ τῆς βασιλείας Βαβυλῶνος ἀρχούσης; cf. Paris Bib. Nat. gr. 37, f. 68r: καταμένοντες ἦσαν καὶ ἐβασίλευον ἀρτὶ τῆς βασιλείας Βαβυλῶνος ἀρχούσης. See also 22.2 (199): “After this, the ark (of Noah) came to rest on the mountain of Ararat near the Bactrians (εἰς τὸ ὅρος τὸ Ἀραρὰτ πλησίον Βάκτρων), between the Assyrians and those known as the Amanites.” Bactria is an area in northern Afghanistan and Pakistan, far removed from Ararat. For the correct reading, see Ven. Bibl. Naz. Marc. gr. 501: πλήρης βάκτρων (“full of sticks”). The author included this information about Ararat to explain his subsequent report about the dove’s discovery of the olive twig on the mountain (22.9 [199]). For a description of Ven. Bibl. Naz. Marc. gr. 501, which contains one of the oldest witnesses to the Greek text of the Palaea Parabiblical Traditions and their Use in the Palaea Historica 39 depicted in the iconography of the Byzantine, Coptic, and Slavic churches are a subject in need of a much more systematic examination. Comparable to the form-critical methods employed in biblical studies, a systematic analysis of passages in the Palaea that have an oral Sitz im Leben (e.g., liturgy or Holy Land folklore) might yield valuable results. Stories connected to well-known sites in the Holy Land—Lot’s tree, Habakkuk’s field, Melchizedek’s grotto, and the place of Moses’s death—suggest a connection with pilgrimage narratives. Given that a substantial part of the Palaea takes the form of a commentary on the Psalter and the Great Canon of Andrew of Crete, there would probably also be something to be learned from liturgical commentaries, many of them yet unpublished. Like the Palaea, some of these commentaries appear to take the form of historical explanations of verses from the Great Canon.106 One should finally not lose sight of the author’s own literary techniques and ideological leanings in “narrativizing” this disparate mass of material. While source criticism of the Palaea may have seen its day, there are still other interpretive methods waiting to be used.

(13th cent.), see E. Mioni, ed. Bibliothecae Divi Marci Venetiarum codices Graeci manu- scripti (7 vols.; Rome: Istituto poligrafico e zecca dello Stato, Libreria dello Stato, 1981– 1985), 2:338. 106 See A. Giannouli, “Die Kommentartradition zum Grossen Kanon des Andreas von Kreta—Einige Anmerkungen,” Jahrbuch der Österreichischen Byzantinistik 49 (1999): 143–59. Giannouli (156–57) notes the existence of an unedited collection of excerpts from a commentary on the Great Canon linking troparia from Andrew’s hymn to “stories from the Old Testament” (ἱστορίας ἐκ τῆς Παλαιᾶς Διαθήκης), among them Cain and the seven divine judgments against him, Lamech, Seth, the sons of Noah, Abraham, Jacob, Esau, Moses, et al. Outsider Impurity: Trajectories of Second Temple Separation Traditions in Tannaitic Literature

Yair Furstenberg

1 Introduction

Rabbinic and Qumranic literature share the notion that Jews who did not adhere to their respective observances were impure. Both corpora include strict measures of separation in order to avoid contracting this impurity. The ʿammei ha-ʾareṣ—literally, “people of the land,” as such persons are called in rabbinic literature—and the regulation of contact with them, feature prominently in rabbinic halakhah, from the earliest glimpses we have of its incep- tion in the Second Temple period.1 At the same time, tannaitic sources do not supply much of a conceptual framework concerning the nature and intent of the separation from the ʿam ha-ʾareṣ.2 What was the ideological import of this separation, and why was the ʿam ha-ʾareṣ considered impure?3 The Qumran sect, in contrast, cast separation from nonmembers in clear ethical and spiritual terms. The Rule of the Community attributes the impurity of the nonmember to the ruling cosmic forces. This impurity was identi-

1 See, for example, m. ʿEd. 1:14; m. Ṭehar. 8:1. 2 The ʿam ha-ʾareṣ poses a substantial challenge to any attempt to extract social history from rabbinic literature. Earlier works assumed this was a recognized social group, but they offered very different accounts of its provenance. See A. Büchler, Der galiläische Am-ha‌‌‌ʾAres des zweiten Jahrhunderts: ‎Beiträge zur innern Geschichte des palästinischen Judentums in den ersten zwei Jahrhunderten (Vienna: Hölder, 1906; repr: Hildesheim: G. Olms, 1968); A. Oppenheimer, The Am Ha-Aretz: A Study in the Social History of the Jewish People in the Hellenistic–Roman Period (Leiden: Brill, 1977). Büchler restricts the references to ʿam ha-ʾareṣ as a defined social group to second-century Galilee, whereas Oppenheimer views ʿammei ha-ʾareṣ as a long-standing social phenomenon. In a paper dedicated to the history of this term, I have suggested that it be understood as a constructed rabbinic label rather than a recognized title of a defined social group. As such its exact meaning and scope underwent substantial transformations with the adaptation of rabbinic policy to changing social contexts. See Y. Furstenberg, “Am Ha-aretz in Tannaitic Literature and its Social Contexts,” Zion 78 (2013): 287–319 (in Hebrew). 3 An extremely harsh attitude towards the ʿam ha-ʾareṣ is expressed in b. Pesaḥ. 49a–b. This unit however is a product of Babylonian rabbinic culture and cannot help us identify the underlying policy towards the ʿam ha-ʾareṣ in Palestine. See J. L. Rubenstein, The Culture of the Babylonian Talmud (Baltimore: Johns Hopkins University Press, 2003), 123–42.

© koninklijke brill nv, leiden, ��5 | doi ��.��63/9789004299139_003 Outsider Impurity 41 fied with the person’s sinful ways under the dominion of the spirits of Belial. Impurity is an inherent quality of sinners who refuse to enter the covenant of God.4 Violators of sectarian precepts become impure and are excluded from Achieving purity is possible only by 5.(טהרת הרבים) ”the purity of the “many joining the sect, as it is a privilege denied to those who reject the sect’s teachings: “For in the filth of wickedness is his plowing and there is contamination in his repentance. . . . He cannot be purified by atonement, nor be cleansed by waters of purification, nor sanctify himself in streams and rivers” (1QS 3:2–5).6 Ritual purification is useless for removing the intrinsic impurity of the outsider who does not yield to the laws of the Yaḥad: “For there is impurity among all those who transgress his [God’s] words” (1QS 5:13–14).7 Rabbinic literature records no such statements. It seems as though the rabbis and their Second Temple predecessors were concerned with distanc- ing themselves solely from what we might call “levitical impurities”—e.g., corpses, creeping things, menstruants, and others who had genital discharges. Outsiders were avoided by the more scrupulous “Pharisee” and “associate”

4 The association of sin with impurity in Qumran is discussed by D. Flusser, “The Dead Sea Sect and Pre-Pauline Christianity,” in idem, Judaism and the Origins of Christianity (Jerusalem: The Hebrew University Magnes Press, 1988), 23–74 (50–53). See also F. García Martínez, “The Problem of Purity: The Qumran Solution,” in The People of the Dead Sea Scrolls: Their Writings, Beliefs and Practices (F. García Martínez and J. Trebolle Barrera; trans. W. G. E. Watson; Leiden: Brill, 1995), 139–57; J. Klawans, Impurity and Sin in Ancient Judaism (New York: Oxford University Press, 2000), 67–91. 5 1QS 6:24–7:25. 6 Translations of Rule of the Community follow E. Qimron and J. H. Charlesworth, The Dead Sea Scrolls: Hebrew Aramaic and Greek Texts with English Translation, Rule of the Community and Related Texts (Tübingen: Mohr Siebeck; Louisville: Westminster John Knox, 1994), 13. 7 The supposed conflation of sin and ritual impurity in Qumran writings is contested by M. Himmelfarb, “Impurity and Sin in 4QD, 1QS and 4Q512,” DSD 8 (2001): 9–37. Himmelfarb contends that the Rule of the Community employs the terminology of impurity and purity only metaphorically and does not consider sins to be ritually defiling. A more nuanced formulation has been suggested by H. Birenboim, “‘For He is Impure among All Those Who Transgress His Words’: Sin and Ritual Defilement in the Qumran Scrolls,” Zion 68 (2003): 359–66 (in Hebrew). In his opinion, sin does not necessarily generate ritual impurity; however, purification from the wretched human state is granted only to the righteous. This view was further elaborated by M. Kister, “On Good and Evil: The Theological Foundations of the Qumran Community,” in The Qumran Scrolls and their World (ed. M. Kister; 2 vols.; Yad Izhak Ben-Zvi: Jerusalem, 2009), 2:497–528 (519, 525) (in Hebrew). 42 Furstenberg

(ḥaver)8 due to their presumed negligence in such matters.9 Aharon Shemesh thus asserts that unlike Qumran law, which treated all outsiders as intrinsically impure, “Rabbinic halakhah would undoubtedly permit an ʿam ha-ʾareṣ to eat with ḥaverim, provided that he was willing to undergo a ritual cleansing of his body and of his clothing.”10 Should we therefore assume that the Qumran system of separation is completely irrelevant for understanding the rabbinic strictures for separation from the ʿam ha-ʾareṣ? Are these two systems only superficially similar? There are indeed clear differences between the two corpora, specifically regarding the association of the “other” with sin. I contend, however, that these differences should not obscure the existence of a foundational legal structure common to the two traditions. An examination of the halakhic details reveals that the earliest rabbinic sources are best understood against the backdrop of Qumranic and other Second Temple traditions regarding the impurity of outsiders. Extracting clusters of halakhic traditions belonging to the Second Temple period from their current setting within tannaitic literature reveals a system

is unclear, although scholars tend to חברים and פרושים The exact relationship between 8 associate the two. See, for example, E. Schürer, The History of the Jewish People in the Age of Jesus Christ (175 BCE–135 AD) (trans. T. A. Burkill et al.; rev. and ed. G. Vermes, F. Millar and M. Goodman; 3 vols. in 4 parts; Edinburgh: T&T Clark, 1973–1987), 2:398. For a more cautious approach see M. Goodman, The Ruling Class of Judaea (Cambridge: Cambridge University Press, 1987), 83. An examination of the earliest tannaitic sources associated explicitly with Beit Shammai and Beit Hillel, and which mention either of these terms, points to a subtle difference between the two. The Pharisees are concerned with purity, and they make sure to distance their foods from an ʿam ha-ʾareṣ (see for example, t. Shabb. 1:13 [S. Lieberman, The Tosefta According to Codex Vienna (4 vols.; New York: Jewish Theological Seminary of America Press, 1955–1988; rev. ed. 1992–1995), 2:4]). The ḥaverim, however, exceed in their scrupulousness. They restrict all contact with an impure ʿam ha-ʾareṣ, so as to be publically deemed trustworthy in these matters (m. Demai 2:3 and t. Demai 2:12 [Lieberman, Tosefta 1:71]; see discussion below). Membership is thus contingent upon adherence to a strict code of social separation. This difference seems to reflect degrees within the Pharisaic movement itself. Although the movement was known for its concern for purity matters, its boundaries were naturally blurry and encouraged the creation of a sub-elite in the form of a special association (see Furstenberg, “Am Ha-aretz in Tannaitic Literature,” 309–10). 9 Klawans, Impurity and Sin, 108–9. 10 A. Shemesh, “The Origins of the Laws of Separatism: Qumran Literature and Rabbinic Halacha,” RevQ 18 (1997): 223–41 (232). The contrast between this rabbinic approach and the notion of spiritual purification according to the Qumran writings is highlighted by J. M. Baumgarten, “The Purification Rituals in DJD 7,” in The Dead Sea Scrolls: Forty Years of Research (ed. D. Dimant and U. Rappaport; STDJ 10; Jerusalem: The Hebrew University Magnes Press and Yad Izhak Ben-Zvi; Leiden: Brill, 1992), 209. Outsider Impurity 43 surprisingly close to the Qumranic laws of “outsider impurity.” Against this shared purity discourse, we can better evaluate specific ideological differences between Qumranic purity legislation and the Pharisaic system embedded in early rabbinic sources. Rabbinic policy toward the ʿam ha-ʾareṣ diverged from Second Temple separation traditions only later, as the early halakhic traditions were transferred to a new social setting.

2 Separation from Outsiders in the Second Temple Period

Shemesh explains that the members of the Yaḥad saw themselves as the true Israel. Statements which seem at first to represent exclusively sectarian ideology—e.g., “to love all that He has chosen, and to hate all that He has rejected . . . to love all the sons of light, each according to his lot in the Council, and to hate all the sons of darkness” (1QS 1:3–10)—are in fact adaptations of earlier expressions of enmity between the people of Israel and Gentiles. In particular, Shemesh points to the resemblance between the laws of separation in the Rule of the Community and the commandments regarding separation from Gentiles given in Jub. 22:16–22.11 Notably, Shemesh’s comparison of the separation themes is based on the Cave 1 version of the Rule of the Community (1QS 5:7–20). At the same time, an examination of the variant versions of this text found in Cave 4 (4QS)12 reveals not only the shared theme of separation, but also additional closely related clusters of expressions embedded in both the Serekh instructions and Jubilees. The rules concerning separation assume a similar role in both texts. In Jubilees, the rules of separation follow Abraham’s deathbed blessing to Jacob. Abraham demands that his grandson accept upon himself the commandments of his grandfather (v. 16), which involve one major concern: separation from Gentiles (vv. 16–22). Similarly, the Rule of the Community locates the rules for separation from nonmembers at the definitive moment of entering the covenant and returning to the Torah. These laws thus constitute the only specific commitment that the novice accepts upon himself:

11 Shemesh, “Origins of the Laws of Separatism,” 234–38. Shemesh argues elsewhere that Jubilees attained canonical status at Qumran; see A. Shemesh, “Scroll 4Q265 and the Status of the Book of Jubilees in the Yaḥad,” Zion 73 (2008): 5–20 (in Hebrew). 12 The relevant fragments, 4Q256 (= 4QSb) frag. 4 and 4Q258 (= 4QSd) frag. 1, and the translation used here (with some modifications), appear in P. S. Alexander and G. Vermes, Qumran Cave 4.XIX: Serekh Ha-Yaḥad and Two Related Texts (DJD 26; Oxford: Clarendon, 1998), 53, 93–94. 44 Furstenberg

Jub. 22:16–17, 20 13 4QSd 1 i 5–10

And you also, my son Jacob, remember And everyone who enters the council of my words and keep the commandments the Community shall take upon himself of Abraham, your father. with a binding oath to return to the Torah of Moses with all his heart and with all his soul [. . .]14

Separate yourself from the Gentiles, to be separated from all the men of injustice

Furthermore, they shall not touch the purity of the men of holiness.15 and do not eat with them, and he shall not eat with him in community. and do not perform deeds like theirs. Furthermore no man of the men of the community shall give answer in accordance with their opinion relating to any Torah or judgment.

And do not become associates of theirs; Furthermore, he shall not be associated with him in possessions or in work, and no man of the men of holiness shall eat from their possessions nor take from their hand anything.

13 The translation follows (with slight modifications) that of O. S. Wintermute, OTP 2:98. 14 At this point the text adds that the obligation is to adhere to the interpretation of the Law of Moses specifically “in accordance with the council of the men of the Yaḥad” (4QSd 1 i 7). The parallel text in 1QS 5:9–10 suggests a different sectarian framework including the sons of Zadok. 15 The syntax of this phrase links it quite poorly to its surroundings. It may plausibly be considered a later addition, intended to emphasize the impurity of the men of injustice. Thus, although the scribe of 1QS starts a new paragraph at this point (5:13) and seems to begin a new subject (“He shall not enter the water to touch the purity of the men of holiness”), it is clear from the 4QS version that this detached phrase complements the original list of strictures. See J. Licht, The Rule Scroll: A Scroll from the Wilderness of Judaea: 1QS, 1QSa, 1QSb: Text, Introduction and Commentary [Jerusalem: Bialik Institute, 1965], 128–29 (in Hebrew). Outsider Impurity 45

And they shall not depend on the people Because their works are defiled, and all of vanity. For vanity are all who do not of their ways are contaminated, and know His covenant, and all those who despicable, and abominable . . .16 and despise his word He shall blot out from all their works are worthless and vain. the world, and all their works are defilement before Him and all their possessions are unclean. Be careful, my son Jacob, that you not take a wife from the seed of the daughters of Canaan.17

4QS preserves the same sequence of laws as Jubilees, but gives each general instruction on separation from Gentiles a sectarian interpretation. Thus, for example, the prohibition against “works like those of the Gentiles,” which originally referred to idolatry and related practices, is reframed within a Jewish context, and read as an injunction to avoid following dissident opinions regarding the interpretation of the Torah. In its concluding statement, 4QS attributes impurity to all the works of the “people of vanity”; i.e., those to whom the foregoing laws of separation apply. This is very similar to Jubilees’ concluding pronouncement on the contamination and vanity of the works of the Gentiles. This same set of prohibitions appears in a longer adaptation in 1QS (5:7–20), the version discussed by Shemesh. Although the exact textual relationship between the two versions of the Serekh texts is obscure,18 they clearly follow the same structure. Yet, 1QS diverges from the parallel 4QS version in two inter- related aspects. First, it elaborates on the dire fate of those not in the covenant, who are condemned at length for their impurity and vanity. Second, 1QS diverges from both Jubilees and 4QS in that it explicitly defines those outside the covenant as themselves impure, not only their actions and possessions: “For there is impurity among all those who transgress his words” (5:13–14). This statement, when read against its textual forerunners, does not derive exclusively from a unique Yaḥad ideology. Rather, this is a subtle and almost natural development of the same impurity notion embedded in the earlier texts. We can easily follow the trajectory of the separation tradition, from its earliest stage (Jubilees) prohibiting association with Gentiles, whose works are impure; through a parallel claim against other Jews (4QS); to the more stringent notion

16 Here, Jubilees specifies the impure deeds; e.g., sacrifices to the dead, eating in tombs. 17 This last prohibition is derived from Gen 28:1. 18 See C. Hempel, “The Literary Development of the S Tradition: A New Paradigm,” RevQ 22 (2006): 389–401. 46 Furstenberg of the inherent impurity of nonmembers (1QS). In 1QS, the outsider, whether a Gentile or a sinning Israelite, is deemed intrinsically impure.19 A parallel development is manifest also in one peculiar element in Josephus’s description of the Essene impurity system:

They are divided into four groups, according to the period of their train- ing; and so far are the juniors inferior to the seniors, that if by chance they touch (ψαύσειαν) them, they (the seniors) wash themselves, as if they had intermixed with a foreigner (ἀλλοφύλῳ συμφυρέντας). (Josephus, J.W. 2.150)

The touch of a newer member contaminates the senior member, as if the former were a Gentile. Josephus assumes that “intermixing” with Gentiles is defiling;20 he adduces separation from Gentiles as the model that the Essenes chose to regulate the impurity of new members. Though junior members accept the same purity regulations as those more senior and are now developing their skill in matters of purity, their level of impurity is still that of a

19 Hannah K. Harrington has recently argued for the impurity of all outsiders, whether Jews or Gentiles, according to Qumran literature. However, her argument rests more on the silence of the sources, which do not distinguish between different kinds of external threats, than on direct textual parallels between Gentiles and Jewish outsiders, as I propose here. See H. K. Harrington, “Keeping Outsiders Out: Impurity at Qumran,” in Defining Identities: We, You, and the Other in the Dead Sea Scrolls. Proceedings of the Fifth Meeting of the IOQS in Groningen (ed. F. García Martínez and M. Popović; STDJ 70; Leiden: Brill 2008), 187–203. 20 Jonathan Klawans and Christine Hayes have attempted to undermine the relevance of Josephus’s statement for assessing the significance of Gentile impurity within Second Temple Judaism. Both claim that Josephus alludes only to a sectarian notion of Gentile impurity, which the sectarians imposed on novices as well. Josephus, however, speaks not of a sectarian stringency against touching Gentiles, but rather of a more generalized notion of avoidance of intermingling with them. The verb συμφύραω concerns association, Ezra ,התערבו) and it is reminiscent of Ezra’s language concerning mixing with Gentiles 9:2). Josephus’s account thus testifies to the evolution from a notion of social separation (from impure Gentiles) to one of defilement through touch, very close to the adaptation of the Jubilees tradition in the Serekh texts. See J. Klawans, “Notions of Gentile Impurity in Ancient Judaism,” AJS Review 20 (1995): 285–312; C. E. Hayes, Gentile Impurities and Jewish Identities: Intermarriage and Conversion from the Bible to the Talmud (New York: Oxford University Press, 2002), 63–67. See also V. Noam, “‘The Gentileness of the Gentiles’: Two Approaches to the Impurity of Non-Jews,” in Halakhah in Light of Epigraphy (ed. A. I. Baumgarten, H. Eshel, R. Katzoff, and S. Tzoref; JAJSup 3; Göttingen: Vandenhoeck & Ruprecht, 2011), 27–41. See also Harrington, “Keeping Outsiders Out,” 192. Outsider Impurity 47

Gentile, on the lowest rung. The system arranges all people within a set of con- centric circles, and any contact with someone in any external circle, no matter how close, communicates the most severe impurity. The concern is clearly not that some Levitical impurity such as semen or corpse impurity might be con- tracted from new members, but rather that less than full-fledged membership itself imparts a degree of impurity. Thus, this system is best understood as a nuanced expression of outsider impurity. According to Josephus, who seems at this point to faithfully represent a widespread notion in sectarian circles, this conception is modeled after the notion of the impurity of the Gentile, the ultimate outsider. In this scheme, impurity is an inherent result of the very distance—short or long—from the Essene elect. Hence, both the Rule of the Community and Josephus’s Essenes blur the differences between Jew and non- Jew and ascribe intrinsic impurity to all outsiders. At the same time, Josephus seems to ascribe to Essene circles a further development of this notion, in which outsider impurity serves as the theoretical foundation for separation and hierarchy in general, even in relation to internal groups.21

3 ʿAm ha-ʾAreṣ Impurity in Early Rabbinic Law

Two fundamental characteristics of Yaḥad/Essene outsider impurity are thus modeled after the notion of Gentile impurity. First, only members can achieve purity. Second, outsider impurity establishes the framework for creating a graded system of impurity. All those outside of one’s own circle are deemed impure on the same level, as outsiders, whether Jews or Gentiles. In this section, I will argue that this very same conception and its implications characterize the earliest rabbinic sources regarding the ʿam ha-ʾareṣ. Although the Pharisaic system embedded in early rabbinic sources differs in detail from the Qumran laws, the two systems share a common conceptual and legal framework based on a notion of outsider impurity.

21 While Josephus presents a graded system based on this one principle only, column 6 of the Rule of the Community employs a more complex system of impurity levels with regard to new members. Elsewhere I argue that the Rule of the Community integrates the sectarian notion of outsider impurity with the biblical model (according to Qumran interpretation) of the gradual reentry of the impure person into the camp. See Furstenberg, “Eating in the State of Purity during the Tannaitic Period: Tractate Ṭeharot in its Cultural and Social Contexts” (Ph.D. diss., The Hebrew University of Jerusalem, 2010), 148–58 (in Hebrew). 48 Furstenberg

The first source pertinent to understanding the Pharisaic notion of ʿam ha-ʾareṣ impurity appears in Mishnah Ḥagigah, which has long been considered part of the earliest stratum of rabbinic literature. The specific unit discussed here, m. Ḥag. 2:5–7, which shows clear signs of being an early halakhah,22 presents a developed system of degrees of purity:23

]2:5[ נוטלים לידיים לחולין ולמעשר ולתרומה, ולקודש מטבילים, ולחטאת אם ניטמו ידיו ניטמא גופו.

]2:6[ הטובל לחולים והוחזק לחולים אסור למעשר. הטובל למעשר והוחזק למעשר אסור לתרומה. הטובל לתרומה והוחזק לתרומה אסור לקודש. הטובל לקודש והוחזק לקודש אסור לחטאת. הטובל לחמר הותר לקל. טבל ולא הוחזק כילו לא טבל.

]2:7[ בגדי עם הארץ מדרס לפרושים. בגדי פירושין מדרס לאוכלי תרומה. בגדי אוכלי תרומה מדרס לקדש. יוסה בן יועזר היה חסיד שבכהונה והיתה מטפחתו מדרס לקודש. יוחנן בן גודגדא היה אוכל על טהרת הקודש כל ימיו והיתה מטפחתו מדרס לחטאת.

[2:5] One must wash his hands for unconsecrated foods, tithes and heave- offerings, but immerse them for hallowed things. With regard to [the preparation of the] purification waters,24 if his hands have been defiled, his whole body is defiled.

[2:6] He who immerses for unconsecrated food, and is “held”25 for unconsecrated food, may not touch tithes. He who immerses for tithes, and is

22 J. N. Epstein, Introduction to Tannaitic Literature: Mishna, Tosephta and Halakhic Midrashim (Tel Aviv: Dvir; Jerusalem: The Hebrew University Magnes Press, 1957), 46–52 (in Hebrew). Epstein concludes from the names mentioned in m. Ḥag. 2:7, Yose b. Yoezer and Yoḥanan b. Gudgada, who were Second Temple sages, that this passage is significantly earlier than the parallel tosefta, which mentions post-70 rabbis, Rabban Gamaliel and Onqelos (t. Ḥag. 3:3, ed. Lieberman, Tosefta, 2:387). In addition, the label “Pharisee” as denoting a level of purity is known only from early tannaitic sources (see n. 8 above). 23 The Hebrew text is given according to Ms. Kaufmann A 50 (The Library of the Hungarian Academy of Sciences); the translation is my own. 24 In this current context, the term ḥattat refers to the preparation of the purification waters, including the treatment of the ashes, which required the highest level of purity. See, for example, m. Parah 10. This usage follows the biblical definition of the red heifer as a sin- offering (Num 19:9), although it is questionable whether the rabbis considered it an actual sacrifice. 25 See discussion below on the meaning of this term. Outsider Impurity 49

“held” for tithes, may not touch the heave-offering. He who immerses for the heave-offering, and is “held” for the heave-offering, may not touch hallowed things. He who immerses for hallowed things, and is “held” for hallowed things, may not touch purification waters. He who immerses for the higher sanctity is allowed to touch what is on a lower sanctity. If he immersed without being “held,” it is as though he had not immersed at all.

[2:7] The clothes of an ʿam ha-ʾareṣ carry midras impurity26 for Pharisees. The clothes of Pharisees carry midras impurity for priests eating heave- offerings. The clothes of heave-offering eaters carry midras impurity for hallowed things. Yose b. Yoezer was the most pious of the priesthood, yet his apron carried midras impurity for hallowed things. Yoḥanan b. Gudgada always ate in accordance with the purity of hallowed things, yet his apron counted as carrying midras impurity with respect to purification waters.

-in m. Ḥag. 2:6, translated literally here as “held,” is regu הוחזק The verb larly understood to denote intention. This understanding of the mishnah is reflected in both Talmuds,27 and possibly already in the Tosefta.28 According to this rendering, purification on all levels depends solely on personal intention and decision.29 In other words, as one immerses he must decide what kind of

26 Clothing and objects upon which one regularly sits contract impurity from those with severe genital impurity (zav, zavah), even through indirect contact. Physical pressure such as “treading” (midras) renders such objects impure; these objects in turn have the power to defile other people through direct pressure and contact. 27 Both Talmuds compare this mishnah to other sources dealing with intention during ablu- tions. Accordingly, both rephrase the mishnah to create a new term denoting intention, “to hold oneself”: “As long as his feet are in the water he may hold himself to any level he wishes” (y. Ḥag. 2:6 [78b]; b. Ḥag. 19a). point of] holding)? Once]) חזקה T. Ḥag. 3:1–2 (Lieberman, Tosefta, 2:386): “What is the 28 he raises his feet from the water. If his feet are still in the water, and he immersed to the lighter degree but was held for the more severe degree, what he has done is done.” For further analysis see Furstenberg, “Eating in the State of Purity,” 237. 29 One of the problems raised by this interpretation is the seemingly redundant phrasing of He who immerses himself for unconsecrated“ .הטובל לחולין והוחזק לחולין :the mishnah -refers only to a person’s intention dur הוחזק food, and is ‘held’ for unconsecrated food.” If ing immersion, why does the mishnah present immersion and intent as two separate conditions? Indeed, the printed editions of the Mishnah and the Babylonian Talmud read the second condition as an apodosis: “If one immerses himself for unconsecrated food, he is held for unconsecrated food etc.” The interpretation suggested below solves this problem. 50 Furstenberg purity he wishes to accept upon himself, and his scrupulousness with regard to that specific level of purity will be determined accordingly. Read in this way, m. Ḥagigah seems to express a distinctively rabbinic notion of purity, which acknowledges a variety of levels but at the same time enables each individual, even an ʿam ha-ʾareṣ, access to all of them—provided that he is aware of the specific obligations they entail. However, this accepted interpretation hardly matches the standard mean- held,” as attested elsewhere. Without exception, in“ ,הוחזק ,ing of the verb other contexts this verb denotes not subjective intentionality but recognized social status and public image, the way a person is perceived by others.30 If we in m. Ḥag. 2:6, the mishnah takes on a whole הוחזק employ this meaning of new meaning: the purifying act of immersion must be coupled with public recognition of one’s status.31 Purification is otherwise incomplete. Although immersion is the same for all levels (in contradistinction to hand purification in m. Ḥag. 2:5), a person is deemed pure only at the level with which he is publically associated. He is barred from any food on a higher level of sanctity. Thus, if someone is publically associated with the practice of eating unconsecrated food in a state of purity, his immersion purifies him only for that practice. His level of purity does not allow him to eat heave-offerings, while the priest who is permitted to eat heave-offerings is considered impure with respect to sacrifices, the next level up. The mishnah does not specify the conditions for belonging to a specific level. However, it takes public recognition of personal status to be a known social fact and assumes that such recognition is a necessary aspect of purification. Indeed, it is unclear from m. Ḥag. 2:6 how the public association with a specific level of purity is achieved, and when this association takes place. Does it follow the immersion or is it a prior or parallel condition?32 The following mishnah, however, gives us a clear idea of the significance of being publicly affiliated with a recognized level of purity. Mishnah Ḥagigah 2:7 assumes a high level of segregation between different groups: the ʿam ha-ʾareṣ, the Pharisee, the priest

30 See, for example, m. Qidd. 4:5; t. Ketub. 5:1; t. B. Bat. 7:3; b. Giṭ. 14a. 31 Eduard Baneth attempted to reinterpret this mishnah in accord with the general sense -He suggested that each case should in fact be divided into two pos .הוחזק of the verb sibilities. Either the person has undergone proper purification in the presence of others who can testify to the fact, or he is presumed to have done so previously. Baneth admits that this interpretation is insufficient to explain the final case in this mishnah. See Mischnaioth: Die sechs Ordnungen der Mischna (ed. A. Sammter et al.; 6 vols. in 7; Berlin: Itzkowski; Wiesbaden: Kanel, 1887–1933), 2:484. and then the imperfect (הטובל) The verbal sequence in this sentence, first the participle 32 .could indicate either possibility ,(הוחזק) Outsider Impurity 51 eating heave-offerings, and one who eats his sacrifices in Jerusalem. Each is considered a carrier of severe midras impurity with respect to the higher levels, and any direct contact between people on different levels is completely proscribed. In other words, a person’s group affiliation determines the scope of his permissible social interactions. This affiliation must therefore be easily recognizable. A person must be publically “held” as pure to a particular degree, exactly as the previous mishnah requires, so that others may avoid contact with him. Although no clear social setting is disclosed in m. Ḥag. 2:6,33 the requirement of public recognition nonetheless links the two mishnayot. In is explicitly set forth as a condition for חזקה the first mishnah (2:6), the public purification, and mishnah 2:7 implies clear social distinctions necessary for maintaining purity. Most instructive, however, is the concluding statement of m. Ḥag. 2:6, which reveals that public recognition is not merely a practical mechanism used for socially managing the manifold levels of purity: “If he immersed without being ‘held,’ it is as though he had not immersed at all.” If someone is not a recognized member of any of the aforementioned pure groups—in other words, an ʿam ha-ʾareṣ—and he attempts to undergo purification, his action is useless. Surprisingly, this assertion is reminiscent of the declarative statement we have already encountered in the Rule of the Community: “He [who refuses to enter the covenant of the Yaḥad] cannot be purified by atonement, nor be cleansed by waters of purification” (1QS 3:4–5). Beyond the boundaries of the group, purification rituals are worthless. Both texts, notwithstanding substantial differences considering the status of the outsider, structure boundaries in the same way. The mishnah does not claim that outsiders are sinners, while the Rule of the Community employs sin as a decisive and exclusive factor. However, both texts assume that no matter how hard they try, outsiders are impure in the highest degree. Purity is a privilege, granted only through membership. An additional but essential aspect of outsider impurity, known also from sectarian practice, can be found in m. Ḥag. 2:7. The clothing of an ʿam ha-ʾareṣ carries severe midras impurity, as though these clothes had been contaminated by someone with a genital discharge.34 This ruling is usually explained as a result of such a person’s lack of concern for purity; e.g., the wife of such

33 M. Ḥag. 2:6, unlike the following mishnah, does not refer to well-defined groups. After all, there is no party of “tithe-eaters.” The mishnah is probably describing the situation during festivals, when all pilgrims came to Jerusalem and took care to observe purity practices. Even in this temporary setting, the requirement to establish an affiliation was indispensable. 34 See above n. 26. 52 Furstenberg a person is presumed not to be diligent about avoiding contact with vessels and clothing during her menstrual period.35 This practical concern, however, is inadequate to explain the mishnah’s system as a whole. After all, why should the clothes of the Pharisee—who is clearly scrupulous in purity matters—be considered midras for others more scrupulous than he? Moreover, how can the mishnah claim that the garments of priests eating heave-offerings, who are certainly pure according to biblical standards, are considered to transmit midras impurity to those who eat Temple sacrifices (“hallowed things”)? Why does the mishnah disregard the scrupulousness presumable at a given level and consider everyone in a given circle as a carrier of midras impurity with respect to those in inner circles? These degrees of impurity reflect the notion that anyone outside one’s own circle of purity is considered impure in the greatest degree, regardless of his actual level of purity. From the priest’s perspective, there is thus no difference between the Pharisee and the ʿam ha-ʾareṣ. The clothing of both imparts severe midras impurity. Here, too, we find a clear parallel to Essene practice, as documented by Josephus. A new member of the sect, although clearly careful in matters of purity, is at the same time considered—at least in matters of purity—the equivalent of a complete outsider, in relation to senior members. Here again, both rabbinic and sectarian rulings share the same fundamental notion of purity as embodying social affiliation. With the recognition that the Pharisaic and Qumran/Essene systems share the same basic underlying conceptual framework of outsider impurity, we can better evaluate the specific differences between the two. The differences dis- close the ideological inclination of each of these halakhic systems. The Yaḥad redrew the boundaries of Israel:36 they identified nonmembers as Gentiles with respect to their level of impurity, and Gentile impurity served as a model for defining all those in circles external to the sect. Mishnah Ḥagigah, in contrast, sets the ʿam ha-ʾareṣ, not the Gentile, as the model for the impurity of the nonmember. Rabbinic literature in general markedly distinguishes between the impurity of a Gentile and that of an ʿam ha-ʾareṣ. The Gentile is considered to be like a zav, who defiles not only through direct contact, but also through spittle, urine, The origin and development of this status 37.(הסט) and even indirect pressure

35 B. Ḥul. 35b, and Rashi ad loc. 36 Shemesh, “The Origins of the Laws of Separatism,” 224–26. 37 On the Gentile as analogous to the zav, see Sifra, Zabim 1:1 (J. H. Weiss, Sifra: Commentar zu Leviticus [Vienna: Schlossberg, 1862], 74d); t. Zabim 2:1 (M. S. Zuckermandel, Tosephta: Based on the Erfurt and Vienna Codices [2d ed.; Jerusalem: Bamberger & Wahrmann, 1937], Outsider Impurity 53 is hard to track, and it seems to have been beyond the knowledge of the rabbis themselves; however, it is already assumed by Beit Hillel and Beit Shammai,38 and may serve to illuminate Pharisaic conceptions of outsider impurity. In contrast to the Gentile, the ʿam ha-ʾareṣ, as we learn from m. Ḥag. 2:7, transmits severe impurity only through his clothes. Practically speaking, he is considered to be a carrier rather than a source, of severe impurity; therefore his own touch does not convey impurity to other people but only to foods and earthen vessels.39 Taken together, early rabbinic sources clearly distinguish the impurity of an ʿam ha-ʾareṣ from that of a Gentile.40 By maintaining three distinct spheres—pure Jews; other Jews whose clothing is considered to be a carrier of severe impurity; and intrinsically impure Gentiles—these rabbinic sources

m. Ṭehar. 5:7 (best understood as :(הסט) On defilement through indirect contact .(677 referring to a Gentile), 7:6 (see n. 39 below); t. Ṭehar. 6:11 (Zuckermandel, Tosephta, 666); t. ʾAhilot 9:2 (606). On urine and spittle: m. Ṭehar. 4:5, 5:8; m. Makhsh. 2:3; t. Kippurim 3(4):20 (Lieberman, Tosefta, 2:248); t. Mikvaot 6:7 (Zuckermandel, Tosephta, 658); t. Ṭehar. 5:2, 5:4 (664), 8:9 (669). 38 M. Niddah 4:3. For a comprehensive discussion of the rabbinic definition of Gentiles as zavim see Hayes, Gentile Impurities, 122–31. Hayes emphasizes that the rabbis understand this definition as a rabbinic decree, and not as a law of the Torah. That is indeed true, but it does not prove that such a status is a rabbinic innovation. More plausibly, the rabbis are attempting to negotiate and define an earlier practice. The story of the high priest in t. Kippurim 3(4):20 points in that direction. See M. Balberg, Purity, Body, and Self in Early Rabbinic Literature (Berkeley: Stanford University Press, 2014), 122–47. 39 The level of ʿam ha-ʾareṣ impurity in comparison to the more severe zav/Gentile impurity is clearly demonstrated in m. Ṭehar. 7:5–6. If an ʿam ha-ʾareṣ is left alone in someone else’s house, and no one else enters, only foods and earthenware vessels are deemed impure; if the house is left unlocked and unattended, we assume that a Gentile might have entered, and the whole house is considered impure in the severest degree (midras), as if contaminated by a zav. However, things placed in the house of an ʿam ha-ʾareṣ incur both midras and corpse impurity (m. Ṭehar. 8:1). 40 Additional evidence for this stringent system of outsider impurity in non-Pharisaic circles may be detected within rabbinic literature itself. In t. Parah 3:8 (Zuckermandel, Tosephta, 632) we read of a Zadokite priest who was about to prepare the red heifer. R. Yoḥanan ben Zakkai rested his hands on the priest’s head and defiled him, so that he had to immerse on the spot and prepare the red heifer in the state of ṭevul yom, in accordance with Pharisaic halakhah. R. Yoḥanan, who was supposedly pure with respect to unconsecrated food, was considered impure in relation to the priest according to both systems of graded purity. Yet the fact that he was able to defile the priest by mere touch does not follow from Pharisaic halakhah, which in such circumstances considered only the clothing to be a source of impurity, but rather from the more stringent non-Pharisaic system. This story thus preserves a memory of Zadokite halakhah, closely resembling the Essene system of impurity as described by Josephus. 54 Furstenberg appear to be opposing Qumran/Essene ideology and offering a structure which seems to reflect a Pharisaic outlook.41 Within a shared conceptual framework of outsider impurity, the dispute concerning the outsider’s exact level of impurity may serve to reflect substantial ideological differences. The fragment of Second Temple halakhic tradition in m. Ḥagigah presents an approach which depends on the public recognition of clearly established and hierarchical group affiliations and which responds to other, stricter approaches to outsider impurity. In later rabbinic interpretations of this mish- was הוחזק nah, however, this framework is radically reinterpreted. The verb understood in a less than literal manner, where personal intention took the place of public recognition. This interpretation restricts purification to the personal sphere and isolates it from any particular social setting. In what follows, I take a closer look at the transformations of additional Second Temple halakhic traditions concerning the impurity of the “nonmember,” as they were cast into new communal contexts.

4 Separation from Outsiders in Early Tannaitic Sources

The close correspondence of early rabbinic sources to the discourse of contemporary texts is also reflected in the specific laws of separation from outsiders, whether the rabbinic ʿam ha-ʾareṣ or the Qumranic “men of injustice.” Despite some fundamental differences, the rabbinic rulings take part in a single, distinctive tradition of separation legislation, which was adapted by different groups during the Second Temple period to fit their differing ideologies. It was only later that the Tannaim, in the course of their reformulation of early traditions, abandoned this well-defined framework. The best illustration of the nature of separation from the ʿam ha-ʾareṣ in early tannaitic sources and the changes undergone by this system in later tannaitic literature is found in the laws relating to the ḥaver, the “associate” publically deemed trustworthy in matters of purity. Although the Mishnah never defines the Pharisees, as this was probably a recognized group, the conditions

41 For a full account of the two social strategies of Second Temple sects as “reformist” and “introversionist,” see A. I. Baumgarten, The Flourishing of Jewish Sects in the Maccabean Era: An Interpretation (Leiden: Brill, 1997), 13. Both Qumran and Essene communities, as “introversionist” sects, completely absorbed their members’ identity and demanded complete separation from outsiders. The Pharisees, on the other hand, are considered “reformist,” since the walls separating them from other Jews were less high and much more permeable. Outsider Impurity 55 for becoming a ḥaver are spelled out in m. Demai 2:3 and reformulated in t. Demai 2:2–14.42 These sources have enabled scholars to compare the rabbinic ḥavurah with the Qumran Yaḥad and to assess the relationship between the two phenomena.43 I will suggest here that tannaitic literature contains in fact two fundamentally different definitions of the ḥaver and his relation to the ʿam ha-ʾareṣ. Only the source found in the Mishnah shows similarity to the separatism of the Yaḥad and other Second Temple sources. Through a comparison of the Mishnah and the Tosefta on this point, we can trace the later rabbinic divergence from the Second Temple setting and its separation traditions. First, however I will describe the specific prohibitions in tannaitic literature against associating with an ʿam ha-ʾareṣ. These instructions are fundamental to the obligations of the ḥaver, according to m. Demai 2:3; and as we shall see below, they bear close resemblances to other Second Temple separation traditions:

המקבל עליו להיות חבר אינו מוכר לעם הארץ לח ויבש ואינו לוקח ממנו לח, ואינו מתארח אצל עם הארץ ולא מארחו אצלו בכסותו.

He who commits to being a ḥaver may not sell to an ʿam ha-ʾareṣ wet or dry [foods], or buy from him wet [food]; and he may not be the guest of an ʿam ha-ʾareṣ, nor may he receive him as a guest in his own garments.44

The mishnah assumes that the candidate is already careful in matters of purity, as the standard Pharisee would be,45 and therefore the conditions relate only to his level of contact with the impure ʿam ha-ʾareṣ. This relationship is shaped through two sets of parallel conditions. The ḥaver may not sell anything to an ʿam ha-ʾareṣ, but he may buy dry, pure foodstuffs. He may not visit the ʿam ha-ʾareṣ, but he may host him if he changes his impure clothes. The relation-

.חבר and פרוש See n. 8 above on the relationship between the 42 43 For a useful critical survey of scholarship on this matter see S. D. Fraade, “Qumran Yaḥad and Rabbinic Ḥăbûrâ: A Comparison Reconsidered,” DSD 16 (2009): 433–53. 44 M. Demai 2:3; further on in the mishnah, R. Judah adds other conditions for becoming an associate, which are dismissed by the rabbis. All Tannaim accept this basic list of conditions. 45 The dispute between Jesus and the Pharisees over hand-washing also assumes that the Pharisees maintained purity. See Y. Furstenberg, “Defilement Penetrating the Body: A New Understanding of Contamination in Mark 7:15,” NTS 54 (2008): 176–200. For a survey of other views see H. K. Harrington, “Did the Pharisees Eat Ordinary Food in a State of Purity?” JSJ 26 (1995): 42–54. 56 Furstenberg ship between the ḥaver and the ʿam ha-ʾareṣ is thus one-sided: the ḥaver may accept the ʿam ha-ʾareṣ and his produce into his own domain, as long as he is careful that purity is maintained, but he may not transfer to the domain of the ʿam ha-ʾareṣ. Another parallel instruction regarding the separation from ʿam ha-ʾareṣ, which complements m. Demai 2:3, should also be considered. Tosefta ʿAbodah Zarah 3:9 reads:46

לוקחין מעם הארץ עבדים ושפחות—בין גדולים בין קטנים, ומוכרין לעם הארץ עבדים ושפחות—בין גדולים ובין קטנים. ולוקחין מהן בנות— קטנות אבל לא גדולות, דברי ר’ מאיר; וחכמים אומרים גדולה ומקבלת עליה. ואין נותנין להן בנות—לא גדולות ולא קטנות, דברי ר’ מאיר; וחכמים אומרים נותן לו גדולה ופוסק עמו על מנת שלא תעשה טהרות על גביו. מעשה ברבן גמליאל הזקן שהשיא את בתו לשמעון בן נתנאל הכהן ופסק עמו על מנת שלא תעשה טהרות על גביו.

We buy slaves from the ʿam ha-ʾareṣ—both mature and young, and sell them slaves—both mature and young. We take daughters from them— young but not mature, the words of R. Meir. The rabbis say: even a mature girl, who accepts upon herself. But we do not give them daughters— neither mature nor young, the words of R. Meir. The rabbis say: he may give him a mature girl on condition that she not be required to prepare purities together with him. It happened that Rabban Gamaliel the Elder married his daughter to Shimon ben Netanel the priest and he stipu- lated with him that she would not be required to make purities together with him.47

46 Zuckermandel, Tosephta, 464. 47 The story of Rabban Gamaliel the Elder is adduced to support the lenient position of the rabbis. A man may marry off his older daughter to an ʿam ha-ʾareṣ, if he stipulates that she not be required to prepare pure foods. However, it should be noted that this early story does not concern marriage to an ʿam ha-ʾareṣ, and it is hard to assume that Rabban Gamaliel would give his daughter to such a priest. A better supposition is that Rabban Gamaliel (like Yoḥanan b. Gudgudah in m. Ḥag 2:7) maintained a high level of purity— i.e., for hallowed things—and the priest, who maintained purity only at the level of eating terumah, was considered relatively impure by comparison (see S. Lieberman, Tosefta ki-feshuṭah: A Comprehensive Commentary on the Tosefta [10 vols.; New York: Jewish Theological Seminary of America Press, 1955–1988], 5:1310). This story cannot prove that as early as the Second Temple period a lenient approach towards marriage with ʿammei ha-ʾareṣ was prevalent. Outsider Impurity 57

This source discusses the purchase of slaves, as well as marriage; the latter is presumably the central concern of the separation laws. The core of this halakhic formulation is clearly early. The second-century rabbis cited in this source refer to a story about the famous Pharisee, Rabban Gamaliel the Elder (early 1st century CE). In addition, the early halakhah may easily be distinguished from the later interpretive glosses of R. Meir and the rabbis. The second- century rabbis dispute the exact scope of the earlier halakhah concerning marriage with ʿammei ha-ʾareṣ, as to whether it includes both young and mature daughters. All however agree on the wording of the early halakhah: “One may take daughters from them [ʿam ha-ʾareṣ], but may not give them daughters.”48 Usually there is no way to identify the roots of these early halakhot disputed by later Tannaim. In this case, however, the striking resemblance of this ruling to the two laws addressed to the ḥaver in m. Demai allows us to restore it to its original context. Here, too, it is forbidden to go to an ʿam ha-ʾareṣ or to bring him produce; but the ʿam ha-ʾareṣ may approach the ḥaver, on the condition that the laws of purity be maintained. These three similar laws, concerning food, hosting, and marriage, articulate a coherent policy towards the non-ḥaver. These issues are also closely related to the laws of separation in Jubilees and the Rule of the Community, discussed above. An examination of early rabbinic legislation against the backdrop of these laws will uncover both the textual and conceptual framework shared by all three corpora and the nature of the Pharisaic variation upon this tradition. Both Jubilees and the Rule of the Community equate the commitment to follow the laws of the Torah with the obligation of complete separation from outsiders. Thus Jubilees: “And you also, my son Jacob, remember my words and keep the commandments of Abraham, your father. Separate yourself from the Gentiles. . . .” This principle also shapes the entrance requirement in 4QS, which is enforced by an oath: “And everyone who enters the council of the Community shall take upon himself with a binding oath to return to the Torah of Moses with all his heart and with all his soul, all which is revealed of the Torah according to the council of the Yaḥad, to be separated from all the men of injustice.” In this respect, these two sources resemble Nehemiah 10, which positions the people of the land, at the very center of the ,עמי הארץ the separation from

48 Significantly, this is a rare case in which the expression ʿam ha-ʾareṣ appears in tannaitic literature as a collective term, and does not indicate an individual. The origins of this halakhah may explain this fact; see discussion below. 58 Furstenberg

through a public oath. Thus we read in ,(9:38) אמנה ”,enactment of the “pledge Neh 10:29–31:

And the rest of the people . . . join with their noble brothers, and take an oath with sanctions to follow the Teaching of God, given through Moses the servant of God, and to observe carefully all the commandments of God our Lord, his rules and laws. Namely, we will not give our daughters -nor take their daugh ,)לעמי הארץ( in marriage to the people of the land ters for our sons.

While Nehemiah only mentions marriage with Gentiles, Jubilees, as we have seen, adds three additional prohibitions: eating with Gentiles, imitating their actions, and associating with them. In the Rule of the Community, we have seen detailed references to these prohibitions (apart from that of marriage) in relation to the men of injustice. Reading the tannaitic material against this tradition, we come upon considerable thematic and literary connections. As in Nehemiah, Jubilees, and the (מקבל עליו) Rule of the Community, the candidate in m. Demai commits himself to the rules of a ḥaver 49 by accepting the laws of separation. He is prohibited from associating with an ʿam ha-ʾareṣ, a term which alludes to Nehemiah. The close relation of the tannaitic material to Nehemiah is even more apparent in regard to the specific issue of marriage discussed in t. ʿAbodah Zarah. The separation tradition in Nehemiah 10 centers on the prohibition against marrying the women of the ʿammei ha-ʾareṣ. Rabbinic halakhah regarding marriage with the ʿam ha-ʾareṣ is conspicuously reminiscent of the wording of Neh 10:31 quoted above:

ואשר לא נתן בנתינו לעמי הארץ ואת בנתיהם לא נקח לבנינו Neh 10:31 We will not give our daughters in marriage to the peoples of the land, or take their daughters for our sons.

לוקחין מהן )= עם הארץ( בנות . . . ואין נותנין להן בנות .t. ʿAbod Zar. 3:9 We take daughters from [ʿam ha-ʾareṣ], but we do not give them daughters.

49 S. Lieberman, “The Discipline in the So-Called Dead Sea Manual of Discipline,” JBL 71 (1952): 199–206, suggests that the ḥaver’s acceptance of obligations includes some kind of an oath, like the pledge in Nehemiah. Outsider Impurity 59

This parallel presentation illustrates that the rabbinic laws which explicitly restrict relations with the ʿam ha-ʾareṣ apparently relate to the Ezra–Nehemiah separation tradition, rooted in the prohibition against marrying women of the ʿammei ha-ʾareṣ.50 The parallel language suggests that the rabbis, like the com- pilers of the Rule of the Community, applied to other Jews the rules governing separation from Gentiles. These Jews were called ʿammei ha-ʾareṣ, although they were clearly distinct from their namesakes in Ezra–Nehemiah, who were assumed to be non-Jewish and were identified with the Canaanites.51 However, unlike the Rule of the Community, which legislates a complete separation from other Jews by applying prohibitions that were originally directed against non- Jews, early rabbinic sources mandate separation in only one direction. In the case of marriage, the early Tannaim rephrased the biblical precept directed against the ʿammei ha-ʾareṣin Nehemiah 10, and applied only half of it against those Jews they themselves considered ʿammei ha-ʾareṣ. The prohibition was thus confined to marrying daughters to ʿammei ha-ʾareṣ; at the same time, the admission of the daughters of ʿammei ha-ʾareṣ into rabbinic circles was permitted. This tendency to curtail the separation laws also served the rabbis in shaping the other arenas of separation: food and hosting. Above we saw that Jubilees prohibits sharing food and associating with Gentiles. The Rule of the Community defines “association” as the act of receiving anything from the “men of injustice.” Both issues, sharing food and hosting, are included among the commitments of the ḥaver in m. Demai 2:3; but here, again, complete separation is supplanted by a patronage system. While maintaining the inferior

50 There are additional indications that the rabbinic usage of the term ʿam ha-ʾareṣis based on the usage of this term in Ezra–Nehemiah (indicated also by Oppenheimer, The Am פרוש Ha-Aretz, 11–12). The terminological opposition in the mishnah between the pure is best understood in light of Ezra’s demand to separate from עם הארץ and the impure in Aramaic פרש the inhabitants of Judaea. The verb ,עמי הארצות the impurity of the פרושים in Biblical Hebrew. Thus, the בדל and Mishnaic Hebrew correlates with the verb ,as in Ezra 9. See S. S. Cohon ,עם הארץ in the Mishnah are those who separate from the “Pharisaism: A Definition,” in Joshua Bloch Memorial Volume: Studies in Booklore and History (ed. A. Berger; New York: New York Public Library, 1960), 65–74. See in more detail Furstenberg, “Am Ha-aretz in Tannaitic Literature,” 291–99. 51 See Ezra 9:1–12. It has long been debated who exactly were these “people of the land” rejected by Ezra and Nehemiah. Were they groups of non-Israelites who settled the land, Israelites who stayed behind and were not sent to exile, or a mixture of these elements, who as such threatened the integrity of the group returning from exile? At any rate, the rabbinic usage of the term ʿam ha-ʾareṣ as definitively Jewish clearly diverges from that of Ezra–Nehemiah. 60 Furstenberg status of the impure, the rules enable some sort of contact. The ḥaver may derive benefit through the ʿam ha-ʾareṣ, as long as this does not entail transgression in matters of purity, but he may not hand anything over to his control. Here, too, early rabbinic literature preserves a unique version of impurity laws governing relationships with nonmembers. The absolute separation of nonmembers, Jews or non-Jews, has evolved into a graded system, meant to influence other Jews.52

5 Association for Purity without Separation

As we have seen, the three separation laws in m. Demai 2:3 and in the early halakhah embedded in t. ʿAbod. Zar. 3:9 correspond to early Second Temple separation traditions, and they offer a unique variation on the earlier material. The early provenance of this legislation may be further substantiated by comparing m. Demai 2:3 with the parallel tosefta (t. Demai 2:12), which suggests that Beit Hillel and Beit Shammai were aware of this halakhah. In other, later, halakhot, however, the Tosefta completely revises the requirements of the ḥaver, in what

52 A set of “laws of separation” directed against minim has been identified by A. Schremer, “Seclusion and Exclusion: The Rhetoric of Separation in Qumran and Tannaitic Literature,” in Rabbinic Perspectives: Rabbinic Literature and the Dead Sea Scrolls. Proceedings of the Eighth International Symposium of the Orion Center For the Study of the Dead Sea Scrolls and Associated Literature, 7–9 January, 2003 (ed. S. D. Fraade, A. Shemesh and R. A. Clements; STDJ 62; Leiden: Brill, 2006), 127–45. Schremer refers (134–35) to t. Ḥul. 2:20–21: “The act of slaughter of a min is idolatry. Their bread is the bread of a Samaritan and their wine is libation wine. . . . People are not to sell anything to them, or buy anything from them. And they may not take wives from them or give children in marriage to them. And they may not teach their sons a craft. And they may not seek medical assistance from them, either healing for property or healing for a person.” This source includes all three elements of separation familiar from Jubilees 22: food, commerce, and marriage. The minim are equated with Gentiles, or considered even worse. Although this source seems to adopt the same approach towards other Jews as that of the Yaḥad, Schremer argues that there is a marked rhetorical difference. Whereas the Qumranites use a language of seclusion (that is, secluding themselves from the larger, impure group), the rabbis employ a self-assured rhetoric of exclusion (that is, marking the minim as complete outsiders). Alternatively, the laws of minim may simply be a mirror image of sectarian self-separation from the rest of Israel. In other words, the severance is mutual; those who separate themselves are in turn pushed away, by completely cutting off contact. In contrast, the rabbinic laws pertaining to separation from the ʿam ha-ʾareṣ are parallel to those of the Yaḥad, in that they are intended to set apart the ḥaverim as a minority group within the rest of Israel. Outsider Impurity 61 may plausibly be seen as a later tannaitic abandonment of Second Temple separation traditions. Thus, the range of halakhot preserved in the Tosefta reveals the transformation of the concept of separation during the tannaitic period. Mishnah Demai 2:3 presents two areas in which separation is required of the potential ḥaver: commerce and hosting. The two houses spell out the periods of probation required for each of these realms before acceptance of the ḥaver:

M. Demai 2:3 T. Demai 2:1253

עד מתי מקבלין? המקבל עליו להיות חבר בית שמיי אומרים למשקין שלשים יום, אינו מוכר לעם הארץ לח ויבש, ואינו לוקח לכסות שנים עשר חדש. ממנו לח ובית הלל אומרים זה וזה לשלשים יום ואינו מתארח אצל עם הארץ, ולא מארחו אצלו בכסותו

He who commits to being a ḥaver For how long do they take upon may not sell to an ʿam ha-ʾareṣ wet themselves [the commitments of or dry [foods], or buy from him the ḥaver]? wet [food]; and he may not be the Beit Shammai says: concerning liq- guest of an ʿam ha-ʾareṣ, nor may uids thirty days, and concerning gar- he receive him as a guest in his own ments twelve months. garments. Beit Hillel says: thirty days for each.

The toseftan passage falls within a series of later halakhot that discuss the process of education once one becomes a ḥaver.54 The mishnah and this tosefta are closely connected, however. Both define the ḥaver through two conditions that distinguish him from the ʿam ha-ʾareṣ: purity of liquids and purity of garments. The order of these matters in the tosefta follows that of the mishnah. Thus, the dispute between Beit Shammai and Beit Hillel is also about the probationary period during which the candidate is to separate himself from the

53 Lieberman, Tosefta, 2:71. 54 This early dispute in the Tosefta is clearly marked off from the surrounding halakhot (2:10–13), which are attributed to later authorities and employ a different terminology concerning the process of initiation into the ḥavurah. Whereas halakhah 11 discusses two distinct stages, “wings” and “purities,” the traditions ascribed to Beit Shammai and Beit Hillel mention “liquids” and “garments.” These halakhot seem therefore to describe different processes of becoming a ḥaver. Thus, the question in 2:12, “For how long do they take upon themselves [the commitments of the ḥaver]?” cannot refer to the previous halakhah; it must therefore be referring to a different source, most probably the mishnah. For attempts to reconcile these sources see the works cited below in n. 57. 62 Furstenberg

ʿam ha-ʾareṣ, the period before the candidate is recognized as a ḥaver. The connections between the mishnah and this early tosefta suggest that the anonymous halakhah in m. Demai is also quite early; it may be considered a fragment of Second Temple halakhah. Furthermore, only in this early halakhah concerning the ḥaver do we find a fixed probationary period, parallel to that of the Essenes or Qumranites. In contrast, another passage from the Tosefta (t. Demai 2:2)55 suggests an alternative set of obligations for the ḥaver, which completely departs from Second Temple separation discourse. A close examination of these obligations reveals a radically different state of affairs than we have hitherto encountered.

המקבל עליו ארבעה דברים מקבלים אותו להיות חבר: שלא יתן תרומה ומעשרות לעם הארץ, ושלא יעשה טהרות לעם הארץ, ושיהא אוכל חולין בטהרה.

He who accepts upon himself four things, is accepted to be a ḥaver: That he will not give heave-offerings and tithes to an ʿam ha-ʾareṣ, that he will not make purities for an ʿam ha-ʾareṣ, and that he will eat unconsecrated food in purity.56

The requirements spelled out in this tosefta differ from the list in m. Demai 2:3 in two significant respects. The tosefta adds an explicit demand to eat ordinary foods in purity, whereas the mishnah takes this practice for granted. Additionally, the tosefta places absolutely no limitation on contact with an ʿam ha-ʾareṣ.Strikingly, the ḥaver is only required to make sure he does not provide the ʿam ha-ʾareṣ with consecrated pure foods, or prepare foods for him that are supposed to be kept in purity. In other words, the ḥaver in this tosefta is to be concerned only with maintaining a higher standard of purity in his own surroundings. Saul Lieberman and subsequent scholars have attempted to reconcile these two lists of requirements, and have reconstructed an extremely complex probationary procedure that included all these various conditions.57 These scholars argue that the tosefta sets out the preliminary conditions imposed upon

55 Lieberman, Tosefta, 1:68. 56 The tosefta requires “four things,” but mentions only three. For a suggested solution, see Lieberman, Tosefta ki-feshuṭah, 1:210. 57 Various suggestions have been put forward as to how to combine the requirements spelled out in the mishnah and in the tosefta (including the substages mentioned in t. Demai 2:10–13) into one continuous process. See Lieberman, Tosefta ki-feshuṭah, 1:209, Outsider Impurity 63 the candidate, prior to initiation into the ḥavurah, whereas the mishnah lists the obligations of those already admitted into the ḥavurah. However, this solution is problematic on many levels, the cumbersome reconstruction aside. It leaves unexplained the remarkable fact that, in marked contrast to the social restrictions imposed upon the ḥaver in the sources discussed above, the Tosefta repeatedly refers to the possibility that a ḥaver would regularly visit an ʿam ha-ʾareṣ, as well as marry off his daughter to one.58 The Tosefta also instructs the practicing ḥaver as to how to draw the line between ordinary foods, which he may hand to an ʿam ha-ʾareṣ, and consecrated foodstuffs, which he is committed to preserving from defilement by not giving them to the ʿam ha-ʾareṣ.59 If the ḥaver were to follow the mishnah’s absolute restrictions, he would never face these issues. Mishnah Demai 2:3 and t. Demai 2:2 thus assume different social settings which determine the nature of the ḥaver’s commitments. As we have seen, the mishnah is deeply grounded in Second Temple separation traditions. The parallel tosefta features a new kind of ḥaver, whose purity does not entail complete separation. This ḥaver chooses to join the supportive ḥavurah, some sort of voluntary association, in order to excel in matters of purity, without severing his connections with other groups.60 The creation of this type of ḥaver and the abandonment of the separation demands are most likely the result of the reception of Second Temple halakhic institutions in late second-century Galilee.61 As the contrast between pure and impure ceased to correlate with recognized groups who identified themselves on some level within the separatist tradition, purity no longer played a constitutive social role; it shifted into the personal realm as a matter of choice and

in the Second Jewish Commonwealth,” HTR 53 (חבורה) J. Neusner, “The Fellowship ;216 (1960): 125–42; Oppenheimer, Am Ha-Aretz, 118–31. 58 On hosting, see t. Demai 3:7; on marriage, t. Demai 2:17. For all of these, see Lieberman, Tosefta, 1:71, 1:74. 59 T. Demai 3:1 (Lieberman, Tosefta, 1:73). 60 The transformation in organizational patterns, from sect to voluntary association, brings this rabbinic phenomenon closer to Greco–Roman civic culture. See A. I. Baumgarten, “Graeco–Roman Voluntary Associations and Ancient Jewish Sects,” in Jews in a Graeco– Roman World (ed. M. Goodman; Oxford: Clarendon, 1998), 93–111. 61 On the prominence of late second-century Galilean rabbis in laws pertaining to the ʿam ha-ʾareṣ see Büchler, Am-ha‌‌‌ʾareṣ. However, whereas Büchler claimed that the laws of the ʿam ha-ʾareṣ were an innovation of the later Tannaim, the evidence rather suggests that they infused an ancient halakhic and social institution with new meaning. For a detailed account of the changing definition of ʿam ha-ʾareṣ in late tannaitic sources see Furstenberg, “Am Ha-areṣ in Tannaitic Literature.” 64 Furstenberg expertise.62 The dominance of this later approach within tannaitic literature created the impression that from its commencement, rabbinic purity had been a matter of personal choice, and not of social affiliation.

6 Trusting the Impure: An Early Mishnah Revised

Early rabbinic ʿam ha-ʾareṣ legislation should be understood against the background of separation traditions, as a Pharisaic reaction against more stringent attitudes held by other groups towards nonmembers. These early sources share the assumption that purity depends on membership; in all traditions, the laws governing the contact with outsiders are derived on some level from traditions pertaining to separation from Gentiles. The Pharisees adjusted this system to differentiate between Jewish and Gentile outsiders: they set the impurity of the Jew on a less stringent level than that of the Gentile (section 3 above) and they applied the laws of separation originating in Ezra–Nehemiah only partially (section 4). These two strategies serve the Pharisaic goal of maintaining the separation tradition concerning purity in the realm of unconsecrated foods, without resigning from society as other, more separatist groups chose to do. This policy, which suits the Pharisaic image as distinct from yet closely associated with the people, has implications for yet another aspect of rabbinic law relating to the ʿam ha-ʾareṣ, regarding the trustworthiness of impure persons. According to another early unit in Mishnah Ḥagigah:

M. Ḥagigah 3:6:

הגבאים שניכנסו לתוך הבית וכן הגנבים שהחזירו את הכלים— נאמנים לומר לא נגענו. ובירושלם נאמנין על הקדש ובשעת הרגל אף על התרומה.

Tax collectors who entered the house, as well as thieves who returned the stolen vessels—are deemed trustworthy if they say: we have not touched. In Jerusalem they are trustworthy regarding hallowed foods, and during the festival even regarding the heave-offering.

62 In later amoraic sources this ability is attributed only to outstanding individuals such as Abraham (b. B. Metziʿa 87a). Rabbi Judah the Prince advised his nephew to try eating in purity for at least a week every year (y. Shabb. 1:3 [3c]). Outsider Impurity 65

This ruling seems intentionally paradoxical: As the impure tax collector and thief invade one’s private space in search of money and goods, turning the house upside down, all defenses erected for the security of purities should supposedly have collapsed. However, the mishnah not only grants the possibility that the intruders were careful to leave the purities intact, it is also willing to believe the testimony of tax collectors and thieves, neither of whom were famous for their honesty in the ancient world.63 It is assumed that these questionable people are knowledgeable about the significance that purity holds for others (for they themselves are clearly impure), and that they would deliberately refrain from contaminating a household. Through this extreme example, the mishnah conveys the idea that purity is maintained only with the cooperation of the impure. Unlike the examples discussed in the previous sections, I am not aware of a direct textual parallel to this rabbinic ruling in nonrabbinic sources. It seems reasonable, however, to assume that such a policy would be considered unacceptable among the Yaḥad, not least because nonmembers were considered the embodiment of deceit and futility: “He should remain at a distance from him in every task, for it is written as follows ‘you shall remain at a distance from every false matter’ (Exod 23:7)” (1QS 5:15); or, “They may not rely on the people of futility, for futile are all those who do not know his covenant” (4QSd 1 i 10). In contrast, m. Ḥagigah assumes the essential cooperation of these impure ʿammei ha-ʾareṣ in maintaining purity. A close reading of the mishnah reveals the range of prevalent attitudes towards the trustworthiness of the ʿam ha-ʾareṣ. Tax collectors, thieves, and all other ʿammei ha-ʾareṣ, are trusted specifically with regard to everyday, unconsecrated foods which the Pharisees would habitually eat in a state of purity.64 The mishnah applies the same policy to consecrated sacrificial foods as well, which, in Jerusalem, were handled and consumed by even wider circles.

63 See, e.g., m. B. Qam. 10:1; t. Demai 3:4 (Lieberman, Tosefta, 1:74). New Testament scholars have discussed our mishnah (and the parallel below) in connection with Jesus’s habit of dining with tax collectors and sinners (Mark 2:15; Matt 9:9; Luke 5:27). See H. Maccoby, “How Unclean were Tax Collectors?” Biblical Theology Bulletin: A Journal of Bible and Theology 31 (2001): 60–63. See also the Greco–Roman literary sources on tax collectors noted by A. Yarbro Collins, Mark: A Commentary (Hermeneia; Minneapolis: Fortress, 2007), 193–94. 64 Most commentators assumed that the first statement of the mishnah refers only to consecrated foods, revered even by the ʿam ha-ʾareṣ, and not to unconsecrated pure foods I contend elsewhere that this unit .(נאמנים לומר .see for example Rashi, ad b. Ḥag. 26a, s.v) assumes that ordinary food was also kept in purity. See Furstenberg, “Eating in the State of Purity,” 278–80. 66 Furstenberg

However, ʿammei ha-ʾareṣ are not considered trustworthy with regard to the priests’ own pure foods, i.e., the terumah (heave-offering), apart from festival periods. Paradoxically, despite the higher level of purity required in relation to hallowed foods, more people were trusted with regard to the purity of terumah. Thus, the mishnah seems to assume that the priests receiving the terumah were less motivated than the Pharisees to cooperate with the impure ʿam ha-ʾareṣ.65 Another, presumably later, version of the tax collector mishnah appears in the seventh chapter of tractate Ṭeharot. This chapter extensively addresses the problem, mentioned in Ḥagigah only in passing, of the status of a house visited by an ʿam ha-ʾareṣ. Whereas Hagigah pictures the violation of the pure house only through hostile invasion, Ṭeharot assumes that the impure ʿam ha-ʾareṣ enters the house of the associate on a regular basis. The discussion, partially attributed to second-century Galilean rabbis (7:1–4), introduces a major principle: supervision is a necessary condition for the maintenance of purity. If one leaves an ʿam ha-ʾareṣ unsupervised, the content of the house (or at least what is within the reach of the ʿam ha-ʾareṣ) is automatically considered defiled. Various considerations determine the degree of threat posed by such a scenario.66 One fact is striking: The option of questioning the ʿam ha-ʾareṣhimself is never offered; the possibility of learning directly from him what he has done and whether he has touched the purities is never considered. He represents an unreliable threat which can be handled only through some kind of supervision. Only once in the chapter do we hear the silenced voice of ʿam ha-ʾareṣ, but a close analysis of this case merely reinforces the general impression of complete distrust. In m. Ṭehar. 7:6 we read, according to the major textual witnesses:

הגבאים שנכנסו לתוך הבית טמא. אם יש עמהן גוי, נאמנים לומר נכנסנו אבל לא נגענו.67

65 I substantiate the Second Temple provenance of this mishnah in Furstenberg, “Eating in the State of Purity,” 280–83. For our discussion the relative primacy of m. Ḥagigah in relation to its parallel in m. Ṭeharot is a sufficient indicator of its relatively early dating. 66 The mishnah distinguishes between cases in which the ʿam ha-ʾareṣ is busy and when he has time to look around (7:4). Another consideration is to what degree the ʿam ha-ʾareṣ is aware he is being watched (7:5). 67 There are two versions of this mishnah in the manuscript tradition. The first, quoted here, appears in the primary manuscripts, including those with Maimonides’ commentary. An alternative and contradictory version has appeared in the printed editions of the הגבאים שנכנסו לתוך :Mishnah since its editio princeps and likewise appears in b. Ḥag. 26a .הבית טמא. אם יש עמהן גוי נאמנים לומר לא נכנסנו אבל אין נאמנין לומר נכנסנו אבל לא נגענו Thus in this version, when a Gentile is present, the tax collectors are trusted when they say that they have not entered, but not trusted if they claim that they have not touched Outsider Impurity 67

Tax collectors who entered the house, it is defiled. If there is a Gentile with them, they are deemed trustworthy to say: We have entered but not touched.

This mishnah clearly contradicts the parallel in Ḥagigah. In this case we can safely assume that it represents a revised version, adjusted to the concept of supervision, prevalent in m. Ṭeharot. By adding an interpretive gloss, “if there is a Gentile with them,” this version overturns the original sense of the mishnah. In its new form, the law rules that the house invaded by the impure tax collector is defiled, as in all other cases in the chapter discussed by the second century Galilean rabbis. The tax collector is trustworthy only if his work was inspected by a Gentile supervisor. Paradoxically the Gentile functions as a control for the ʿam ha-ʾareṣ. This law completes the adjustment of Second Temple textual traditions regarding the ʿam ha-ʾareṣ impurity to later contexts. Unlike the ʿam ha-ʾareṣ of m. Ḥagigah, in m. Ṭeharot he lacks the facilities for comprehending purity. Cooperation with such a person is impossible, and those who attempt to maintain purity are obliged to mark out manageable spheres beyond his reach. This change of policy towards the ʿam ha-ʾareṣ corresponds precisely to the process described above, which facilitated the abandonment of the separation tradition. Within the clearly defined structure, assumed by earlier Second Temple sources, the separation between the groups was maintained and necessarily acknowledged by nonmembers as well as Pharisees and other pure groups. This is the system assumed in m. Ḥag. 2:7, which evolves into a shared modus vivendi of mutual awareness and assumed cooperation, as can be seen in m. Ḥag. 3:6. Separation then supplied the basis for a confident management of purity within a diverse surrounding. As the separation tradition fell by the wayside in later tannaitic sources, purity ceased to play a role in determining group affiliation and in setting the limits of social contact. Thus, in t. Demai as well as m. Ṭeharot, ḥaverim were inseparable from their impure ʿam ha-ʾareṣ surroundings. Within such an environment, those who chose to be diligent in

the housewares. We can therefore conclude from this version that tax collectors are to be trusted completely when there is no Gentile around. In consideration of the manuscript evidence and the Babylonian provenance of the second version, I conclude that this is a later reworked version and therefore irrelevant to our current discussion. See S. Lieberman, Tosefet Rishonim (4 vols.; Jerusalem: Bamberger & Wahrmann, 1939), 4:82–83. 68 Furstenberg purity were compelled to resort to a more suspicious attitude towards their neighbors who were ʿammei ha-ʾareṣ, considered unreliable in these matters.68

Second Temple halakhah is often described as consisting of two competing traditions: the Sadducean tradition shared by the Qumran Yaḥad, and the Pharisaic tradition later developed by the rabbis.69 Our comparison of tannaitic sources concerning ʿam ha-ʾareṣ impurity with parallel notions of outsider impurity has supplied this scheme with an additional diachronic component. Early rabbinic traditions embedded in tannaitic compilations have been found to share with the Qumran sources a common discourse of separation. Both early rabbinic halakhah and Yaḥad and Essene teachings reflect a common conception of purity as a defining category, and each group shaped its specific practices according to its unique version of this tradition. In contrast, later tannaitic passages sought to integrate the early halakhic traditions concerning the ʿam ha-ʾareṣ into new settings; these passages incorporated the old traditions into a new nonseparatist framework, which ultimately undermined the very traditions on which it drew. Thus, the evolution of Pharisaic separation traditions in rabbinic halakhah illuminates the diversity of their changing contexts.

68 Interestingly, this later development, far removed from the intricacies of Second Temple ideological disputes, brought rabbinic policy much closer to the sectarian stance. 69 Y. Sussmann, “The History of the Halakha and the Dead Sea Scrolls: Preliminary Talmudic Observations on Miqṣat Maʿase Ha-Torah (4QMMT),” Tarbiz 59 (1989–1990): 11–76 (in Hebrew); V. Noam, From Qumran to the Rabbinic Revolution: Conceptions of Impurity (Jerusalem: Yad Izhak Ben-Zvi, 2010), 353–59 (in Hebrew). No Angels before the World? A Preexistence Tradition and Its Transformations from Second Temple Literature to Early Piyyuṭ

Yehoshua Granat

1 Introductory Notes

Early Piyyuṭ1 is undoubtedly one of the greatest textual monuments that evolved in the Jewish culture of late antiquity. Yet, though it has aroused considerable scholarly interest from the Wissenschaft des Judentums era onwards, this strikingly rich corpus is relatively seldom discussed from a thematic perspective. This state of affairs is probably influenced, to a significant extent, by the assumption that in regard to subject matter, early Piyyuṭ constantly follows authoritative rabbinic sources and merely paraphrases them, rendering their wording so as to fit specific stylistic preferences, poetic structures, and liturgical circumstances. If that were indeed the case, there would not be much sense, then, in directing scholarly attention to the thematic arena of early Piyyuṭ; one would naturally focus on these themes’ original formulations in rabbinic literature, rather than on their mere secondary derivatives in early Piyyuṭ. Yet such a portrayal is misleadingly simplistic rather than precise. Even though early Piyyuṭ typically does rely on rabbinic sources, these poems tend to render and reshape the “borrowed” elements in distinctly creative and sometimes transformative ways. Furthermore, themes and motifs occurring in these poems cannot in every instance be traced back to the rabbinic texts available to us. At times it is possible to reconstruct lost rabbinic traditions or homilies on which they were based.2 In some instances of themes found in early Piyyuṭ, but not in line with rabbinic sources, a possible link to a tradition

1 The term “early Piyyuṭ” designates here Hebrew liturgical poetry composed in Palestine (Eretz Israel) and the periphery, from approximately the fourth century to the eighth century CE. This label is preferable to the more general designation, “Piyyuṭ,” as the latter may also encompass Hebrew liturgical poetry through the Middle Ages and beyond; these later works substantially differ from the earlier corpus of poems, which belongs to the world of late antiquity. “Early Piyyuṭ” is also preferable to terms such as “Byzantine-era Piyyuṭ,” since a significant part of the corpus may be dated to the first two centuries after the Arab conquest. 2 See, e.g., S. Elizur, Elʿazar Birabi Kiliri: Hymni Pentacostales (Jerusalem: Mekize Nirdamim, 2000), 76–79 (in Hebrew).

© koninklijke brill nv, leiden, ��5 | doi ��.��63/9789004299139_004 70 Granat recorded in extrarabbinic, “noncanonical,” sources may be suggested. Such occurrences, though not very frequent, are of particular interest from a Traditionsgeschichte point of view; sometimes a few passages or even a single, unique text may strike an unexpectedly resonant chord, echoing an ancient, rarely recorded tradition.3 The present article illustrates this latter phenomenon through the discussion of one noteworthy thematic example. According to some early piyyuṭim, to be discussed below, the angels were created well before the world itself, a view that clearly contradicts unequivocal rabbinic statements on the matter. However, these references to the preexistence of angels can be paralleled to several texts of varied provenance (Jewish as well as Christian) from late antiquity, and their origins may arguably be traced back, as we shall see, to Second Temple sources. Indeed the standard rabbinic view regarding the time of the angels’ creation (which denies angelic preexistence) should be regarded as a polemical response to potentially challenging theological implications that could be drawn from this ancient “preexistence tradition”;4 the specific version of this tradition presented in the piyyuṭim discussed here may actually reflect comparable theological concerns, though it differs in its strategy of addressing them. This case study may thus contribute to our recognition of the dynamic and sometimes surprising afterlife of early Second Temple traditions in the Judaism of late antiquity within diverse contexts, including the sphere of early Piyyuṭ.

2 Dating the Creation of Angels: Diverse Views

The account of the creation of the world that opens the book of Genesis keeps silent about the angels. “Although all sorts of other biblical texts . . . make

3 See, e.g., M. Kister, “Observations on Aspects of Exegesis, Tradition, and Theology in Midrash, Pseudepigrapha, and Other Jewish Writings,” in Tracing the Threads: Studies in the Vitality of Jewish Pseudepigrapha (ed. J. C. Reeves; SBLEJL 6; Atlanta: ‎Scholars Press, 1994), 24–25. Several such cases relating to preexistence (including an earlier discussion of the issue examined more extensively in this article) are presented in Y. Granat, “Before ‘In the Beginning’: Preexistence in Early Piyyuṭ, Against the Background of Its Sources” (Ph.D. diss., The Hebrew University of Jerusalem, 2009), 119–99 (in Hebrew). 4 Various theological and scholarly usages of the term “preexistence” are surveyed in R. G. Hamerton-Kelly, Pre-existence, Wisdom, and the Son of Man: A Study of the Idea of Pre- existence in the New Testament (SNTSMS 21; Cambridge: Cambridge University Press, 1973), 1–13; in the present study the term signifies existence prior to the creation of the world. No Angels Before the World? 71

mention of angels, nothing is said here about when they were first created.”5 This is not so, however, in the case of the early payṭanim, who “were fond of the theme of creation and loved to recount the work of the six days in a highly ornate and dramatic style.”6 Accounts of the creation of the world found in early Piyyuṭ typically do mention the angels, most often as having been created on the second day.7 This may indeed be regarded as “the rabbinic standard view” of the matter.8 Thus in Gen. Rab. 11:9 we read that on the second day God 9.(בשני רקיע וגיהנם ומלאכים) ”created “the firmament, Gehenna, and the angels Similarly, Targum Pseudo-Jonathan on Gen 1:26 mentions the angels, “who דאיתבריין ביום תניין) were created on the second day of the creation of the world the same view is expressed in numerous additional rabbinic 10;”(לבריית עלמא texts.11 Yet certain early piyyuṭim do not conform to this apparent consensus, distinctly stating that the angels came into being at an earlier point in the creation process. These statements, as we shall see below, may be compared, however, to various sources from Second Temple literature and onwards. The Book of Jubilees, which features the earliest extant “paraphrase” of the six-day cosmogony according to the first chapter of Genesis, mentions several

5 See J. L. Kugel, Traditions of the Bible: A Guide to the Bible as It Was at the Start of the Common Era (Cambridge, Mass.: Harvard University Press, 1998), 58. 6 P. S. Alexander, “‘In the Beginning’: Rabbinic and Patristic Exegesis of Genesis 1:1,” in The Exegetical Encounter between Jews and Christians in Late Antiquity (ed. E. Grypeou, H. Spurling; Jewish and Christian Perspectives 18; Leiden: Brill, 2009), 6. The various genres of early Piyyuṭ in which creation accounts occur are surveyed in Granat, “Before ‘In the Beginning,’” 12–18. 7 See, e.g., E. Fleischer, The Yoẓer: Its Emergence and Development (Jerusalem: The Hebrew University Magnes Press, 1984), 713 (in Hebrew); S. Elizur, The Liturgical Poems of Rabbi Pinḥas ha-Kohen: Critical Edition, Introduction and Commentaries (The David Moses and Amalia Rosen Foundation: Sources for the Study of Jewish Culture 8; Jerusalem: World Union of Jewish Studies, 2004), 731 (in Hebrew). 8 Although it has precedents in the Second Temple period; see p. 82 below. 9 J. Theodor and C. Albeck, Midrash Bereshit Rabba: Critical Edition with Notes and Commentary (Jerusalem: Shalem, 1996), 96; Midrash Rabbah (ed. H. Freedman and M. Simon; 10 vols.; London: Soncino, 1939), 1:86. 10 M. Ginsburger, Pseudo-Jonathan (Thargum Jonathan ben Usiël zum Pentateuch): Nach der Londoner Handschrift (Brit. Mus. Add. 27031) (Berlin: Calvary, 1903), 3; Targum Pseudo- Jonathan: Genesis (trans. M. Maher; The Aramaic Bible 1B; Edinburgh: T&T Clark, 1992), 19–20. 11 See the references in Theodor–Albeck, Midrash Bereshit Rabba, 5; L. Ginzberg, The Legends of the Jews (trans. H. Szold; 7 vols.; Philadelphia: Jewish Publication Society of America, 1909–1938), 5:20–21. 72 Granat types of angels among the “seven great works” made by God on the first day (Jub. 2:2–3).12 The text highlights the creation of angels at this point, both by the lengthy enumeration of their various classes, and by “making them the fourth and thus the middle one of the seven created entities.”13 A parallel tradition itemizing the creations of the first day occurs in several later sources, both Jewish and Christian.14 In contrast to Jubilees, however, none of these sources list the angels among the creations of the first day; indeed the view that the angels were created on the first day does not seem to be recorded in the extant rabbinic sources.15 On the other hand, the angels are included in a list, comparable to the aforementioned “Jubilees tradition,” of seven “primordial natures who came into being in silence” (i.e., without an explicit divine command) in the very beginning of creation, presented by some Syriac authors (Narsai, Theodore bar Koni).16

In early Christian literature the view that the angels were created on the first day is, in fact, fairly widespread.17 Of special interest here is Augustine’s discussion in his De Civitate Dei (11.9), on the question, “What conclusion we should

12 The translation used here is that of O. S. Wintermute, OTP 2:55. 13 See J. C. VanderKam, “Genesis 1 in Jubilees 2,” in idem, From Revelation to Canon: Studies in the Hebrew Bible and Second Temple Literature, (JSJSup 62; Leiden: Brill, 2000), 500–521 (esp. 505–10; quotation on p. 506). 14 See M. Kister, “Tohu wa-Bohu: Primordial Elements and Creatio ex Nihilo,” JSQ 14/3 (2007): ”,אחור וקדם : אגדות ודרכי מדרש בספרות החיצונית ובספרות חז"ל“ ,and idem ;(44–242) 56–229 in Higayon L’Yona: New Aspects in the Study of Midrash, Aggadah and Piyuṭ, in Honor of Professor Yona Fraenkel (ed. J. Levinson, J. Elbaum, and G. Hasan-Rokem; Jerusalem: The Hebrew University Magnes Press, 2006), 231–59 (241–45). See also Kister’s article in the present volume, “Hellenistic Jewish Writers and Palestinian Traditions: Early and Late,” n. 69. 15 Admittedly in Tanḥuma Vayeshev 4, it is stated that “when the Holy One, blessed be He, created His world, He fashioned the angel of death on the first day :see S. A. Berman, Midrash Tanḥuma–Yelammedenu ;”(מיום הראשון ברא מלאך המות) An English Translation of Genesis and Exodus from the Printed Version of Tanḥuma– Yelammedenu with an Introduction, Notes, and Indexes (Hoboken, N.J.: Ktav, 1996), 232. However this tradition has to do with the predestination of death rather than with the creation of angels in general (contra Ginzberg, Legends of the Jews, 5:21). 16 T. Jansma, “Investigations into the Early Syrian Fathers on Genesis,” OtSt 12 (1958): 99–101 (the citation, from Bar Koni, is found on p. 99); A. Guillaumont, “Genèse 1,1–2 selon les commentateurs syriaques,” in In principio: Interprétations des premiers versets de la Genèse (Paris: Études augustiniennes, 1973), 119. 17 See, e.g., E. G. Mathews Jr., Jacob of Sarug’s Homilies on the Six Days of Creation: The First Day (Piscataway: Gorgias, 2009), 28 n. 32. No Angels Before the World? 73 reach, relying on scriptural evidence, about the creation of angels.”18 Augustine argues there that “surely if the angels are included among the works of God on those six days, they are the light that received the name ‘day’ (ipsi sunt illa lux, quae diei nomen accepit).” Augustine’s identification of angels with primordial light, though described as coming from his own independent deliberations,19 seems to echo much earlier traditions.20 A correlation of angels and the primordial light appears to be implied already in Job 38:4–7, where the angelic figures of “the sons of God” are described as present at the very start of cre- איפה היית ביסדי-ארץ . . . ברן-יחד ּכוכבי בקר) ”ation, alongside the “morning stars The view that the angels witnessed the initial creation of .(ויריעו כל-בני אלהִ ים light is expressed even more markedly in the Hymn to the Creator of the Psalms Scroll from Qumran (11QPsa), where “all of God’s angels” immediately respond in song to God’s separation of light from darkness and creation of dawn מבדיל אור מאפלה / שחר הכין בדעת לבו / אז ראו כול מלאכיו וירננו / כי הראם את) 21.(אשר לא ידעו It is noteworthy that an anonymous early piyyuṭ, a Shivʾata based upon Gen 1:1–7, similarly associates the creation of angels with the creation of light. The Shivʾata’s third strophe treats Gen 1:3; the creation of angels is implicitly assigned here to the first day of the week of creation, as it immediately follows the divine command “let there be light”:

וַ יֹאמֶראֱ ֹלהִ ים /‘זְרַ חאֹור’ חוְזָרַ ּבְעֹולָ מֹו חֶלֶד הֵאִיר מִּכְבֹודֹו וְהּוא בִמְ רֹומֹו טַפְסְרִ ים יָצַר לְהַקְּדִ יׁשֹו, לְ רֹומְ מֹו ּובָא יְיָ אֱ יֹלהַ וְכָל-קְ דֹׁשִ יםעִ ּמֹו

And God said (Gen 1:3) “shine, light,” and it shone in His world / The universe was illumined by His glory while He was in his dwelling on high / He created ministering angels to sanctify and praise Him / And the Lord my God came, and all the holy ones with Him (see Zech 14:5).22

18 Saint Augustine, The City of God against the Pagans (7 vols.; LCL; Cambridge, Mass.: Harvard University Press, 1957–1972), 3 (trans. D. S. Wiesen): 460–61. 19 See E. Muehlberger, Angels in Late Ancient Christianity (New York: Oxford University Press, 2013), 48–49. 20 M.-T. d’Alverny, “Les anges et les jours,” Cahiers Archéologiques 9 (1957): 271–300 (282). 21 11Q5 26:11–12; J. A. Sanders, The Psalms Scroll of Qumran Cave 11 (11QPsa) (DJD 4; Oxford: Clarendon, 1965), 89–90. See also Kugel, Traditions of the Bible, 49. 22 See S. Elizur, Shivʾatot for the Weekly Torah Readings (Sources for the Study of Jewish Culture 1, Jerusalem: World Union of Jewish Studies, 1993), 163–64 (all translations are my own unless otherwise specified). 74 Granat

It might possibly be argued that the reference to glorifying angels at this junc- ture derives from the specific liturgical “niche” of this strophe, introducing as it does the third benediction of the ʿAmidah, which concerns the angels’ sanctifi- cation of God. However these circumstances did not obligate the poet to mention here the creation of angels, or indeed the angels as such; any occurrence ,in the concluding line of the strophe could suffice. Indeed ק־ד־ש of the root angels of a specific sort, the cherubim, are mentioned again, also in connection with the creation of light, in the beginning of the next strophe, relating to Gen 1:4:

‘וַיַרְא אֱ ֹלהִ ים’ יֹופִי הָאֹור ּבְ זָרחֹו מֵחַ ּלֹונֹות ּכְ]נִ[מְ לְַךּבִקְ דּומָ ה וְסָחעַל ּכְ רּובִ ים לַחֲ נֹות

And God saw (Gen 1:4) the beauty of light, shining out of (heaven’s) windows / As He consulted the ancient one (the Torah) and commanded to dwell upon cherubim.

Three motifs are intertwined in this interesting passage. The first is God’s approval of the created light, as related in Gen 1:4. The second is God’s con- -the ancient one”) before cre“ , ד קְ ּו מָ ה) sultation with the preexistent Torah ation. This is a clear reference to the preexistence tradition most widespread in rabbinic literature as well as in early Piyyuṭ: the primordial existence of Torah before creation, serving as God’s confidant and advisor.23 The third motif is that of God’s dwelling upon cherubim. This motif was interpreted by Shulamit Elizur as referring to the (future) Divine Presence in the tabernacle, between the two cherubim placed upon the ark of the testimony (Exod 25:22; Num 7:89).24 However, as this interpretation is rather remote from the given context (i.e., the story of creation), it may be preferable to see here a reference to God’s enthronement above the celestial cherubim (Ezek 10:4), which took place, according to this piyyuṭ, at the time of the creation of light. Interestingly, in the Slavonic Book of Enoch (2 Enoch) 25:3–4, God is described, in a remarkably similar setting, as enthroned immediately following the primordial creation of light, and admiring the goodness of it: “And I was in the midst

23 See, e.g., M. Kister, “Some Early Jewish and Christian Exegetical Problems and the Dynamics of Monotheism,” JSJ 7/4 (2006): 548–93 (575–79), and the literature mentioned there. 24 Elizur, Shivʾatot, ad loc. No Angels Before the World? 75 of the great light. And light out of light is carried out. . . . And I saw how good it was. And I placed for myself a throne, and I sat on it.”25 Unlike the preceding strophe, which may imply that the angels were created simultaneously with or immediately after the light, a careful reading of these two lines from the Shivʾata suggests that at least some angelic beings (the cherubim) may have existed even beforehand, at the very threshold of creation. This idea is explicitly expressed in the opening section of the late midrash Tanna DeBe Eliyahu, a short homily based on Gen 3:24, which describes the that the cherubim“) שהכרובים קודמים למעשה בראשית :cherubim as preexistent are anterior to the work of creation”).26 An additional occurrence of this tradition can be found in section 31 of the same midrash, where we find two alternative accounts of the early stages of creation. The second account includes a standard list of preexistent entities (the Torah, the Temple etc.), which has parallels in several rabbinic sources.27 The first, on the other hand, is an description of the heavenly realm, depicting God as sitting upon different groups of angels (including the cherubim), prior to the creation of the world and at the beginning of it.28 The thematic and stylistic indebtedness of Tanna DeBe Eliyahu to the esoteric corpus of Hekhalot literature is quite evident here.29 And indeed, in a Genizah fragment of a little known Hekhalot work, we find a clearly comparable description of different groups of angels, including the cherubim, upon which God’s throne was placed “before the world was created” In the Shivʾata under discussion here, then, two early 30.(קודם שיברא העולם) traditions seem to be echoed, both atypical from the rabbinic perspective, but recorded in other late antique sources: 1) the angels as created alongside light

25 The translation follows F. I. Andersen, OTP 1:144–45; the citation is according to version J. The presence of preexistent quasi-angelic figures in this account is discussed below.‬ 26 M. Friedmann, Seder Eliyahu Rabbah and Seder Eliyahu Zuta (Tanna DeBe Eliyahu) (Jerusalem: Wahrmann, 1969), 3. 27 B. Pesaḥ. 54a, b. Ned. 39b, Gen. Rab. 1:4, and numerous later sources; see Granat, “Before ‘In the Beginning,’” 75–82. 28 Friedmann, Seder Eliyahu Rabbah, 160. 29 J. Elbaum, “The Midrash Tana devei Eliyahu and Ancient Esoteric Literature,” in Early Jewish Mysticism: Proceedings of the First International Conference on the History of Jewish Mysticism (= Jerusalem Studies in Jewish Thought 6) (ed. J. Dan; Jerusalem: The Hebrew University of Jerusalem, 1987), 139–50 (in Hebrew); regarding Tanna DeBe Eliyahu 31, see Elbaum’s comments on pp. 141–42. 30 See P. Schäfer, Geniza-Fragmente zur Hekhalot-Literatur (TSAJ 6; Tübingen: Mohr, 1984), 181; G. Bohak, “The Hidden Hekhalot: Towards Reconstructing an Unknown Hekhalot Composition from the Cairo Genizah,” Tarbiẓ 82 (2014): 423 (in Hebrew). 76 Granat on the first day; and (possibly) 2) their presence even before the creation of light, at the very beginning of the cosmogony.

3 The Preexistence of Angels in Early Piyyuṭ

The remarkable notion that the angels had already come into being before the beginning of the creation process is clearly detectable in a few early payeṭanic texts which now deserve our attention. -When All Was Not), an anonymous and monu) אז באין כל The first section of mental Seder Avodah, comprises twenty-four massive quadripartite lines, all This is a lengthy introduction to the creation 31.א beginning with the letter account proper, presented in the following sections. The “prologue” deals mostly with God’s obedient angelic servants, described here passionately and in great detail. The seventh line recounts God’s deed of creating His angels out of primal, celestial fire:

רְאֶ אֶ י ּלִ ם חָ צַ בְ ּתָ ה / מִ ּלַהֲבֹות אֵ ׁש // וְחֹולַלְּתָה חַ יֹות / מִ ּנְהַר מִרְ ּכֶבֶ >ת<

You carved Erelim / from flames of fire; You engendered the creatures / out of the river of the Chariot32

This statement is followed by a detailed characterization of the angels’ elevated constitutions and supernatural qualities. Only significantly later, towards the end of the section, does the creation of the world come to be mentioned. The poet declares that, even though the angels are God’s confidants, He did not take any advice from them when proceeding to create the world:

אֵילֵי אֶרֶץסֹודֶ ָך / םּובָ לֹא נִמְ לַכְּתָ ה // ּבְ נָושְאְָךלִּבָ ְך / לִבְרֹות עֹולָ ם

The mighty of the world are Your confidants / yet You did not consult them When You made up your mind (or: thought in Your heart) / to create the world33

31 M. D. Swartz, J. Yahalom, Avodah: An Anthology of Ancient Poetry for Yom Kippur (The Penn State Library of Jewish Literature; University Park, Pa.: Pennsylvania State University Press, 2005), 97–103 (Hebrew text and English translation). 32 Swartz and Yahalom, Avodah, 88–89. 33 Swartz and Yahalom, Avodah, 102–3; Kister, “Some Early Jewish and Christian Exegetical Problems,” 583. No Angels Before the World? 77

section) is dedicated to God’s consultation ב The next part of the poem (its with the preexistent Torah,34 and the actual cosmogony is recounted in the following seven sections. This sequence of events distinctly suggests that according to this poet the angels were preexistent, i.e., created well before the world. A comparable sequence can be identified in a creation account which ,(I Shall Loudly Cry) אנהם ברב כח ,opens yet another anonymous early piyyuṭ the only unrhymed—and hence probably the earliest—extant Seder Olam from a Shivʾata for Shemini Atzeret.35 Here as well, the angels, created from fire are described at ,(חֲ צּובֵ יאֵ ׁש אֹוכֵילָ ה; ּגְדּודֵ י לַהֲבֵ י לַהַ ט; בְ נִ רָ אִ ם י מִ ּגַ חֲ י לֵ אֵ ׁש :lines 11–12) the very beginning of the narrative, within the utterly primordial, chaotic state of affairs, and even before the preexistent Torah and the preexistent Temple are mentioned. In four additional early piyyuṭim the preexistence of angels is stated in a straightforward and explicit manner, rather than merely being implied by the narrative order. Our first example is a Qedusha piyyuṭ, probably by the well- known classical payṭan Elʿazar Birabi Kilir (Qillir, Kalir). The poem begins with the statement that the “holy living creatures” were created before the earth; i.e., presumably, before the world, since this statement precedes the reference to the creation of heaven and earth:

וְחַ יֹות הַ ּקֹו> דֶ ׁש< אֲ ׁשֶ >ר< ּבַ ּׁשָ מַ ִ י ם / ּבְרּואֹות הֵם עַ ד לֹא אֶרֶ ץ . . . הַּדָר למַעַ יְמִינֹו טִּפְחָה ׁשָמַ >יִ ם< / תּומִּתַחַ יָדֹו יָסְדָה אָרֶ ץ

And the holy living creatures which are in heaven / are (were) created before the earth. . . . He who resides on high (God), His right arm spanned the heaven / and below His hand founded the earth36

The remaining three texts in this category belong to the Seder Beriyot genre.37 Functioning as preludes to Seder Avodah poems, which traditionally begin with a creation account, the Seder Beriyot poems mostly celebrate God’s preexistence but also present additional preexistence and creation themes.

34 Regarding this section, see Kister, “Some Early Jewish and Christian Exegetical Problems,” 583–85; Granat, “Before ‘In the Beginning,’” 188–99. 35 S. Elizur, “‘Visit Your Land with Rain’: Poetic Fragments of Early Shivʾatot for Rain,” Ginzei Qedem 1 (2005): 31–78 (54) (in Hebrew). 36 Elizur, Elʿazar Birabi Kiliri, Hymni Pentacostales, 201 (in Hebrew); regarding the attribution of the poem, see ibid., 34. 37 See M. Rand, “The Seder Beriyot in Byzantine-Era Piyyuṭ,” JQR 95 (2005): 667–83. 78 Granat

The only unrhymed—and hence probably the earliest extant—examples of this genre comprise a sequence of two poems, both markedly rich in Hekhalot With Whom did You Take Counsel?), is) את מי נועצת ,phraseology.38 The first one labelled in the manuscript with the noteworthy heading, unknown from other Creation of Angels”). This poem is dedicated mainly to“) בריות-מלאכים ,sources a detailed description of the creation of the heavenly palaces and their “guard- Then, Upon the Mighty) אז על אדירים ,whereas the second poem ;( ׁש ֹו מְ רִ י ם) ”ians Ones), focuses on the creation of the heavenly chariot and divine throne, including the fiery living creatures that inhabit it. Both poems conclude with a closing line declaring that only after all of the abovementioned did God make All the various .( חַ וְאַ ר ּכָ ך נִ מְ לַ כְ ּתָ ה / לִבְ רֹות עֹולָ ם) up His mind to create the world angelic creatures described in these two poems, then, antedate the world. God of Old), is a rhymed, and) אלוהי קדם ,The last poem in this category hence probably somewhat later, Seder Beriyot piyyuṭ.39 Here it is stated most straightforwardly that the angels had been created and were praising God already before the world was made:

טֶרֶםלְׁשִ יכְ נְָך אִ יּמַצְּתָה ׁשְחָקִ ים / יָּה, נֶעֱרַ צְּתָ הּבְרִ יבְבֹות מַּׁשִיקִ ים . . . ּפְעֻוּלָתָם ּכָאֵ ׁש וְהֵם הִיקְּדִ יׁשּוָךּבְ סֹוד / צּור, עַ ד לֹא נִימְ לַכְּתָ העֹולָם לִ יסֹוד

Before You strengthened the skies for Your dwelling / God, You were adored by myriads of calling ones. . . . Their making is firelike, and they sanctified You in their assembly, / Lord, before You made up your mind to create the world.

4 Angels before or after Cosmogony? Rabbinic Polemics and Contemporary Debates

4.1 The Rabbinic Denial of Angelic Preexistence The attribution of preexistence to angels may be regarded as atypical, perhaps even heterodox, from a rabbinic perspective, as is made clear by the following passage from Gen. Rab. 1:3:

38 M. Rand, “More on the Seder Beriyot,” JSQ 16 (2009): 198–209. 39 J. Elbogen, Studien zur Geschichte des jüdischen Gottesdienstes (Berlin: Mayer & Müller, 1907), 186. I intend to republish this text, with a full commentary, along with two other texts from the same Genizah fragment, preserved in the Bodleian Library, Oxford (Ms. Heb. 2740.14). No Angels Before the World? 79

אימתי ניבראו המלאכים? ר’ יוחנן אמר, בשני נבראו. . . . ר’ חנינא אמר, בחמישי נבראו. . . . ר’ לוליני בר טברי בשם ר’ יצחק: בין על דעתיה דר’ חנינא ובין על דעתיה דר’ יוחנן הכל מודים שלא נברא ביום ראשון כלום, שלא תאמר מיכאל היה מותח בדרומו שלרקיע וגבריאל בצפונו והקב"ה ממדד באמצע. אלא—‘אני י"י עושה כל נוטה שמים לבדי רוקע הארץ מאתי’: מי אתי כתיב, מי היה שותף עימי בבריית העולם.

When were the angels created? R. Joḥanan said: They were created on the second day. . . . R. Ḥanina said: They were created on the fifth day. . . . R. Luliani b. Tabri said in R. Isaac’s name: Whether we accept the view of R. Ḥanina or that of R. Joḥanan, all agree that none were created on the first day; lest you say, Michael was stretching out the southern side of the sky and Gabriel the northern side, while the Holy One, blessed be He, measured it in the middle. But [rather], “I am the Lord, that makes all things; that stretched forth the heavens alone; that spread abroad the -who was with Me) is writ) מי אתי :(Isa 24:24) (מאתי)—”earth by Myself ten; who was associated with Me in the creation of the world?40

Two alternative dates for the creation of angels are offered here: the second and fifth days of the week of creation. Then R. Luliani b. Tabri (in R. Isaac’s name) declaratively rules out attributing their creation to the first day, on the following grounds: “all agree that none were created on the first day; lest you say, Michael was stretching out the southern side of the sky and Gabriel the northern side, while the Holy One, blessed be He, measured it in the middle.” The polemical tone, directed against a theologically problematic view, is unmistakable: angels should not be thought to have been created before the second day, lest one think that they had assisted God in subsequent acts of creation. Though phrased as a hypothetical possibility (“lest you say . . .”), Rabbi Luliani’s statement in our passage has been interpreted as “testifying that some

40 Theodor–Albeck, Midrash Bereshit Rabba, 5; Midrash Rabbah, 5 (alt.). The passage is discussed, e.g., by Kister, “Some Early Jewish and Christian Exegetical Problems,” 585; J. E. Fossum, The Name of God and the Angel of the Lord: Samaritan and Jewish Concepts of Intermediation and the Origin of Gnosticism (WUNT 36; Tübingen: Mohr, 1985), 194, 236, 265; C. Markschies, Valentinus Gnosticus? Untersuchungen zur valentinianischen Gnosis mit einem Kommentar zu den Fragmenten Valentins (WUNT 65; Tübingen: Mohr, 1992), 18; P. Schäfer, The Jewish Jesus: How Judaism and Christianity Shaped Each Other (Princeton: Princeton University Press, 2012), 162–65. I am grateful to Prof. Menahem Kister for bringing the latter two studies to my attention. 80 Granat people alleged that Michael and Gabriel were associates of God in creation.”41 (המינים) ”Similarly, in y. Ber. 9:1 (and parallel sources) we read of “the heretics כמה אלוהות) ”?who asked Rabbi Simlai, “How many gods created the world the implication that several divine agents were involved in the ;(בראו את העולם cosmogony is firmly refuted by the rabbi.42 Beliefs concerning demiurgic angels who were supposed to have created the universe (rather than God himself) were indeed fairly widespread among esoteric and Gnostic or quasi-Gnostic sects in late antiquity,43 and it is worthy of mention here that certain Gnostic and Mandaic texts regard both Michael and Gabriel (the two angels mentioned by R. Luliani in Gen. Rab. 1:3) as principal demiurgic angels.44 The currency of such beliefs in certain contemporaneous Jewish circles45 is probably the background to the evident polemic nature of the two rabbinic passages mentioned above. But the discussion regarding the role of angels in creation has much earlier roots.

4.2 Angels and Cosmogony in Second Temple Literature and Related Texts These later Gnostic views of preexistent angels actively engaged in the creation process are probably related to early Jewish traditions already found in Second Temple writings. In 1 En. 69:13–25 the creation of the world and the maintenance of cosmic order are achieved by means of an oath, using a secret divine name; this oath was “placed in the hand of Michael, . . . and through his

41 A. F. Segal, Two Powers in Heaven: Early Rabbinic Reports about Christianity and Gnosticism (SJLA 25; Leiden: Brill, 1977; repr. 2002), 137. 42 See Segal, Two Powers, 121–34; Kister, “Some Early Jewish and Christian Exegetical Problems,” 588–89, esp. n. 132. 43 See, e.g., Fossum, The Name of God, 211–20; C. B. Smith II, No Longer Jews: The Search for Gnostic Origins (Peabody, Mass.: Hendrickson, 2004), 125–49; F. Williams, The Panarion of Epiphanius of Salamis, Book I (Sects 1–46) (NHS 35; Leiden: Brill, 1987), 55–57; Markschies, Valentinus Gnosticus? 21–24. 44 See Fossum, The Name of God, 261–64 (Gabriel); 323–29 (Michael); T. Rasimus, “The Archangel Michael in Ophite Creation Mythology,” in Jewish and Christian Cosmogony in Late Antiquity (ed. L. Jenott, S. Kattan Gribetz; TSAJ 155; Tübingen: Mohr Siebeck, 2013), 107–25. 45 In at least one Gnostic text such beliefs are attributed to “some Jews” See Fossum, The Name of God, 211. Indeed, some Hekhalot texts may hint at such a concept (J. Dan, “ʿAnafiʾel, Metatron, and the Creator,” Tarbiz 52 [1982–1983]: 447–58 [in Hebrew]), including the Genizah fragment mentioned above (p. 75 and n. 30; see Granat, “Before ‘In the Beginning,’ ” 132–34). Versions of the same idea occur later in Karaite circles; see H. A. Wolfson, “The Preexistent Angel of the Magharians and al-Nahāwandī,” JQR n.s. 51/2 (1960): 89–106. No Angels Before the World? 81 oath, heaven was suspended from before the world was created and forever” (1 En. 69:15–16).46 Though this is not explicitly stated, angelic agency at the very beginning of creation is probably implied here; bearing in mind the view rejected in Gen. Rab. 1:3 (“Michael was stretching out the southern side of the sky . . .”), the reference to Michael’s involvement in the suspension of heaven is particularly interesting. Second Enoch 25–26 also describes the initial stages of the creation of the world, implemented through mediating agents.47 The work of creation begins here with the appearance, through the divine command, of two primordial beings. In response to God’s call to “let one of the invisible things descend visibly,” the first such being appears: “and Adoil descended, extremely large” (2 En. 25:1). God commands him to “disintegrate” himself, and a great light comes out; this light becomes “the foundation of the highest things” (25:2–4). God then calls similarly “into the very lowest things”; in response to this call “Arkhas came out, solid, and heavy and very red” (26:1). Following God’s orders he disintegrates himself as well, and becomes “the foundation of the lowest things” (26:3). In this dramatic account of the beginning of creation, two personified entities, originating from a preexistent realm of “invisible things,” become, as they obey God’s orders, the foundations of the upper and lower realms of the universe. Also noteworthy is the account of the six days of creation in 4 Ezra 6. On the second day (recounts the prophet, addressing God), “You created the spirit of the firmament, and commanded him to separate the waters, that one part might move upward and the other part remain beneath” (4 Ezra 6:42).48 According to this description, as Michael Stone puts it, “God creates a ‘spirit of the firmament’ and delivers his commands to this personalized firmament.”49 Again we see here an angel-like “spirit,” which functions as God’s agent in creation (and specifically in the creation of the sky, as in 1 Enoch and Genesis Rabbah!). Interestingly, in the Armenian and Arabic versions of 4 Ezra the reference to “spirit” does not occur; the reason for this apparent omission is

46 See J. Ben-Dov and E. Ratzon, “The Oath and the Name in 1 Enoch 69,” JSS (forthcoming). I am grateful to Dr Ben-Dov for sending me the text of this article, which includes a new English translation of the passage under discussion. See also G. W. E. Nickelsburg and J. C. VanderKam, 1 Enoch 2: A Commentary on the Book of 1 Enoch, Chapters 37–82 (Hermeneia; Minneapolis: Fortress,‎ 2012), 304; Fossum, The Name of God, 257–59.‬ 47 Andersen, OTP 1:144–45, version J. The details of this passage are discussed by Fossum, The Name of God, 287–90; and A. A. Orlov, “Secrets of Creation in 2 (Slavonic) Enoch,” Henoch 22 (2000): 45–62.‬ 48 M. E. Stone, Fourth Ezra: A Commentary on the Book of Fourth Ezra (Hermeneia; Minneapolis: Fortress, 1990), 178. 49 Stone, Fourth Ezra, 185 (alt.). See also d’Alverny, “Les anges,” 283–84. 82 Granat

“perhaps dogmatic,” as Stone notes (ad loc.); probably due to the theological sensitivity of giving a role in the initial act of creation to angelic figures along with God. Indeed, Second Temple sources already regard traditions of this sort as distinctively problematic from a theological point of view; i.e., as potentially undermining God’s status as sole creator. Thus various passages in Jubilees, 1 Enoch, 2 Enoch, and Ben Sira all highlight the superiority of God’s wisdom to that of the angels, as well as God’s essential independence in creation, “so decisively and emphatically as to leave no doubt as to its importance,” as Menahem Kister has demonstrated.50 The question of when the angels came into being becomes especially pertinent in this context. Several of the Second Temple sources we have explored here set the timing of the creation of the angels as a counterpoint to the portrayal of God as sole creator. Jubilees 2:2–3 dates the creation of the angels to the first day, following God’s initial creative acts, and limits their role to praising those acts; J. C. VanderKam rightfully points out that although this timing is “the view against which the scholars cited in Genesis Rabbah argue,” yet, “the author [of Jubilees] took pains to avoid the very danger to which such a view was considered vulnerable”51—i.e., the danger of undermining God’s exclusiv- ity in creation. The very statement that the angels were created on the first day, and hence were not present beforehand, may be interpreted as reflecting the same theological motive. Second Enoch 29:3 assigns the creation of angels to the second day (R. Joḥanan’s view in Gen. Rab. 1:3): “and from the fire I created the ranks of the bodiless armies—ten myriad angels”; notwithstanding the prior appearance of Adoil and Arkhas in 2 Enoch 25–26 and their own crucial role. The second day comes to the fore in 4 Ezra as well; 4 Ezra 6:42, as mentioned above, relates the creation of God’s “assistant,” “the spirit of the firmament,” on that day, although angels in general are not mentioned. In L.A.B. 60:1–3,52 “The Song of David the exorcist,”53 it is stressed that spirits came into being later than the initial stage of cosmogony, as an indication of

50 Kister, “Some Early Jewish and Christian Exegetical Problems,” 579–82 (quotation pp. 579–80). 51 VanderKam, “Genesis 1 in Jubilees 2,” 508–9. 52 H. Jacobson, A Commentary on Pseudo-Philo’s Liber antiquitatum biblicarum, ‎with Latin Text and English Translation (2 vols.; AGJU 31; Leiden:‎ Brill,‎ 1996), 1:82, 188–89; 2:1173–78. ‬‬ 53 So titled by Daniel J. Harrington; see D. J. Harrington, “Pseudo-Philo, Liber Antiquitatum Biblicarum,” in Outside the Old Testament (ed. M. de Jonge; Cambridge Commentaries on Writings of the Jewish and Christian World, 200 BC to AD 200 4; Cambridge: Cambridge University Press, 1985), 24. No Angels Before the World? 83 their lower status; as such this text is somewhat comparable to Gen. Rab. 1:3, even though it is the evil spirits rather than the angels which are concerned here. This short (and rather obscure) apotropaic “song,” sung by David to drive out the spirit that had possessed Saul (1 Sam 16:14–23), opens with a succinct account of the creation of heaven and earth. Then David points out, addressing the spirit, that only after the creation of all things did the spirits come into being: “after this was the tribe of your spirits made” (“Et post haec facta est tribus spirituum vestrorum”). This is the ground for the following rebuke: “Now do not be troublesome, since you are a secondary creation (secunda creatura).” The unusual term secunda creatura has been interpreted by some scholars as referring to the creation of the spirits “on the second day,”54 though the Latin phrase is more plausibly explained as to “a second/secondary creation.”55 At any rate, it is illuminating for our discussion that an inferior status is assigned here to the spirits, in correlation with their (relatively) late time of creation.

4.3 Divine Agency in Cosmogony and Angelic Preexistence: The Theological Risk and the Rabbinic Response In view of the various sources surveyed above, it may be useful to clarify the distinction between two issues which, though often intertwined, are essentially independent: 1) the question of whether the angels preceded God’s initial act that created the world; and 2) the question of whether angels took part in creating the world. One can observe a general tendency, in most of the Jewish sources so far surveyed, to differentiate God from the angels by claiming either that a) the latter did not exist before the world; or that b) they did not take any part in its creation. The linkage of these two concerns (made explicitly by R. Luliani in Gen. Rab. 1:3) is possible, but not necessary. Fourth Ezra 6:42, as noted above, refers to the creation of “the spirit of the firmament” on the second day, not beforehand; yet it does ascribe to this spirit a crucial role in performing the work of that day. Additionally, in several rabbinic sources it is said that God consulted the angels as He turned to create the human being on the sixth ,נעשה אדם day; a consultation implied in the plural form of God’s statement “let us make man” (Gen 1:26). Menahem Kister has interpreted this statement as reflecting “efforts at refining a tradition according to which angels had an active role in the creation of man, i.e., in the creation of the human body.”56

54 Harrington, “Pseudo-Philo,” 25. 55 Jacobson, Commentary, 2:1177. I am grateful to Dr. Eran Almagor for his comments on this point. 56 See Markschies, Valentinus Gnosticus? 19–21; also Kister, “Some Early Jewish and Christian Exegetical Problems,” 569–71, and the literature mentioned there. 84 Granat

That the rabbis were ready to accept this tradition, albeit in a “refined” version, despite their determined objection to undermining God’s exclusive status as the sole creator, may be explained by the specific, unavoidably conspicuous in Gen 1:26.57 It also נעשה textual challenge of the aforementioned plural verb should be taken into account, however, that the creation of man occurs only at the very last stage of the creation narrative, taking place within a nearly complete universe. On the other hand, angelic participation in the initial creation of the world is explicitly repudiated in rabbinic sources, a repudiation which is linked with the denial of angelic preexistence. We have already seen, though, that some early piyyuṭim take a different tack in their affirmation of angelic preexistence, and it is to their approach that I now turn.

5 Praise, not Creation: The Raison d’etre of Preexistent Angels in Piyyuṭim and Parallels

5.1 The “Work” of Preexistent Angels in Early Piyyuṭ As mentioned above, according to Gnostic views of cosmogony, angels were the active agents in the creation of the universe. Note, for example, Epiphanius of Salamis’s reports of Basilides, who believed that

A highest first heaven has been made by . . . principalities, authorities, and angels . . . and other angels were made by them. Again, the angels they made have created a second heaven, and made angels themselves in their turn. . . . And thus, by preparing another heaven and other angels in turn, the angels which go with each heaven have produced a total of 365 heavens, from the highest to ours.58

The specific reference here to angels as creating the heavens brings to mind some of the texts discussed above (1 En. 69:15; 4 Ezra 6:42; Gen. Rab. 1:3). Moreover, the description of different heavens and “the angels which go with each heaven” is somewhat reminiscent of Hekhalot texts which describe

57 It is noteworthy that, based on the special phrasing of Gen 1:26, Philo of Alexandria (De Opficio Mundi, 73–75) suggests that “the enlistment of others as collaborators” is indeed relevant only in the case of human beings, which he explains as a consequence of the uniqueness of human nature. For the translation, see Philo of Alexandria, On the Creation of the Cosmos according to Moses (ed. and trans. D. T. Runia; PACS 1; Leiden: Brill, 2001), 65–66. 58 Pan. 24.1.8-9; Williams, Panarion, 70–71. No Angels Before the World? 85 in detail the heavenly palaces vis-à-vis specific angels who are in charge of With Whom did You Take) את מי נועצת them.59 More specifically, we may recall Counsel?), discussed above, which refers to preexistent angels and is imbued with Hekhalot motifs,60 and which opens by describing the creation of the heavenly palaces and the appointment of guardian angels for each of them. Yet in clear contrast to Basilides’ view of angels creating both other angels like all the aforementioned ,את מי נועצת ,and the heavens in which they reside early piyyuṭim, unequivocally portrays God Himself as creating the preexistent angels; none of these texts attributes to the angels any creative function what- When All) אז באין כל soever. Furthermore, as noted above, the Seder Avodah Was Not) narrates the creation of the angels well before the creation of the world. Nevertheless, relates the poem, God did not (even) consult them when He decided to create the world.61 Significantly, the poem highlights the angels’ utter humility and absolute obedience to God. Of particular interest is the following statement:

‘אָמֵן’ מְ לַאכְּתָ ם /‘ּבָרּוך’ עֲבֹודָתָ ם

“Amen” is their vocation (or: craft) / “Blessed” is their labor62

The angels’ “craft” and “labor” are no more (and no less) than the recitation of hymns, which they dedicate to God, and which are marked by the liturgical Blessed”). Whereas R. Luliani in Gen. Rab. 1:3“) ברוך Amen”) and“) אמן formulae opposes the attribution of preexistence to angels as a precaution against ascribing to them a role in cosmogony, the piyyuṭim discussed here openly describe angels as having been created before the world; yet they accentuate these angels’ absolute dedication to the praise of God in song, to the exclusion of any other role. As such they can be regarded as continuing the same ancient tradition of primordial angelic praise recorded in Job 38:7; Jub. 2:3 (“Then we saw his works and blessed him and offered praise for him on account of all

59 The number 365 which appears here has a special significance in some Hekhalot texts; see J. Dan, History of Jewish Mysticism and Esotericism: Ancient Times (3 vols.; Jerusalem: The Zalman Shazar Center for Jewish History, 2008), 2:786 (in Hebrew). 60 Rand, “More on the Seder Beriyot,” 198. 61 See p. 76 and n. 33 above. 62 Swartz and Yahalom, Avodah, 98–99. An interesting Syriac parallel occurs in Jacob of Sarug’s verses concerning the angels created on the first day: “He established them . . . to bless, to give praise and to sing “Alleluia” . . . He set “Holy” into the mouth of the seraph when He created him” (Mathews, Jacob of Sarug’s Homilies, 30–31). 86 Granat his works”); and the Hymn to the Creator in 11Q563—although here the angelic praise occurs not as the angels’ spontaneous response to God’s creation but as the performance of their essential “vocation,” even before the creation had begun.

5.2 Preexistent Angels of Praise in Early Christian Texts Angelic preexistence, even of such an auxiliary, “inactive” sort, is only rarely presumed by Jewish sources of late antiquity. A comparable view concerning the preexistence of angels is often expressed and embraced by early Christian writers, however, alongside references to the preexistence of Christ.64 Statements on the subject occur, for example, in Basil of Caesarea’s Hexameron 1.5,65 and in Ambrose of Milan’s Hexameron 1.5.19 (“The Angels, Dominations, and Powers, although they began to exist at some time, were already in existence when the world was created”).66 An illuminating parallel to these early piyyuṭim may be seen in a prayer from the Apostolic Constitutions.67 This compilation of prayers in Greek, dated to the second or third century CE but included in a church order composed in the fourth century, has been much discussed by scholars with regard to the presence of Jewish elements contained within it.68 Apostolic Constitutions 8.12.6–27 comprises a lengthy liturgical hymn recounting the creation story.69 It begins with a declaration of the eternal nature of God and the preexistence of Christ before all else; he was begotten “before all ages” (πρὸ πάντων αἰώνων), and it

63 See p. 73 n. 21 above. 64 See, e.g., Guillaumont, “Genèse 1, 1–2,” 118–19. 65 Basile de Césarée, Homélies sur l’Hexaéméron: Texte grec, introduction et traduction (ed. S. Giet; 2d rev. and enl. ed.; SC 26; Paris: Cerf, 1968), 104–7; Saint Basil, Exegetic homilies (trans. A. C. Way; FC 46; Washington, D.C.: Catholic University of America Press, 1963), 8–9. 66 Saint Ambrose, Hexameron, Paradise, and Cain and Abel (trans. J. J. Savage; FC 42; Washington D.C.: Catholic University of America Press, 1961), 18. 67 See the editions of D. A. Fiensy, Prayers Alleged to be Jewish: An Examination of the Constitutiones Apostolorum (BJS 65; Chico, Calif.: Scholars Press, 1985); and P. W. van der Horst and J. H. Newman, Early Jewish Prayers in Greek: A Commentary (CEJL; Berlin: de Gruyter, 2008), 1–93. 68 See the review of scholarship in van der Horst and Newman, Early Jewish Prayers, 9–22; they set out their own approach on 27–28; and see also the recent discussion of P. W. van der Horst, “Mystical Motifs in a Greek Synagogal Prayer?” in With Letters of Light: Studies in the Dead Sea Scrolls, Early Jewish Apocalypticism, Magic and Mysticism in Honor of Rachel Elior (eds. D. V. Arbel and A. A. Orlov; New York: de Gruyter, 2011), 254–64. 69 See Fiensy, Prayers, 98–101. No Angels Before the World? 87 was through him that all things came into being. Then it is stated that “before all things” (πρὸ πάντων), ten different classes of angels were created, including “the cherubim and seraphim” (τὰ Χερουβὶμ καὶ τὰ Σεραφίμ), and that only “after all these” (μετὰ ταῦτα πάντα) did the visible world come into being. The angels’ preexistence is framed here from two perspectives, that of their anteriority to the world, and that of the world’s creation as subsequent to them; both statements have parallels in the piyyuṭim quoted above. After a detailed account of creation and the sacred history of humankind and Israel, from Adam to Joshua, the hymn returns to the angelic sphere, listing again the various classes “of angels, archangels, thrones, dominions . . . the Cherubim and the six-winged Seraphim,” now described as worshipping and glorifying God. The opening and closing sections, the first dealing with the angels’ preexistence and the second describing angelic worship, clearly complement one another and function as an inclusio for the account of the creation and the history of human beings within it. Though it is not explicitly said here that the angelic praise took place already before creation, the conspicuous chiastic symmetry of the two sections makes this at least a plausible understanding of the passage. One may recognize thus in this early Christian Greek liturgical text a (mildly) “Christianized” version of the very same tradition we identified in our Hebrew liturgical poems: the angels praising God already before the world was made.

5.3 Preexistent Angels of Praise in Two Rabbinic Passages: Traces of an Earlier Tradition Interestingly enough, and despite the decisive dictum of Gen. Rab. 1:3, the concept of preexistent angels praising God before the creation does survive, albeit indirectly, in two rabbinic passages. The first is Abot de-Rabbi Nathan (A), 31:

תשע מאות ושבעים וארבעה דורות קודם שנברא העולם הייתה תורה כתובה ומונחת בחיקו של הקב"ה ואומרת שירה עם מלאכי השרת

Nine hundred and seventy four generations before the world was created, the Torah was (already) written: it lay in the bosom of the Holy One, blessed be He, and recited the Song along with the ministering angels.70

70 Avoth de-Rabbi Nathan (ed. S. Schechter; introduction by M. Kister; New York: Jewish Theological Seminary of America, 1997), 91; The Fathers according to Rabbi Nathan (trans. J. Goldin; New York: Schocken, 1974), 126–27. 88 Granat

The second is Tanḥuma (Buber), Qedoshim 7:2:

אמר להן הקב"ה לישראל עד שלא בראתי עולמי היו מלאכי השרת מקלסין אותי בכם ומקדשין את שמי בכם והיו אומרים: “ברוך ה' אלהי ישראל מן העולם ועד העולם”

The Holy One, blessed be He, said to Israel: Before I created My world, the ministering angels praised Me through you and sanctified My name through you by saying “Blessed is the Lord God of Israel from everlasting to everlasting.” (1 Chr 16:36)71

Unlike the piyyuṭim discussed here, neither of these rabbinic passages highlights the preexistence of praising angels in its own right; in both of them the presence of angels is tangential to another preexistence theme. The main point of ʾAbot de-Rabbi Nathan (A) 31 is that the Torah was preexistent and itself took part in primordial praise; whereas the second passage, from Tanḥuma Qedoshim 7:2, celebrates the fact that even before creation, it is (only) through Israel that the angels are able to praise God. Interestingly, the biblical verse cited as the basis of this statement, 1 Chr 16:36, opens with the formulaic word Blessed’ is their‘ “) ּבָ ‘רּוְך ’עֲבֹודָתָ ם Blessed”); this brings to mind the line“) ּבָ רּוְך labor”), which describes the function of the preexistent angels in the Avodah .When All Was Not), discussed above) אז באין כל These two passages are briefly discussed by E. E. Urbach.72 Aware of the discrepancy between the explicit repudiation of angelic preexistence in Gen. Rab. 1:3 and the clear implication of angelic preexistence in these passages, Urbach casts doubt on the reliability of the latter. Regarding ʾAbot de-Rabbi Nathan (A) 31, Urbach points to parallel versions of this text in which “the song of the angels is not mentioned” (implying that these offer the preferable phrasing). As to Tanḥuma Qedoshim 7:2,73 he suggests that “‘world’ is used instead of ‘man’; . . . the homilist forgot, in his fervour, that there were no angels before the creation of the world” (italics mine, Y. G.). However in view of the sources surveyed above, it is much likelier that, rather than random slips of a homilist or a scribe, we have here echoes of an ancient, authentic tradition. Indeed, both of these passages preserve the earlier tradition about angelic preexistence

71 Midrash Tanḥuma [S. Buber Recension] (trans. J. T. Townsend; 3 vols.; Hoboken: Ktav, 1989–2003), 2:300–301. 72 E. E. Urbach, The Sages, Their Concepts and Beliefs (trans. I. Abrahams; 2 vols.; 2d enl. ed.; Jerusalem: The Hebrew University Magnes Press, 1987), 2:778 n. 92. 73 The passage is quoted in the footnote but the reference is missing (in the English translation of Urbach’s work as well as in the Hebrew original). No Angels Before the World? 89 expressed in the piyyuṭim discussed above; yet unlike these texts, they attach it to (and subsume it under) a “mainstream” rabbinic preexistence tradition relating to the Torah or to Israel. Interestingly, both these rabbinic statements may have analogues in comparable passages from Second Temple literature. The book of Ben Sira portrays the personified figure of wisdom/Torah as she “sings her own praises”:

In the assembly of the Most High (ἐν ἐκκλησίᾳ ὑψίστου), she opens her mouth; in the presence of his hosts (ἔναντι δυνάμεως αὐτοῦ), she declares her worth. (Sir 24:2)74

Wisdom’s self-praise is staged here among the assembly of the Most High (ἐν ἐκκλησίᾳ ὑψίστου), that is, “the angelic attendants at God’s throne, where Wisdom personified is said to reside.”75 The theme of Wisdom’s preexistence (relying on Prov 8:22–31) is central to the self-praise introduced by these verses (cf. Sir 24:3–5). In that respect (though the scene does not take place before creation but rather in a timeless present), this passage is comparable to the עם מלאכי) description of the preexistent Torah praising God among the angels .in ʾAbot de-Rabbi Nathan (A) 31 (השרת In a different vein, the Prayer of Joseph presents Jacob the patriarch as Israel, an angelic and preexistent figure:

I, Jacob, who am speaking to you, am also Israel (καὶ Ἰσραὴλ), an angel of God (ἄγγελος θεοῦ) and a living spirit. . . . Abraham and Isaac were created before every work. But I, Jacob, whom men call Jacob but whose name is Israel, am he who God called Israel, i.e., a man seeing God, because I am the firstborn of every living thing to whom God gives life (πρωτόγονος παντὸς ζῴου ζωουμένου ὑπὸ θεοῦ).76

This cluster of motifs brings to mind the notion of the angels praising God “through Israel” already before creation, as described in Tanḥuma Qedoshim 7:2; albeit the Prayer of Joseph concerns Jacob/Israel the patriarch/angel

74 The Wisdom of Ben Sira: A New Translation with Notes (trans. P. W. Skehan; introduction and commentary by A. A. Di Lella; AB 39; Garden City, N.Y: Doubleday, 1987), 327. 75 Ibid., 331; the motif of Wisdom in heaven, sitting beside God or residing among the angels, occurs in other Second Temple sources as well; see Kister, “Some Early Jewish and Christian Exegetical Problems,” 586. 76 J. Z. Smith, “The Prayer of Joseph,” in Religions in Antiquity: Essays in Memory of Erwin Ramsdell Goodenough (ed. J. Neusner; SHR 14; Leiden: Brill, 1968), 256. 90 Granat whereas the Tanḥuma passage discusses Israel the nation. It is noteworthy that the Prayer of Joseph attributes preexistence to Abraham and Isaac as well. They are described as “created before any work” (προεκτίσθησαν πρὸ παντὸς ἔργου), although it is implied that Jacob/Israel preceded them as “firstborn” (notwithstanding the biblical genealogical sequence!). This is comparable to Gen. Rab. 1:4, “The intention to create Israel preceded everything else” this statement describes Israel (the nation) as 77;(מחשבתן שלישראל קדמה לכל) preceding all the other preexistent entities enlisted in Gen. Rab. 1:4, including 78.(האבות) ”the patriarchs“

5.4 Preexistent Angels and the Preexistent Throne of Glory Finally, a vestige of the ancient notion of preexistent angels can actually be detected at the core of the standard, “canonical” rabbinic list of seven preexistent entities (e.g., Gen. Rab. 1:4),79 in which God’s throne, the “throne of glory” ,(is always included. In the (relatively late) Midrash Tehillim (90:12 ,(כסא הכבוד) it is stated that

By the loving-kindness of the Holy One, blessed be He, the throne of glory was set upright on the firmament which is above the head of the celestial creatures—the celestial creatures, however, did not yet exist. . . . Only after the Holy One, blessed be He, created His world and created the sacred creatures, did He establish the firmament . . . upon the horns of the creatures.80

This, in all probability, is a “retrospective” attempt to harmonize two seemingly discrepant traditions: the preexistence of God’s throne, on the one hand; and the denial of angelic existence prior to the second day of creation, on the other.81 The very attempt uncovers the fact that the throne of glory as it is normally perceived is inextricably connected with the angelic creatures that carry and surround it. It is not surprising, then, that Second (Syriac Apocalypse of ) Baruch 21:6, describes “the countless holy living creatures that you created from everlasting

77 Theodor–Albeck, Midrash Bereshit Rabba, 6; Midrash Rabbah, 7. 78 Theodor–Albeck, Midrash Bereshit Rabba, 6; Midrash Rabbah, 6. 79 See p. 75 and n. 27 above. 80 W. G. Braude, The Midrash on Psalms, Translated from the Hebrew and Aramaic (2 vols.; Yale Judaica Series 13; New Haven: Yale University Press, 1959), 2:94. 81 Granat, “Before ‘In the Beginning,’ ” 143. No Angels Before the World? 91

[ܡܠܥ ܢܡ] with flame and fire, which stand around your throne.”82 The phrase ܡܠܥ ܢܡ (“from everlasting”) denotes here the primordiality of the holy beings surrounding the divine throne; it most probably echoes Ps 93:2, which sets the notion of the antiquity of God alongside that of the antiquity of His throne: Your throne is established from of old; You are from“ ,נכון כסאך מאז, מעולם אתה in the Peshitta). Since according to its famous ܘܡܢ ܥܠܡ ,מעולם) ”everlasting biblical descriptions (e.g., Isaiah 6; Ezekiel 1), God’s throne is populated all around by various crowds of angelic beings, to describe these beings as created “from everlasting,” following the reference to the antiquity of the throne in Ps 93:2, is indeed to draw an obvious conclusion.83 Psalm 93:2 is frequently cited by the rabbis (e.g., in Gen. Rab. 1:4) as a scriptural proof text for the preexistence of God’s throne. However, in those other rabbinic sources (excluding Midrash Tehillim 90:12) no reference at all is made to the holy living creatures; it is as if their presence is silenced implicitly, in order to avoid incongruity with the repudiation of the idea of angelic preexistence. In the piyyuṭim studied here, on the other hand, the ancient notion of preexistent angels was preserved in a clear, unequivocal form, which enables us to detect its early roots and parallels and to reconstruct the reservations and responses it aroused among the rabbis.

6 Summary and Concluding Observations

Though the question of when exactly the angels came into being might appear rather remote and “scholastic,” the various sources discussed above demonstrate the considerable attention it attracted in late antiquity. This interest was evidently enhanced by the issue’s (potential) implications for significant theological concerns, such as the nature of hierarchy in the heavenly sphere; the distinction between God and his celestial messengers; and the status of these messengers as intermediaries between God on the one hand and the created world on the other.

82 Based (with some changes in the English translation) on D. M. Gurtner, Second Baruch: A Critical Edition of the Syriac Text, with Greek and Latin Fragments, English Translation, Introduction, and Concordances (New York: T&T Clark, 2009), 54–55. 83 And this is most probably also the background of the references to angelic preexistence in the Tanna DeBe Eliyahu passages and the Hekhalot text from the Genizah, discussed above (section 2). 92 Granat

On the one hand, the concept of angels as primordial, preexistent beings is an ancient one, echoed in biblical as well as in Second Temple literature. The very absence of the angels from the creation account that opens the Bible may itself be interpreted as reflecting the theological sensitivity of this matter. On the other hand, the explicit reference to the angels’ creation in Jubilees’ paraphrase of the biblical account draws a clear distinction between God, as the sole creator, and the angels as among several constituents of the universe created by God on the first day. The prevailing rabbinic view, which delays the creation of angels to the second day, further accentuates the distinction between God and the angels; according to Gen. Rab. 1:3, this view is explicitly directed against the notion of the angels as active agents at the beginning of the creation process, a conception that was highlighted and developed in Gnostic circles. Early Christian writers, whose theological sensitivities obviously differed from those of the rabbis, combatted similar gnostic views by denying the angels a role in creation itself; even while locating their creation on the first day or earlier. Against this backdrop, it is therefore quite interesting that certain early piyyuṭim also present the angels as existing before God commenced his creation of the world proper. Of particular significance is the fact that the activity of these angels is circumscribed by and limited to that of praising and glorifying God; in fact, that is the reason for their creation. Praising God is presented as a role of the angels in diverse biblical and postbiblical sources, but in our piyyuṭim this ancient, widespread tradition is distinctively appropriated through the straightforward association of the angels’ praise with their preexistence. In contrast to Gen. Rab. 1:3, and in a different way than the passages from ʾAbot de-Rabbi Nathan and Tanḥuma discussed above, these piyyuṭim also defuse the potential threat presented by the preexistent angels tradition—but by adapting and adopting this tradition, rather than subsuming or refuting it. It should again be emphasized that the piyyuṭim discussed here are rather atypical in their deviation from the common rabbinic view normally followed in early Piyyuṭ creation accounts, which refer to the angels as created on the second day. However, these unusual texts accentuate the fact that early Piyyuṭ does not always merely restate standard rabbinic material, and that the instances in which it deviates from adherence to “mainstream” rabbinic traditions deserve special scholarly attention. This study thus illustrates well the potential contribution of early Piyyuṭ texts to mapping the transmission and transformation of ancient traditions in the Jewish culture of late antiquity. Pious Long-Sleepers in Greek, Jewish, and Christian Antiquity

Pieter W. van der Horst

1 Introduction

In the middle of the thirteenth century, the Dominican monk James de Voragine (Iacopo da Varazze, c. 1230–1298) compiled his famous Legenda aurea (the “Golden Legend”). This immensely influential work, of which almost a thousand medieval Latin manuscripts survive and which was translated into many vernaculars, consisted of a collection of saints’ lives and short treatises on the Christian festivals, in 175 chapters. In ch. 24, James tells us the famous story of the Seven Sleepers of Ephesus.1 Briefly summarized it runs as follows: During the persecution of Christians by the emperor Decius (ca. 250 CE), seven pious young men took refuge in a cave near Ephesus, where they fell asleep and were walled up by Decius. When they woke up, they initially thought that they had slept only for a short time and sent one of their number, Iamblichus, to the market to get some food. But as he came into the city, everything appeared strange to him: the buildings were different, Jesus Christ was being talked about freely by the people, and crosses were inscribed on all the city gates. He couldn’t believe that this was his Ephesus. Finally he realized that it was 372 years after they had fled: Theodosius, not Decius, was now the Emperor.2 The appearance of the seven young men became the occasion for great ecclesiastical festivities, in which the Emperor also participated. All who saw the young men thanked God for the miracle. The cave became a much- visited pilgrim site for many centuries.3

* A version of this paper now appears in the volume of my collected essays, Studies in Ancient Judaism and Early Christianity (AJEC 87; Leiden: Brill, 2014). 1 R. J. Pillinger, “Siebenschläfer,” RGG 7:1306. The literature on this subject is vast. 2 It is curious that the span of time is explicitly given as 372 years (which would set the story in about 622 CE). The other narrative details clearly locate the story in the reign of the Emperor Theodosius, either I or II (379–395 and 408–450 CE respectively). The number goes back to Jacob of Sarug, but is not in the Latin synopsis of Gregory of Tours (see below). 3 An English translation is that of W. G. Ryan, Jacobus de Voragine: The Golden Legend (2 vols.; Princeton: Princeton University Press, 1993). The Latin text of the Seven Sleepers legend is easily accessible (with a German translation) in R. Nickel, Jacobus de Voragine: Legenda aurea

When James of Voragine penned this legend of the miraculously long sleep of pious persons, the story, or rather such stories, had already had a long prehistory of more than one-and-a-half thousand years. We find it in pagan Greek, Jewish, Christian, and Muslim versions. Due to the enormous influence of the Legenda aurea, the story became widely known in medieval Europe. It is the purpose of my paper, however, to trace this motif—of persons falling asleep and finding the world completely changed when they wake up—from the Legenda aurea back into time as far as we can get, and to try to reconstruct its Werdegang.

2 Graeco–Roman Sources

The early third-century CE account by Diogenes Laertius of the fifty-seven-year sleep of Epimenides (1.109; we will come back to this text) is the best known and most often quoted Greek witness to this motif. However, as the author of the first major scholarly monograph on this subject,4 John Koch, already observed as far back as 130 years ago, it is Aristotle who is the first to allude briefly to stories about long-sleepers. In his Physics, he engages in a very subtle discussion of what exactly time is. In that context he says,

Time does not exist without change, for when the state of our mind does not change at all, or we have not noticed its changing, we do not think that time has elapsed any more than those who are fabled to sleep among the heroes of Sardinia do when they are awakened; for they connect the earlier “now” with the later and make them one, cutting out the interval because of their failure to notice it. (Physics 4.11, 218b23–26)

Unsatisfactory though this remark may be for us because of its tantaliz- ing briefness—we want to know the precise contents of this legend5—it

(Stuttgart: Reclam, 1988), 250–63. On the cave and the basilica that was later built on the site, see H. Leclercq, “Sept dormants d’Éphèse,” DACL 15.1:1251–62. 4 J. Koch, Die Siebenschläferlegende, ihr Ursprung und ihre Verbreitung: Eine mythologisch- literaturgeschichtliche Studie (Leipzig: Reissner, 1883). In spite of the fact that the author overlooked some of the ancient sources, especially Jewish ones, and offers some hard-to-follow explications of the evidence, this study is still unsurpassed as the first critical survey of the ancient evidence. This book was recently offered for reprint in the United States (Charleston, S.C.: Nabu Press, [print on demand: ISBN 9781148395272]). 5 As H. Wagner remarks, it is impossible to know in which form Aristotle knew the legend of the Sardinian long-sleepers; see Wagner, Aristoteles: Physikvorlesung (Darmstadt: Pious Long-Sleepers 95 is important for our purposes in that it clearly shows that by the end of the fourth century BCE, stories were being circulated about people who slept long enough (apparently!) to be useful for Aristotle’s argument about the impos- sibility of the passage of time without the occurrence of change. He is evidently not talking about regular sleep here. And, as we shall presently see, the fact that the story about Epimenides in Diogenes Laertius goes back to much earlier sources makes it very probable that Greek stories about long-sleepers existed already in pre-Christian, possibly even pre-Hellenistic times. Although Diogenes only says vaguely that the sources for his chapters on Epimenides were “Theopompus and many other writers” (1.109), it may be taken as a fact that most of his sources were from the Hellenistic period (Theopompus lived in the fourth century BCE).6 What he tells us about Epimenides, who supposedly lived in the decades around 600 BCE, is the following:

[Epimenides] was a native of Cnossos in Crete, although, because of his long hair, he did not look like a Cretan. One day he was sent into the country by his father to look for a stray sheep, and at noon he turned aside out of the way and went to sleep in a cave, where he slept for fifty- seven years. After this he got up and went in search of the sheep, thinking he had been asleep only for a short time. And when he could not find it, he came to the farm and found everything changed and another owner in possession. Then he went back to the town in utter perplexity; and there, on entering his own house, he fell in with people who wanted to know who he was. At length he found his younger brother, now an old man, and learnt the truth from him. So he became famous throughout Greece and was believed to be especially loved by the gods (theophilestatos).7

Wissenschaftliche Buchgesellschaft, 1967; repr. 1972), 571. The commentator Philoponus, however, describes the myth as follows: “certain sick people went to the heroes in Sardinia and were treated, and slept for five days, of which they had no recollection when they awoke” (thus W. D. Ross, Aristotle’s Physics [Oxford: Clarendon, 1936], 597, with reference to Simplicius, Comm. in Arist. 4.338b). See also E. Rohde, “Sardinische Sage von den Neunschläfern,” in his Kleine Schriften (2 vols.; Tübingen: Mohr, 1901), 2:197–208; M. Huber, Die Wanderlegende von den Siebenschläfern: Eine literarische Untersuchung (Leipzig: Harassowitz, 1910), 384–87; J. H. Waszink, Tertulliani de anima (Amsterdam: Meulenhof, 1947), 516. 6 On the sources for our knowledge of Epimenides, see H. Diels and W. Kranz, Die Fragmente der Vorsokratiker (3 vols.; 6th ed.; Berlin: Weidmann, 1951 [repr. 1996]), 1:27–37 (no. 3). Our text is labeled 3A1 D–K. Diogenes’ latest source was probably Phlegon of Tralles, of the early second century CE. 7 Lives 1.109 (Diogenes Laertius, Lives of Eminent Philosophers [trans. R. D. Hicks; 2 vols.; LCL; Cambridge, Mass.: Harvard University Press; London: Heinemann, 1972], 1:115). 96 van der Horst

The final word, theophilestatos, is very important. It indicates that Epimenides’ ability to sleep for an extremely long time and survive without food and drink was a special divine favor, a motif that will recur time and again, as we shall see. That Diogenes did not invent the story of Epimenides’ long sleep is proved by the fact that half a century earlier, the geographer Pausanias states very briefly in passing that “people say” (legousin) that Epimenides slept for forty years in a cave (Descr. 1.14.4).8 Thus, the motif is older than Pausanias and most probably dates back to the Hellenistic period. We can be rather sure of this because in the middle of the first century CE, Pliny the Elder states that he learned of a tradition concerning Epimenides’ long sleep. He regards it as a fabulous invention (fabulositas); but he nevertheless reports that Epimenides, “when a boy, being weary with the heat and with travel, slept in a cave for fifty-seven years, and when he woke up, just as if it had been on the following day, was surprised at the appearance of things and the change in them.”9 The origins of the motif remain shrouded in darkness, but it is not completely incomprehensible why it should have been attached to Epimenides. Like the semi-legendary and mysterious Greek traveler, poet, and miracle-worker Aristeas of Proconnesus,10 Epimenides, a poet, a holy man, supposed to have been called in to purify Athens after a sacrilegious event, is a very shadowy figure of the late seventh century BCE. The traditions about him were “quickly obscured by legends and miraculous tales”11 concerning his out-of-body experiences, his oracular capacities, his extreme old age (157 or 299 years), his amazing asceticism, his purifying activities, etc.12 Since tradition assimilated him to the type of a shaman, the story about his long sleep at the beginning of his saga suggests that “the Greeks had heard of the long ‘retreat’ which is the shaman’s novitiate

8 At roughly the same time, Maximus of Tyre alludes to “a tale hard to credit if taken at face value . . . that he [Epimenides] had lain for many years in a deep sleep in the cave of Dictaean Zeus” (Or. 10.1; cf. 38.3); translated in Maximus of Tyre: The Philosophical Orations (trans. and ed. M. B. Trapp; Oxford: Clarendon, 1997), 85. 9 Nat. hist. 7.175; (Pliny, Natural History [trans. H. Rackham; 10 vols.; LCL; Cambridge, Mass.: Harvard University Press; London: Heinemann, 1938–1962], 2[1942]:623). See the note in R. König, Plinius Secundus d. Ä.: Naturkunde, Buch VII (Zürich: Artemis, 1996), 223. 10 See J. D. P. Bolton, Aristeas of Proconnesus (Oxford: Oxford University Press, 1962). On p. 156, Bolton calls Epimenides “the prince of cataleptics.” 11 A. H. Griffiths, “Epimenides,” OCD 546. 12 See E. Rohde, Psyche: Seelencult und Unsterblichkeitsglaube der Griechen (2 vols.; 2d ed.; Freiburg: Mohr, 1898; repr: Darmstadt: Wissenschaftliche Buchgesellschaft, 1961), 2:96–99; E. R. Dodds, The Greeks and the Irrational (Berkeley: University of California Press, 1951), 141–46. Pious Long-Sleepers 97 and is sometimes largely spent in a condition of sleep or trance.”13 Be that as it may, for our purposes it suffices to establish that the first time the motif of an excessively long sleep surfaces in classical sources, it is in connection with a person who lives in close contact with the supernatural world and is apparently favored by the divinities who dwell there.

3 Jewish Sources

We see this theme again when we turn to the Jewish material. The earliest occurrence of our motif is in the so-called Paralipomena Ieremiae (or 4 Baruch), a text that most scholars agree was written in the early decades of the second century CE.14 Here, the long-sleeper in question is “Abimelech the Ethiopian,” portrayed as a sort of servant of Jeremiah. The Lord promises Jeremiah, before the destruction of Jerusalem, to protect and save Abimelech “until I bring back the people to the city” (3:11). Jeremiah sends Abimelech out of the city with the directive to “take a basket and go to the estate of Agrippa by the mountain trail; bring a few figs in it and give them to the sick among the people” (3:15). Abimelech does as he is told, but in the meantime Jerusalem is destroyed by the Chaldeans. What then occurs deserves to be quoted in full, since it is the pivotal text for our purposes.15

Ch. 5 (1) Abimelech carried the figs in the heat of the day; and coming upon a tree, he sat down in its shade to rest a while. And leaning his head on the basket of figs, he fell asleep and slept for sixty-six years, and he was not awakened from his sleep. (2) After these things he awoke from his sleep and said, “I would gladly have slept a little longer; my head is heavy

13 Dodds, ibid., 142, with references in n. 46 to literature on the lengthy sleep of the shamans. There he also notes that Diels ingeniously thought that the long sleep of Epimenides was invented to reconcile the chronological inconsistencies in the various tales about Epimenides, but Dodds rightly remarks that if that were the case, lengthy sleep should have been a very common motif in early Greek history. 14 See, e.g., J. Herzer, 4 Baruch (Paraleipomena Jeremiou): Translated with an Introduction and Notes (SBLWGRW 22; Atlanta: Society of Biblical Literature, 2005), xxx–xxxvi. The very close relationship of Par. Jer. to the Syriac Apocalypse of Baruch (2 Baruch) also supports an early second century CE dating; see Herzer, 4 Baruch, xviii; and B. Schaller, Paralipomena Jeremiou (JSHRZ 1.8; Gütersloh: Gütersloher Verlagshaus, 1998), 672, 678–81. 15 The translation is from Herzer, 4 Baruch, 13–19, with some slight modifications. See also the translation by S. E. Robinson, “4 Baruch,” in OTP 2:413–25. There is a good bibliography in Schaller, Paralipomena Jeremiou, 696–710. 98 van der Horst

because I did not get enough sleep.” (3) And when he uncovered the basket of figs, he found them dripping with their milky sap. (4) And he said, “I want to sleep a little (more) because my head is heavy. (5) But I am afraid that I might fall asleep again and wake up too late and Jeremiah, my father, would have a low opinion of me. For if he were not in a hurry, he would not have sent me today at dawn. (6) So I will get up and proceed in the heat and go to where there is neither heat nor toil every day.” (7) So he got up, took the basket of figs and placed it on his shoulders. And he entered Jerusalem, but he did not recognize it, neither the house nor the place nor his own family, and he said, (8) “Blessed be the Lord, for a great trance has come upon me: This is not the city. (9) I lost my way because I came by the mountain trail when I awakened from my sleep. (10) And since my head was heavy because I did not get enough sleep, I lost my way. (11) This is an astonishing thing to say to Jeremiah, ‘I lost my way!’ ” (12) And he went out of the city and when he looked carefully, he saw the landmarks of the city and said, “Indeed, this is the city, but I lost my way.” (13) And again he went back into the city and searched, but he found no one of his own people. (14) And he said, “Blessed be the Lord, for a great trance has come upon me.” (15) And again he went out of the city, and he remained there grieving, for he did not know where to go. (16) And he laid down the basket, saying, “I shall sit here until the Lord lifts this trance from me.” (17) And while he was sitting, he saw an old man coming from the field. And Abimelech said to him, “I say to you, old man, what city is this?” And he said to him, “It is Jerusalem.” (18) And Abimelech said to him, “Where is Jeremiah the priest, and Baruch the reader, and all the (other) people of this city? For I could not find them.” (19) And the old man said to him, “You are from this city, aren’t you? (20) You just remem- bered Jeremiah, seeing that you are asking about him after such a long time. (21) For Jeremiah is in Babylon with the people, for they were taken captive by King Nebuchadnezzar, and Jeremiah is with them to announce to them the good news and to teach them the word.” (22) As soon as Abimelech heard this from the old man, he said, (23) “If you were not an old man, and if it were not improper for a person to upbraid one older than oneself, I would laugh at you and say that you are crazy because you say, ‘The people have been taken captive to Babylon.’ (24) Had the heavenly torrents descended to them, there would not yet have been time to go to Babylon. (25) For how long has it been since my father Jeremiah sent me to the estate of Agrippa for a few figs so that I might give them to the sick among the people? (26) And I went and brought them, and when I came upon a tree in the scorching heat of the day, I set down to rest a Pious Long-Sleepers 99

little and leaned my head on the basket and fell asleep. And when I awoke I uncovered the basket of figs supposing that I was late, and I found the figs dripping with their milky sap, just as I had picked them. And then you say that the people have been taken captive to Babylon? (27) But that you might know, take the figs and see!” (28) And he uncovered the basket of figs for the old man. (29) And he saw them dripping with their milky sap. (30) And when he saw them, the old man said, “O my son, you are a righteous man and God did not want to show you the desolation of the city, so God brought this trance upon you. Behold, it has been sixty-six years today since the people were taken captive to Babylon. (31) But that you may learn, child, that it is true, look at the field and see that the growth of the crops has just begun. Notice also the figs, that their time has not yet come, and understand.” (32) Then Abimelech cried out in a loud voice, saying, “I will bless you, O Lord, God of heaven and earth, the rest of the souls of the righteous in every place.” (33) And to the old man he said, “What month is this?” And he said, “Nisan, and it is the twelfth day.” (34) And taking a few of the figs, he gave them to the old man and said to him, “God will lead you by his light to the city above, Jerusalem.”

A number of features of this account call for special consideration. Firstly, Abimelech16 is a biblical name, but our Abimelech does not have anything to do with the biblical persons called by that name (see Genesis 20–21; 26; Judges 9). There is no doubt that this Abimelech is to be identified with the biblical Ebed-Melech,17 the Ethiopian courtier who saved Jeremiah’s life and received God’s promise, via the prophet, that he himself would be saved during the destruction of Jerusalem (see Jer 38:7–13 and 39:15–18). Curiously enough, in the Bible God promises Ebed-Melech that even though he is to be rescued, God will fulfill his words against the city in Ebed-Melech’s presence, literally, “before your face” (lephanêkha [Jer 39:16]);18 the text of Par. Jer. 5, however, has

16 Unfortunately, I have not been able to consult J. Riaud, “Abimélech, personnage-clé des Paralipomena Jeremiae?” Dialogues d’histoire ancienne 7 (1981): 163–78. 17 The LXX renders the name as Abdemelech (only minuscule 534 has Abimelech); see J. Ziegler, Ieremias, Baruch, Threni, Epistula Ieremiae (Septuaginta 15; Göttingen: Vandenhoeck & Ruprecht, 1957), 406. 18 The sentence with lephanêkha is omitted in the LXX; see J. Herzer, Die Paralipomena Jeremiae (TSAJ 43; Tübingen: Mohr Siebeck, 1994), 91 n. 257. 100 van der Horst drastically changed that promise into a rescue scene in which Abimelech does not have to witness anything of the Destruction.19 Secondly, while the stories of Epimenides and Abimelech overlap in the striking motif of supernaturally long sleep and the unrecognizable world in which they find themselves after their awakening, almost everything else is different. But these differences (e.g., the nondramatic setting of the Epimenides story over against the highly dramatic setting of the Abimelech story) should not make us overlook the fact that in both cases the long sleep is regarded as a divine gift, as a sign of favor on the part of heaven. Thirdly—and now we come to the main problem—the Abimelech story has been presented so far as our first Jewish specimen of the motif of long-sleepers; but is it? Or rather, is it Jewish at all? Let us face the facts: The Paralipomena Jeremiae as we have it is a Christian text that has been handed down to us via Christian channels. Like all or most other Jewish pseudepigrapha, our text, too, has undergone Christian editing and redaction. How sure can we be that the Abimelech story does not derive from a Christian hand? This question becomes all the more pressing when we see that there seems to be undeniably Christian usage in this passage; e.g., in 5:21, where it says that Jeremiah is in Babylon with the people “in order to announce to them the good news (literally, to preach the Gospel, euangelisasthai), and to teach them the Word” (katêchêsai ton logon). And this is exactly what happens in ch. 9, where Jeremiah preaches the Gospel of Jesus Christ before he is stoned to death. Is this not enough to disqualify our story as a source for Jewish ideas about long-sleepers?20 Since we need to answer this question before we can tackle the problem of a possible influence of the Greek Epimenides (or a similar) story on a Jewish tradition, we will first briefly have to survey the evidence. We have a longer and a shorter version of Paralipomena Jeremiae. Of both versions we possess many dozens of manuscripts, in Greek, Coptic, Ethiopic, Armenian, and Old Church Slavonic.21 All scholars agree that all of these manuscripts are of Christian provenance. However, most students of the text regard only the final chapter on Jeremiah’s performance as a preacher of the Gospel as a patently Christian addition (and perhaps some other phrases elsewhere in

19 For the means by which this motif found its way into later tradition in 3 Baruch 1:1 and elsewhere see A. Kulik, 3 Baruch: Greek-Slavonic Apocalypse of Baruch (CEJL; Berlin: De Gruyter, 2010), 98–99. 20 Note that P. Bogaert regards the entire Abimelech episode as a Christian interpolation; see his L’Apocalypse syriaque de Baruch (2 vols.; SC 144, 145; Paris: Cerf, 1969), 1:192–95. 21 For a good survey see, inter alios, A.-M. Denis et al., Introduction à la litterature religieuse judéo-hellénistique (2 vols.; Turnhout: Brepols, 2000), 1:681–718. Pious Long-Sleepers 101 the work as well); they regard the main body of the work as definitely Jewish. For instance, Michael Stone stresses that the work is undoubtedly part of the wider Jewish Baruch and Jeremiah literature represented also by the Greek Book of Baruch, the Epistle of Jeremiah, and the Syriac and Greek Apocalypses of Baruch, as well as fragments from Qumran (4Q384–385).22 Stone adds that the Jewish nature of the original is apparent from many distinctive features: “Thus the approval of sacrifice, the rejection of foreign women, and the attitude to circumcision, to mention only the most prominent, clearly disprove the theory of a Christian original.”23 This is convincing to my mind. Even though there is good reason to regard the passage in the Abimelech story about Jeremiah as a preacher of the Gospel (5.21) as a Christian interpolation, there is equally good reason in this case not to regard the story as a whole as a Christian addition to a Jewish Grundschrift.24 Abimelech’s long sleep does not serve any Christian purpose; and even though we do know of some ancient Christian writings that do not seem to serve such a purpose,25 in this case we have no reason at all to think that the story is such an instance. The strong emphasis on typically Jewish halakhic elements in the Paralipomena Jeremiae as a whole makes it an unlikely candidate for composition as Christian propaganda. As a Christian interpolation, the Abimelech story would simply be pointless. For that reason we must assume that it is from the pen of the Jewish author. Perhaps the fact that none of the Church Fathers makes any reference to the Paralipomena Jeremiae26 may be seen as corroborative evidence. Is there any plausibility to the suggestion that the Jewish author here has taken over a Greek motif? Even though in the cases of both Epimenides and

22 The originally Jewish Jeremiah Apocryphon that has been preserved only in Coptic should also be mentioned here; see for text and translation K. H. Kuhn, “A Coptic Jeremiah Apocryphon,” Le Muséon 83 (1970): 95–135 and 291–326. The third- or fourth-century text, translated from a now lost Greek original, seems to have been dependent on Par. Jer. Abimelech’s miraculous sleep is narrated in three stages in the Coptic work (chs. 12, 22, 38–39). 23 M. E. Stone, “Baruch, Rest of the Words of,” EncJud 4:276. Further arguments for a Jewish provenance may be found in Schaller, Paralipomena Jeremiou, 677–78, who concludes: “Die ParJer sind, wenn nicht alles täuscht—abgesehen von dem Schlußkapitel—im wes- entlichen ein Text genuin jüdischer Herkunft und Prägung.” 24 The situation is very different from the case of, e.g., the Testaments of the Twelve Patriarchs; see M. de Jonge, Pseudepigrapha of the Old Testament as Part of Christian Literature (SVTP 18; Leiden: Brill, 2003), 71–177. 25 See J. R. Davila, The Provenance of the Pseudepigrapha: Jewish, Christian, or Other? (JSJSup 105; Leiden: Brill, 2005). 26 See Denis, Introduction, 1:690. 102 van der Horst

Abimelech, we meet men who are described as especially favored by a deity, there is no reason to think in terms of Greek influence. Stories of long-sleepers are known from a wide variety of cultures27 and it is reasonable to think that the motif could have sprung up independently in any setting. And that someone who is able to sleep for a miraculously long time can be enabled to do so only by a deity is something that any person in an ancient culture could think up. Moreover, as we shall see, there are a few other Jewish stories about long- sleepers that do not give us much reason to surmise Greek influence. The texts in question concern the long sleep of Ḥoni the Circle-Drawer (Ḥoni ha-meʿaggel) and/or his grandfather.28 There are two different stories about Ḥoni as a long-sleeper, each of which occurs in two not too different versions. The earliest attestation of one of these stories is found in y. Taʿan. 3:9 (66d). There we read that Ḥoni’s grandfather (or he himself)29 lived shortly before the destruction of the (first!) Temple. He went out to a mountain with his workmen, and when it began to rain, he went into a cave where he fell asleep. He remained asleep for seventy years, during which the Temple was destroyed and rebuilt for the second time. When after seventy years he awoke and left the cave, he found out that the world had completely changed. What had been vineyards now were olive groves, and where there had been olive groves there now grew other plants.

He asked the people, “What is going on in the world?” And they said, “You don’t know what is going on in the world?” He said, “No.” They asked him, “Who are you?” And he said, “Ḥoni, the Circle-Drawer.” Thereupon they said, “We have heard that whenever you entered the Temple court, it would be illuminated.” Then he entered (the Temple court), and it was illuminated. Thereupon he applied to himself the following verse, “When the Lord brought back those who returned to Zion, we were like those who dream.” (Ps 126:1) [paraphrase].

27 See the impressive survey in Huber, Wanderlegende. Many references to the motif in other cultures are also noted in S. Thompson, Motif-Index of Folk-Literature (rev. and enl. ed.; 6 vols.: Bloomington, Ind.: Indiana University Press, 1955–1958), motif D1960 (vol.2). 28 See J. L. Rubenstein, Stories of the Babylonian Talmud (Baltimore: Johns Hopkins University Press, 2010), 62–76 (in n. 1 he mentions older secondary literature). On pp. 74–76, Rubenstein discusses some important textual variants to the talmudic stories, but these do not affect my argument. 29 I will not discuss the question of whether the stories concern Ḥoni or his grandfather, for it is apparent that the transfer to the grandfather is only a later development intended to solve chronological problems in the story, to no avail. Pious Long-Sleepers 103

Whatever the exact reading and translation of the original words of Ps 126:1,30 the point of the quote is clear: Ḥoni interprets the psalm as referring to the returnees from the Babylonian exile and applies the words “those who dream” to himself; if he has been dreaming, he must have slept long enough to see that “the Lord brought back those who returned to Zion”; i.e., for seventy years (see Jer 25:11; 29:10). That Ḥoni did not live during the Babylonian exile but five centuries later apparently did not matter.31 The other story about Ḥoni as a long-sleeper is told in b. Taʿan. 23a, and may be paraphrased as follows:

R. Yoḥanan said: “The righteous Ḥoni was troubled all of his life about the meaning of Ps 126:1. He said, ‘Is it possible for a man to lie dreaming for seventy years?’ One day, when he was walking on the road, he saw a man planting a carob tree and asked him, ‘How long does it take for this tree to bear fruit?’ The man answered, “Seventy years.’ Ḥoni then asked, ‘Are you sure that you will live for seventy years more?’ The man replied, ‘I found [full-grown] carob trees already planted in the earth, and so, like my forefathers planted these for me, I, too, plant these for my children.’ Then he sat down to eat and sleep overcame him. As he slept, a rocky hedge enclosed him and hid him from sight. He continued to sleep for seventy years. When he awoke, he saw a man gathering fruit from the carob tree and he asked him, ‘Are you the man who planted this tree?’ The man replied, ‘I am his grandson.’ Thereupon Ḥoni said, ‘Now it is certain that I slept for seventy years!’ ”

The story then continues by relating how Ḥoni went to his home, but nobody recognized him or believed that he was Ḥoni Ha-Meʿaggel; and the same happened again when he went to the Beit Ha-Midrash: he was not recognized and consequently not given the honor due to him. This hurt him so deeply that he asked God to let him die, and he died: “Thus people say, ‘Either fellowship or death.’” It is clear that here the motif of a sleep of seventy years is pinned to a specific exegesis of Ps 126:1: The Lord has caused his people to return from exile to Zion, and we know from Jeremiah that the exile lasted seventy years. Those who see the returnees feel like they must have been dreaming; thus it

30 See the discussion in H.-J. Kraus, Psalmen (BKAT 15.2; Neukirchen: Neukirchener Verlag, 1972), 853–55. 31 See Huber, Wanderlegende, 418: “Vor dem Anachronismus schreckte die Sage eben nicht zurück.” 104 van der Horst follows—according to this exegesis—that they must have been dreaming, i.e., sleeping, for seventy years. So the motif of a miraculously long sleep here receives a biblical basis. That Ḥoni is here the one who receives this divine gift is not remarkable in view of the fact that in the traditions about him, Ḥoni is always regarded as someone who was God’s favorite (theophilês). As is well- known, his prayers were famous for being extremely powerful and effective.32 Now one might argue, of course, that the text of Psalm 126 does not necessarily imply that it is possible to sleep for seventy years, and that is correct. So it might be the case that the exegetical quandary (what does “those who dream” refer to?) was thought by the rabbis to be solvable by recourse to a motif from Greek legends about divinely favored long-sleepers. But that is hard to prove. It is not certain in what period this legend about Ḥoni arose. The historical figure of Ḥoni most probably lived in the first half of the first century BCE, and it is not improbable that legends about him began to flourish soon after his death. But the legend of his long sleep probably does not predate the destruction of the Temple in 70 CE, and it is apparently not known to Josephus, who wrote in the quarter of a century thereafter.33 We can only guess that the Abimelech legend and the Ḥoni legends developed at more or less the same time, around the turn of the first to the second century CE. Whether or not they developed independently of one another is hard to say. Maybe the story began with Ḥoni as the protagonist and was later corrected by making the more probable figure of Abimelech the prime actor, in order to avoid blatant anachronism. Or maybe originally Abimelech was the protagonist and Ḥoni attracted the story as a magnet because of his reputation for miracles. We do not know. There are both major differences and major agreements between the talmudic stories and that of 4 Baruch 5. Let us briefly compare the three accounts.34 In both the Palestinian Talmud and 4 Baruch the setting is the destruction of the Jerusalem Temple, while in the Babylonian Talmud it is not. All three stories, however, feature prominently the motif that the world has completely changed, albeit more so the Palestinian than the Babylonian account. Both y. Taʿaniyot and 4 Baruch stress that the sleeper was convinced that he had slept only briefly. While in y. Taʿaniyot there is a mountain with a cave, there is no cave in 4 Baruch (although a mountain is mentioned in 3:10 and 5:9); neither is present in the Babylonian version. In the rabbinic stories, Psalm 126 plays an important

32 For references see A. Büchler, Types of Palestinian Jewish Piety from 70 B.C.E. to 70 C.E.: The Ancient Pious Men (London: Oxford University Press, 1922: repr.: New York: Ktav, 1968), 196–264; G. Vermes, Jesus the Jew (London: Collins, 1973), 69–72. 33 Josephus writes about Ḥoni (in his version called Onias) in Ant. 14.22–24. 34 See also the remarks by G. B. Sarfatti, “Pious Men, Men of Deeds and the Early Prophets,” Tarbiz 26 (1956/57): 126–53 (149–53) (in Hebrew). Pious Long-Sleepers 105 role—in y. Taʿaniyot it is featured only at the end, while in b. Taʿanit it is the point of departure; but in 4 Baruch it is lacking (at least explicitly; the combined motifs of exile, return, sleeping/dreaming, 66/70 years might suggest, however, that some form of exegesis of this psalm lurks in the background). Only in the Babylonian story does Ḥoni die of grief.35 On the whole, the Palestinian Talmud’s narrative is much closer to 4 Baruch than is the Babylonian Talmud’s, but the agreements do not warrant a judgment of literary dependence. That there certainly is a common dependence on a tradition of Jewish stories that were created after the fall of the Temple in 70 CE is, however, obvious, and that a miraculously long sleep during and after the Destruction was part of that tradition is beyond doubt. Pierre Bogaert sees it as follows:

L’auteur des Par. Jer. avait besoign de souligner le long intervalle de temps qui sépare la prise de Jérusalem de sa reconstruction. Ne pouvant introduire Honi dans son recit sous peine de commettre un anachronisme grossier, il a remplacé ce personage par un contemporain de Jérémie et de Baruch, Abimélech.36

That sounds plausible. An ancient story about Ḥoni the miracle worker is taken into service for the explanation of Ps 126:1, which is interpreted as referring to the return of the exiles after seventy years, but that makes Ḥoni end up in the sixth century BCE. In the Paralipomena Ieremiou the more plausible figure of Abimelech, who did live in that century, replaces Ḥoni in order to solve this problem. But admittedly, there is no way to reach any certainty in this matter. In any case, it is clear that the Palestinian Talmud’s story of Ḥoni’s long sleep has a much closer relationship to the Abimelech narrative than does the Epimenides story. Jens Herzer sees y. Taʿaniyot’s story as “an intermediary step linking the two [i.e., Greek and Jewish] traditions.”37 He states,

By comparing the three versions of the motif, it is possible to identify the process by which the narrative was revised to conform to the individual interests of the authors of y. Taʿan. 3:9 and 4 Baruch. In the Yerushalmi version of the tradition, for example, Epimenides’ fifty-seven years is lengthened to the seventy years of the exile, a length again changed by the writer of 4 Baruch to sixty-six years. . . . Thus one can follow an interesting process of reworking a tradition that also provides evidence

35 For a more detailed comparison of the three stories see J. Herzer, 4 Baruch, 86–87; also Huber, Wanderlegende, 418–26; and Sarfatti, “Pious Men.” 36 Bogaert, L’Apocalypse syriaque, 197–98. 37 Herzer, 4 Baruch, 88. 106 van der Horst

for knowledge of Greek classical traditions and their reuse in Jewish circles.38

This is probably a bit too speculative, but an attractive speculation it is.

4 Christian Sources

We now turn to the Christian material.39 I will deal only with the two earliest witnesses to the story of the Seven Sleepers of Ephesus, from the early and late sixth century, respectively.40 The first author to tell us the story of the Seven Sleepers is the Syrian bishop Jacob of Sarug (ca. 450–521),41 although he bases himself upon an older source. The other early witness, Gregory of Tours (ca. 538–594),42 states that his own knowledge of the story comes from a written source, perhaps in Syriac.43 Whether the story originated in Syriac- or

38 Herzer, 4 Baruch, 88. In n. 34 Herzer draws attention to the fascinating fact that interest in Epimenides among Christian circles of the late first century CE is evident from the famous quotations in Acts 17:28 and Titus 1:12. 39 By far the most extensive survey (though somewhat outdated) is still Huber, Wanderlegende (1910). 40 I mention only briefly in passing the travelogue De situ terrae sanctae of Theodosius (about 525 CE), which mentions that in Ephesus there are septem fratres dormientes (the text is found in Itineraria et alia geographica [ed. P. Geyer and O. Cuntz; CCSL 175; Turnhout: Brepols, 1965], 123). See the comments by H. Donner, Pilgerfahrt ins Heilige Land: Die ältesten Berichte christlicher Palästinapilger (1.–7. Jahrhundert) (Stuttgart: Katholisches Bibelwerk, 1980), 220 n. 96. On the reason why the seven sleepers are called “brothers” here see Koch, Siebenschläferlegende, 85. On the possibility that this passage in Theodosius is a later interpolation see E. Honigmann, “Stephen of Ephesus (April 15, 448–Oct. 29, 451) and the Legend of the Seven Sleepers,” in his Patristic Studies (Studi e Testi 173; Vatican City: Bibliotheca Apostolica Vaticana, 1953), 125–68 (135). According to some scholars, the earliest witness to the story of the Seven Sleepers is the source used by Pseudo-Zachariah Rhetor in his Chronicle (ca. 570); see G. Greatrex et al., eds., The Chronicle of Pseudo-Zachariah Rhetor (Translated Texts for Historians 55; Liverpool: Liverpool University Press, 2011), 49 and 83–84. 41 See F. Rilliet, “Jakob von Sarug,” RAC 16 (1994): 1217–27. 42 See B. K. Vollmann, “Gregor IV (Gregor von Tours),” RAC 12 (1983): 895–930. 43 Actually, Gregory says (in the final line of his De gloria martyrum 95) that this is a story “quam Syro quodam interpretante in Latinum transtulimus.” This does not necessarily imply that the story itself was in Syriac. The Syrian might have translated it for Gregory from Greek. A strong case for a Greek original was made by P. Peeters, “Le texte original de la Passion des Septs Dormants,” AnBoll 41 (1923): 369–85. That Gregory would have known a Syrian interpreter is not strange—there were many contacts between Syria and Western Europe in his time; see Huber, Wanderlegende, 371–76. Pious Long-Sleepers 107

Greek-speaking circles is not easy to decide, but it is more than reasonable to assume that both Jacob’s story and that of Gregory’s source are based upon a Greek original from the latter half of the fifth century which, unfortunately, is now lost.44 Be that as it may, in one of his poetic homilies (or homiletic poems), Jacob of Sarug tells the story of the Seven Sleepers in what is already basically the form in which we learned about it at the beginning of this paper, from the much later Legenda aurea.45 For that reason it is not necessary to quote his and other late antique versions of the story in extenso. In these versions one finds an enormous variety in regard to the names (and even the number) of the long-sleepers, the name of the mountain where they hid in the cave, the name of the city, the number of years that their sleep lasted etc.; but the basic storyline remains the same.46 Much abbreviated, Jacob’s version may be summarized as follows:

The Emperor Decius comes to Ephesus and orders everyone to sacrifice to the pagan gods. Some boys of the leading families refuse and go into hiding, but they are denounced. Decius orders that they be flogged and kept until he returns. The boys escape and hide in a cave near Ephesus. They take some of their parents’ money with them. In the cave they pray to God; God raises their spirits into heaven and sends a watcher to guard their bodies.47 On his return, Decius orders the cave’s entrance to be blocked. When, after the pagan era, God wants to awaken them, a man in need of building materials reuses the stones at the cave’s entrance, and the boys are awakened by the daylight. Then they decide to send one of their number, Iamlikha (=Iamblichus), to the city, in order to see if Decius has already returned; they give him some small change to buy bread. Iamlikha is utterly surprised to see crosses above the city gates

44 Thus also Koch, Siebenschläferlegende, 2–3, 84–87; and Honigmann, “Stephen of Ephesus,” 131. 45 See the edition of the Syriac text in Homiliae selectae Mar-Jacobi Sarugensis (ed. P. Bedjan, with additional material by S. P. Brock; Piscataway: Gorgias, 2006), 6:324–30. Brock’s English translation can be found in his article, “Jacob of Serugh’s Poem on the Sleepers of Ephesus,” in “I Sowed Fruits into Hearts” (Odes Sol. 17:13): Festschrift for Professor Michael Lattke (ed. P. Allen, M. Franzmann, R. Strelan; Early Christian Studies 12; Strathfield, NSW: St. Paul’s Publications, 2007), 13–30. I owe many thanks to Prof. Sebastian Brock for sending me a copy of this article. 46 All major and minor variants are discussed in detail in Koch, Siebenschläferlegende. 47 The notion that the bodies are kept intact is strongly reminiscent of a remark of Alexander of Aphrodisias (as reported by Simplicius) on Aristotle’s story of the Sardinian long- sleepers: holoklêra diamenein ta sômata; see Rohde, “Sardinische Sage,” 198, for references. 108 van der Horst

and wonders whether this is really Ephesus. He tries to buy bread but among the bread-sellers his archaic coins raise the suspicion that he has found a treasure. He denies it but is taken to the bishop, who questions him. He says that he is the son of one of the leading citizens, but he fails to recognize anyone in the crowd who might rescue him. When he asks where Decius is, people think he has gone mad since that would make Decius 372 years old. Then the boy tells the bishop how he and his companions escaped to the mountain to hide in a cave. The people go up to the mountain, and the bishop enters the cave, where he greets the boys. He sends a message to the Emperor Theodosius, who immediately comes to Ephesus. Theodosius offers to build a shrine on the spot, but the boys decline and say all this has happened to prove the truth of the resurrection. They lie down, the Emperor covers them with his mantle, and again they sleep peacefully; i.e., they die.

The version of Gregory of Tours, written about 590 CE, is much shorter (De gloria martyrum 94[95]).48 The story is the basically same as that told by Jacob of Sarug, but there are some differences: the number of years the young men sleep is given as “many” (not 372); it is explicitly said that, just before the men wake up again, “the impure heresy of the Sadducees, who denied that there was a resurrection, was spreading”; and after the Emperor enters the cave, the men speak at length with him about the resurrection. We will now first look briefly at the circumstances which gave rise to this legend, at least according to a probable historical reconstruction (partly based upon sources later than Jacob and Gregory).49 The forties of the fifth century witnessed a revival of the Origenist controversy, which had been more or less

48 Text in PL 71:787–89; a better text in B. Krusch, ed., Gregorii episcopi Turonensis liber in Gloria martyrum (Monumenta Germaniae Historica: Scriptorum Rerum Merovingicarum 1: Gregorii episcopi Turonensis Miracula et opera minora, Hannover: Hahn, 1885), 550–52. Translation in Gregory of Tours: Glory of the Martyrs (trans. R. van Dam; Translated Texts for Historians; Liverpool: Liverpool University Press, 1988), 147–49. 49 In the following I rely partly on the study of Honigmann, “Stephen of Ephesus,” 142–68. Honigmann’s historical reconstruction is mainly based upon the form of the legend found in Zacharias Rhetor, Photius, and the medieval Greek version of Symeon Metaphrastes. This last text is printed as Hypomnemata in PG 115:437–45; Honigmann conjectures that it may go back to Stephen himself because of the (self?-)glorification of “the most holy bishop Stephanus” (see cols. 444–45). See also A. Samellas, Death in the Eastern Mediterranean (50–600 A.D.: The Christianization of the East, an Interpretation (STAC 12; Tübingen: Mohr Siebeck, 2002), 63–67. Pious Long-Sleepers 109 at rest since the beginning of that century.50 There was a new agitation in favor of the “heretical” ideas of Origen concerning the bodily resurrection, much to the sorrow of the Emperor Theodosius II. In the year 448 CE, Stephen, a priest in Ephesus, usurped the episcopal see of the city after having thrown the previous bishop into prison. The legitimacy of Stephen’s episcopate was very doubtful; at the council of Chalcedon in 451, Stephen was once again rebuked and finally deposed for this unlawful action. Shortly after his usurpation, he seems to have attempted a stratagem in order to strengthen and consolidate his position: He invented and spread an impressive tale about seven pious sleepers and interpreted the whole “event” as a corroboration on God’s part of the orthodox doctrine of the bodily resurrection. In this way, he sought to make his bishopric the scene of a spectacular discovery, of a miracle which would be unparalleled in history, casting himself as the bishop who thus refuted the new Origenists and put an end to doctrinal incertitude. It is hard to see how he could have convinced his own people of the reliability of the story, as an account of contemporary events; but something “miraculous” must have happened because archaeologists have proved that the building of the great church on the spot of the cave had begun by the middle of the fifth century, i.e., immediately after the “event,” whatever that may have been.51 Was it a chance find of some well- preserved bodies in a cave, following which Stephen exploited this “miracle”? Or should we follow Ernest Honigmann, who argues: “It seems incontest- able that about the middle of the fifth century seven young Ephesians really believed or tried to make others believe that they had been persecuted at the time of Decius.”52 If so, it is impossible to say whether they had been instructed to do so by Bishop Stephen, but that he used the story to secure his own position seems certain. Much else, however, remains uncertain here.53

5 Conclusions

If we assume that the Abimelech story, with its setting in the First Temple period, is of Jewish origin, which seems certain, we cannot but conclude that

50 See E. A. Clark, The Origenist Controversy: The Cultural Construction of an Early Christian Debate (Princeton: Princeton University Press, 1992). 51 Moreover, the earliest surviving Syriac manuscript containing the story dates to a period no later than around 500 CE; see Brock, “Jacob of Serugh’s Poem,” 14. 52 Honigmann, “Stephen of Ephesus,” 142. 53 That Stephen’s name does not occur in most preserved versions of the story is because it was edited out after his condemnation at the council of Chalcedon of 451. 110 van der Horst the Christian originator(s) of the legend of the Seven Sleepers of Ephesus knew this story and borrowed heavily from it.54 What both stories have in common is, first, the setting in a period of great upheaval: the destruction of the Jerusalem Temple and the deportation of the Jews to Babylon, on the one hand; and the persecution of the Christians by Decius, on the other. Second, there is the central motif of falling asleep, in or near a mountain. Third, there is the element of the return to the city after the long sleep; followed—fourth—by utter amazement in confronting the total change of the world, which has become well-nigh unrecognizable. Fifth, there is a dialogue between the sleeper and the inhabitants of the city. Sixth, both Abimelech and Iamblichus begin to wonder whether they have lost their wits. Finally, it seems that even the names of the protagonists are related: Abimelech and Iamblichus (or Malchus) have names (m-l-kh) that are too similar to go unnoticed. These agreements are too many and too striking to be coincidental! They cannot be explained as deriving from general folklore. We must assume that the creator of the Christian legend knew the Jewish story of the pious long-sleepers. That Christians did know this Jewish story is certain. The proof is that the Abimelech story has been handed down to us only in the christianized form of the Paralipomena Jeremiae, and that the same story is also found in the Christian Coptic translation of the originally Jewish Jeremiah Apocryphon.55 Both writings certainly predate, in their christianized form, the origin of the story of the Seven Sleepers, which is to be dated to about 450 CE. Thus, knowledge of this Jewish story in Christian circles before the middle of the fifth century is demonstrable.56 That its adoption and adaptation in a Christian setting was possibly facilitated by the fact that these Greek Christians perhaps also already knew the story about the miraculously long sleep of Epimenides cannot be proven but cannot be excluded either. After all, the Seven Sleepers were, like Epimenides (and, of course, Ḥoni and Abimelech), theophilestatoi, especially loved by God.57 But that matter must remain uncertain. It is certain only that in the middle of the fifth century CE, Christians in Ephesus saw fit to use a motif they knew from Jewish sources which had been preserved among Christians, and to put it into new service for

54 Thus also B. Heller, “Éléments, parallèles et origine de la légende des Sept Dormants,” REJ 49 (1904): 190–218 (214). 55 See Kuhn, “A Coptic Jeremiah Apocryphon,” chs. 12, 22, 38–39. 56 Moreover, there was also a Jewish community in Ephesus in late antiquity; see W. Ameling, Inscriptiones Judaicae Orientis II: Kleinasien (TSAJ 99; Tübingen: Mohr Siebeck, 2004), 147–62. 57 Koch, Siebenschläferlegende, 51, calls the pious long-sleepers “gottbegnadete Wesen.” Pious Long-Sleepers 111 their own purposes. The legend of the Seven Sleepers is nothing but the Jewish Ḥoni/Abimelech story in Christian dress.58

58 This is also the conclusion of Heller, “Éléments,” 217: “Nous signalons derrière le décor chrétien le fond juif.” Cf. also Huber, Wanderlegende, 422. There is also a long and complicated reception history of the legend of the Seven Sleepers in the Islamic world, beginning as early as the Qurʾān (Sūra 18.8–25), but that falls outside the scope of this paper. Most of the Islamic material is dealt with extensively in the monographs by Koch and Huber. Typical for Islam, the Qurʾān makes the boys Muslims, not Christians. M. Vogt, “Die Siebenschläfer—Funktion einer Legende,” Hallesche Beiträge zur Orientwissenschaft 38 (2004): 223–47, discusses, inter alia, the Qurʾānic material. I owe thanks to the editors of this volume for many useful critical remarks. Remnants of a Pharisaic Apologetic Source in Josephus and in the Babylonian Talmud

Tal Ilan and Vered Noam

The question of parallels between rabbinic texts and the writings of the historian Flavius Josephus has been the subject of longstanding debates and is currently the topic of a comprehensive project which the authors of this article have conducted.1 Presuppositions of major significance for this field were articulated by Shaye Cohen as early as the World Congress of Jewish Studies in 1985. In Cohen’s own words:

In many cases in A.J. the paralleled story seems to be an addition from some other source into the basic narrative. . . . On the rabbinic side, some of the parallel narratives were originally written for a purpose other than the one for which they were adduced in their present literary contexts. [For example] The story about Simon the Righteous. . . . The long narrative about the break between Yannai and the Pharisees . . . The narrative about the siege of Aristobulus II . . . by Hyrcanus II. . . . These are the three longest parallels between Josephus and rabbinic tradition. All three are found in the Bavli, not the Yerushalmi, and all three seem to be cited by the rabbis from some written source, a sort of Ur-Yosippon. . . . Two of the three . . . seem to be accretions to Josephus’s source, implying that he had access to this history as well.2

In other words, Cohen implies (a) that the most prominent parallels between Josephus and the rabbis are found between the Jewish Antiquities and the Babylonian Talmud; (b) that they are foreign to their context in both corpora; and (c) that these two corpora had independent access to the same repository of early Jewish traditions.3 However, in his discussion of the aforementioned

1 T. Ilan and V. Noam, in collaboration with M. Ben Shahar, D. Baratz and Y. Fisch, Between Josephus and the Rabbis (Jerusalem: Yad Izhak Ben-Zvi, forthcoming) (in Hebrew). 2 S. J. D. Cohen, “Parallel Historical Traditions in Josephus and Rabbinic Literature,” in Proceedings of the Ninth World Congress of Jewish Studies, Division B.1: History of the Jewish People (ed. D. Assaf; Jerusalem: World Union of Jewish Studies, 1986), 7–14 (13). 3 The term “Ur-Yosippon” used by Cohen appears to hint at some written adaptation of Josephus’s works which the rabbis had allegedly used, a late antique forerunner of the

© koninklijke brill nv, leiden, ��5 | doi ��.��63/9789004299139_006 Remnants of a Pharisaic Apologetic Source 113 story of the encounter between Simon the Righteous and Alexander the Great, Cohen simultaneously suggested a direct dependence of the rabbinic texts that tell this story on Josephus’s work.4 This ambivalence between the notions of an independent reliance of both corpora on a theoretical third source, and the dependence of the rabbis on Josephus, is typical of current scholarship, which is embarrassed by the intriguing riddle regarding the potential reasons for the similarities between the diverse texts. We have discussed elsewhere the foreignness of the stories to their Josephan/rabbinic contexts and have argued that the incongruity and secondary character of the parallel traditions, in the rabbinic texts and especially in Josephus’s works, constitute some of the most compelling evidence that the rabbis used neither Josephus nor any reworked version of his writings, but rather a repository of traditions from the Land of Israel known also to the historian.5 This issue will also be addressed briefly below. Our main goal in the current paper is to engage Cohen’s first assumption in the paragraph cited above, concerning the very location of these parallels in Josephus’s works, on the one hand, and in the large corpus of rabbinic literature, on the other: namely, that the parallels between Josephus and the rabbis are to be found mainly between the Jewish Antiquities and the Babylonian Talmud. In the project we are conducting we have examined this supposition, and our conclusions are as follows:

1. The vast majority of parallels between Josephus and the rabbis from the period between the Hasmonean Revolt and the eve6 of the war with Rome are indeed found in Josephus’s Jewish Antiquities, while missing

medieval Yosippon. However, Prof. Cohen has assured us (by written communication) that what he meant was that both Josephus and the rabbis had access to a written collection of stories that resembled the book that would emerge a thousand years later and be known as Yosippon. 4 Cohen, “Parallel Historical Traditions,” 13; idem, “Alexander the Great and Jaddus the High Priest according to Josephus,” AJS Review 7–9 (1982–1983): 41–68. 5 See Ilan and Noam, Between Josephus and the Rabbis; and recently V. Noam, “Did the Rabbis Know Josephus’s Works?” Tarbiz 81 (2013): 367–95 (in Hebrew). 6 Parallel stories concerning the events of the Great Revolt itself naturally appear in Josephus only in B.J., since A.J. does not cover this period. This includes such vignettes as the cessation of the sacrifice on behalf of the Roman Emperor (B.J. 2.409–417; Lam. Rab. 4:2; b. Giṭ. 55b–56a); the priests who committed suicide while the Temple burned (B.J. 6.278–280; ʾAbot R. Nat. B, 7; b. Taʿan. 29a); the mother who ate her son during the siege of Jerusalem (B.J. 6.199–218; b. Yoma 38b and parallels); and more. The similarity of these stories in Josephus to the rabbinic versions stems mainly from memories of the same events, atmosphere, and even theology, rather than from a shared literary source—oral or written. 114 Ilan and Noam

from the earlier Jewish War.7 This fact is not surprising, considering the difference of goal, genre, and scope between these two Josephan works. A.J. is much more comprehensive and inclusive by nature than B.J. and encompasses a variety of sources which Josephus had not utilized in B.J., including folkloristic Jewish traditions. Whereas the period preceding the Great Revolt serves only as background in B.J. (see B.J. 1.18), in the Antiquities Josephus was interested in the history of the Jewish people at large (A.J. 1.5). The inclusion in A.J. of Jewish traditions that Josephus had chosen to ignore in B.J. is probably also due to the pro-Jewish inclination which characterizes A.J., as many scholars have observed.8 2. The Babylonian Talmud does indeed preserve the largest number of parallels to the writings of Josephus of all the rabbinic compositions. Our abovementioned research has identified eleven such parallels.9 Scholars

7 With one significant exception—the story of the death of Herod. See B.J. 1.659–660; 666 (= A.J. 17.173–181); the same story is told (about King Yannai!) in the Scholion on Megillat Taʿanit for the second of Shevat. On this story, see M. Stern, “Nicolaus of Damascus as a Source of Jewish History in the Herodian and Hasmonean Age,” in Studies in Jewish History: The Second Temple Period (ed. M. Amit, I. M. Gafni, and M. D. Herr; Jerusalem: Yad Izhak Ben-Zvi, 1991), 445– 63 (especially 453) (in Hebrew); V. Noam, Megillat Taʿanit: Versions, Interpretations, History (Jerusalem: Yad Izhak Ben-Zvi, 2003), 282–80 (in Hebrew); and Ilan and Noam, Between Josephus and the Rabbis. 8 See R. Laqueur, Der Jüdische Historiker Flavius Josephus: Ein biographischer Versuch auf neuer quellenkritischer Grundlage (Giessen: von Münchow, 1920), chapter 5; on the reception of his opinion in later research see D. R. Schwartz, “Josephus on Hyrcanus II,” in Josephus and the History of the Greco–Roman Period: Essays in Memory of Morton Smith (ed. F. Parente and J. Sievers; Leiden: Brill, 1994), 212–16. See also idem, “Josephus on the Pharisees as Diaspora Jews,” in Josephus und das Neue Testament: Wechselseitige Wahrnehmungen. II. Internationales Symposium zum Corpus Judaeo–Hellenisticum, 25.–28. Mai 2006, Greifswald (ed. C. Böttrich, J. Herzer, and T. Reiprich; Tübingen: Mohr Siebeck, 2007), 137–46; and see recently M. Tuval, From Jerusalem Priest to Roman Jew: On Josephus and the Paradigms of Ancient Judaism (Tübingen: Mohr Siebeck 2013). 9 These are 1) The High Priest and Alexander the Great (A.J. 11.302–347; b. Yoma 69a); 2) The victory over Nicanor (A.J. 12.402–412; b. Taʿan. 18b); 3) John Hyrcanus and a heavenly voice (A.J. 12.282–283; b. Soṭah 33a); 4) The rift with the Pharisees (A.J. 13.288–299; b. Qidd. 66a); 5) Jannaeus’s will (A.J. 13.398–404; b. Soṭah 22b); 6) The fraternal war between Hyrcanus II and Aristobulus II (A.J. 14.22–28; b. Soṭah 49b; b. B. Qam. 82b; b. Menaḥ. 64b); 7) Herod’s trial (B.J. 1.203–211; A.J. 14.158–184; b. B. Bat. 3b–4a); 8) The rise of Herod (A.J. 14.488–491; 15.2–7, 262– 266, 366–370, 380–381, 388–391; b. B. Bat. 3b–4a); 9) The rain miracle in the days of Herod (A.J. 15.425; b. Taʿan. 22b–23a); 10) A priest named Ben Ilem (A.J. 17.165–166; b. Yoma 19b; b. Meg. 9b; b. Hor. 12a); 11) The census by Cestius Gallus/Agrippa II (B.J. 6.423; b. Pesaḥ. 64b). This list includes the three examples mentioned by Cohen (see above). Parallels which probably stem from shared history rather than from a common formulated source are not enumerated here. Remnants of a Pharisaic Apologetic Source 115

have wondered at the reappearance of Palestinian traditions dating from Second Temple times in late amoraic literature in far-off Babylonia. Some have suggested that the writings of Josephus were transported from Palestine or from somewhere in the Roman Empire directly to Babylonia during the fourth century.10 However, our investigations have shown that the prima facie impression of a multitude of parallels between Josephus and the Babylonian Talmud is misleading. More than a half of these parallels were initially absorbed into Palestinian tannaitic literature and are only secondarily cited by the Babylonian Talmud.11 Therefore, the encounter between these traditions and rabbinic culture must have occurred during the first two centuries CE rather than in the fourth century, and in the Land of Israel rather than in Babylonia. 3. Thus, the parallels which are found solely in Josephus’s A.J. and in the Babylonian Talmud are five only: 1. b. Qidd. 66a; 2. b. Soṭah 22b; 3. b. Soṭah 49b [= b. B. Qam. 82b; b. Menaḥ. 64b]; 4. b. Sanh. 19a–b; 5. b. B. Bat. 3b–4a. 4. These traditions can be divided chronologically and thematically into two groups:

a. b. Qidd. 66a and b. Soṭah 22b deal with the relationship between King Yannai (Alexander Jannaeus) and the Pharisees and are parallel to A.J. 13; b. b. Soṭah 49b [= b. B. Qam. 82b; b. Menaḥ. 64b], b. Sanh. 19a–b, and b. B. Bat. 3b–4a deal with the fall of the Hasmonean dynasty and the rise of Herod, and are parallel to traditions in A.J. 14–15.

Note traditions concerning, e.g., the temple of Onias (B.J. 1.31–33; 7:423–432; A.J. 12.237– 240, 387–388; 13:62–74; b. Menaḥ. 109b); Caligula’s statue in the Temple (B.J. 2.184–203; A.J. 18:256–309; b. Soṭah 33a); the cessation of the sacrifice on behalf of the Emperor (B.J. 2.409–417; b. Giṭ. 55b–56a); and many more. 10 See R. Kalmin, Jewish Babylonia between Persia and Roman Palestine (Oxford: Oxford University Press, 2006), 149–68. 11 Thus, for example, the story of the encounter of the High Priest “Simon the Righteous” with Alexander the Great, which Cohen cites as a typical Babylonian tradition (see above), is in fact quoted in the Talmud from an early version of the Palestinian Scholion on Megillat Taʿanit; this is evident from the fact that the Babylonian Talmud recounts it as appended to a passage from Megillat Taʿanit (b. Yoma 69a; compare the Scholion on Kislev 21; see Noam, Megillat Taʿanit, 262–65). The story of John Hyrcanus and the heavenly voice is known from the Tosefta (t. Soṭah 13:5); the story of Yosef Ben Ilem also appears in t. Kippurim 1:4. The rain miracle is first described in the Sifra (Beḥuqotai 1:1). 116 Ilan and Noam

We believe that the two groups—a. and b.—derive from separate sources. In this article we will try to reconstruct the character and provenance of the source that presumably stands at the background of the first group of stories, namely the two traditions concerning Alexander Jannaeus and the Pharisees.12 In regard to the first of these traditions, the story of the rift with the Pharisees, Vered Noam has recently concluded that this legend is in fact a rare remnant extracted from a Pharisaic polemical work; this conclusion is based on the story’s major themes, special terminology, unique hybrid Hebrew style, and integration of biblical allusions (as preserved in the Babylonian Talmud, see below)—all characteristics shared by sectarian texts.13 In the current paper we wish to show that the second tradition, which recounts Jannaeus’s advice to his wife regarding the Pharisees, stems from the same work or collection of works, and serves the same goals. In what follows we will first discuss the way Josephus incorporated these two traditions into his running account. We will then analyze each tradition separately, comparing Josephus’s treatment of each with that of the Babylonian Talmud. In our conclusion we will argue that these two stories concerning Alexander Jannaeus and the Pharisees have a common theological and political aim; this supports our thesis that they derive from the same source, or at the very least were produced by the same circles.

1 The Incorporation of the “Yannai” Sources in Josephus’s Jewish Antiquities

The two talmudic traditions concerning Jannaeus and the Pharisees are both paralleled in Josephus A.J. 13 (288–299; 398–404), but are absent from the earlier recounting of this history in B.J. 1 (67–68; 106). The first passage in Josephus tells the story of John Hyrcanus’s rift with the Pharisees. The second relates the advice that Alexander Jannaeus gave to his wife, concerning the Pharisees. The

12 The second group, we believe, stems from a source or a collection of traditions that told the story of the downfall of the Hasmoneans, which was brought about by an act of sacrilege (b. Soṭah 49b [= b. B. Qam. 82b; b. Menaḥ. 64b]). This act accelerated the ascent of Herod, himself a murderer who had refused to stand trial (b. Sanh. 19a–b; though he is not explicitly named here; see n. 26 below); when he seized the throne, he massacred many of the Jewish people. He was able to repent, however, by building the Temple (b. B. Bat. 3b–4a). Our reconstruction of this source may be found in Ilan and Noam, Between Josephus and the Rabbis. 13 V. Noam, “The Story of King Jannaeus (b. Qiddushin 66a): A Pharisaic Reply to Sectarian Polemic,” HTR 107 (2014): 31–58. The Josephan context of this story and its relationship to the talmudic version, to be discussed below, were not elaborated there. Remnants of a Pharisaic Apologetic Source 117 following two tables show, by comparison of the relevant passages in B.J. and A.J., how these traditions were incorporated into the original framework of A.J.:

Table 1 Hyrcanus’s rift with the Pharisees

B.J. 1.67–6814 A.J. 13.288–29915

67. The prosperous fortunes of John and 288. As for Hyrcanus, the envy of the his sons, however, provoked a sedition Jews was aroused against him by his own among his envious countrymen, large successes and those of his sons. numbers of whom held meetings to oppose them and continued to agitate, until the smoldering flames burst out in open war Particularly hostile to him were the Pharisees, who are one of the Jewish schools, as we related above. And so great is their influence that even when they speak against a king or high priest, they immediately gain credence. 289. Hyrcanus too was a disciple of theirs, and was greatly loved by them. And once he invited them to a feast. . . . [290–296 relates the story of the rift.] Out of this of course grew the hatred of the masses for him and his sons, but of this we shall speak hereafter. 297. For the present I wish merely to explain that the Pharisees had passed down to the people certain regulations . . . for which reason they were rejected by the Sadducees. . . . 298. . . . But of these two schools and of the Essenes a detailed account has been given in the second book of my Judaica.

14 Translation taken from Josephus in Nine Volumes (trans. H. St. J. Thackeray et al.; LCL; London: Heinemann, 1956–1965), 2:33–35. 15 Translation taken from Josephus, 7:373–77. 118 Ilan and Noam

Table 1 Hyrcanus’s rift with the Pharisees (cont.)

B.J. 1.67–68 A.J. 13.288–299 and the rebels were defeated. 68. For the 299. And so Hyrcanus quieted the rest of his days John lived in prosperity, outbreak, and lived happily thereafter; and after excellently directing the and when he died after administering government for thirty one whole years, the government excellently for thirty one died leaving five sons; truly a blessed years, he left five sons. individual and one who left no grounds Now he was accounted by God worthy of for complaint against fortune as regards three of the greatest privileges: the rule himself. He was the only man to unite in of the nation, the office of high priest, his person three of the highest privileges: and the gift of prophecy. the supreme command of the nation, the high priesthood, and the gift of prophecy.

Table 2 Alexander Jannaeus’s advice to his wife

B.J. 1.106 16 A.J. 13.398–40417

Afflicted by a quartan ague, he hoped 398. After these conquests King to shake off the malady by a return to Alexander fell ill from heavy drinking, active life. He accordingly plunged into and for three years he was afflicted with ill-timed campaigns and, forcing himself a quartan fever, but still he did not give to tasks beyond his strength, hastened up campaigning until, being exhausted his end. from his labors, he met death in the territory of Ragaba, a fortress in Transjordan. 399. And when the queen saw that he was on the point of death, and no longer held to the hope of recovery, she wept and beat her breast, lamenting the bereavement that was about to befall

16 Josephus, 2.53. 17 Josephus, 7.427–431. Remnants of a Pharisaic Apologetic Source 119

B.J. 1.106 A.J. 13.398–404

her and her children, and said to him, “To whom are you thus leaving me and your children? . . . 401. And then, he said, on her return to Jerusalem as from a splendid victory, she should yield a certain amount of power to the Pharisees. . . . These men, he assured her, had so much influence with their fellow Jews that they could injure those whom they hated and help those to whom they were friendly. . . . 403. “. . . Promise them also that you will not take any action, while you are on the throne, without their consent. 404. If you speak to them in this manner. . . you will reign securely.”

He died, at any rate, amid stress and With this exhortation to his wife he died, turmoil, after a reign of twenty seven after reigning twenty-seven years, at the years. age of forty-nine.

It is easy to see in the above tables how Josephus has inserted new material into the ongoing narrative he had already composed when he wrote B.J. some two decades earlier. If we remove these blocks of material from their surroundings in A.J., we are left with a coherent narrative, free of lacunae and contradictions, a narrative almost identical to its counterpart in B.J. This literary treatment, as well as additional considerations,18 confirms that the added material represents a previously independent source (or sources) upon which Josephus drew.

18 Many have noted, for example, the visible seams between Josephus’s frame narrative for the “rupture with the Pharisees” tradition, which displays hostility toward the Pharisees, and the pro-Pharisaic tone of the story itself. See S. Mason, Flavius Josephus on the Pharisees: A Composition–Critical Study (Leiden: Brill, 1991), 213–45; Kalmin, Jewish Babylonia, 56. The rupture tradition, as shown by these and other scholars, runs counter to Josephus’s general tendencies and terminology, and thus must have stemmed from a different source. For the scholarly dispute regarding the sources used by Josephus for the 120 Ilan and Noam

We now turn to a comparison of the way each of these stories is deployed in Josephus and the Babylonian Talmud. First, we will briefly discuss each of these parallel stories individually. We will then discuss the common features of the two stories and argue that they both derive from the same source (or group of sources), which had its own unique ideological–political message.

2 The Rift with the Pharisees

The first of these stories has been placed by Josephus towards the end of his account of the reign of John Hyrcanus. It relates how this Hasmonean ruler quarreled with the Pharisees:

(A.J. 13:289) . . . And once he invited them to a feast and entertained them hospitably. . . . (291) However, one of the guests, named Eleazar, who had an evil nature and took pleasure in dissension, said, “Since you have asked to be told the truth, if you wish to be righteous, give up the high priesthood and be content with governing the people.” (292) And when Hyrcanus asked him for what reason he should give up the high priesthood, he replied, “Because we have heard from our elders that your mother was a captive in the reign of Antiochus Epiphanes.” But the story was false, and Hyrcanus was furious with the man, while all the Pharisees were very indignant. (293) Then a certain Jonathan, one of Hyrcanus’s close friends, belonging to the school of Sadducees, who hold opinions opposed to those of the Pharisees, said that it had been with the general approval of all the Pharisees that Eleazar had made his slanderous statement; and this, he added, would be clear to Hyrcanus if he inquired of them what punishment Eleazar deserved for what he had said. (294) And so Hyrcanus asked the Pharisees what penalty they thought he deserved—for, he said, he would be convinced that the slanderous statement had not been made with their approval if they fixed a penalty commensurate with the crime, and they replied that Eleazar deserved

frame narrative, see D. R. Schwartz, “Josephus and Nicolaus on the Pharisees,” JSJ 14 (1983): 157–71; Mason, Flavius Josephus on the Pharisees, ibid., and the additional references in both. See also L. I. Levine, “The Political Struggle between Pharisees and Sadducees in the Hasmonean Period,” in Jerusalem in the Second Temple Period: Abraham Schalit Memorial Volume, (ed. A. Oppenheimer, U. Rappaport and M. Stern; Jerusalem: Yad Izhak Ben-Zvi, 1980), 61–83, (esp. 70–74) (in Hebrew); J. Efron, Studies on the Hasmonean Period (Leiden: Brill, 1987), 161–65, 176–86. Remnants of a Pharisaic Apologetic Source 121

stripes and chains; for they did not think it right to sentence a man to death for calumny, and anyway the Pharisees are naturally lenient in the matter of punishments. (295) At this Hyrcanus became very angry and began to believe that the fellow had slandered him with their approval. And Jonathan in particular inflamed his anger, and so worked upon him (296) that he brought him to join the Sadducaean party and desert the Pharisees, and to abrogate the regulations which they had established for the people, and punish those who observed them. Out of this, of course, grew the hatred of the masses for him and his sons, (297) but of this we shall speak hereafter.

The Babylonian parallel to this Josephan tradition is found in Tractate Qiddushin (66a). The text tells in unusual, pseudobiblical Hebrew—unknown from elsewhere in rabbinic literature in general and the Babylonian Talmud in particular—a similar story, related here not about Hyrcanus, but about Alexander Jannaeus:19

[I]t was taught: It once happened that King Yannai went to Koḥalith in the desert and conquered sixty towns there. On his return he (כוחלית) rejoiced exceedingly and invited all the sages of Israel. Said he to them, when they were engaged in the (מלוחים) Our forefathers ate mallows“ building of the Temple; let us too eat mallows in memory of our forefathers.” So mallows were served on golden tables, and they ate. לץ רע) Now, there was a man there, frivolous, evil, and a scoundrel Elʿazar b. Poʿirah to King (ויאמר) named Elʿazar b. Poʿirah. Said ,(ובליעל Yannai: “O King Yannai, the hearts of the Pharisees are against you Then what shall I do?” He said to him: ‘Make them“ ”.(לבם של פרושים עליך) between your eyes.” [So] he made (בציץ) by the plate (הקם להם) swear them swear by the plate between his eyes. Now an elder named Yehudah b. Gudgeda was present there. Said he to King Yannai: “O King Yannai! Let the royal crown suffice you (ויאמר) and leave the crown of priesthood to the seed of ,(רב לך כתר מלכות) Aaron.” For it had been rumored that his mother had [once] been taken captive in Modiʿim [Modiʿit]. The matter was investigated but not sus- -and the sages of Israel separated them ,(ויבוקש הדבר ולא נמצא) tained Then said Elʿazar b. Poʿirah to .(ויבדלו חכמי ישראל בזעם) selves in anger

19 Translation as in Noam, “King Jannaeus,” 32–34 (slightly revised), based on Kiddushin (ed. I. Epstein; trans. H. Freedman; new ed.; The Hebrew–English Edition of the Babylonian Talmud 15; London: Soncino, 1966). For linguistic and philological observations see Noam, ibid. 122 Ilan and Noam

King Yannai: “O King Yannai! That is the law even for a commoner in Israel, and you, a King and a High Priest, shall that be thy law [too]?” “Then what shall I do?” He told him: “If you will take my advice, trample But what shall happen with the Torah?” “Behold, it“ ”.(רומסם) them down whoever ,(כרוכה ומונחת בקרן זוית) is bound up and lying in the corner wishes to study, let him come and study!” through Elʿazar b. Poʿirah (ותוצץ הרעה) Straightway, the evil burst forth and through Yehudah b. Gudgeda. All the sages of Israel were massacred and the world was desolate until Shimʿon b. Shetaḥ ,(ויהרגו כל חכמי ישראל) came and restored the Torah to its pristine [glory].

The striking similarities between the story as told by Josephus, and the story as told in the Babylonian Talmud, combined with the implications of this account for our picture of the political tensions in Jewish society under Hasmonean rule, have given rise to countless scholarly discussions.20 The independence of the two versions of the story, and their provenance in a source which preceded both Josephus and rabbinic literature, is evident on several counts. First, the obvious nature of the passage as an insertion in A.J., as shown above, proves that we are not dealing here with a rabbinic citation from a coherent and continuous Josephan narrative, but rather with a unit which is from the outset exceptional and secondary within A.J. as well. Second, as mentioned above, the talmudic version displays exceptional vocabulary and syntax, typical of Jewish literature of the Second Temple period but otherwise unattested throughout the entire rabbinic corpus. This could not have been invented by any talmudic-era rabbi or redactor as a translation of Josephus’s Greek for a much later Hebrew/Aramaic context.21 Third, notwithstanding their broad similarities, the Josephan and the talmudic accounts also differ in many details.22 The original core of this legend should therefore be traced to some lost repository of Jewish traditions which predates Josephus just as it predates the Babylonian Talmud. One intriguing difference between the versions concerns the reference to the captivity of the ruler’s mother. The rabbinic version of the story gives the spatial location of the event (Modiʿim/Modiʿit), while Josephus prefers the temporal definition of a specific historical period (the time of Antiochus Epiphanes, 175–164 BCE). It is noteworthy that Modiʿim is seldom mentioned

20 For references see Noam, “King Jannaeus,” 36–37, nn. 20, 22. 21 The same is true for the use of Second Temple sectarian terminology, and for the employment of biblical allusions in a manner quite foreign to rabbinic literature. See Noam, “King Jannaeus,” 39–44, 46–51. 22 See Noam, “King Jannaeus,” 34–35. Remnants of a Pharisaic Apologetic Source 123 in rabbinic literature, and never identified elsewhere with the Hasmoneans or the Hasmonean revolt.23 Antiochus is also hardly ever mentioned by the rabbis, and certainly not as the great prosecutor of the Jews at the time of the Hasmonean revolt.24 Thus, neither of these terms evokes the historical circumstances of Hasmonean rule within the rabbinic frame of reference, and they would certainly not function as interchangeable within rabbinic writings. Therefore, it does not makes sense to assume that a rabbinic storyteller citing from Josephus would alter (to Modiʿim) the Josephan reference to Antiochus Epiphanes. This is a clear indication of rabbinic utilization of an independent, prerabbinic source, rather than the writings of Josephus. It is obvious, however, that the most significant difference between the two versions of the story in question is the name of the Hasmonean leader who is being attacked. While Josephus designates this ruler as John Hyrcanus (reigned 135–104 BCE), the rabbis identify him with Alexander Jannaeus (the son of the former; reigned 103–76 BCE). Most scholars hold that Josephus has preserved the original name; they presume that the editors of the Babylonian Talmud have deliberately changed, here and elsewhere, the names of those they perceive as wicked Jewish rulers to “King Yannai.”25 However, this

23 Modiʿim is mentioned mainly in m. Pesaḥ. 9:2; m. Ḥag. 3:5, and several other tannaitic and amoraic passages (appearaing as Modiʿit in the Palestinian versions of the Mishnah and in the Palestinian Talmud, as in the better versions of our story in the Babylonian Talmud). In all of them, the issue is its distance from Jerusalem for halakhic purposes, and not the Hasmonean uprising. is mentioned only three times in rabbinic literature. One reference (אנטיוכוס) Antiochus 24 is in Megillat Taʿanit for the 28th of Shevat, commemorating the departure of Antiochus from Jerusalem. Scholars are divided about the identity of this Antiochus, and neither scholia on this passage mention Antiochus at all (see Noam, Megillat Taʿanit, 115, 291–92). This is further proof that the Megillah itself is not a product of rabbinic circles, for—in contrast to rabbinic literature at large—it is familiar with the Seleucid Antiochides; but the Scholion is rabbinic, and thus is not informed about them (see Noam, Megillat Taʿanit, 19–22). The second reference is found in Gen. Rab. 23:1, where it is stated that towns are named after rulers: “Tiberias for Tiberius, Alexandria for Alexander, Antiochia for Antiochus.” Here, too, a historical narrative concerning Antiochus is absent. See, in addition, Seder Olam Rabbah 30 (Seder Olam: Critical Edition, Introduction and Commentary (ed. C. Milikowsky; 2 vols.; Jerusalem: Yad Izhak Ben-Zvi, 2013), 1:323–24, and Milikowsky’s notes there). It is remarkable that in another parallel between Josephus and the rabbis, Josephus rightly mentions a King Antiochus (Cyzicenus 116–96 BCE), whereas the rabbinic version in the Tosefta erroneously replaces this name with that of the city of Antiochia, probably due to ignorance (see A.J. 13.282–283; t. Soṭah 13:5). This is another case where Josephus and the rabbis are obviously dependent on the same shared earlier source. 25 See J. Derenbourg, Essai sur l’histoire et la géographie de la Palestine d’après les Thalmuds et autres sources rabbiniques (Paris: Imprimerie Impériale, 1867), 79–81; I. Friedlander, “The 124 Ilan and Noam argument is circular, because it is based first and foremost on this very story (b. Qidd. 66a).26 Scholars have also argued that the rabbis, too, ascribe to John Hyrcanus a change of heart in the direction of Sadduceanism towards the end of his life, a fact that supports Josephus’s version.27 Another argument in favor of Hyrcanus as the political leader originally depicted in the legend is the slander that the ruler’s mother was taken captive in Modiʿim, the home town of the Hasmoneans (see 1 Macc 2:1), as the Babylonian version suggests. A Jewish woman who is married to a priest (such as the Hasmoneans), is forbidden to return to her husband if she has been held captive (cf. m. Ketub. 2:9; 4:8); if she does return, this renders any children born after her captivity unfit for the priesthood. The alleged captivity, according to Josephus, must have happened in 168–166 BCE (the time frame when Antiochus Epiphanes’ soldiers might have held captives in Modiʿim). This would fit the case of Simon the Hasmonean’s wife, who was the mother

Rupture between Alexander Jannai and the Pharisees,” JQR 4 (1914): 443 n. 3; V. Aptowitzer, Parteipolitik der Hasmonaeerzeit im rabbinischen und pseudoepigraphischen Schrifttum (Veroffentlichungen der Alexander Kohut Memorial Foundation 5; Vienna: Kohut Foundation, 1927), 13–17; C. Rabin, “Alexander Jannaeus and the Pharisees,” JJS 7 (1956): 8–9; V. Tcherikover, Hellenistic Civilization and the Jews (Philadelphia: Jewish Publication Society, 1959), 254–61; J. Neusner, The Rabbinic Traditions about the Pharisees before 70 (3 vols.; Leiden: Brill, 1971), 1:175; Efron, Studies, 178 n. 151; Cohen, “Parallel Historical Traditions,” 14; D. Goodblatt, “The Union of Priesthood and Kingship in Second Temple Judea,” Cathedra 102 (2001): 7–28 (13) (in Hebrew). According to some opinions it was the sage Abaye, the transmitter of the legend, who was responsible for the switch from Hyrcanus to Jannaeus, since he is the one who erroneously identified “Yannai” with “Yoḥanan” in b. Ber. 29a. See, e.g., Derenbourg, Essai, 80–81 n. 1; Efron, Studies, 143–218; D. R. Schwartz, “Κατὰ τοῦτον τὸν καιρόν: Josephus’ Source on Agrippa II,” JQR 72 (1982): 241–68 (266–67); Cohen, “Parallel Historical Traditions,” 11; and the references in Goodblatt, “Union,” 13 n. 21. 26 The other texts usually cited to support this contention are the following: 1) b. Sanh. 19a, where Jannaeus supposedly replaces Herod; a closer reading of the passage shows that the evidence here is not straightforward (see Ilan and Noam, Between Josephus and the Rabbis, chapter 6, appendix b, by Tal Ilan). 2) b. Ker. 28b, where Jannaeus supposedly replaces King Agrippa II; this is a late scribal addition, since in the parallel tradition (b. Pesaḥ. 57a), and in some mss., the king is anonymous. 3) b. Yebam. 61a, where Jannaeus again supposedly replaces Agrippa II. This last case may indeed indicate Babylonian editorial activity, placing Jannaeus in the position of a king who is bribable. On the other hand, the Babylonian Talmud knows Herod very well as a wicked king in his own right, see b. B. Bat. 4a, which undermines the assumption that all wicked kings become “Yannai” in the Babylonian Talmud. 27 b. Ber. 29a. See, e.g., M. Stern, Hasmonean Judaea in the Hellenistic World: Chapters in Political History (Jerusalem: Merkaz Shazar, 1995), 199 (in Hebrew). Remnants of a Pharisaic Apologetic Source 125 of John Hyrcanus, but not that of Jannaeus's mother, who might not even have been born yet.28 However, Modiʿim, otherwise unknown to the rabbis in the Hasmonean context (see above), must have appeared in their original source. If we presume that Josephus used the same source utilized by the rabbis, the replacement of Modiʿim with Antiochus Epiphanes could have been his own handiwork.29 Although most scholars maintain that the mention of Modiʿim is the stron- gest argument in favor of Hyrcanus as the hero of this tradition, it is noteworthy that 1 Macc 16:4 describes a battle that took place in Modiʿin (Μωδεϊν) in 135 BCE, between John Hyrcanus and a general of Antiochus Sidetes.30 Such a battle could just as likely have been the historical occasion for a “captivity in Modiʿim” as referred to in the original text on which both Josephus and the Babylonian story are based. In such circumstances (during the reign of Antiochus VII Sidetes), John Hyrcanus’s wife, the future mother of Jannaeus, could well have been a captive for a brief while, and the rumor would consequently refer to her. Josephus, as already noted by others, relates elsewhere a similar rumor that had circulated concerning Jannaeus himself; i.e., that he was the son of a captive, and thus unfit to serve at the altar (A.J. 13.372).31 A minority opinion does argue, however, that the Talmud’s association of the tradition with Jannaeus must be closer to the original account.32 According to this opinion, a central motif in the story is the royal crown and the double authority of kingship and high priesthood. Since John Hyrcanus, unlike his son Jannaeus, was not a king, it is improbable to assume that he was the

28 See Goodblatt, “Union,” 12 n. 20. 29 Like 1 Maccabees, Josephus identifies Modiʿin as the site of the outbreak of the Hasmonean Revolt (implied in B.J. 1.36 and narrated more fully in A.J. 12.265–271). Josephus, like 1 Maccabees, also identifies Modiʿin as the site of the Hasmonean family sepulcher; see 1 Macc 2:70; 9:19; 13:25; A.J. 12.285; 432; 13.210–212). 30 Josephus, however, fails to mention this reference to Modiʿin, because, as is well known, he did not use the last chapters of 1 Maccabees for his account in A.J. 13 (see J. Sievers, Synopsis of the Greek Sources for the Hasmonean Period: 1–2 Maccabees and Josephus, War 1 and Antiquities 12–14 [Rome: Pontifical Biblical Institute, 2001], 200; and see also S. J. D. Cohen, Josephus in Galilee and Rome [Leiden: Brill, 1979], 45 n. 79 and bibliography there). 31 See, e.g., G. Alon, Jews, Judaism and the Classical World (Jerusalem: The Hebrew University Magnes Press, 1977), 26 and n. 22; and M. J. Geller, “Alexander Jannaeus and the Pharisee Rift,” JJS 30 (1979): 202–11 (209–10). 32 Friedlander, “Rupture,” 443–48; G. Alon, Jews, 26 and n. 22; Geller, “Alexander Jannaeus”; E. Main, “Les Sadducéens et l’origine des Partis Juifs de la Période du Second Temple” (Ph.D. diss., The Hebrew University of Jerusalem, 2004). See also the references in Efron, Studies, 177 n. 150. 126 Ilan and Noam protagonist of the original story.33 Moreover, Josephus’s general description of John Hyrcanus as a successful ruler who was loved by his people (A.J. 13.230– 300, esp. 299–300), which is supported by rabbinic testimony,34 does not cohere with the legend’s depiction of the ruler’s harsh attitude toward the Pharisees, or with the people’s hatred towards him. The Babylonian Talmud’s version does, however, correspond quite well with what we know from Josephus, Pesher Nahum and rabbinic literature concerning the murderous Alexander Jannaeus and the rebellion against him.35 These arguments are no less compelling than those that speak for the supremacy of Josephus’s version. In what follows we will argue for an inherent connection between this legend and the other tradition about Jannaeus and the Pharisees that we are discussing here. We believe that this connection may serve as further evidence in favor of the second conjecture; namely, that the story was originally recounted about Alexander Jannaeus. However, before we discuss the next episode, we wish to draw the reader’s attention to another difference between the story of the rift in Josephus and that in the Babylonian Talmud. Because the latter version is in Hebrew, the original language of composition, it has retained subtle biblical allusions that have all but disappeared from Josephus’s retelling. As has been shown by Noam, the biblical inspiration for these allusions is clearly the story of Korah’s rebellion in Numbers 16, which is the source of some of its unique vocabulary.36

33 Geller, “Alexander Jannaeus,” 208; Main, “Les Sadducéens,” 406. 34 See Alon, Jews, 26 n. 22. 35 Several scholars have identified the ruler in the story with Jannaeus on the basis of Alexander Jannaeus’s coinage, pointing to coins on which the title “King” was originally imprinted but then replaced with the title “High Priest,” presumably due to Pharisaic pressure; for references to this argument and the counter-arguments, see Goodblatt, “Union,” 9–10. Dan Barag suggested that another group of coins, on which both “Priest” and “King” were minted, are proof of Jannaeus’s wish to support his own double status against the demand to give up one of the two positions (D. Barag, “Alexander Jannaeus: Priest and King,” in “Go Out and Study the Land” (Judges 18:2): Archaeological, Historical and Textual Studies in Honor of Hanan Eshel [ed. A. M. Maeir, J. Magness, and L. H. Schiffman; Leiden: Brill, 2011], 1–5. Barag himself accepted Josephus’s connection of the story with Hyrcanus; but he suggested that the Pharisaic demand to give up the high priesthood gathered strength during the reign of Jannaeus, as is evident from the fact that accusations were voiced against the latter’s pedigree as well. -priest“ ,כהונה let [the royal crown] suffice for you”); the noun“) רב לך Such as the phrase 36 and the sages of“ ,ויבדלו חכמי ישראל בזעם hood”; the motif of separation in the phrase ,burst forth”; see Noam“ ,ותוצץ Israel separated themselves in anger”; and the rare verb “King Jannaeus,” 40–44. Remnants of a Pharisaic Apologetic Source 127

3 Alexander Jannaeus’s Advice to His Wife

The second tradition in which we are interested is incorporated into Josephus’s narrative just before the account of the death of King Alexander Jannaeus, when the latter bequeaths the throne to his wife, Shelamzion Alexandra (A.J. 13.398–404). As in B.J. (1.106), we are informed that Jannaeus was afflicted by a fever and died fighting. In A.J., just before he dies, he advises his wife to share power with the Pharisees and to allow them to violate his corpse. The Babylonian Talmud presents a very short tradition on some advice that Yannai gives his wife regarding the Pharisees.37 The advice sounds like a sort of poetic witticism in Hebrew, but it is introduced by a sentence in Babylonian Aramaic (the Aramaic phrase is marked by italics):

King Yannai said to his wife: Fear not the Pharisees, nor those who are not Pharisees, but fear rather the hypocrites who are like the Pharisees, who do the deeds of Zimri but demand the reward of Pinḥas. (b. Soṭah 22b)

The similarity of this text to Josephus’s narrative of Jannaeus’s testament is minimal. There we are informed that the king is dying; here there is no spatial or temporal context for the story. There the king gives his wife practical advice about how to deal with the Pharisees; here he merely describes them. There the woman becomes queen upon her husband’s death and shares power with the Pharisees; here we do not even know that the wife is the queen honored elsewhere by the rabbis.38 She is not named, but merely appended to her husband.39 The only reason why scholars assume that this rabbinic tradition parallels the Josephus narrative is because Jannaeus’s wife and the Pharisees are mentioned in both. It appears that both traditions answer the same (unstated) complaint of the king’s wife: “I fear the Pharisees.” As we saw for the tradition about the rift with the Pharisees, in this case, too, only the Babylonian text has preserved a biblical allusion, this time an explicit one. “Zimri” and “Pinḥas (Phineas)” of the king’s rejoinder are well known from the biblical episode about the temptation of the Israelites by the daughters of Midian. This story, like the narrative of Korah’s rebellion

מכות) ”The context is a discussion of m. Soṭah 3:4, on the “afflictions of the Pharisees 37 On this tradition see T. Ilan, Integrating Women into Second Temple History .(פרושים (Tübingen: Mohr Siebeck, 1999), 19–20; idem, Silencing the Queen: The Literary Histories of Shelamzion and Other Jewish Women (Tübingen: Mohr Siebeck, 2006), 76–77. 38 See Ilan, Silencing the Queen, 40. 39 On the Babylonian Talmud’s attitude to the queen see Ilan, Silencing the Queen, 36–41. 128 Ilan and Noam mentioned above, is located in the Book of Numbers (25). It is also remarkable Zimri,” who is tempted“ 40.כהונה that both biblical episodes use the rare noun by the Midianite woman, is the chief of the tribe of Simeon; “Phineas” is the priest, Aaron’s grandson, who slays the Midianite woman and her paramour. He is rewarded for his action by being granted an eternal covenant of priesthood for himself and his offspring (Num 25:10–13). Since the king’s rivals (who are not Pharisees but appear to be like them) seek Phineas’s reward, namely the gift of eternal priesthood, the issue behind the biblical episode alluded to here is probably the high priesthood and who holds it by right, as in the story of the rift with the Pharisees. Within the terms of the king’s allusion, these rivals who claim the high priesthood are not the “good guys” of this story (Phineas) but rather the “villains” (Zimri). The reward offered to the hero (Phineas) tallies well with the political position of Alexander Jannaeus (as both secular ruler and high priest). However, in Second Temple times, the deeds of Phineas, and indeed the very verb used to act with zeal”), carried“ :קנא) to describe his actions by the biblical author broader connotations.41 First Maccabees describes Mattathias, the forefather of Jannaeus, who slew a Jew sacrificing to idols during the Antiochean perse- cutions, with the words: “Thus he showed his zeal for the Law, as Phineas had done toward Zimri son of Salom”42 (1 Macc 2:26; see also 2:51–54). There is little doubt that the Hasmoneans identified with Phineas.43

,as a verb or a noun describing a person is quite common in the Pentateuch כה"ן The root 40 כהנת ,One is the reward of Phineas .כהונה but there are only few occurrences of the noun .see below. For one occurrence in an appendix to the Korah story see Num 18:1 ,עולם 41 For a review of the image of Phineas in postbiblical literature see for example L. H. Feldman, “The Portrayal of Phinehas by Philo, Pseudo-Philo, and Josephus,” JQR 92 (2002): 315–45; C. Batsch, “Priests in Warfare in Second Temple Judaism: 1QM, or the ‘Anti- Phinehas,’ ” in Qumran Cave 1 Revisited: Texts from Cave 1 Sixty Years after Their Discovery. Proceedings of the Sixth Meeting of the IOQS in Ljubljana (ed. D. K. Falk et al.; STDJ 91; Leiden: Brill, 2010), 165–78 (173). -but the Greek versions suggest alternative read ,סלוא The Hebrew of Num 25:14 reads 42 ings of the name. 1 Macc 2:26 reads Σαλωμ (= Salom); LXX to Num 25:14 reads Σαλω (= Salo). 43 The Second Temple ideal of zeal has been discussed by many. See, e.g., J. D. G. Dunn, The New Perspective on Paul (rev. ed.; Grand Rapids: Eerdmans, 2008), 11–12; and in reaction, D. Ortlund, “Phinehan Zeal: A Consideration of James Dunn’s Proposal,” JSP 20 (2011): 299–315; and see also G. Aran, “The Other Side of Israelite Priesthood: A Sociological– Anthropological Perspective,” in Was 70 C.E. a Watershed in Jewish History? On Jews and Judaism before and after the Destruction of the Second Temple (ed. D. R. Schwartz and Z. Weiss; Leiden: Brill, 2012), 43–58. We are grateful to Dr. Ruth Clements for these references. Remnants of a Pharisaic Apologetic Source 129

However, they were not alone in this. For example, Ben Sira, in his “Praise of Great Men,” identifies Phineas as one of the ancestors of Simon the Righteous, whose praise is the climax of his panegyric (50:24). In Qumran too, Phineas is mentioned first and foremost as the ancestor of the Second Temple Zadokite dynasty.44 In other words, not only the Hasmoneans but also their worst enemies embraced Phineas as their hero. It appears that the Pharisees also wished to be associated with him. In an early and enigmatic tannaitic midrash on Numbers, when Phineas pretends to enter the tent of the Midianite woman, as though intending to engage in forbidden sexual relations with her, onlookers assume that “the Pharisees have permitted this” (Sifre Numbers 131). In other words, the midrash depicts Phineas as a Pharisee. The pesher-like ascription of a biblical hero’s identity to a Second Temple political personality is irrelevant to, and atypical of, the world of the sages, and therefore appears to be a resi- due of a lost, early midrashic layer which applied the biblical scene to current political needs. The sinner Zimri, who was justifiably slain, is no doubt the antithesis of Phineas, the blessed and pious priest. He is considered an ultimate villain in Qumran literature as well. In a work called by its editors “[a] wisdom psalm with historical model” (4Q372),45 the author praises the Lord for taking revenge on Israel’s enemies, mentioning the punishment afflicted on the Midianites and “Zimri son of Salu.”46

44 See 6QPriestlyProphecy (6Q13) ll. 4–5, in Les ‘Petites Grottes’ de Qumran (ed. M. Baillet, J. T. Milik, and R. De Vaux; DJD 3; Oxford: Clarendon, 1962), 127. J. D. G. Dunn first suggested (New Perspective, 343–44) that the last words of 4QMMT, which are the closing words of the halakhic letter the Qumranites sent to their opponents (presumably Pharisees), are -And it shall be reck) ונחשבה לך לצדקה בעשותך הישר והטוב :also of relevance here oned to you for righteousness, in doing what is upright and good). These are reminiscent ותחשב :of Ps 106:28–31, which mentions Phineas’s deeds, and concludes with the words and it was reckoned to him as righteousness”). We thank Menahem Kister“) לו לצדקה for this reference. Ruth Clements has noted (orally) that the psalm itself brings Phineas under the rubric of Abraham, the earlier target of this phrase (Gen 15:6), and that in both cases it has implications of covenant and chosenness. 45 See M. J. Bernstein, “Poetry and Prose in 4Q371–373 Narrative and Poetic Compositiona,b,c,” in Liturgical Perspectives: Prayer and Poetry in Light of the Dead Sea Scrolls. Proceedings of the Fifth International Symposium of the Orion Center for the Study of the Dead Sea Scrolls and Associated Literature, 19–23 January 2000 (ed. E. G. Chazon et al.; STDJ 48; Leiden: Brill, 2003), 33. 46 4Q372 3 10–12. See E. M. Schuller and M. J. Bernstein, “4Q372. Narrative and Poetic Compositionb,” in Wadi Daliyeh II: The Samaria Papyri from Wadi Daliyeh and Qumran Cave 4.XXVIII: Miscellanea, Part 2 (ed. D. M. Gropp, M. J. Bernstein et al.; DJD 28; Oxford: Clarendon 2001), 181–2. Ruth Clements drew our attention to the editors’ note that line 12 130 Ilan and Noam

However, the “deeds” of this biblical Zimri, to which “Yannai” appears to allude—namely, engaging in sexual contact with non-Jewish women, and possibly idolatry—do not seem to fit the historical-political situation of Alexander Jannaeus.47 It should therefore be noted that Josephus presents Zimri, not primarily as a sexual transgressor or idolater, but rather as a rebel against the legitimate leadership of Moses (A.J. 4.145–149). Zimri stands before Moses and accuses him of hypocrisy and tyranny.48 He claims that Moses acts like a tyrant (τυραννικῶς); uses criminal means (κακουργεῖς) in order to prepare for himself a position of absolute sovereignty under the guise of establishing God’s laws (προσχήματι νόμων τοῦ θεοῦ) (A.J. 4.146); and that he is himself punish- able (τιμωρίαν) under the same laws (147). Interestingly, Josephus places these same arguments in the mouth of Korah (A.J. 4.15–19): Moses, claims Korah, is a criminal who pursues his own glory under the pretense that this is God’s will (κακουργοῦντα . . . ἐπὶ προφάσει τοῦ θεοῦ), as is the way of tyrants (τυράννων τρόπῳ); and such people should be punished (κολάζειν). Another intriguing point in the Talmud’s account is Yannai’s warning, formulated as a short proverb, against hypocrites who look like Pharisees but are actually not Pharisees. Accusations against the hypocrisy of the Pharisees are attested in the literature of Qumran,49 in the New Testament,50 and in Josephus’s writings elsewhere (A.J. 17.41–42).51 According to this commonly held opinion, the Pharisees are a most influential group, but at the same time hypocritical and bent on intrigue. They can undermine the authority of kings and high priests and turn the hearts of the populace against them, and they are the ones responsible for the rift with the ruling Hasmonean dynasty (whether

brings together allusions to the Zimri episode in Numbers 25 and to the slaying of the five Midianite kings in Numbers 31, in both of which Phineas plays a prominent role (p. 184). 47 Yannai might also be referring here to another biblical Zimri, namely the regicide, seven- day king of Israel of 1 Kgs 16:8–20, whose claim to fame became his murder of his master; as nicely put a generation later by Jezebel, the widow of King Ahab, to her killer-to-be, Yehu, “Peace, Zimri, murderer of his master” (2 Kgs 9:30). Perhaps Yannai is warning his wife against potential assassins, like Zimri (and like Yehu, who murdered Jezebel, the king’s widow). 48 And cf. also b. Sanh. 82a–b. ,i.e., slippery ,דורשי החלקות 4QpNah (4Q169) 3–4 ii 9–10, where they are designated 49 untrustworthy. 50 Primarily Matthew 23. 51 Josephus’s descriptions of the Pharisees here are probably based on the writings of Herod’s court historian Nicolaus of Damascus. See Schwartz, “Josephus and Nicolaus,” 157–71. Remnants of a Pharisaic Apologetic Source 131 represented by John Hyrcanus or Alexander Jannaeus). The author of our passage was probably trying to refute these accusations by claiming that 1) those who object to Jannaeus believe that they deserve the high priesthood like Phineas (or conversely that he, Jannaeus, is not entitled to it), but are in fact sinners and rebels like Zimri; 2) these people appear to be Pharisees, but in truth they are hypocrites who only pretend to be Pharisees. This message is surprisingly similar to the lesson of the rift tradition discussed above, which argued that the “real” Pharisees objected to the Korah-like instigators who denied the Hasmoneans’ right to the High Priesthood. All these rich details are completely absent from the parallel story in Josephus. Is it likely that rabbis in talmudic Babylonia found a version of the story identical to the one integrated into A.J., and added the explicit biblical allusions and political accusations which are hallmarks of Second Temple history—references to zealotry in the Hasmonean family; designation of the Pharisees (or a faction resembling them) as hypocrites; warning the queen- to-be against potential enemies; creating a shrewd, polished Hebrew proverb? Or is it more probable that all these were part and parcel of the original tale, and were removed by Josephus in order to accommodate his story for his non- Jewish audience? We think the latter is much more plausible.

4 A Common Source on Jannaeus and the Pharisees?

The two Babylonian traditions that were presented above have several features in common. The most striking is that both refer to the Pharisees and to King Yannai. In what follows we wish to show an even deeper internal similarity between them. In the tradition concerning the rift between the Pharisees and Alexander Jannaeus, we identified features of a sectarian dispute. According to Josephus’s account, the opponent of the Pharisees was a Sadducee; the rabbinic version of the story mentions a villain who schemes against the Pharisees, probably motivated by sectarian leanings. As shown by Noam, the story employs terminology, arguments, and rhetorical strategies which were widely used in political disputes extant from the Second Temple period, especially in Qumran. One of these strategies is the use of a “pesher-like” political “midrash” which connects biblical characters and events to the circumstances of the Hasmonean state. The rift tradition sets out to discredit a common belief, of which we know from other Second Temple texts: namely, that the Pharisees were by nature hypocritical and the worst enemies of the Hasmoneans. Noam has concluded that, contrary to this popular impression, the Jannaeus legend argues that the 132 Ilan and Noam

Pharisees/sages of Israel never opposed the king and did not believe in the wicked libel concerning his doubtful pedigree. The rift with the Hasmoneans was not spurred by the Pharisees; neither was it the fault of the ruler.52 On the contrary, the enemies of the Pharisees were the ones who had spread the wicked rumor, and it was they who had brought about the conflict and the catastrophe.53 In this context a villain is described who incites the ruler against the Pharisees and accuses them of the attempt to undermine the king’s position. The storyteller implies that this scoundrel is the equivalent of the biblical Korah, whereas the Pharisees are cleared of all guilt, since they “separated themselves in anger” on hearing the slander pronounced against Jannaeus’s pedigree. In other words, the storyteller strives to distinguish between the libel, which derives from the Pharisees according to the slanderer, and the real Pharisees, who completely disown this troublemaker. By doing so, the storyteller shifts the accusations previously leveled at the Pharisees onto others, who have been, in his opinion, mistakenly identified with them. A very similar attitude can be identified in the Talmudic tradition concerning Yannai’s advice to his wife. In this tradition, the king instructs his wife on how to deal with the Pharisees. This advice is formulated as a short proverb on hypocrites, people who look like Pharisees, but are actually not Pharisees. Those are the equivalents of the biblical Zimri. Both passages deal with the Pharisees’ attitude to Alexander Jannaeus. In both of them this attitude is ambivalent. There is no hostility between the king and the Pharisees at the outset. Others are to blame for an eventual rift (the person who insulted the ruler, from whom the Pharisees dissociate themselves, in the first story; the hypocrites who only “look like” Pharisees in the second). It appears as if the authors of both stories were seeking conciliation with the Hasmoneans, claiming that the conflict was in fact a misunderstanding, caused by a malicious third party. At the background of both stories a biblical context is implied: The tradition about the rift with the Pharisees echoes the story of the rebellion of the biblical Korah; the talmudic story of Yannai’s advice to his wife evokes another

52 Menahem Kister has identified a similar argument in 4QMMT. According to his inter- C 8–9), “And you) ואתם י]ודעים שלוא[ ימצא בידנו מעל ושקר ורעה pretation, the words [know that no] treachery or deceit or evil can be found in our hand,” declare that the sect separated itself “from the multitude of the people” (see line 7), not due to rebellion against Hasmonean authority, but rather because of religious reasons. See M. Kister, “Studies in 4QMiqṣat Maʿaśe Ha-Torah and Related Texts: Law, Theology, Language, and Calendar,” Tarbiz 68 (1999): 317–71 (321 n. 12) (in Hebrew). 53 Noam, “King Jannaeus,” 44–46. Remnants of a Pharisaic Apologetic Source 133

rebellion, that of Zimri—both are found in the biblical book of Numbers and both have to do with the covenant and authority of priesthood. Perhaps associating these two rebels of the Book of Numbers one with the other was a common Second Temple topos (see above on Josephus’s similar treatment of both figures) that is evident also in the two passsages we are investigating here, if we consider them as derived from a common source. Additionally, in both cases the association with the Bible is not based on the explicit citation of verses, as is usual in classical rabbinic writings, but is rather established through a comparison of the biblical situation with the political situation of the Second Temple (Jannaeus’s advice), or through allusions to the relevant biblical verses (the rift with the Pharisees). Both strategies are familiar to us from the literature of Qumran, but are atypical of rabbinic midrash. We thus conclude that these two traditions are in fact two authentic passages of Pharisaic apologetics, reacting to commonly held accusations that they were undermining Hasmonean legitimacy. It appears that these two traditions are the handiwork of moderate Pharisees, who wished to prevent a complete break with the Hasmonean dynasty; they sought to distance themselves from more extreme groups, who actively opposed Hasmonean rule. These, so our traditions argue, are not “real” Pharisees, and “we” should not be accused of their problematic activity. It is interesting to note that, in contrast to the vast majority of parallels between Josephus and rabbinic literature, these two stories arrived in Babylonia without any intermediary tannaitic reframing. This fact enhances our suspicion that they derive from one common source, and that it was this source (or a variation of it) that Josephus used in his own composition. Windy and Fiery Angels: Prerabbinic and Rabbinic Interpretations of Psalm 104:4

Yaakov Kaduri

What goes on in highest heaven was a subject of some curiosity among the ancient interpreters of Scripture. Surprisingly, however, rather few passages in the Hebrew Bible provide details about angels and other heavenly creatures; perhaps for this reason, the few that do so were the subject of continued scrutiny. In addition to Isaiah’s vision of the heavenly throne (Isaiah 6) and Ezekiel’s vision of the divine throne-chariot (Ezekiel 1), one psalm in particular was the subject of much speculation concerning God’s heavenly servants, Psalm 104. There, describing God’s mighty works at the creation, the Psalmist relates that God—

עשה מלאכיו רוחות משרתיו אש להט

These words (104:4) seemed to be saying something about the very nature of but precisely what ,(משרתיו) and servants (מלאכיו) God’s angels or messengers they were saying was a matter of some debate in ancient times, since the syntax of this verse is altogether ambiguous. One could read it as asserting that God or—an ,עשה מלאכיו רוחות ,makes (or at one time made) His angels into winds equally possible translation—that He makes/made the winds (or “spirits” for carries both senses) His angels or messengers. As for רוחות that matter, since /this could mean that God makes ,משרתיו אש להט ,the second half of the verse made His servants into flaming fire, or vice versa, that He makes/made flaming fire into His servants—or, for that matter, it could be read as an independent assertion, “His servants are a flaming fire.” Sorting through these various alter- natives thus posed a major challenge to interpreters. It is striking how these different interpretations seem to have coexisted for centuries. Even today, modern translators are quite divided as to how to trans- late this verse. On the one hand, the New Jewish Publication Society renders it, “He makes the winds His messengers, fiery flames His servants.” Similarly, the New Revised Standard Version translates: “You make the winds your messengers, fire and flame your ministers.” On the other hand, the English Standard Version offers, “He makes his messengers winds, his ministers a flaming fire.” Similarly, the World English Bible reads: “He makes his messengers winds; his servants flames of fire.” These both echo the King James Version, which reads:

“Who maketh His angels spirits, His ministers a flaming fire.” (Note, however, (”.is rendered here as “spirits” rather than “winds רוחות that Greek and Latin syntax easily duplicated the ambiguous syntax created by the juxtaposed direct objects of the Hebrew version. Thus, the Septuagint translation reads:

ὁ ποιῶν τοὺς ἀγγέλους αὐτοῦ πνεύματα και τοὺς λειτουργοὺς αὐτοῦ πῦρ φλέγον1

This certainly might be read as if it were saying that God turns His angels into spirits/winds and His servants into fire, but one might equally well trans- late the Greek as: “He who makes spirits His messengers and flaming fire His ministers.”2 The same ambiguity accompanies Jerome’s Vulgate (it is virtually the same in his Gallican Psalter):

Qui facis angelos tuos spiritus et ministros tuos ignem urentem

The ambiguous syntax of these translations may well duplicate that of the original Hebrew, but it hardly answered the question raised by this verse: Are God’s servants made out of wind and fire, or are they turned into wind and fire—or does the verse hint at yet some other meaning?3 Faced with this sort of ambiguity, a common recourse of ancient scholars was to look elsewhere

1 Ps 103:4 LXX. Another ms. form reads πυρòς φλόγα; this is the form cited in Heb 1:7. See also G. Steyn, “ ‘Notes on Ps 101 (LXX) and Ps 103 (LXX) in Hebrews 1 in the Light of Evidence from the Dead Sea Scrolls and Papyrus Bodmer XXIV,” Acta Patristica et Byzantina 20 (2010): 341–59. 2 Cited from: A. Pietersma, A New English Translation of the Septuagint: The Psalms (New York: Oxford University Press, 2000), 101. 3 As noted, the two halves of the verse need not necessarily be treated in the same way: one could maintain that the “default” state of God’s servants is to serve before Him as fire—that they are, in that sense, made of fire—but when sent on a mission they take the form of winds. This might well accord with Isaiah’s vision of the seraphim that surround the heavenly throne (6:2, 6). While this word elsewhere seems to describe a kind of serpent (see Numbers 21 as well as Deut 8:15 and Isa 14:29 and 30:6)—perhaps because of the burning sensation of its bite—in Isaiah 6 one of the seraphim touches the prophet’s lips with a burning coal, which no doubt led some to conclude that they were “seraphim” in the sense of “burners” -burn”). This would fit as well with Ezekiel’s sight of the “flashing fire” (1:4) sur“ ,שר''פ from) rounding the divine throne-chariot, as well as the further references to “the fire” (1:5), “burning coals of fire, like torches” (1:13), and “the fire’s glow, with lightning coming out from the fire” (1:13). Note in this connection 3 En. 26: “Serapiel is the name of that angel. . . . Why is his name Serapiel? Because he is in charge of the seraphim and the seraphim of flame are committed to his care. . . . Every day they take the tablets from Satan’s hand and burn them 136 Kaduri in Scripture for adjudication. Here, the researches of rabbinic and earlier exegetes frequently coincide: not only do they agree on the general principle that explaining a biblical verse often involves considering more than the words of the verse in question, but they often refer to the same, apparently unrelated passages from Scripture in order to clarify that verse’s meaning.

1 The Eden Connection

So it was that, among both groups of exegetes, the reference in Psalm 104:4 to a flaming fire came to be connected to something apparently quite unrelated, God’s action after Adam and Eve had sinned in the Garden of Eden: “And He [God] banished the man and stationed east of the garden of Eden the cheru- to guard the way [להט החרב המתהפכת] bim and the flame of a turning sword to the tree of life” (Gen 3:24). No doubt the appearance of the relatively rare flame”) in both this passage and Ps 104:4 suggested the connection“) להט word of the two. If the meaning of Ps 104:4 is (to cite one of the modern translations above), “He makes His messengers winds, His ministers a flaming fire,” then perhaps that is what God did after banishing Adam and Eve from the Garden: some cherubim (understood to be a kind of angel) were stationed at the entrance to the Garden, along with another of God’s servants, this “flame of a turning sword”; both were charged with blocking the entrance to the Garden. Such an understanding may be hinted at in 1QHa Thanksgiving Hymns 16:11–12:

And You, God, have protected its [i.e., Eden’s] fruit with the mystery of ,and holy spirits and the flame of a turning fire [גבורי כוח] mighty ones lest anyone enter the source of life and [be] with the trees of eternity.4

רוחות קודש ולהט) ”The combination “holy spirits and the flame of a turning fire of Ps 104:4. The phrase אש להט and רוחות clearly derives from the (אש מתהפכת “mighty ones” is borrowed from a different psalm, Ps 103:20, where it refers to angels: “Bless the Lord, O His angels, mighty ones who do His bidding.” If the author of this verse from the Thanksgiving Hymns asserted that these “mighty

in the blazing fire that stands opposite the high and mighty throne” (translation is that of P. S. Alexander, OTP 1:281 [alt.]). See further below. 4 The text is given according to Qumran Cave 1.III: 1QHodayota, with Incorporation of 1QHodayotb and 4QHodayota–f (ed. H. Stegemann with E. Schuller; trans. C. Newsom; DJD 40; Oxford: Clarendon, 2009). Here and throughout the paper, translations are my own unless otherwise indicated. Windy and Fiery Angels 137 ones” were stationed at the entrance to Eden, and he then went on to mention it would ,אש להט and רוחות ,specifically the two items referred to in Ps 104:4 seem that these two terms, “spirits” and “flame,” refer to the form that God’s angels had at the time that they were charged with guarding the entrance to Eden (Gen 3:24): some of them were spirits, and at least one had the form of a flame of turning fire. In further support of this understanding is the fact that of (להט) the author of this passage has transformed the Genesis phrase “flame a turning sword” into “flame of a turning fire”—no sword at all! This turning fire, it would seem, is none other than an angel whose fiery form would dis- courage intruders from entering the garden. Such an approach is supported by two passages from rabbinic works that comment on this same Genesis narrative:

“The flame of the turning sword,” is so called in keeping with [the verse] ”turning“ המתהפכת [His servants a flaming fire” (Ps 104:4) [It says here“ because they [the angels] do indeed turn: sometimes [they are] men, sometimes women, sometime winds/spirits, sometimes messengers. (Gen. Rab. 21:9)5

In Exod. Rab. 25:2 this list is expanded, and the element of transformation is extended in the process:

Sometimes He makes them [the angels] in the likeness of women, as it says (Zech 5:9), “Behold two women came out with the wind in their wings”; sometimes in the likeness of men, as it says (Gen 18:2) “And behold three men”; sometimes He makes them into winds, as it says (Ps 104:4) “He makes His angels into winds” and sometimes He makes them into fire, as it says “And his servants a flaming fire” (Ps 104:4).6

Thus, for some interpreters, the connection of Ps 104:4 to Gen 3:24 supported the idea that the former verse means that some of God’s angels have, or take, -might have contrib המתהפכת the form of wind or fire. In addition, the word uted to the notion, explicit in Gen. Rab. 21:9, that angels are essentially protean beings whose form is readily changed.7 This understanding accords well with another Second Temple text, 4 Ezra:

5 Bereschit Rabba (ed. J. Theodor and C. Albeck; 3 vols.; Jerusalem: Wahrmann, 1965), 1:203. 6 Text: Midrash Rabbah (Vilna: Romm, 1878); my translation. 7 Further on this theme: Gen. Rab. 78:1. 138 Kaduri

O Lord . . . whose throne is beyond measure and whose glory is beyond comprehension, before whom the hosts of angels stand trembling and at whose command they are changed to wind and fire. (8:22)8

2 The Fiery Throne

fire was, of these two, more ,אש להט and רוחות While Ps 104:4 speaks of both commonly found in the depiction of angels and other heavenly beings. This priority was due in part to the natural world: the heavens are full of stars, and from earliest times these were conceived to be celestial burners, divine beings whose light issued forth from the flaming radiance of their own bodies. Thus, -is construed, the notion that its “flam משרתיו אש להט however the syntax of ing fire” refers to these heavenly burners was no doubt widely assumed—and these divine beings, visible to the naked eye as stars, acquired a certain prominence thereby. In addition, the seraphim (a word that was itself understood as burners)9 in Isaiah 6 had offered a detailed picture of these heavenly beings and further located them in a specific place: they surrounded God’s heavenly throne. One early source to elaborate on such a picture of the heavenly throne was the Book of the Watchers section of 1 Enoch. When Enoch ascends to heaven in chapter 14, he sees “shooting stars and lightning flashes” on his way, and proceeds through “tongues of fire” until at last he approaches a “great house”:

And the ceiling was like shooting stars and lightning flashes, and among them were fiery cherubim, and their heaven was water, and a flaming fire encircled all their walls, and the doors blazed with fire. And I went into that house—hot as fire and cold as snow, and no delight of life was in it. . . . And I saw in my vision, and behold, another open door before me, and a house greater than the former one, and it was all built of tongues

8 Michael E. Stone has noted the connection of this passage to the late rabbinic text, Pirqe de-Rabbi Eliezer (which, however, contains clear traces of the earlier biblical apocrypha and pseudepigrapha). PRE 4 relates: “The angels, who were created on the second day, when they are sent by His word they become winds, but when they serve before Him, they are turned to fire, as it is written, ‘Who makes the angels winds and His servants flaming fire.’ ” See Stone, Fourth Ezra: A Commentary on the Book of Fourth Ezra (Hermeneia; Minneapolis: Fortress, 1990), 273. Translation of 4 Ezra is from Stone, ibid. The translation of PRE is my own. Cf. Midrash Tehillim on Psalm 104. 9 See above, n. 3. Windy and Fiery Angels 139

fire. . . . Its floor was of fire, and its upper part was flashes of lightning and shooting stars, and its ceiling was a flaming fire. And I was looking and I saw a lofty throne; and its appearance was like ice, and its wheels were like the shining sun, and its 10 were cherubim, and from beneath the throne issued streams of flaming fire. And I was unable to see. And the Great Glory sat upon it [i.e., the throne]. . . . Flaming fire encircled Him and a great fire stood by Him, and none of those about Him approached Him. (1 En. 14:11–13, 14–15, 17–20, 22)

The relentless reference to various manifestations of fire in this passage is quite remarkable.11 Somewhat later, the book of Daniel described a similarly fiery divine throne:

And as I looked on, thrones were set in place, and the Ancient of Days took His seat. His garment was like white snow, and the hair of His head its ;[שביבין די נור] was like lamb’s wool. His throne was tongues of flame A river of fire streamed forth before .]נור דלק[ wheels were blazing fire Him. Thousands upon thousands served Him. (Dan 7:9–10)12

Perhaps as a result of such passages (along with the natural-world causes already mentioned), the fiery angels came to enjoy a kind of “default” status: whatever else they are or may become, God’s heavenly creatures are in essence burners. Indeed, fire sometimes appears both as the material from which angels are made and, moreover, that which continues to characterize their appearance and function:

10 See on this reading G. W. E. Nickelsburg, 1 Enoch: A Commentary on the Book of 1 Enoch (2 vols.; Hermeneia; Minneapolis: Fortress, 2001, 2012; vol. 2 coedited by J. C. VanderKam), 1:258; all translations of 1 Enoch are from this edition. 11 See also the vision in 1 En. 71:1–2, 5–7, as well as further fire references in 17:1, 5 (“river of fire”), 39:7. In 1 Enoch and other sources, fire has a dual role: it not only surrounds the heavenly throne, but it serves to punish sinners (such as the Watchers). This aspect of fire is quite prominent in 1 Enoch; see 1 En. 10:6, 13; 18:11–13, 15; 21:7–10; 54:1–2, 5; 67:4–7; 90:24– 27; 98:3; 100:7, 9; 102:1; 103:8; 108:3–6. Among numerous manifestations of fiery punishment elsewhere, see CD 2:5; 1QpHab 10:5, 13; 1QS 2:8; 4:13; 1QM 11:10, and numerous other Qumran texts; also 2 Bar. 48:39, 43; 4 Ezra 7:36. One of the scriptural sources for this idea is Isa 30:33; see further my Traditions of the Bible (Cambridge, Mass.: Harvard University Press, 1998), 58. 12 Cf 4Q530 EnGiantsb ar 2:16–18, where, however, there is no apparent mention of fire. 140 Kaduri

And from the rock I [God] cut off a great fire, and from the fire I created the ranks of the bodiless armies—ten myriad angels—and their weapons are fiery and their clothes are burning flames. (2 En. 29:3 [version J])13

This passage does not specifically mention the fiery angels in the context of the heavenly throne, but that combination appears (inter alia) in the Slavonic Apocalypse of Abraham. There, Abraham ascends to heaven at the time of the “covenant between the pieces” (Gen 15:8–21). He sees a multitude of human- like creatures “changing in appearance and likeness, running and being transformed.”14 While his angelic guide is speaking to him, Abraham then sees what appears to be the place of the heavenly throne:

And while he [the angel] was still speaking, behold, a fire was coming toward us round about, and a sound was in the fire like a sound of many waters, like a sound of the sea in its uproar. . . . And I heard a voice like the roaring of the sea, and it did not cease because of the fire. And as the fire rose up, soaring higher, I saw under the fire a throne [made] of fire and the many-eyed Wheels, and they were reciting the song. And under the throne [I saw] four singing fiery Living Creatures of fire, singing. (Apoc. Ab. 17:1; 18:2–3)

Those Wheels and Living Creatures evoke the wheels and the four creatures seen by Ezekiel (Ezek 1:5–12), but along with them are multiple references to fire. The narrative continues:

And I saw on the seventh firmament, upon which I stood, a fire spread out, and light, and dew, and a multitude of angels, and a power of the invisible glory of the Living Creatures, which I had seen from above. . . . And I looked from the height where I stood to the sixth expanse; and there I saw a multitude of incorporeal spiritual angels, carrying out the orders of the fiery angels. (19:4–6)

13 This translation of version J is by F. I. Andersen, OTP 1:148. It may be that these angels are in which case the two halves of ,רוחות ,bodiless” in that they are spiritual beings, that is“ Ps 104:4 would seem to describe a single type of heavenly being. 14 This description is somewhat reminiscent of that of the protean angels seen earlier. The translation from ApAbr here is based on that of A. Kulik, Retroverting Slavonic Pseudepigrapha: Toward the Original of the Apocalypse of Abraham (SBLTCS 3; Atlanta: Society of Biblical Literature, 2004), 24–25, with minor modifications. Windy and Fiery Angels 141

Here, two types of angels are mentioned, the “spiritual” ones and the fiery ones. אש and רוחות This combination seems to owe its existence to the pairing of in Ps 104:4. But the two types are not equal: the “multitude of angels” on להט the seventh firmament, surrounded by fire and called “the fiery angels” in the last words of this passage, are quite literally superior to the spiritual angels, who are stationed one floor below them and subservient to the fiery angels’ commands.15 mentioned in tandem in Ps 104:4 also appear together in אש and רוחות The some Qumran texts, with both types surrounding the heavenly throne. Their relationship varies, however: they are either two distinct classes of angel or, as seen earlier, a single type of angel that sometimes takes one form, sometimes the other:

the spirits of the Holy of Holies [. . .] the Holy of Holies, the spirits of [בדני] in the form16 [רוחות אלוהים] God, the appearance of . . . the spirits round about. (4Q403 Songs of the Sabbath [להבת אש] of flashing fire Sacrifice 1 ii 7–9)

Likewise:

The holy angels come forth from between its glorious wheels in the like- ;of the Holy of Holies roundabout [רוחות] ness of fire, and the spirits streams of fire in the image of electrum.17 (4Q405 20 ii 9–10)

Note also:

No strength can stand before it [apparently, God’s anger], nor any hope before the rage [of our God’s wrath]. And who can endure standing

15 Note also the fiery angels seen above in 2 En. 29:3. In PRE 4 (above, n. 8), the angels are fiery when they serve before the heavenly throne, but turn into winds/spirits when they are sent on some mission. 16 On this meaning see C. Newsom, Songs of the Sabbath Sacrifice: A Critical Edition (HSS 27; Atlanta: Scholars Press, 1985), 235 (translation mine); E. Qimron, The Hebrew of the Dead Sea Scrolls (HSS 29; Atlanta: Scholars Press, 1986), 105; and N. Mizrahi, “A Body Refigured: The History and Meaning of Hebrew BDN,” JAOS 130 (2010): 541–49. in Ps שבלת may be related to שובולי The term .שובולי אש בדמות חשמל :The text reads 17 69:3 (in some Bibles 69:2), usually translated contextually as “stream,” “flood,” and the like. paths,” though that might seem a little less appropriate in“ ,שבילי The word may represent context. 142 Kaduri

[. . .] He will punish [אש להבה] before His angels, for with a flashing fire 4Q185 Sapiential Work 1–2 8–9)18) . . . ]רוחתיו[ His spirits

In this last passage, as frequently in 1 Enoch and later texts,19 the flashing fire of these heavenly creatures has a function quite separate from their presence around the heavenly throne: down on earth, their fire serves to punish wrong- doers.20 How the isolated reference to “His spirits” fits into the overall picture is, unfortunately, unclear in this fragment.

3 Present at the Creation

It is noteworthy that the passages alluding to Ps 104:4 that have been examined thus far make no explicit reference to the wider context of this verse, and this is quite in keeping with ancient interpretive practice. Usually, the interpreters’ gaze was set exclusively on a single verse or phrase, sometimes even when their proposed interpretations seemed to conflict with what immediately preceded or followed the verse or phrase in question.21 But this was not always the case with Ps 104:4. Sometimes, even without mentioning the wider context,

18 On the text see H. Lichtenberger, “Der Weisheitstext 4Q185—Eine neue Edition,” in The Wisdom Texts from Qumran and the Development of Sapiential Thought (ed. C. Hempel, A. Lange, and H. Lichtenberger; BETL 159; Leuven: Leuven University Press and Peeters, 2002) 127–50 (translation mine). The combination of “flashing fire” and “spirits” echoes Ps 104:4, but the lacunae leave the text tantalizingly ambiguous. After “He will punish” ,the text breaks off, so that this word may actually have been plural in form (ישפט) they will punish,” and thus refer to the angels. But on balance it would seem“ ,ישפטֹ]ו[ that the reference is to God, who punishes by means of “His angels,” who consist of both “flashing fire” and “His winds/spirits.” If this is correct, then it would suggest that Ps 104:4 is being interpreted, in keeping with the larger context of this verse, as referring to the creation of the angels: some were, and still are, made of fire, while others were, and still are, made of wind/spirit. See below. 19 Above, n. 11. 20 Note that 2 En. 29:3 (version J), cited above, presents these fiery angels as armed to punish with fire, “. . . and their weapons are fiery and their clothes are burning flames.” However, version A lacks reference to these weapons: “From the rock I cut off a great fire, and the fire I created all the armies of the bodiless ones, and all the armies of the stars and cherubim and seraphim and ophanim, and all these from the fire I cut out.” 21 See on this my “Two Introductions to Midrash” Prooftexts 3/2 (1983), 131–55; reprinted in Midrash and Literature (ed. G. Hartman and S. Budick; New Haven: Yale University Press, 1985), 77–103. Equally striking is the fact that two interpretations, one reading the verse in isolation and the other in context, sometimes coexisted. I have discussed the example of Windy and Fiery Angels 143

עשה ancient interpreters did clearly look to the verses surrounding the words and understood that context as referring to how מלאכיו רוחות משרתיו אש להט God created His messengers and servants at the time of the creation:

Bless the Lord, O my soul; O Lord, my God, You are very great; You are clothed in glory and majesty, wrapped in light as in a garment. [You] who stretched out the heavens like a curtain, roofed His upper chambers with the waters, who has made clouds His chariot, walking about on the wings of the wind, who made winds His angels and flaming fire His servants— He established the earth on its foundations, so that it shall never be shaken. (Ps 104:1–5)22

This whole passage alludes to God’s glorious deeds in making the world, stretching out the heavens and establishing the earth’s foundations. Reading Ps 104:4 in this context inevitably sheds some light on the verse’s apparent ambiguity since, in spite of some of the interpretations seen above, the broader context would suggest that at the time of the creation, the winds were made God’s angels and not vice versa. The reason is that the parallel syntax of the previous verse, “who has made clouds His chariot,” seems to assert that an element of the natural world (clouds) was taken by God and shaped into something of service to Him (His chariot). So too in verse 4, two more elements of the natural world, winds and flaming fire, are likewise said to be shaped into, or made as, something of service to Him, namely, His angels and servants. Thus, to quote another modern translation cited earlier, “He makes the winds His messengers, fiery flames His servants” (NJPS). The idea that God’s angels were created out of fire has already been seen in a passage previously cited:

Lev 19:17; see my “On Hidden Hatred and Open Reproach: Early Exegesis of Lev. 19:17,” HTR 80 (1987): 43–61. 22 Translation mine. 144 Kaduri

The same theme appears in a passage in 2 Baruch: 23

You who rule with great design the hosts [i.e., of angels] who stand before You, and command with reproof24 the countless holy beings of flame and fire, whom You created from the beginning, who stand around Your throne. (2 Bar. 21:6)

Neither of these brief passages mentions the spirits/winds of Ps 104:4. However, these appear together with the fiery angels in another part of 2 Baruch, amidst a prayer of Baruch that one scholar has recently characterized as a “doxology in praise of God the Creator of all,” which describes “how God created the universe and apportioned to all the elements their own place and number”:25

You remember the beginning that You made and do not forget the destruction that is to be. With frightening and formidable signs You command the flames, and they change into spirits/winds. And with a word You bring to life that which was not, and with mighty power You hold back that which has not yet come. (2 Bar. 48:6–8)

It would seem that the first line of this passage is a general assertion that is then exemplified in the two verses that follow. That is, “With frightening and formidable signs You command[ed] the flames . . .” is part of what God remembers in thinking of the “beginning,” His creation of the world, which included the creation of His own angelic servants. As for God’s not forgetting the destruction that is to be, this presumably is the Big Bang that is to mark the Endzeit, whose positive side will be exemplified by the resurrection of the dead, “You bring to life that which was not”; these end-time events are, however, being held back for now by God’s “mighty power.” If so, then 2 Baruch would seem here (as in 2 Bar. 21:6, cited above) to situate God’s making of the fiery angels in Ps 104:4 at the time of the creation—quite in keeping with the overall context of this biblical verse.26 But according to

23 Note the discussion of the dating of this text; see M. Henze, Jewish Apocalypticism in Late First Century Israel (TSAJ 142; Tübingen: Mohr Siebeck, 2011), 25–32. Translation mine. 24 R. H. Charles suggested replacing zeʿifuta (“indignation”) with zuʿamuta (“reproof”); APOT 2:493. Truly, neither seems quite right. 25 The description is taken from Henze, Jewish Apocalypticism, 49. 26 The tense-less participles of Ps 104:2–4 notwithstanding, these verses clearly describe God’s acts at the creation; cf. the remark of Henze cited above. As for the phrase in Windy and Fiery Angels 145

2 Baruch, the creation of the fiery and windy angels was not simultaneous. The .רוחות flames existed first; only later are/were they turned into

4 Fiery and Windy Angels in Genesis 1

If Psalm 104:4 was to be interpreted as referring to God’s shaping of His angels and servants at the time of the creation, interpreters had a problem: the great account of the creation of the world that begins the book of Genesis makes no mention of angels being created. This was a classic crux interpretum.27 Ancient readers were in agreement that the angels must have already been in existence by the end of that first week, since chapter 2 of Genesis begins, “And the heav- The last phrase .(וכל צבאם) ens and the earth were completed and all their host was taken to refer to the army of angels that serve before God on high; apparently, they had been created on one of the preceding six days. But when?28 One clue as to the time of their creation is found in Gen 1:2, which speaks ”.hovering over the surface of the waters (רוח אלהים) of the “spirit of God is understood as a collective noun, then it might well refer to all רוח If the word the spirits that were created by God to serve Him—they were all created on the first day. Such an understanding seems to underlie the account of the creation in the Book of Jubilees:

2:1 On the Lord’s orders the angel of the presence said to Moses: “Write all the words about the creation—how in six days the Lord God completed all his works, everything that He had created, and kept sabbath on the seventh day. He sanctified it for all ages and set it as a sign for all His works. 2:2 For on the first day He created the heavens that are above, the earth, the waters, and all the spirits who serve before Him, namely: the angels of the presence; the angels of holiness; the angels of the spirits of fire; the angels of the spirits of the winds; the angels of the spirits of the clouds, of darkness, snow, hail, and frost; the angels of the sounds, the

2 Bar. 48:6–8, “You command the flames, and they change into spirits/winds,” its wording עשה מלאכיו רוחות, משרתיו אש להט is apparently intended to combine the two halves of into a single assertion, although the syntax is somewhat sketchy: “He made His angels, spirits [who are] His servants, [out of] fire.” 27 See my discussion in Traditions of the Bible, 48–51, 74–76. 28 See on the development of this and related ideas in later sources, Yehoshua Granat’s contribution to this volume, “No Angels before the World? A Preexistence Tradition and Its Transformations from Second Temple Literature to Early Piyyuṭ.” 146 Kaduri

thunders, and the lightnings; and the angels of the spirits of cold and heat, of winter, spring, autumn, and summer, and of all the spirits of His creatures which are in the heavens, on earth, and in every (place). [There were also] the depths, darkness and light, dawn and evening which he prepared through the knowledge of his mind. 2:3 Then we saw His works and blessed him. We offered praise before him regarding all his works because he had made seven great works on the first day.29

Here, the “spirit of God hovering over the surface of the waters” in Gen 1:2 is רוחות בריותו] apparently interpreted to include “all the spirits of His creatures in 4Q216 Jubilees 5:9] which are in the heavens, on earth, and in every (place).” The angels, in other words, were all created on the first day, and among them, according to Jubilees, were the “angels of the spirits of fire; the angels of the עשה מלאכיו רוחות, משרתיו ,spirits of the winds.” This is a reference to Ps 104:4 these heavenly beings of fire and of spirit/winds were among those ;אש להט created on the very first day of the creation. But what is the significance of those puzzling, and apparently redundant, compounds in Jubilees, “the angels the “angels of the spirits ,(מלאכי רוחות האש of the spirits of fire” (presumably in 4Q216 5:6) and so forth? What could מלאכי רוחות הע]ננים[) ”of the clouds “angels of spirits” be—are not “spirits” themselves simply a kind of angel or divine emissary (as is abundantly illustrated in the various Qumran and other texts cited above)? Jubilees’ pleonastic phrases seem to reflect a particular interpretation of Jubilees understands this to mean: “He .עשה מלאכיו רוחות ,our psalm’s wording

29 The translation is from J. C. VanderKam, The Book of Jubilees (2 vols.; CSCO 510–511; Scriptores Aethiopici 87–88; Louvain: Peeters, 1989), 2:7–8. Note that 4Q216 5:9 probably did not read, as in the Ethiopic text, “all the spirits of his creatures which are in the heavens, on earth, and in every (place),” because that would not sufficiently fill out the line in Hebrew. VanderKam and Milik propose: “all the spirits of His creatures which He made in the heavens and which He made on earth and in all the abysses. . . .”; see J. C. VanderKam and J. T. Milik, “A. Jubilees,” in Qumran Cave 4.VIII: Parabiblical Texts, Part 1 (ed. Harold Attridge et al.; DJD 13; Oxford: Clarendon, 1994), 13–16. On this passage see also R. Sollamo, “The Creation of Angels and Natural Phenomena Intertwined in the Book of Jubilees: Angels and Natural Phenomena as Characteristics of the Creation Stories and Hymns in Late Second Temple Judaism,” in Biblical Traditions in Transmission: Essays in Honour of Michael A. Knibb (ed. C. Hempel and J. M. Lieu; JSJSup 111; Leiden: Brill, 2006), 273–290 (273). Windy and Fiery Angels 147

spirits”; that is, at the very beginning [מלאכיו] made30 [some of] His angels of creation, some of the angels that He made were designated to be the spirits that will control fire, the winds, the clouds, and so forth. The existence of such spirits controlling the weather and other earthly matters was particularly important to Jubilees’ author, since it explained how lightning and thunder, along with winds, rain, and other phenomena of nature could occur on the Sabbath, when, according to Jubilees, God and the highest angels were taking their Sabbath rest.31 God arranged for these lower angels, the spirits of fire and so forth, to be in charge of the forces of nature while the world was, as it were, running on automatic pilot. Here, it seems, the spirits that control these functions were created as part of that great army of angels that God created on the in Ps 104:4 plays a double role in Jubilees: these רוחות first day. If so, the word spirits are the angelic emissaries who control elements of the natural world, but are controlled by ,אש להט they are also the winds which, along with the psalm’s the angelic emissaries, who also are in charge of the “the clouds, . . . darkness, snow, hail, and frost,” and so forth. A somewhat similar approach is attested in a later rabbinic text, Exodus Rabbah:

soldiers for himself, strong and [מכתיב] A king] of flesh and blood levies] stout men wearing helmets and armor and weaponry, but the Holy One levied as His soldiers ones that are invisible, as it is said, “Making winds His messengers”; the wind goes forth and the lightning follows it, as it is said, “His servants a flaming fire.” (Exod. Rab. 15:22)32

as ʿasah instead of MT ʿoseh, though either would do עשה It may be that Jubilees is reading 30 to describe past action. 31 See L. Doering, “The Concept of the Sabbath in the Book of Jubilees,” in Studies in the Book of Jubilees (ed. M. Albani, J. Frey, and A. Lange; TSAJ 65; Tübingen: Mohr Siebeck 1997), 179–205; J. Kugel, “4Q369 ‘Prayer of Enosh’ and Ancient Biblical Interpretation,” DSD 5/2 (1998): 119–48; and idem, Traditions of the Bible, 88–90. 32 Note the slightly different version in Midrash Tanḥuma (printed edition), Ḥayye Sarah 3 (Jerusalem: Merkaz ha-Sefer Publishing, p. 79): “A king of flesh and blood surrounds himself with soldiers and girds them with weaponry and armor; but God’s soldiers are invisible, as it is said, ‘He makes His angels winds,’ and the wind brings along lightning, as it is said, ‘His servants a flaming fire.’” Note that the angels in 2 En. 29 are likewise “bodiless,” hence invisible (above, n. 13), as are the “incorporeal spiritual angels” in Apoc. Ab. 19:6. 148 Kaduri

This divine act of recruitment involved enlisting invisible soldiers, that is, the it, too, is ,להט אש winds, which God will use for His own purposes; as for the part of the natural world, the lightning that tags along with the winds.

5 Conclusions

This look at some of the interpretive traditions arising from Ps 104:4 leaves open the straightforward question, “But what does it really mean?” That particular conundrum was never solved. On the contrary, from ancient times to the present, the verse’s six words have been interpreted in a dizzying variety of ways: 1) God makes (or at one time made) His angels winds, or spirits, and His servants blazing fire, such as the blazing fire that guarded the entrance to Eden; or else 2) God makes/made the winds/spirits into His messengers and fire into His servants—this last concurring with the age-old picture of fire as the primary form of God’s heavenly servants surrounding His throne, in fact, the material out of which they were originally fashioned; along with these 3) the action described in Ps 104:4 was sometimes understood to be an oft-repeated one, as God chooses to transform these protean beings when the need arises; or, on the contrary, 4) this verse was held to describe a one-time act of creation, one that was alluded to in the phrase “spirit of God” mentioned in of this verse were sometimes understood as רוחות Gen 1:2; all the while, 5) the ;as a brilliant, heavenly fire אש להט spiritual beings, “bodiless angels,” and the or perhaps, on the contrary, 6) these terms refer to earthly entities, the winds and the lightning that are part of the natural order down here. If this survey has left the basic question of meaning unresolved, it is hoped that these excerpts from the biblical apocrypha and pseudepigrapha, the Qumran writings, and various midrashic collections will serve to concretize one aspect of the overall connection between rabbinic writings and those that preceded them. Although rabbinic midrash was committed to writing long after the Qumran documents and the biblical apocrypha and pseudepigrapha were composed, it often demonstrates the same exegetical methods as well as the same, or similar, understandings of a biblical text. Such shared motifs usually point to a common store of exegetical traditions that was passed on orally from generation to generation and parts of which appeared first in one set of texts and then, centuries later, in another. In a few rare cases, one may hypoth- esize a more direct, literary relationship: the later text betrays a familiarity with an existing written text that preceded it. In yet other instances, it may be that resemblances between two texts are entirely coincidental: two exegetes, Windy and Fiery Angels 149 writing in different times and circumstances, arrived independently at the same interpretation of a verse or passage. Whatever the case, however, the comparison of midrashic motifs to those found earlier, in the Qumran scrolls and the biblical apocrypha and pseudepigrapha, often proves enlightening in both directions: while the earlier texts can sometimes tell us something about the origins and development of rabbinic interpretations, the rabbinic texts frequently make clear the full sense of a passing remark in an ancient pseude- pigraphon, or enable us to fill out the sense of a fragmentary line in a Qumran manuscript. Hellenistic Jewish Writers and Palestinian Traditions: Early and Late

Menahem Kister

One of the most puzzling and perplexing problems for the scholarship of postbiblical Jewish literature is the relationship between Hellenistic Jewish literature and rabbinic literature. The former, written in Greek, was produced primarily in the Diaspora; the latter is a successor to the Palestinian Jewish literature known to us from the writings preserved at Qumran, as well as from some apocryphal and pseudepigraphic literature. The problem of the relationship between the two corpora is crucial not only for students of authors like Philo, or for a general understanding of ancient Jewish culture; it is also of vital interest for students of ancient Jewish interpretation like myself, whose expertise is in midrashic literature. The compilations of rabbinic literature in which the midrashim occur were composed hundreds of years after the Hellenistic Jewish works that have come down to us; and yet, because the midrashim contain many ancient traditions, the possibility that rabbinic traditions may reflect interpretations such as those found in the writings of Philo cannot be dismissed offhand. Although both Philo and his Palestinian contemporaries, as well as their rabbinic successors, were engaged in interpreting the Bible and were faced with some similar exegetical problems, their writings often reveal different modes of thought, manners of expression, and cultural values. While Philo’s Greek education is easily recognizable, he probably did not know Hebrew, whereas it is still an open question whether the Greek of many of the rabbis was good enough to read Philo (or other Greek authors). While the existence of any direct relationship between Hellenistic Jewish authors and the rabbinic literature is doubtful, there are evident links between Jewish traditions preserved in Hebrew and Aramaic in rabbinic tradition and those included in the literature written in Greek.1 The philosophical interpretation of biblical passages by Philo and his predecessor Aristobulus, which owes so much to Greek philosophy, is quite remote from the concrete and fanciful world of the midrashim; yet the links

1 Such links are evident for works such as the Testament of Abraham, Joseph and Aseneth, and the Testament of Job; but this should be dealt with elsewhere.

© koninklijke brill nv, leiden, ��5 | doi ��.��63/9789004299139_008 Hellenistic Jewish Writers and Palestinian Traditions 151 between these Jewish philosophers and the rabbis are worthy of scrutiny, as we shall see.2 What might account for the affinities between Philo’s oeuvre and the writings of the rabbis? I do not venture to offer a solution to this problem. Indeed, I do not think there is a single solution: some of these affinities could be the results of exegetical problems in the biblical passages; sometimes it seems likely that rabbinic aggadah drew on a Hellenistic source; more often, however, it seems likely that Philo and other Hellenistic Jewish writers made use of ancient exegetical traditions that eventually found their way to rabbinic and Samaritan literature as well as to early Christian writings.

Although the huge differences among the diverse Jewish works of the Second Temple period and rabbinic literature should by no means be ignored, the common material—the traditions, motifs and trajectories that underlie much of these corpora—call for an integrative and panoramic study of the whole range of the exegetical and theological material included in these corpora. The following analysis of some statements related to creation in Hellenistic Jewish writings and in rabbinic literature illustrates the benefits of such a study of traditions that transcend the bounds of chronology, languages and cultures.3 Section 1 discusses two rabbinic traditions: according to one of them, Adam was created out of dust taken from the four corners of the world, whereas according to the other, Adam’s creation is related to the Temple. The first tradition may well be derived from a Hellenistic Jewish source; in the case of the second, a Hellenistic Jewish tradition appears to be a counterpart, or rather a spiritualized form, of the midrashic tradition. Section 2 deals with traditions that explain the significance of Adam’s creation subsequent to that of all the other creatures, in relation to the authority of humans over other creatures. While Philo confines human dominion to noncelestial creatures, an

2 Needless to say, affinities with the Hellenistic world and with midrash do not necessarily preclude one another; on the contrary, a Jew accustomed to Hellenistic concepts and culture would more willingly adopt traditions that could be expressed in Hellenistic terms, or that looked familiar to him; on the other hand, midrashic traditions emerged in the Hellenistic and Roman periods, and should certainly not be conceived of as unrelated to the surrounding cultures. 3 To be sure, this view is not novel; it is the premise of such enterprises as L. Ginzberg’s The Legends of the Jews (trans. H. Szold; 7 vols.; Philadelphia: Jewish Publication Society, 1909– 1938); J. L. Kugel’s Traditions of the Bible: A Guide to the Bible As It Was at the Start of the Common Era (Cambridge, Mass.: Harvard University Press, 1998); and other studies. I hope that the detailed scrutiny of specific traditions offered in the present article will illuminate some unexplored aspects of the interrelationships among these corpora. 152 Kister early piyyuṭ asserts that everything was created for the rule of Adam, including the angels and the luminaries; the latter view reflects an ancient tradition, as can be demonstrated by the Life of Adam and Eve, according to which angels are subordinate to Adam and his offspring. Section 3 treats passages from the Hellenistic Jewish writers Aristobulus and Philo where the primeval light of creation is identified with preexistent wisdom, as well as with the light of the Sabbath. The first idea occurs in a variety of Jewish works, while the association of the Sabbath with light (sometimes primeval light) has interesting midrashic parallels. The integrative study of these passages raises general and fundamental questions concerning the transformation of traditions both before and after their incorporation in rabbinic compilations, as well as their relationship to other, more spiritualizing traditions.

1 Adam, Universalism, and the Temple: Hellenistic, Rabbinic, and Samaritan Traditions

1.1 Adam and the Four Corners of the World According to a well-known rabbinic aggadah, Adam’s stature extended from east to west and from north to south.4 As has been noted,5 this tradition is apparently dependent on the Greek anagram of the name Adam as indicating east, west, north, and south: ἀνατολή, δύσις, ἄρκτος, μεσημβρία; this Greek tradition is documented in Sibylline Oracles 3:24–27, as well as in 2 En. 30:13.6 Hellenistic Jewish sources apparently do not understand this tradition to refer to Adam’s stature;7 rather, the notion that in Adam’s name the four cor-

4 Gen. Rab. 8:1 [55] = 21:3 [199] = 24:2 [230]; Lev. Rab. 14:1 [297] = 18:2 [401]. See also b. Ḥag. 12a. 5 A. A. Orlov, The Enoch–Metatron Tradition (Tübingen: Mohr Siebeck, 2005), 239–42. See next note. 6 In Enoch–Metatron, Orlov interprets the statements in 2 Enoch and in the Sibylline Oracles as referring to Adam’s huge dimensions (with which he deals on pp. 236–39). In my opinion, however, the context of these two sources does not speak for such a reading, nor is it an expected way to express Adam’s gigantic stature. But see also below, n. 7. 7 In the rabbinic midrash, “east, west, south, and north” indicate the dimensions of Adam’s stature. This midrash apparently combines the tradition from the Sibylline Oracles and 2 Enoch with a separate ancient motif, the immense size of Adam. Note, however, the wording of a Gnostic myth according to Irenaeus, Haer. 1.30.6: “[they] formed a man of immense size, both in regard to breadth and length” (the translation is that of A. Roberts, ANF 1:355; for the text, see Contre les Hérésis Livre I [ed. A. Rousseau and L. Doutreleau; SC 264; Paris: Cerf, 1979;] 2.370), cited by Orlov (238 n. 124). The emphasis on “breadth” might point to a combination of traditions similar to that found in rabbinic literature, at an early date. Hellenistic Jewish Writers and Palestinian Traditions 153 ners of the earth are combined underlines his universality. The same idea is expressed in rabbinic literature by the assertion that Adam’s body was created of dust from the four corners of the earth.8

1.2 Adam’s Creation and the Temple A rival tradition, namely that Adam was created from the dust, not of all the earth, but rather of the most particular place on earth—the place of the Temple—occurs in rabbinic literature as well, in the name of an Amora: “Adam was created from the place of his atonement.”9 According to a Samaritan homily from late antiquity, Adam’s body was created from the dust of holy place on the “Good Mountain,” i.e., Mount Gerizim:10

8 Compare: “In some of these traditional texts, Adam is created from sods of earth taken from different parts of the world; in others, he is made from different elements. Often these two ideas are connected to the further idea that Adam’s name is derived from the cardinal points of the compass. . . . Adam is associated . . . with the totality of the land occupied by human cultures” (G. Macaskill, “The Creation of Man in 2 [Slavonic] Enoch and in Christian Tradition,” in Congress Volume Ljubljana 2007 [ed. A. Lemaire; VTSup 133; Leiden: Brill, 2010], 399–422, [401–2]). Macaskill is referring primarily to b. Sanh. 38a–b. According to a statement there, attributed to Rav (early third century CE), the soil for Adam’s head was taken from Eretz Israel; that for his trunk was taken from Babylon; etc. According to another saying in the same sugya, attributed to Rabbi Meʾir (second century CE), “Adam’s dust was gathered from all the world” (b. Sanh. 38a). This aggadah is also attested in later sources, primarily Pirqe de-Rabbi ʾEliʿezer and Targum Ps.-Jonathan (both merge this aggadah with a different one, according to which Adam’s dust was taken from the place of the Temple (to be discussed in the second unit of the present section); see L. Ginzberg, Legends of the Jews, 5:73 n. 16). According to Pirqe de-Rabbi ʾEliʿezer ch. 11, the proof text for Adam’s creation from the dust of the whole earth is the verse “. . . until you return to the earth from which you were taken” -Gen 3:19); all human beings, wherever they are bur ;עד שובך אל האדמה כי ממנה לקחת) ied, are buried in the land from which Adam’s dust was taken (concerning this proof text see below, n. 20). See also M. Harl’s careful remark on the assertion of Philo in Alleg. Interp. 1.31 that “the earthly man is of scattered matter” (M. Harl, La Bible d’Alexandrie: La Genèse [Paris: Cerf, 1994], 101). Unless otherwise noted, translations of Philonic works quoted in this article follow F. H. Colson and G. H. Whitaker, Philo, with an English Translation (10 vols.; LCL; London: Heinemann; Cambridge, Mass.: Harvard University Press, 1929–1962). 9 Y. Nazir 7:2 (56b); Gen. Rab. 14:8 (132). 10 Tibat Marqe 2.44 (ed. Z. Ben-Ḥayyim; Jerusalem: The Israel Academy of Sciences and Humanities, 1988), 142–43 (in Hebrew; English translations of the work in this essay are my own). According to Ben-Ḥayyim, the Aramaic in this section of Tibat Marqe indicates that it was composed either prior to the Arab conquest of Palestine or in the period immediately following the conquest (“Introduction,” p. 23). The passage is recorded in Ben-Ḥayyim’s Tibat .ש and ק two very different recensions, represented in manuscripts 154 Kister

“The Lord God formed man of dust from the ground” (Gen 2:7)—from the good dust of the Good Mountain; and Adam’s house (i.e., humanity)11 of all the creatures; and (דמע) is praised by the Lord, for it is terumah every terumah is delivered to holiness, and unholiness has no portion in it. Every terumah that is set aside by humans to the Holy One gains in So also Adam’s body, (which was) the .(מתחיל ומתיקר) power and in honor terumah of all Creation. For that reason it was taken from a *place* that was *sanctified.*12

Philo’s assertion concerning the creation of Adam may be considered a spiritualized form of this tradition; he states:

It is not likely that God took clay from any part of the earth which He happened to come across . . . but rather . . . out of pure matter, the purest, . . . which was especially suited for the construction. For it was built as a home (οἶκος) or holy temple for the rational soul, which it was to carry around as the most god-like of images.13

From where did God take Adam’s clay? It was not taken randomly from just any physical location, Philo says; in the following words, however, he refers (unlike the midrash) not to a physical locale, but rather to the substance (“purest matter”) out of which Adam’s body was fashioned.14 Finally, the text designates

was separately published and translated into Hebrew by ק while ,ש Marqe is based on Ben-Ḥayyim: “A Fragment of the Memar Marqe in an Unknown Version,” Teʾudah 3 (1983): 121–37, esp. 126, 131 (in Hebrew). Adam’s creation from the dust of Mount Gerizim is stressed in both of them. literally: “Adam’s house”) as signifying “humanity” see A. Tal, A Dictionary) ביתה דאדם For 11 ,ק of Samaritan Aramaic (Leiden: Brill, 2000), 88. Note, however, in contrast to recension Adam’s house is praised by the Lord”), the“) ביתה דאדם מתרבי ביד קשטה :which reads Adam’s body was“) גויתה דאדם אתברת ביד אלה :reads ש parallel passage in recension created by God”; was the “house” interpreted as the “body”?). Both recensions refer to Adam’s body in the following sentence. The two words in asterisks are emended according .ק My translation, following recension 12 to the editor’s note; see Ben-Ḥayyim, “A Fragment,” 126, 131 n. 3. 13 Philo, On the Creation of the World §137; trans. D. T. Runia, Philo of Alexandria: On the Creation of the Cosmos according to Moses (PACS 1; Leiden: Brill, 2001), 83. 14 On the Creation of the World §137 is the continuation of the argument in §136: “Since the earth was newly established, having just appeared when the great body of water which was named sea was separated out, it was the case that the material for those things that came into being was unmixed and undefiled and pure, as well as receptive and easy to work with.” Note, however, that according to Genesis 2 God formed the human being from Hellenistic Jewish Writers and Palestinian Traditions 155 the human body as the temple of the soul, thus incorporating a “spiritualized” version of the midrashic Temple motif.15 Similarly, though in a very different religious context, the Samaritan passage cited above combines the idea of the purity of Adam’s body with the notion of his creation from the dust of the holy place.16 In a late midrash,17 Pirqe de-Rabbi ʾEliʿezer, the “purity” of the holy וגבל ולש עפרו של אדם הראשון במקום טהור place of the Temple is emphasized:18 שבראו במקום He kneaded the dust of Adam[ʼs body] in a holy place” (ch. 11);19“ that He created him [= Adam] in a holy place, namely“ ,טהור ובמקום בית המקדש in the place of the Temple” (ch. 12).20

the ground right after “the whole face of the ground” had been soaked in the water of the Gen 2:6–7; see also Gen. Rab. 14:1 [Bereschit Rabba (ed. J. Theodor and) אד mysterious C. Albeck; 3 vols.; Berlin: Popeloyer, 1912–1929), 1:126], where the figure of kneading the is cited in the name of a sage probably from ,אד first human being, using the water of the the end of the second century CE). Philo, however, interpreted these verses differently; (in 2:6 אד based on the Septuagint, he interpreted the “spring” (Greek πηγή for Hebrew as localized in the earth, not as covering its surface (§§131–133); thus, when he envisioned the earth as saturated with water, he had to return to the third day of creation, when the earth was covered by the waters of the sea. The possibility that Philo’s explanation in this case is a reworking of an already existing tradition should be considered. 15 “The imagery emphasizes the high value that Philo attaches to the body in this context . . . To my knowledge this is the only text in Philo where the body is called a temple” (Runia, On the Creation, 335). As briefly noted by Runia, similar imagery is known from Paul: “Do you not know that your body is a temple of the Holy Spirit within you, which you have from God” (1 Cor 6:19). 16 According to a passage in the Palestinian Talmud, parallel to Gen. Rab. 14:1 (above, n. 14), y. Shabb. 2:6 [5b]), an image) ”(חלה טהורה) Adam is spoken of as “a pure dough set aside .in the Samaritan passage cited above (דמע) similar to terumah 17 On the merging of two distinct traditions in this midrash, see above n. 7. 18 As mentioned above, the idea that Adam was created on the site of the (future) Temple is attested in much earlier midrashim. The remarkable feature of PRE is the wording, “a holy place.” 19 The Hebrew text cited is the result of my philological considerations on the basis of a synopsis of the work by Dr. Eliezer Treitl (in preparation). For the textual tradition of Pirqe de-Rabbi ʾEliʿezer, see the groundbreaking work of E. Treitl, Pirke de-Rabbi ʾEliʿezer: Text, Redaction and a Sample Synopsis (Jerusalem: The Hebrew University Magnes Press and Yad Izhak Ben-Zvi, 2012) (in Hebrew). is (הר המוריה) In Pirqe de-Rabbi ʾEliʿezer, ch. 20, Adam’s creation on the Temple Mount 20 Gen 3:23). This verse is also) לעבד את האדמה אשר לקח משם inferred from the verse adduced in the longer recension of the passage of Tibat Marqe discussed above (n. 10), as a proof text for the view that Adam was created in Mount Gerizim (the verse is absent from the shorter recension of this work). As noted above (n. 8), a similar verse (Gen 3:19) 156 Kister

Reading together these Philonic, rabbinic, and Samaritan passages thus reveals a common tradition; apparently all three make use of the same ancient interpretive complex. Significantly, Philo, the earliest of the three, reshapes his material in a distinctively different direction.21 Thus, integrative study leads here to the surprising (but not unusual) observation that texts in later corpora may guide us to earlier forms of interpretive traditions. The benefit of this observation for the study of each of these texts is evident.

2 Adam’s Superiority in Philo, the Life of Adam and Eve, Celsus, Midrash and Piyyuṭ

Anthropocentric views of the cosmos were a matter of dispute between Jews and pagans in antiquity. Tiberius Julius Alexander, a nephew of Philo who had abandoned Judaism, had argued: “[It is] not only the human mind [that] was made after the (divine) image and received a great honor separate and distinct from that of other (creatures)”; but rather, “God gave common advantage to all (creatures).”22 The contention of Tiberius Julius Alexander is based on the argument that animals have rational mind no less than human beings. Philo dedicates an entire treatise (Whether Animals Have Reason)23 to respond to his nephew’s argument and prove the exclusive status of human beings. The pagan Celsus (probably second century CE) argued at length against the belief shared by Judaism and Christianity that human beings, made in God’s image, have dominion over all the cosmos, which exists for their benefit.24 Origen’s rejoinder to him was that “The Creator has not made these things

serves as a proof text for the competing view, namely that Adam’s dust was gathered from the whole world. 21 Ginzberg (Legends of the Jews, 5:73 n. 16) views both Philo and the midrash as evidence for ancient legends. Contrast J. Levison, Portraits of Adam in Early Judaism (JSPSup 1; Sheffield: JSOT Press, 1988), 74: “That Op. 134–50 expresses Philo’s own concerns so explicitly is evidence that he does not here appeal to a general Adam myth, but creates his own interpretation by combining biblical elements with Greek concepts. . . .” 22 Philo, Animals §16 (A. Terian, Philonis Alexandrini de Animalibus: The Armenian Text with an Introduction, Translation, and Commentary [SHJ 1; Chico, Calif.: Scholars Press, 1981, 73]). This phrase seems to be the only (negative) allusion to a biblical verse (Gen 1:26) in the whole treatise. 23 See n. 22. 24 Origen, Contra Celsum 4.23; see also the lengthy discussion of Celsus and Origen in 4.74– 99. See Contra Celsum: Translated with an Introduction and Notes (trans. H. Chadwick; Cambridge: Cambridge University Press, 1965), 199–200; 242–63. Hellenistic Jewish Writers and Palestinian Traditions 157 for the lion or eagle or dolphin, but has created everything for the rational being.”25 Students of Philo and Origen have noted that the pagan opponent in each of these texts has used the arguments of the Platonic New Academy, whereas the Jewish or Christian respondent has marshaled Stoic arguments in reply.26 Pagan–Christian polemics of the first centuries CE continue more ancient polemics between Jews and pagans.27 The problem at stake was obviously the status of human beings, an issue which occupied thinkers in antiquity. However, fundamental views of Judaism and Christianity might also be involved: Celsus’s argument is explicitly linked with a vehement attack on Jewish and Christian notions of the election of their own groups.28 Although Tiberius Julius Alexander’s argument against anthropocentricity is expressed as a purely philosophical debate, one might infer from Celsus’s argument that similar issues could be at stake in this case as well.

Philo’s treatise, Whether Animals Have Reason, was utilized by Peder Borgen for the interpretation of the arguments concerning the status of Adam in On the Creation of the World,29 arguments that have close affinities with statements in rabbinic literature. Both Philo and the rabbis emphasize that the fact that— according to Genesis chapter 1—Adam was created last of all should not be taken as an indication of the inferiority of human beings. Philo gives four reasons for Adam’s being created last:

25 Origen, Contra Celsum 4.99. 26 See Chadwick’s notes, and his introduction (Contra Celsum, x–xi), as well as H. Chadwick, “Origen, Celsus, and the Stoa,” JTS 48 (1947): 34–47, especially 36–37; see also Terian, De Animalibus, 46–53. 27 Terian (De Animalibus, 47–48) suggests that this treatise might have had an apologetic dimension. This view is corroborated by comparison with Celsus’s arguments. For the similarity between the Jewish view that the world was created for the sake of human beings and the view of the Stoa, see P. Borgen, “Man’s Sovereignty over Animals and Nature according to Philo of Alexandria,” in Texts and Contexts: Biblical Texts in their Textual and Situational Contexts: Essays in Honor of Lars Hartman (ed. T. Fornberg and D. Hellholm; Oslo: Scandinavian University Press, 1995), 369–89 (379). 28 This is quite clear in Contra Celsum 4.23. 29 See Borgen, “Man’s Sovereignty,” 376-77. Note, however, that the idea that Adam, as the image of God, is a rational being, is central for the polemic with pagans (see above), but marginal for Philo’s arguments concerning his being created last (although this idea is mentioned in On the Creation of the World §77). 158 Kister

(1) For [Adam] He made everything ready in advance . . . Just as those who give a banquet do not invite their guests to the meal until all the preparations for the feast have been completed.30 (2) When the human being first came into existence, he found all the equipment required for living . . . if they follow the example of the original ancestor of their race, they will lead a life without toil.31 (3) God . . . proceeded to make heaven as its beginning and human being as its end [i.e., symmetry in creation].32 (4) The human being had to emerge as last of all creatures so that, when he . . . appeared before the other living beings, he would instill astonishment in them.33 On first seeing him, they would be amazed and would worship him as their natural director and master . . . (God) appointed him king of all creatures in the sublunary realm. . . . All the mortal creatures in the three elements—earth, water and air—he sub-

30 On the Creation of the World §§77–78. The translation is based on that of Runia, On the Creation, 66–70. 31 On the Creation of the World §79. This is clearly a variation of the preceding reason. A similar idea is expressed in a poem written in late antiquity, ʾAz Βe-ʾEin Kol, in which the creation of Adam is described: “He willingly made a human being in (His) image, and set him to govern all creation, and commanded him to observe (His commandments). Fat and milk, cream and honey, he shall eat and enjoy and say praise to His creator. No wrath shall come upon him, or destruction trouble him; ‘I will protect his life if he observes My חפץ ועשה אדם בצלם והשליטו בכל מעש וצוה הוא ]=וצוהו[ לשמור, חלב וחלב ” ’,laws חמאה ודבש יאכל ויתענג ויאמר הלל לבוראו, חרון בל יבואינו שוד בל ידאיגנו אעצים חייו :ʾAz Βe-ʾEin Kol, lines 191–196; Y. Yahalom, ed., Priestly Palestinian Poetry) אם חוקיי ישמור A Narrative Liturgy for the Day of Atonement [Jerusalem: The Hebrew University Magnes Press, 1996], 83 [in Hebrew]; the translation is based on M. D. Swartz and J. Yahalom, Avodah: An Anthology of Ancient Poetry for Yom Kippur [The Penn State Library of Jewish Literature; University Park, Pa.: Pennsylvania State University Press, 2005], 126). For the general idea, compare, e.g., t. Qiddushin 5:15 (ed. S. Lieberman, The Tosefta [5 vols.; New York: Jewish Theological Seminary of America, 1955–1988], 4:298); as well as the statement of Celsus in Origen, Contra Celsum 4.76; and see D. Flusser, “ ‘Have You Ever Seen a Lion Toiling as a Porter?’ ” in idem, Judaism of the Second Temple Period: Volume 2: The Jewish Sages and their Literature (trans. A Yadin; Grand Rapids: Eerdmans; Jerusalem: The Hebrew University Magnes Press, 2009), 338–39. 32 On the Creation of the World §82. 33 “Be fruitful and multiply, and fill the earth and subdue it; and have dominion over the fish of the sea and over the birds of the air and over every living thing that moves upon the earth.” (Gen 1:28). Compare: “Be fruitful and multiply, and fill the earth. The fear of you and the dread of you shall be upon every beast of the earth, and upon every bird of the air, upon everything that creeps on the ground and all the fish of the sea” (Gen 9:2); and see below, n. 38. Hellenistic Jewish Writers and Palestinian Traditions 159

jected to his rule,34 excluding the creatures in heaven because they have obtained a portion that is more divine . . . It is important, however, to be aware that the human being was not created last of all as an indication of inferior rank . . . The maker thus proceeded to fashion the human beings . . . like a governor acting on behalf of the first and great King.35

At least the first reason (conceivably some of the other reasons as well) is attributed by Philo to “those who have studied more deeply the laws of Moses and who examine their contents with all possible minuteness”; one may infer from this statement that by the first century the problem of Adam’s superiority had become an issue for many Jewish exegetes. The rabbis were also concerned with this question and tried to solve it. The multiple solutions suggested in rabbinic literature prove that this was a sore point for them as well. As has been noted, Philo’s first solution has an almost literal parallel in rabbinic literature; according to rabbinic sources, Adam was created last “so that he might enter the banquet immediately,” similarly to guests who are invited only after the building is ready and the meal is prepared.36 Philo’s fourth solution seems to have a parallel in Fourth Ezra, where it is stated that God “placed Adam as ruler over all the works that You have made previously” (4 Ezra 6:54).37 If the reading is original and if the wording is accurate, Adam was the lord of everything that had been created prior to him. No less interesting, and to the best of my knowledge hitherto unnoticed, is a parallel to Philo in a poem by Yose ben Yose (perhaps fifth–sixth century CE):

And when the world was built in wisdom, and when the table was set, and its bounty, You resolved to invite a guest, and to feed him Your choice food, And to make him dominant over the work of Your hands, to be like God, a judge and a ruler,38

34 I.e., the fish of the sea, the birds of the air, the beasts and animals living on earth (see the preceding note). 35 On the Creation of the World §§83–88. 36 T. Sanhedrin 8:9 (ed. M. S. Zuckermandel, Tosephta [Jerusalem: Wahrmann, 1963], 428) and parallels. See Borgen, “Man’s Sovereignty,” 377–79. 37 See M. E. Stone, Fourth Ezra: A Commentary on the Book of Fourth Ezra (Hermeneia; Minneapolis: Fortress, 1990), 179 for variae lectiones. Compare: “He made him similar to .ולהרדותו במעשה ידך, היות כאלהים שוטר ומושל 38 His image, engraved him in His shape, so that on heaven and earth he would be feared.” Yose ben Yose, ʾAzkir Gevurot, line) דימהו בצלם חקקו בתבנית היות בדוק וחלד פחד מוראו 38, in Yose ben Yose: Poems (ed. A. Mirsky [2d ed.; Jerusalem: Bialik Institute, 1991], 133); 160 Kister

And to prevail over an angel, to rend the waters, to make dim the lights39 and to revive the sleepers [= resurrect the dead].40

In these lines we find two of the solutions suggested by Philo, (1) and (4). There is, however, a clear difference between Philo and the tradition embedded in the poem of Yose ben Yose: While Philo stresses that human beings have sovereignty over mortal creatures only (i.e., those created on the fifth and sixth day)41 but excludes heavenly creatures, Yose ben Yose’s tradition is entirely consistent: Adam is lord of the universe, including the angels, the luminaries and the forces of nature (the sea): Adam, being created last, is superior to all his antecedents, without exception.42 According to the view attributed to the Jews and the Christians by Celsus, “there is God first, and we are next after him in rank, since he has made us

cf. Swartz and Yahalom, Avodah, 232). The wording follows Gen 9:2 (see above, n. 33); see also the piyyuṭ quoted below in n. 46. These passages refer not only to the fear of human beings manifested by all creatures, but also to a similar fear manifested by the angels. Compare the following examples: “The servants of the Most High feared him [= Adam], for everything was created by (God’s) speech, whereas he was created by God’s hand” (ʾAz Be-ʾEin Kol, my translation; cf. Swartz and Yahalom, Avodah, 132–33); see also the Syriac work Cave of Treasures, which reads: “When the angels heard the Voice (‘Let us make man in our image after our likeness’) they feared” (ch. 2.4; ed. S.-M. Ri, La caverne des trésors: Les deux recensions syriaques [CSCO 486 = Scriptores Syri 207; Louvain: Peeters, 1987], 14–15; see also 2.13, ed. Ri, 16–17). The fear of the angels when they created the First Man is mentioned in a text allegedly by Valentinus (Clement, Stromata 2.36.2–4: “As if the angels felt fear because of this creation . . . as human actions became fears to those who performed them, statues, for example, and images, and all that human hands produce in the name of God.” [Stromateis: Books One to Three (trans. J. Ferguson; FC 85; Washington, D.C.: Catholic University of America, 1991), 183]; compare the pseudo-Clementine homily cited below, n. 54). For an exhaustive discussion of Valentinus’s passage see C. Markschies, Valentinus Gnosticus? Untersuchungen zur valentinischen Gnosis mit einem Kommentar zu den Fragmenten Valentinus [Tübingen: Mohr Siebeck, 1992], 24–29; the Hellenistic background suggested there for the motif of the angels’ fear seems to me less compelling than the developments of this motif discussed in this note. 39 Cf. Midrash on Psalms 8:7 (ed. S. Buber [2 vols.; Vilna: Romm, 1891], 1:78; see also n. 46). 40 Yose ben Yose, ʾAtta Konanta, lines 21–24 (ed. Mirsky, 181); the translation is based on that of Swartz and Yahalom, Avodah, 297–99. Mirsky notes in his commentary the midrashic parallel in Gen. Rab. 5:5 (ed. Theodor–Albeck, 1:35); this passage lacks, however, the emphasis on absolute human dominion that is shared by Yose ben Yose and Philo. 41 See above, nn. 33, 34. 42 Similarly Gen.Rab. 19:4 (ed. Theodor–Albeck, 173). For this parallel as well as for a midrashic portrayal of Adam as “king of the lower world” see Borgen, “Man’s Sovereignty,” 373–74. Hellenistic Jewish Writers and Palestinian Traditions 161 entirely like God, and all things have been put under us, earth, water, air, and stars; and all things exist for our benefit, and have been appointed to serve us.”43 This wording is probably derived from Ps 8:6–7 (LXX; = MT 8:5–6);44 the words Ps 8:5 MT) must have been interpreted as meaning that) ותחסרהו מעט מאלהים God created Adam “little less than God.”45 That is to say, because Adam was created in God’s image, he is second to God and superior to all other creatures. A similar interpretation of these verses is attested in Yose ben Yose’s poem: “to make him dominant over the work of Your hands, to be like God a judge and a ruler.”46 Philo’s statement “All the mortal creatures . . . He subjected to his rule” (§84) seems also to be an allusion to Ps 8:7 (LXX).47 The biblical text of Philo was, however, that of the Septuagint, which reads (8:6): “You made him a little less than angels”; and this dovetails with his assertion that Adam was given dominion only over the sublunar creatures.48 The Jewish views as presented by Celsus and Yose ben Yose share not only the same ideology of Adam’s superiority to the angels, but plausibly also the same interpretation of Psalm 8. Adam’s sovereignty over heavenly beings and its relation to his creation as last of all God’s works is a central issue in the story of Satan’s rebellion in the apocryphal work the Life of Adam and Eve, according to the Latin, Armenian and Georgian versions.49 When God orders the angels to adore Adam, because he was created in His image, Satan says: “I shall not adore him who is inferior to me and created later than me. I am prior to that creature. Before he was

43 Origen, Contra Celsum 4.23. 44 The psalmist says: “When I look at your heavens, the work of your fingers, the moon and the stars which you have established; what is man that you are mindful of him. . . ? You have made him little less than God. . . . You have given him dominion over the works of Your hands, You have put all things under his feet, all sheep and oxen, and also the beasts of the field, the birds of the air, and the fish of the sea.” While the plain reading of this psalm might be that human beings are inferior to the heavens and the luminaries although they have dominion over the plants and the animals, other interpretations of the sequence of Psalm 8 are also possible. 45 It should be stressed that the wording of Celsus’s Jew, “we are next after him in rank, since he has made us entirely like God, and all things have been put under us,” does not resemble the Septuagint. 46 See also another piyyuṭ attributed to Yose ben Yose, ʾAz Be-Daʿat Ḥaqar, line 24 (ed. Mirsky, to shape clay in the image of“ לרקום גולם בצלם מוראו, לגדלו, לחסרו מעט מאלהים :(225 His fear, to exalt him, to make him little less than God,” where Ps 8:6 is explicitly cited. 47 See D. Runia, On the Creation, 14, 256. 48 Indeed, Origen (Contra Celsum 4.24) states emphatically: “we know that angels are . . . superior to men” (similarly ibid., 4.29). Elsewhere he denies that Christians think that “stars have been put under us” (ibid., 4.30). 49 See the discussion of this passage in Sergey Minov’s article in this volume. 162 Kister made, I had been made. He ought to worship me” (14:3).50 It has been demonstrated that this motif has its counterpart in 2 Enoch and elsewhere.51 The motif occurs, in somewhat different forms, in Jewish literature of late antiquity. Thus, in a dispute between Moses and the Red Sea in (late) midrashim52 and in an Aramaic poem of late antiquity,53 the sea refuses to let Moses divide it so that the Israelites might cross, because, as it says, “I was created three days prior to you.” Needless to say, the creation of Adam on the sixth day of creation (three days after the creation of the sea) is referred to here; Moses represents the human race. Satan’s argument coheres with Philo’s conception that heavenly beings are superior to human beings; whereas the resolution of the plot in the Aramaic poem reflects the notion that human beings are superior to all previously created beings, the idea we have seen expressed in the polemic of Celsus and the poem of Yose ben Yose. The notion of Adam’s universal sovereignty despite his inferiority is also found in the Jewish-Christian Pseudo-Clementine literature. There, it is stated that all created things—including “the ether, the sun, the moon, the stars, the air, the water, the fire,” which are superior to Adam—“willingly endure to serve the inferior in substance because of the shape of their superior”;54 i.e., because Adam is “the image of God.” This text is probably independent of the tradition concerning the conflict of Adam and the angels, for the latter are notably not mentioned in it. It is fascinating to note the common basic tradition reflected in works so far removed in their nature, time of attestation, genre, and background—as are Philo, the Life of Adam and Eve, the view attributed to Jews and Christians

50 This is the numbering of the Latin version; see G. A. Anderson and M. E. Stone, A Synopsis of the Books of Adam and Eve (2d rev. ed. SBLEJL 17; Atlanta: Society of Biblical Literature, 1999), 16–17 (pericope 5). The translation is drawn from this edition, with a slight altera- tion. Philo describes the adoration of Adam by the living creatures of earth (On the Creation, §83), while the Life of Adam and Eve describes his adoration by the angels. 51 M. E. Stone, “The Fall of Satan and Adam’s Penance,” JTS NS 44 (1993): 145–48; G. A. Anderson, “The Exaltation of Adam and the Fall of Satan,” in Literature on Adam and Eve: Collected Essays (ed. G. A. Anderson, M. E. Stone, and J. Tromp; Leiden: Brill, 2000), 83–110. 52 Exod. Rab. 21:6 (this section of Exodus Rabbah belongs to the Tanḥuma literature). 53 ʾEzel Moshe, an Aramaic poem written in Palestinian Aramaic; see M. Sokoloff and Y. Yahalom, Jewish Palestinian Aramaic Poetry from Late Antiquity (Jerusalem: The Israel Academy of Sciences and Humanities, 1999; in Hebrew), 82–87. See also Y. Yahalom, “ ‘ʾEzel Moshe’—According to the Berlin Papyrus,” Tarbiz 47 (1978), 173–84 (in Hebrew). 54 Pseudo-Clementine Homilies 16.19.4–7 (Die Pseudoklementinen I: Homilien [ed. B. Rehm; GCS 42; Berlin: Akademie-Verlag, 1953], 227); translation is that of T. Smith, ANF 8:317. Hellenistic Jewish Writers and Palestinian Traditions 163 by Celsus, early Christian writings,55 passages in earlier and later midrashim, as well as Hebrew and Aramaic poems of late antiquity—and to observe the lively ideological interplay that they represent, along with its influence on the shaping of the specific traditions. This discussion concerning the dominion and superiority of human beings and their rank in comparison to celestial bodies and angels, in connection with the interpretation of Psalm 8, continued into Jewish thought of the Middle Ages.56

3 Primeval Light and Sabbath Light

One the few passages that have survived from the Hellenistic Jewish writer Aristobulus (probably second century BCE)57 reads:

God, who made and furnished the whole universe, also gave us a day of rest—because of the toilsome life everyone has—the seventh day; but which, according to the truth of things (φυσικῶς), might also be called the first, that is, the genesis of light through which all things are seen together. And the same thing could be applied metaphorically to wisdom as well, for all light (issues) from it. And some of the Peripatetic school have said that it [i.e., wisdom] occupies the position of a lamp; for, by following it continually, they will remain undisturbed their entire life. Solomon, one of our ancestors, said . . . that it [i.e., wisdom] was there before heaven and earth. And this is actually in harmony with what is said above.58

55 See also Sergey Minov’s article in this volume, for later Christian transformations of this motif. 56 See S. Stroumsa, “ ‘What is Man?’: Psalm 8:4–5 in Jewish, Christian, and Muslim Exegesis in Arabic,” in “Open Thou My Eyes . . .”: Essays on Aggadah and Judaica Presented to Rabbi William G. Braude on His Eightieth Birthday and Dedicated to His Memory (ed. H. J. Blumberg et al.; Hoboken, N.J.: KTAV, 1992), 295–302. The passages cited in the present article demonstrate that both the notion that the rank of human beings is higher than that of the angels, and the interpretation of Psalm 8 in accordance with this view, need not indicate Christian influence on Al-Muqammiṣ (tenth century CE), as is suggested by Stroumsa; both the ideology and the prooftext may well continue the ancient Jewish trajectory and biblical interpretation discussed here. 57 For a general introduction to this writer see A. Yarbo Collins, “Aristobulus,” OTP 2:831–42 (832–33); C. R. Holladay, Fragments from Hellenistic Jewish Authors: Volume III: Aristobulus (SBLTT 39; SBLTTPS 13; Atlanta: Scholars Press, 1995), 74–75. 58 Aristobulus, frg. 5 (Eusebius, Praep. ev. 13.12.10-11); translation is that of Holladay, Fragments, 3:176–81. 164 Kister

This passage, preserved in both Clement of Alexandria and Eusebius,59 identifies the primeval light with wisdom on the one hand and with the light of the Sabbath on the other hand. I will trace these two conceptions in a variety of sources. The first unit of this section will deal with the interpretation of Gen 1:3 in accord with a general motif shared by ancient authors of divergent worldviews and cultural backgrounds—namely, that God’s light is the source of our light. On the one hand, God is seen as the source of perceptible light; however, in Second Temple literature, as well as in the Hellenistic world, perceptible light is connected with metaphysical light, which itself precedes from the divine.60 This metaphysical light is associated with understanding and knowledge in a large variety of sources,61 from late biblical passages62 to Qumran scrolls63 and rabbinic literature.64 In this section, I will discuss both of these motifs and the interplay between them. The second unit will deal with a specific conception, the association of Sabbath and light, which may be traced in Aristobulus, Philo, and rabbinic midrashim. Tracing this conception and its exegetical implications, strikingly shared by Philo and the rabbis, enables us to discern more lucidly not only the

59 For a discussion of the Christian appropriation of Aristobulus’s ideas as expressed in this passage, and of similar Philonic ideas, see A. Choufrine, Gnosis, Theophany, Theosis: Studies in Clement of Alexandria’s Appropriation of His Background (Patristic Studies 5; New York: Peter Lang, 2002), 123–48. 60 I. L. Seeligmann, “ΔΕΙΞΑΙ ΑΥΤΩΙ ΦΩΣ,” in idem, Studies in Biblical Literature (Jerusalem: The Hebrew University Magnes Press, 1992), 411–26 (in Hebrew). 61 For similar Hellenistic concepts see R. Bultmann, “Zur Geschichte der Lichtsymbolik im Altertum,” Philologus 97 (1948): 1–36. For much information on concepts of light see H. Conzelmann, “φῶς,” TDNT 9:310–58. For the problem of the relationship between Hellenistic and Jewish ideas, see G. Vermes, “The Torah is a Light,” VT 8 (1958): 436–38; and Seeligmann, “ΔΕΙΞΑΙ ΑΥΤΩΙ ΦΩΣ,” esp. 423–26. .(Dan 5:11) נהירו ושכלתנו וחכמה ,.E.g 62 63 “And David the son of Ishai was wise, and shining like the light of the sun, and a scribe and prudent . . . and God gave him prudent spirit and ,(חכם ואור כאור השמש וסופר) .belong together (cf חכם וסופר light; and he wrote psalms” (11QPsa 27:2–4). The words in J. C. Greenfield, M. E. Stone and E. Eshel, The ,ספר ומוסר וחכמה ;Dan 1:17 ,ספר וחכמה Aramaic Levi Document: Edition, Translation, Commentary [SVTP 19; Leiden: Brill, 2004], -in Syriac Ahiqar [ibid., 208–9 (commen ספרא וחכמתא and the expression ;[13:4] 102 b. Giṭ. 67a = ʾAbot de-Rabbi] חכם וסופר ;[b. B. Batra 22a] קנאת סופרים תרבה חכמה ;[(tary Nathan A, 18 [(ed. S. Z. Schechter; Vienna: Knöpfmacher, 1887), 68]; the meaning of the .(varies in these passages חכם וסופר collocation 64 See, e.g., S. Aalen, Die Begriffe “Licht” und “Finsternis” im Alten Testament, im Spätjudentum, und im Rabbinismus (Oslo: Dybwad, 1951), 272–80. Hellenistic Jewish Writers and Palestinian Traditions 165 similarities, but also the dissimilarities between Aristobulus and Philo on the one hand and rabbinic literature on the other hand, and to raise questions concerning the form and significance of traditions underlying rabbinic literature.

3.1 Primeval Light and God’s Light, Perceptible and Spiritual, in Works of Second Temple Judaism, Early Christianity, Midrash, Piyyuṭ, and Targum According to Genesis chapter 1, light was created on the first day by God’s utter- ,Gen 1:3); thus, in the conception of Genesis 1) ”(אור) ance, “let there be light light is a created phenomenon like all of God’s other creations, unrelated to God’s essence.65 The same account, however, narrates the separate creation on the fourth day. The nature of the light created (מאורות) of the luminaries on the first day was therefore puzzling to later generations of readers. In other passages of the Hebrew Bible God’s essence is associated with light (e.g., Isa 67.(עטה אור כשלמה :in an eschatological context;66 possibly Ps 104:2 ,20–60:19 In some texts from Jewish literature in late antiquity, God’s splendor is said to be the source of the light of the world. In the Apocalypse of Abraham we read: “You, the light that shines before the light of morning on Your creation, from Your face it is day on earth, and in Your heavenly dwelling places (there is) another inexhaustible light of an inexpressible dawning from the lights of Your face.”68 According to these verses, to which there is a striking parallel in

65 “The idea of creation by the word preserves first of all the most radical essential distinction between Creator and creation. Creation cannot be even remotely considered an emanation from God; it is not somehow an overflow or reflection of his being, i.e., his divine nature, but is rather a product of his personal will” (G. von Rad, Genesis: A Commentary [trans. J. H. Marks; London: SCM Press, 1961), 49–50. In contrast, Mark Smith infers from other biblical passages and from ancient interpretations of Genesis 1 that the light of Genesis 1:3 was not created (M. S. Smith, “Light in Gen 1:3—Created or Uncreated: A Question of Priestly Mysticism?” in Birkat Shalom: Studies in the Bible, Ancient Near Eastern Literature, and Postbiblical Literature Presented to Shalom M. Paul on the Occasion of his Seventieth Birthday [ed. C. Cohen et al.; Winona Lake: Eisenbrauns, 2008], 125–34; idem, The Priestly Vision of Genesis 1 (Minneapolis: Fortress, 2010), 71–79 and 240–48 [notes]). To my mind von Rad’s interpretation of Genesis 1 is correct; different views in other biblical passages and in postbiblical literature will be discussed below, but these views should not be read into Gen 1:3. 66 God’s light is to replace the light of the sun and the moon in the time of Israel’s deliver- -prob אור שבעת הימים ance, as it was in the beginning; cf. Isa 30:26, where the expression ably attests to an ancient legend concerning the special light of the seven days of creation. 67 See Smith, “Light in Genesis 1:3,” 130. 68 Apocalypse of Abraham (17:18–19). The translation offered here is based on that of A. Pennington in The Apocryphal Old Testament (ed. H. F. D. Sparks; Oxford: Clarendon, 166 Kister a late midrash,69 God’s face is considered the point of origin of the light. The same idea is stated in a poem by Yose ben Yose: “Bohu70 and darkness covered the face of the earth, (then) it shone with light from the face of the King.”71 According to these passages God’s splendor is the source of the light of our world.72 This notion is also expressed in a midrash—based upon Ps 104:2—

1984), 381, with alterations made by my friend, Prof Constantine Zuckerman; see also the translation and discussion of A. Kulik, Retroverting Slavonic Pseudepigrapha: Toward the Original of the Apocalypse of Abraham [SBLTCS 3; Leiden: Brill, 2005], 23. I would suggest אתה האור המאיר לפני אור :the following tentative retroversion to the original Hebrew הבקר על בריאתך, ומפניך היות יום על הארץ. ובמעונותיך ]אשר בשמים[ אור אחר לאין more literally, the last words might be) מחסור )לאין תכלית?( מִשַ חר אורת פניך לאין חקר -see 1QHa 7:29; similar negative expres אין מחסור For the expression .(לבלי סַּפֵ ר :rendered sions occur in a similar context in a Qumranic parallel that will be discussed below (n. ,could perhaps account for the plural expression (אורות if misread) אורת A form like .(74 “lights of Your face” in the Slavonic. 69 As Alexander Kulik and I discovered during a conversation, a very late midrashic work, Midrash Alpha-Bethot contains the passage: “The brilliance of the morning light goes out from Him (literarily: “from his face”); he makes the world shine for them [= the righ- נוגה אור השחר יוצא מלפניו בכל בקר ;”teous] from the light of his glory every morning This recension of .להאיר להם את העולם, מזיו כבודו מאיר להן את העולם בכל בקר ובקר Midrash Alpha-Bethot is published in the online edition of Ma‌ʾagarim, the database of the Historical Dictionary of the Hebrew Language, the Academy of the Hebrew Language, (http://hebrew-treasures.huji.ac.il/). This recension contains some ancient traditions; for example, it thus describes the eschatological state when the universe returns to the primordial situation of tohu wa-bohu: “the upper water and the lower water will intermingle as they were in the beginning, the whole world will be filled with darkness, and the min- ומלאכי השרת מעופפים בכנפיהן) istering angels will fly with their wings over the water רוח אלהים מרחפת על פני ,This paraphrase of Gen 1:2 interprets the phrase ”.(על פני המים .as referring to angels, an interpretation documented in Jub. 2:2 ,המים 70 The meaning of this word here is probably “water”; see M. Kister, “Tohu wa-Bohu: Primordial Elements and Creatio ex Nihilo,” JSQ 14 (2007): 229–56, esp. 234–35. ;ʾAzkir Gevurot, line 15 (ed. Mirsky, 129 ,בוהו ואפלה כסו פני חלד ותבהק אור מאור פני מלך 71 translation in Swartz and Yahalom, Avodah, 225–27). The expression “the face of the King” is based on Prov 16:15, and is interpreted elsewhere in rabbinic literature as referring to God (e.g., Lev. Rab. 20:10 [ed. M. Margulies, Midrash Wayyikra Rabbah (5 vols.; Jerusalem: American Academy for Jewish Research, 1953–1960), 2:466]); note the poetic wordplay with “the face of earth.” 72 See also Ginzberg, Legends of the Jews, 5:8–9 nn. 19–20. For the possibility that Sir 43:1 (according to the Greek translation and the Masada manuscript) states that the bright- ness of the sky reflects the shining of God’s glory, see M. Kister, “A Contribution to the Interpretation of Ben Sira,” Tarbiz 59 (1990): 303–78, esp. 358 (in Hebrew). See also the discussion of Sir 43:4–5, ibid., 360. Hellenistic Jewish Writers and Palestinian Traditions 167 according to which the origin of the light of the world is the garment with which God covered Himself.73 The wording of the Apocalypse of Abraham 17:19, “in Your heavenly dwelling places (there is) another inexhaustible light,” is similar to that of a text from Qumran: “In His dwelling there is the light of ortom [i.e., perfect light], and all darkness is nullified in His presence, and with Him there is no distin- guishing between light and darkness . . . and with him there is limitless light.”74

73 According to Gen. Rab. 3:4 (ed. Theodor–Albeck, 19–20), God “covered Himself with light as a garment, and all the world was enlightened by the splendor of His Glory.” We are told that this tradition was transmitted “in a whisper,” and that an Amora of the third century CE wondered why something that is stated explicitly in Ps 104:2 had been transmitted “in a whisper.” Some scholars have suggested that a Logos-emanation theory (similar to Philo’s) is implied (A. Altmann, “A Note on the Rabbinic Doctrine of Creation,” JJS 7 [1956]: 195–206, esp. 201–2; see E. E. Urbach, The Sages: Their Concepts and Beliefs [trans. I. Abrahams; 2 vols.; Jerusalem: The Hebrew University Magnes Press, 1975], 1:209–10, 2:780–81, where earlier scholarship is reviewed). It should be noted that a teaching concerning the divine “garment of light” played an important role in Gnostic circles; see especially the Gnostic treatise The Paraphrase of Shem, discovered at Nag Hammadi: “So that the darkness of nature might be put to shame, I put on my garment, the garment of the light of Majesty” (trans. M. Roberge, Paraphrase of Shem 8, in The Nag Hammadi Scriptures [ed. M. Meyer; New York: HarperCollins 2008], 452); “The garment the Savior puts on is a word or logos that calls, since it is also a voice” (454 n. 23). See also Altmann, “A Note,” 202. It has been assumed that the transmission of the tradition “in a whisper” is an indication of some esoteric teaching (whatever the latter may be); see also G. Scholem, Jewish Gnosticism, Merkabah Mysticism, and Rabbinic Tradition (New York: Jewish Theological Seminary of America, 1965), 58. If indeed “saying in a whisper” is related to esoteric teachings, saying aloud that which is to be whispered has a paradoxical effect; to highlight the similarity of this teaching to the plain sense of Ps 104:2 may be a subtle way to undermine any “hazardous” implications, transforming the teaching itself into a harm- less piece of biblical exegesis. Note, however, that according to the story in y. Beṣ. 1:11 (61a) a purely halakhic ruling is uttered “in a whisper,” because “it was transmitted in a whisper” (as noted by Scholem); it is quite difficult to see why this halakhah should be transmitted “in a whisper.” The esoteric nature of saying something “in a whisper” is therefore uncertain. Be that as it may, both the verse cited and the question, “whence did light come into the world?” indicate that the physical light was conceived of as reflecting the celestial light, not the light of God’s own face, but rather the light of His garment (contrast Altmann, “A Note,” 199); for a similar interpretation of the rabbinic material, see Aalen, Die Begriffe “Licht” und “Finsternis”, 264, 315. This solution to the question of the origin of light significantly reformulates the account in Genesis 1: physical light is derived from the splendor of God himself, which is immanent in our world (see Altmann, “A Note,” 197). ובמעונתו אור אורתם וכל אפלה לפניו נחה ואין עמו להבדיל בין האור לחשך לכי לבני ]אד[ם 74 4Q392 1 5–7); see D. K. Falk, “Works of God,” in Qumran Cave) הבדילם . . . ועמו אור לאין חקר 168 Kister

While this text refers to the creation of (physical) light in Genesis 1,75 the term ortom (= perfect light) is used elsewhere in Qumran to indicate the light of both knowledge and deliverance.76 The association of the shining of light of the first day (according to Gen 1:3) and the illumination of the heart with the divine light of knowledge occurs also in Paul: “For it is the God who said, ‘Let light shine out of darkness,’ who has shone in our hearts to give the light of the knowledge of the glory of God in the face of Christ” (2 Cor 4:6). This verse combines the notion of God as the source of illumination by knowledge,77 and the conception of light as the effulgence of the glory of God’s face.78 The debt of Paul to Palestinian Jewish phraseology becomes increasingly clear when his wording in this verse, ὅς ἔλαμψεν ἐν ταῖς καρδίαις ἡμῶν πρὸς φωτισμὸν τῆς γνώσεως τῆς δόξης τοῦ θεοῦ, is compared to the strikingly similar wording in God has illuminated“ ,האיר אלוהים דעת בינה בלבבי ,another Qumran fragment

4.XX: Poetical and Liturgical Texts, Part 2 (ed. E. Chazon et al.; DJD 29; Oxford: Clarendon, and their אורתם The translation is mine. For a study of the various senses of .44–25 ,(1999 theological implications see M. Kister, “4Q392 1 and the Conception of Light in Qumran ‘Dualism,’” in Meghillot: Studies in the Dead Sea Scrolls 3 (ed. M. Bar-Asher and D. Dimant; Haifa: Haifa University Press; Jerusalem: Bialik Institute, 2005), 125–42 (in Hebrew). 75 4Q392 indicates that God distinguishes between light and darkness (Gen 1:4) only for the -is beyond such distinc ,אורתם ,sake of human beings, while the endless celestial light tions. Compare Targum Pseudo-Jonathan to Gen 1:5: “And God called the light day, and He made it so that the inhabitants of the world might labor during it”; Yose ben Yose, ʾAzkir Gevurot, line 16: “He gave radiance for the day for humans to go out to work, and made darkness for night, for the creeping of the beasts of the forest (ed. Mirsky, 129); and ʾAz Be-Daʿat Ḥaqar, 7 (ibid., 223): “He spoke and brought light and divided by measure light for work and darkness for rest” (my translations). see ;(דעת) ”is clearly related to “knowledge אורתם In Shirot ‘Olat ha-Shabbat, the term 76 4Q403 1 i 45; 4Q404 5 4–5 (C. Newsom, “Shirot Olat haShabbat,” in Qumran Cave 4.VI: Poetical and Liturgical Texts, Part 1 [ed. E. Eshel et al.; DJD 11; Oxford: Clarendon, 1998], is the light of (אורתם In the Hodayot, the eternal light (sometimes called there .(297 ,269 the hereafter and of salvation (in keeping with the eschatological passages, Isa 30:26 and 60:19); see 1QHa 12:23; 21:14; and perhaps 12:6. See further Kister, “Conception of Light,” 133–42. 77 J. A. Fitzmyer, “Glory Reflected on the Face of Christ (2 Cor 3:7–4:6) and a Palestinian Jewish Motif,” TS 42 (1981): 630–44. 78 J. D. Worthington, Creation in Paul and Philo (WUNT 2.317; Tübingen: Mohr Siebeck, 2011), 89–98 (for a comparison with Philo see ibid., 83–89). The glory of the face of God is, according to Paul, the face of Christ. Hellenistic Jewish Writers and Palestinian Traditions 169 the knowledge of wisdom in my heart.”79 As usual, Paul gives this phrase a christological twist.

The Palestinian targumim seem to imply that the light of the first day is related to God’s splendor. The targum on Exod 12:42 contains an excursus on the night and ,(תהי ובהי) of creation, when “the world was in the state of tohu va-bohu darkness was spread over the surface of the deep (Gen 1:2); the memra of the והוה עלמא תהי ובהי וחשוכא על Lord was light and illumination.”80 The words are an obvious paraphrase of Gen 1:2; thus, it seems likely that the אפי תהומא רוח) ”memra of the Lord in this passage is identified with the “Spirit of God ”in this same verse. Since in Genesis 1, the words “let there be light (אלהים (Gen 1:3) immediately follow Gen 1:2, I find it very likely that the targumic passage here is associating the creation of light with the effulgence of the Spirit of God in Gen 1:2.81 In diverse passages belonging to different genres and eras, the preexistent “wisdom” of Proverbs 8 is described as preexistent light; these passages presume the identification of wisdom with the preexistent Torah.82 Thus we read in another poem by Yose ben Yose: “While the earth was still desert and darkness, You amused Yourself with the light of the Law.”83 As noted by Yehoshua

79 4Q511 [4QShirb] 18 8; ed. M. Baillet, Qumrân Grotte 4.III (4Q482–4Q520) (DJD 7; Oxford: Clarendon, 1982), 230–31. לילא קדמא כד אתגלי >מימריה ד<י’י על עלמא למברי יתה והוה עלמא תהי ובהי וחשוכה 80 The passage occurs in all the Palestinian ;פריס על אפי תהומא ומימריה די’י הוה נהור ומנהר targumim (Neofiti, Fragment Targum, and Genizah fragments) to Exod 12:42. 81 One does not need to press the memra terminology in this passage, as does D. Boyarin, “ ‘He Who Spoke and the World Came into Being’: On the Logos and Its Rejection, a Chapter in the Formation of the Religion of the Rabbis,” in Higayon L’Yona: New Aspects in the Study of Midrash, Aggadah and Piyut in Honor of Prof. Yona Fraenkel (ed. J. Levinson, J. Elbaum, and G. Hasan-Rokem; Jerusalem: The Hebrew University Magnes Press, 2006), 151–64, esp. 151–54 (in Hebrew). The similarity of this passage to the Prologue of the gospel of John has been noted (see R. Le Déaut, La nuit pascale: ‎Essai sur la signification de la Pâque juive à partir du Targum d’Exode XII 42 [AnBib 22; Rome: Pontifical Biblical Institute, 1963], 216). The problems involved in attributing Logos theology to the memra in the (Palestinian) targumim in general are beyond the scope of the present article, and should be dealt with elsewhere. 82 This midrashic notion should by no means be read into Genesis 1 (contrast Smith, “Light in Genesis 1:3,” 131–33). ,ʾAtta Konanta, 3 (ed. Mirsky ;אדמה בעודה ציה וצלמות באור דת שעשעתה ורגלך שחקה 83 178; translation from Swartz and Yahalom, Avodah, 293, translation slightly altered; I do .(”by “wasteland צלמות not concur with their rendering of 170 Kister

Granat,84 a passage in the Tanḥuma similarly reads: “When He was creating the world, the Torah was—if one may say so—enlightening for Him, the world being tohu va-bohu (Gen 1:2); as it is said, ‘For the commandment is a lamp and the Torah is light’” (Prov 6:23).85 In a poem by Rabbi Elʿazar berabbi Qillir (seventh century CE), the Torah speaks: “When the world was waste and void, I was like a candle, and when darkness was spread on the face of abyss, I was like a lamp”; “While there was darkness upon the face of abyss, I shone and gave light.”86

The preceding quotations from rabbinic and paytanic literature represent exceptions. According to most traditions from these corpora the light created on the first day is supernatural but perceptible; it is neither “noetic” nor spiritual, nor is it associated with knowledge, understanding, or the Torah. In Hellenistic Jewish literature and in early Christian writings, the light of the first day is more often associated with spiritual light and connected with the divine light. These ideas are influenced by Platonic notions, but they also have counterparts in Palestinian sources in which no philosophical traces can be detected.87 Philo says that the “invisible and intelligible light has

84 Y. Granat, “Before ‘In the Beginning’: Preexistence in Early Piyyuṭ against the Background of Its Sources” (Ph.D. thesis, The Hebrew University of Jerusalem, 2009), 82 (in Hebrew). וכשהיה בורא את העולם כביכול היתה התורה מאירה לפניו שהיה העולם תוהו ובוהו, שנ’ כי 85 –this passage was published by E. E. Urbach, “Fragments of Tanḥuma ;נר מצוה ותורה אור Yelamdenu,” Kobez Al Yad n.s. 6/1 (1966): 20 (in Hebrew). 86 S. Elizur, Qedushta‌ʾot le-Yom Mattan Torah by Elʿazar berabbi Qillir (Jerusalem: Mekitze Nirdamim, 2000), 99, lines 103–4, line 137. Granat (245–60) discusses the remarkable idea reflected in this poem that the Torah is the essence of everything in our world (for other occurrences of this notion in this text, see Elizur, Qedushta‌ʾot, 99, lines 97–105; 102, lines 143–48; and most explicitly 103–4 lines 162–85). The Torah is the τύπος for the entire world Elizur, Qedushtaʾot, 103, line 161); the image is similar to בטבעתי ארץ ואדניה הוטבעו) Philo’s conception of the functioning of the Logos (such abstract philosophical terminology is, of course, alien to this text). In this context the Torah is to some extent, like the Logos, undifferentiated from God (see M. Kister, “Some Early Jewish and Christian Exegetical Problems and the Dynamics of Monotheism,” JSJ 37 [2006]: 584–85; see also Elizur, Ḳedushta‌ʾot, 102, lines 141, 150; see also Y. Baer, “Eschatological Doctrines in the Second Temple Period,” Zion 23/24 [1958/1959]: 143 [in Hebrew], whose overarching theory, however, I do not accept). Note also the similar imagery in Theophilus of Antioch discussed below. 87 Compare G. H. van Kooten, “The ‘True Light which Enlightens Everyone’ (John 1:9): John, Genesis, the Platonic Notion of the ‘True, Noetic Light,’ and the Allegory of the Cave in Plato’s Republic,” in The Creation of Heaven and Earth: Re-interpretations of Genesis 1 in the Context of Judaism, Ancient Philosophy, Christianity, and Modern Physics (ed. G. H. van Kooten; TBN 8; Leiden: Brill, 2005;), 149–94. Hellenistic Jewish Writers and Palestinian Traditions 171 come into being as the image of the divine Logos.”88 Theophilus of Antioch (second century CE) associates the Spirit of Gen 1:2 with light: “The unique spirit, the (arche)type (τύπος) of light, was situated between the water and the heaven . . . the command of God, his Logos, shining like a lamp in a closed room, illuminated the region under the heaven, making light separately from the world.”89 The cosmological theory of this passage is rather complex, but apparently it reflects a view strikingly similar to that of the targumic passage cited above. In the well-known Prologue to the Gospel of John (1:1–5), the phraseology of which refers to Gen 1:2, the Logos is described as the primeval light.90 Although this passage does not deal with the light created on the first day (Gen 1:3), Christian writers who read the Prologue together with Genesis 1 connected the “light” of Gen 1:3 with the Logos.91

88 Philo, On the Creation of the World §31. Elsewhere Philo states concerning Gen 1:3: “God is light (ὁ θεὸς φῶς ἐστι), for there is a verse in one of the psalms, “the Lord is my illumination (φωτισμός) and my Savior” (Ps 26:1 LXX). And He is not only light, but the archetype (ἀρχέτυπον) of every other light, nay, prior to and high above every archetype, holding the position of a model .” (Philo, On Dreams 1.75; for the emendation, see Colson and Whitaker, Philo, 5:336 n. 1). 89 Theophilus of Antioch, Ad Autolycum 2.13 (trans. R. M. Grant, Theophilus of Antioch: Ad Autolycum [Oxford: Clarendon, 1970], 48–49, altered). My translation of the phrase, “the (arche)type of light,” follows the reading of M. Marcovich: Ἕν μὲν <οὖν> τὸ πνεῦμα, φωτὸς τύπον ἐπέχον (M. Marcovich, ed., Tatiani Oratio ad Graecos, Theophili Antiocheni Ad Autolycum [PTS 43; Berlin: de Gruyter, 1995], 59). Grant prefers the reading, Ἕν μὲν τὸ πνεῦμα φωτὸς τόπον ἐπέχον, and translates, “the unique spirit occupied the place of light.” 90 It has been suggested that these verses (in which the Logos is, of course, Jesus) make use of a Jewish text in which the Logos was identified with Torah; see P. Borgen, “Logos Was the True Light: Contributions to the Interpretation of the Prologue of John,” NovT 14 (1972): 115–30, esp. 125: “there are very good grounds for concluding that the conception of logos-light . . . has as a model the conception of Torah-light. . . .” See also T. H. Tobin, “The Prologue of John and Hellenistic Jewish Speculation,” CBQ 52 (1990): 252–69; R. Bauckham, “The Qumran Community and the Gospel of John,” in The Dead Sea Scrolls: Fifty Years after their Discovery (ed. L. H. Schiffman, E. Tov and J. C. VanderKam; Jerusalem: Israel Exploration Society, 2000), 104–15. The last study specifically deals extensively with light imagery in divergent Jewish texts of the Second Temple period. Van Kooten’s suggestion (“True Light”) that John (both in the Prologue and elsewhere) should be read as using Platonic ideas seems to me facile, and should be balanced by the material brought together in this section from a variety of Jewish sources. This is not to deny, of course, that Jewish and Christian writers familiar with Platonic ideas easily intermingle the two sets of ideas. 91 For a rich discussion of the interpretation of Gen 1:3 in John 1:1–5, see A. Orbe, “A propósito de Gen. 1, 3 (fiat lux) en la exegesis de Taciano,” Gregorianum 42 (1961): 401–43, esp. 430– 40; note, e.g., the statement of Tertullian in Adversus Praxean, 12: “At first, when the Son is not yet on the scene (nondum filio apparente), ‘And God said, Let there be light, and it was 172 Kister

Thus far we have seen that a variety of sources, Jewish and Christian, Hellenistic and Palestinian, share the notion that the primeval light is either the light of God himself or the light of a divine entity (such as Wisdom, the Logos, the Torah). This notion inevitably leads to a different reading of Gen 1:3: that is, the light created on the first day reflects the divine, primeval light that existed before the created order. According to some of the sources, there is a continuum between the perceptible and the spiritual light; according to others, there is a continuum between God and His creation: perceiving the light of day, or attaining wisdom, are, in fact, seeing God’s face and experienc- ing some limited comprehension of His glory. While the notion of some sort of primeval divine light is shared by all these sources, the nature of this light is necessarily related to the theological pre- sumptions of each of them. The similarities among these diverse reworkings of this tradition should not cause us to undervalue their dissimilarities; an integrative study enables us on the one hand to discern the shared theme and on the other hand to be more attentive to the different functions of this motif in various religious and cultural contexts.

3.2 The Light of the Sabbath: Aristobulus, Philo, and the Midrash Aristobulus (above, p. 163) identifies the primeval light with the spiritual light of Sabbath, the seventh day which is “the genesis of light through which all things are seen together,” and which is both the seventh and the first day. Scholars have noted that Aristobulus here follows in the footsteps of Pythagorean writers, and that Aristobulus’s teaching concerning the Sabbath’s light is significantly close to that of Philo, who also adapted Pythagorean material in his discussions of the seventh day.92 In Pythagorean thought, “seven” is related to “one,” and signifies “mind (nous), sanctity, and light.”93 Similarly, as has been noted, when Philo discusses the Sabbath, he calls the number seven “the light

made’ (Gen 1:3).The Word himself is constantly (statim) ‘the true light that enlightens the person that comes into the world’ (cf. John 1:9), and through him also the mundane light ” (E. Evans, Tertulliani Adversus Praxean Liber [London: SPCK, 1948]), 102; quotation from page 145 [with alterations]; see note on 262. 92 For a survey of scholarship, see L. Doering, Schabbat: Sabbathalacha und -praxis im antiken Judentum und Urchristentum (TSAJ 78; Tübingen: Mohr Siebeck, 1999), 309–15. See also J. Riaud, ”Pâque et Sabbat dans les fragments I et V d’Aristobule,” in Le Temps et les Temps dans les littératures juives et chrétiennes au tournant de notre ère (ed. C. Grappe and J.-C. Ingelaere; JSJSup 112; Leiden: Brill: 2006), 107–23, esp. 117–23. 93 Philolaus; see H. Diels and W. Kranz, Die Fragmente der Vorsokratiker (3 vols.; 5th ed.; Berlin: Weidmann, 1934–1937), 1:400, §12. Other numbers between one and ten also represent abstract notions. Hellenistic Jewish Writers and Palestinian Traditions 173 of the six, for seven reveals as completed what six has produced”;94 he says that “in accordance with truth of things (φυσικώτατα) . . . the one is the same as the seven.”95 Clearly, both Aristobulus and Philo are here using Pythagorean language as well as concepts. As we shall see, however, an interpretive trajectory that links the Sabbath with light, and especially with the primeval light, exists not only in Hellenistic, but also within Palestinian Judaism.96 Philo associates the Sabbath with light, describing it as the day of spiritual light, in his treatise Allegorical Interpretation:

“And God blessed (εὐλόγησην) the seventh day and hallowed it” (Gen 2:3). God both blesses and forthwith makes holy the dispositions set in motion in harmony with the seventh and truly divine light. . . . But the reason why the man that guides himself in accordance with the seventh and perfect light (ἕβδομον καὶ τέλειον φῶς) is both of good understanding (εὐλογιστός) and holy, is that the formation of things mortal ceases with this day’s advent. For, indeed, the matter stands thus: when that most brilliant and

94 Philo, On the Special Laws, 2.59. 95 Philo, That God is Unchangeable, 11; On the Posterity of Cain, 64. See H. R. Moehring, “Arithmology as an Exegetical Tool in the Writings of Philo of Alexandria,” in The School of Moses: Studies in Philo and Hellenistic Religion in Memory of Horst R. Moehring (ed. J. P. Kenny; BJS 304 = SPhM 1; Atlanta: Scholars Press, 1995), 141–76. See also Doering, Schabbat, and Riaud, “Pâque et Sabbat.” 96 Aristobulus’s designation of the Sabbath as existing “before heaven and earth” corresponds to a cyclical conception of the first week which occurs in both later Jewish and Samaritan material. In these contexts, Sabbath’s rest is considered a return to the eternal rest which obtained before creation: (1) an addition to the targum of Exod 20:8: “Before therefore you, my ;[הוה כל עלמא שבת] I created the world, the whole world was Sabbath son, rest [on the Sabbath] and do not do any work. For the heaven and the earth that I created were before me [i.e., created by me] in a single hour, and I made the heaven and first ; עד דאודע :read, perhaps ;עד דאירע] *the earth in six days *in order to let you know the day of the Sabbath, [ ] so that you might rest in it from [?ידע person afʿel future of all your labors” (M. L. Klein, Genizah Manuscripts of Palestinian Targum to the Pentateuch [Cincinnati: Hebrew Union College Press, 1986], 1.277; my translation is based on that of Klein, with some alterations). (2) According to a relatively late passage of Tibat Marqe (4.34; ed. Ben-Ḥayyim, 258–59): “the New [i.e., Creation] is built on the foundation of silence, to which it returns; therefore God created all the creatures in six days and rested ”,on the seventh day.” See Ben-Ḥayyim’s note ad loc.; see also Kister, “Tohu wa-Bohu (שבת) 237 n. 37. The cyclical pattern of the Jewish week probably contributed to the emergence of these circular concepts. 174 Kister

truly divine light of virtue has dawned, the creation of that whose nature is of the contrary kind comes to a stop.97

Curiously, a similar association between light and the hallowing of the Sabbath occurs in rabbinic literature; although this interpretation differs significantly from that of Philo in its content. The tannaitic midrash Mekhilta de-Rabbi Ishma‌ʾel records two different transmissions of Rabbi Shimʿon’s interpretation of Exod 20:11, a verse very similar to Gen 2:3:

“Wherefore the Lord blessed the Sabbath day and hallowed it” (Exod 20:11) [A] Rabbi Shimʿon ben Yoḥai says: He blessed it with manna and hal- .(במאורות) lowed it with the luminaries [B] Rabbi Shimʿon ben Yehuda of Kefar Akko says in the name of Rabbi Shimʿon: He blessed it with the manna and hallowed it with the shining light of one’s face.98

In this passage there are two assertions (sections A and B) attributed already in the tannaitic period (prior to the third century CE) to Rabbi Shimʿon ben Yohai (second century CE). While the two assertions differ strikingly from one another, they are evidently two early variants of a single saying of Rabbi Shimʿon, asserting that the Sabbath was hallowed with “light.” Other variants of this tradition occur in Amoraic compilations, in which the statements cited above are reshaped. One such example is a lengthy passage interpreting Gen 2:3 (“And God blessed the seventh day and hallowed it”), found in two Amoraic compilations: 1) Genesis Rabbah;99 and 2) the Midrash on the Ten Commandments included in Pesiqta Rabbati.100 I will here summarize this

97 Philo, Alleg. Interp. 1.16–18. 98 MekhRI Baḥodesh, 7 (H. S. Horovitz and I. A. Rabin, eds., Mechilta D’Rabbi Ismael על כן ברך ה’ את יום השבת ויקדשהו . . . רבי :(Frankfurt a. M.: Kauffmann, 1928–1931], 231] שמעון בן יוחאי אומר: ברכו במן וקדשו במאורות; רבי שמעון בן יהודה איש כפר עכו אומר משום רבי שמעון: ברכו במן וקדשו במאור פניו של אדם 99 Gen. Rab. 11:2 (ed. Theodor–Albeck, 87–89). 100 Pesiqta Rabbati (M. Friedmann [Ish-Shalom], ed., Pesiqta Rabbati [Vienna: Kaiser, 1880]), 117b–118b. Prof. Yaakov Sussmann, who studied this midrash for many years, considers this unit of Pesiqta Rabbati to belong to the so-called “classical Amoraic midrashim,” produced prior to the Islamic conquest of Palestine (oral communication). Menahem Kahana has demonstrated that the version of this unit found in Genesis Rabbah Ms. Vatican 60 is an adaptation of the text of the Midrash on the Ten Commandments; see M. Kahana, “Genesis Hellenistic Jewish Writers and Palestinian Traditions 175 complex passage, calling attention to the major differences between the two versions:

[A] Rabbi Liʿezer101 interprets Gen 2:3: “He [= God] blessed it with a [Sabbath] lamp; and this happened [says Rabbi Liʿezer] in my case: I once lit a lamp for the Sabbath night, and when I came at the termination of the Sabbath I found it still burning and not at all diminished.” [B] The statement of Rabbi Shimʿon ben Yehuda in MekhRI section B is interpreted. Genesis Rabbah understands it to mean that one’s shining ’on the Sabbath is different from one (אור פניו של אדם) countenance countenance during the six working days; i.e., more relaxed. The Midrash on the Ten Commandments, on the other hand, interprets R. Shimʿon’s statement as referring to the supernatural splendor of Adam and Eve which was not taken from them (after they had ,(באור פניו של אדם ואשתו) sinned) until the exit of the Sabbath.102 [C] The statement attributed to Rabbi Shimʿon in MekhRI section A is interpreted as referring to the postponement, until the exit of the Sabbath, of the decrease in the light of the luminaries103 that was the consequence of Adam’s sin. [D] An Amora, R. Levi104 says in the name of R. Nezira that the light of the Sabbath served for thirty-six hours. According to the Midrash on the Ten Commandments, this was the light of the first day: “the light that had been created on the first day served for thirty-six hours: On Sabbath’s eve, on Sabbath’s night and on the day of the Sabbath. (It served) until the Sabbath’s exit, and by means of it Adam saw [everything] from one end of the world to the other.”105

Rabbah MS Vatican 60 and Its Parallels,” Teʾudah 11 (1996): 1–60 esp. 23–26 (in Hebrew), where Genesis Rabbah Ms. Vatican 60 and its parallel in Pesiqta Rabbati are discussed. 101 An Amora who flourished in the third century CE. 102 This interpretation is also necessary for understanding the larger flow of Genesis Rabbah (material not given here); it was probably inadvertently omitted during the transmission of the midrashic material. 103 Based tacitly on Isa 30:26. 104 Early third century CE. 105 According to the original reading of Ms. Leiden of the Palestinian Talmud (y. Ber. 8:5 [12b]) “light served for thirty-six hours on the first Sabbath”; this reading was altered by another scribe to “the light that had been created on the first day served for thirty-six hours;” see Talmud Yerushalmi according to Ms. Or. 4720 (Scal. 3) of the Leiden University Library, with Restorations and Corrections (ed. Y. Sussmann; Jerusalem: Academy of the Hebrew Language, 2001), 62. 176 Kister

All of these statements are elaborations of and variations upon a peculiar interpretation of the verse, “And God (or: the Lord) blessed the seventh (or: the Sabbath) day and hallowed it” (Gen 2:3/Exod 20:11); the implied interpretation is, “He hallowed the Sabbath with light.”106 What is the nature of this light: is it like the light of a Sabbath lamp, the light of the luminaries, the supernatural splendor of Adam, the “shining light of one’s face,” or just a more peaceful and relaxed countenance? The last alternative [D], according to the Midrash on the Ten Commandments, regards the light of the Sabbath as the special light of the first day. This thus constitutes yet another occurrence of the cyclical creation motif (see above). The content of this interpretation is entirely different from that of Aristobulus: the rabbinic statement refers to physical light107 rather than to the light of virtue and knowledge. However, there is some (coincidental?) similarity between the two, even in the wording (for Aristobulus, this light is that “through which all things are seen together”; whereas, according to the midrash, by means of this light, Adam saw everything in the world). How should the phrase “the shining light of one’s face” in the Mekhilta [B] be construed? Possibly (but not necessarily) it may simply mean the peaceful countenance of a person, as understood in Genesis Rabbah [B]. Alternatively, it could mean the enlightening of one’s mind. This usage is well documented אודכה אדוני כי>א< האירותה פני לבריתכה . . . אדורשכה :in the Qumran scrolls 1QHa 12:6–7) “I praise you, Lord, for you have) וכשחר נכון לאור]תו[ם הופעתה לי enlightened my face for your covenant . . . I have looked for you, and like dawn whose light increases to [per]fect light (ortom), you have shone upon me.” God enlightens the face of the elect, probably by revealing his own light (note the final words of the citation); the elect one gains some of the glory of God’s face by which he is illumined. According to the Enochic Book of Parables (in a clear allusion to Moses’s shining face in Exod 34:29), “they will not be able to look at the face of the holy, for the light of the Lord of Spirits will have appeared on the face108 of the holy and righteous elect” (1 En. 38:4).109 In Sifre Numbers the

106 This has been noted by W. Bacher, Die Agada der Tannaiten (Strassburg: Trübner, 1903), 2:95 n. 2. 107 This light of the first day, though supernatural, is also thought of as concrete in rabbinic tradition. 108 Geʾez: lagassa, which should not necessarily be emended; see G. W. E. Nickelsburg and J. C. VanderKam, 1 Enoch 2 (Hermeneia; Minneapolis: Fortress, 2012), 95. 109 “This light is at once spiritual and physical: the nearness of God’s presence transfigures the countenance and person of His saints” (R. H. Charles, The Book of Enoch or 1 Enoch Translated . . . and Edited with Introduction, Notes and Indexes [Oxford: Clarendon, 1912], 71–72; Charles goes on to give many references to “light” in his notes on the Similitudes Hellenistic Jewish Writers and Palestinian Traditions 177 words “may God shine His face upon you” (Num 6:35) are interpreted as meaning “may God give you a shining face.”110 A more spiritual interpretation of “the shining light of one’s face” cannot be excluded, although more mundane interpretations (such as the one in Genesis Rabbah) are equally possible. The reconstruction of a rabbinic midrash based upon this association— “‘God hallowed the Sabbath’: with light”—seems rather safe. The meaning of this reconstructed midrash, as well as that of its interpretations, is opaque. How should we understand the statement that God hallowed the Sabbath “with the luminaries”? What is the meaning of “the shining light of a person’s face”: does it signify being illumined by God’s light or simply being peaceful and relaxed? The elusiveness of the answers to these questions is characteristic of rabbinic literature: it is not easy to penetrate the original meaning of the brief and often enigmatic sayings that were orally transmitted for generations, sayings that acquired new meanings through the long process of transmission. The original meaning of the reconstructed midrash concerning the light of the Sabbath remains obscure. It is conceivable, however, that the original content of this midrash might have been more “spiritual” than the various elaborations of this tradition that have come down to us in rabbinic literature; we might be seeing here another instance of concretization of traditions in rabbinic literature.111 To conclude: Philo and rabbinic tradition share an ancient foundational interpretation that connects the biblical statement that God hallowed the Sabbath (Gen 2:3, Exod 20:7), with the light of the Sabbath. This concept is

section of the Book of Enoch, but these are by and large irrelevant for the interpretation of this verse). :Sifre Num. 41; see M. I. Kahana, ed., Sifre on Numbers ,יאר ה’ פניו אליך—יתן לך מאור פנים 110 An Annotated Edition (3 vols.; Jerusalem: The Hebrew University Magnes Press, 2011), 1:110); in his commentary (2:320 n. 2), Kahana notes the parallel in 1QS 2:3: “May He illuminate your may מאור פנים The expression .(ויאר לבך בשכל חיים) ”heart with the discernment of life be taken either as anti-anthropomorphic (as suggested by Kahana, ibid.); metaphorical (as in the late midrashic parallels adduced by Kahana); nonmetaphorical, as in 1 Enoch; or even as expressing the consequence of a “mystical” experience (as in the case of Moses in Exod 34:29. I plan to deal elsewhere with ancient interpretations this passage). 111 On the dynamics of allegorization and concretization in rabbinic literature, see M. Kister, “Allegorical Interpretations of Biblical Narratives in Rabbinic Literature, Philo, and Origen: Some Case Studies,” in New Approaches to the Study of Biblical Interpretation in Judaism of the Second Temple Period and in Early Christianity (ed. G. A. Anderson, R. A. Clements, and D. Satran; STDJ 106; Leiden: Brill, 2013), 133–83. I do not see a theological motivation for rabbinic tradition to avoid a metaphorical use of “the light of the Sabbath,” since light metaphors are common in rabbinic literature, especially in designating the Torah. 178 Kister also attested in Aristobulus’s association of the Sabbath with the primeval light. In the writings of the two Hellenistic Jewish authors this interpretation has been expressed using Pythagorean terms, and construed in allegorical fashion. Hengel’s assertion, “it is remarkable how . . . Jewish–Palestinian and Pythagorean–Platonic and Stoic concepts are intermingled in Aristobulus”112 is corroborated by the present detailed study.

4 Conclusion

In these several studies we have seen various aspects of the connection between Hellenistic and Palestinian Jewish sources (especially rabbinic compilations). I have tried to demonstrate the common traditions and motifs that underlie these sources. Although much of the Palestinian material is documented in works dated to centuries after the Hellenistic Jewish material (rabbinic literature, piyyuṭ, targum, Samaritan literature), it can be shown that quite often the late material reflects ancient traditions, and that both late and early works share similar trajectories or even specific traditions. The late Palestinian material and the much earlier Hellenistic Jewish material mutually illuminate each other. I by no means underestimate the huge differences between the Hellenistic Jewish writers, such as Aristobulus and Philo; Qumranic and pseudepigraphic works; rabbinic literature of the first centuries CE and late antiquity (midrash, targum and piyyuṭ); Samaritan material; and early Christian writings. Notwithstanding the diversity of these corpora in terms of their periods and languages of composition, religious worldviews, and cultural settings, the studies here presented show the merits of an integrative study of these sources for the illumination of puzzling elements in both earlier and later texts; for tracing and reconstructing trajectories of biblical interpretation and aggadic traditions from the Second Temple period to late antiquity; for clarifying the many elements shared by seemingly quite disparate sources; and also for a better understanding of the interpretive “road[s] not taken.”

112 M. Hengel, Judaism and Hellenism (2 vols.; London: SCM Press, 1974), 1.169. The Severus Scroll Variant List in Light of the Dead Sea Scrolls

Armin Lange

This is among the words which are written in the book of the Law that was removed from Jerusalem with the captivity and taken to Rome. It was stored away in the Synagogue of Severus.1

These verses were written in the book of the Law that was found in Rome. It was stored away and sealed up in the Synagogue of Severus. With differences in letters and words.2

These two remarkable statements introduce four different copies of a list of thirty-three variant readings from the MT text of the Pentateuch.3 Before the Dead Sea Scrolls were found, these reports concerning the scroll in the Synagogue of Severus provided the only insight into the prerabbinic Hebrew textual history of the Torah aside from the Masoretic text, the Samaritan Pentateuch, and the Nash Papyrus. While continuing work on the Dead Sea manuscripts has dramatically changed the way we perceive the prerabbinic textual history of the Hebrew Bible, the Severus Scroll variant list remains an important textual witness to the early textual history of the Pentateuch, and its study should not be neglected. Furthermore, the publication of virtually all of the biblical Dead Sea Scrolls provides important new comparative evidence

* I am much obliged to Ruth Clements and Michael Segal who not only polished the English of this article but helped me to develop my arguments through their critiques. Further helpful suggestions were provided by Hillel Newman, to whom I am indebted as well. 1 Midrash Bereshit Rabbati on Gen 45:8. The above translation is based on J. P. Siegel, The Severus Scroll and 1QIsaa (SBLMasS 2; Missoula, Mont.: Scholars Press, 1975), 50. See for the Hebrew text, Midraš Berešit Rabbati ex libro R. Mosis Haddaršan collectus e codice Pragensi cum adnotationibus et introductione (ed. C. Albeck; Jerusalem: Mekitze Nirdamim, 1940), 209. 2 Farḥi Bible; Ms. Sassoon # 368, folios 146 and 403; and Ms. Hébreu 31 of the Bibliothèque Nationale, Paris, f. 399. 3 In this paper, I designate the copy of the variant list preserved in Midrash Bereshit Rabbati as SevMBR. The two copies of the list attested by folios 146 and 403 of the Farḥi Bible I label SevD146 and SevD403 respectively. The copy preserved in Ms. Hébreu 31 of the Bibliothèque Nationale in Paris, I designate as SevP. I will use the siglum SevComp to designate the text of the Pentateuch with which the compiler of the variant list compared the Severus Scroll.

© koninklijke brill nv, leiden, ��5 | doi ��.��63/9789004299139_009 180 Lange for the textual evaluation of the Severus Scroll variant list in its turn. The purpose of this paper, then, is to offer a preliminary, critical reassessment of the variants found in the list, in light of the Scrolls as well as other manuscript evidence, in the hope that this will provide a basis for further study.4

The two introductions quoted above state that the copy of the Torah from which the variants of the Severus Scroll list derive was stored in the genizah of a “Synagogue of Severus.” SevD146, D403 and SevP locate the manuscript in Rome; SevMBR states that this Torah scroll was brought from Jerusalem to Rome after the first Jewish war. Since Abraham Epstein’s initial publication of SevMBR,5 it has been assumed that this scroll was taken by Titus to Rome following the looting and destruction of the Jerusalem Temple in 70 CE. Josephus mentions such a scroll in his history of the first Jewish war ( J.W. 7.150 and 7.162). Epstein furthermore established the idea that the Severus Scroll was later donated to a Jewish community in Rome by the Roman emperor Alexander Severus (222– 235 CE), when he had a synagogue built for this community. The textual history of the Severus Scroll variant list demonstrates that biblical manuscripts from the Second Temple period might have survived until late antiquity. The compilation of the list sometime after the reign of Alexander Severus leaves little doubt that such manuscripts were studied and read carefully. In the eleventh century, this variant list was reproduced in Midrash Bereshit Rabbati (SevMBR),6 attributed to Rabbi Moshe ha-Darshan of Narbonne.7 In Midrash Bereshit Rabbati, the list follows a remark giving a

4 When I began my work on the Severus Scroll variant list, I thought I could address all unresolved issues connected with the list in a single, if somewhat extensive, scholarly article. I soon recognized, however, that a whole monograph is needed for a thorough study of the list. While I hope to publish such a monograph in due course, the constraints of the present volume allow for a summary of but a part of my research. For a more detailed study and all remaining issues I refer the reader to my longer study, The Severus Scroll: A Late Ancient Variant List of an Ancient Torah Manuscript in Light of the Dead Sea Scrolls, which is currently in preparation. 5 A. Epstein, “Ein von Titus nach Rom gebrachter Pentateuch-Codex und seine Varianten,” MGWJ 34 (1885): 337–51. 6 For Midrash Bereshit Rabbati, in addition to Albeck, Midraš Berešit Rabbati, 209, see also G. Stemberger, Einleitung in Talmud und Midrasch (9th rev. ed.; Munich: Beck, 2011), 394 and the literature listed there. 7 H. Mack, The Mystery of Rabbi Moses Hadarshan (Jerusalem: Bialik Institute, 2010), has recently argued that Midrash Bereshit Rabbati is the product of a longer history of literary growth, and that only parts of it may reliably be attributed to Moshe ha-Darshan. Severus Scroll Variant List 181 reading of the Torah scroll of Rabbi Meir for Gen 45:8.8 Three additional, and almost identical, copies of the list were included in two biblical manuscripts finalized some three hundred years later. Two of these copies are to be found on ff. 146 and 403 of the Farḥi Bible (Ms. Sassoon 368), a manuscript from the Catalonian part of Provence, dated to 1382; this manuscript was owned for many years by the Farḥi family of Damascus but is today one of the last remaining items in the Sassoon collection.9 The third copy is written on f. 399 of Ms. Hébreu 31 of the Bibliothèque Nationale in Paris, produced in Saragossa in 1402.10 The use of the Severus Scroll variant list by Moshe ha-Darshan of Narbonne in the eleventh century is just one example of this scholar’s employment of Second Temple traditions in Midrash Bereshit Rabbati. Both Martha Himmelfarb11 and Israel M. Ta-Shma12 have suggested that traditions stem- ming from Second Temple Jewish literature reached Narbonne and Moshe ha-Darshan in the Middle Ages via the Jewish communities of southern Italy, which had existed from Byzantine times. One may speculate that copies of the Severus Scroll variant list took similar routes to southern France and Moshe ha-Darshan, on the one hand; and to the closely related Jewish communities of Catalonia, the milieu of Elisha Crescas and Vidal Sattore (the scribes of the Farḥi Bible and Ms. Hébreu 31 respectively), on the other. The stemmatic relationship of these four copies can best be deduced from the differences in the sequence and repertoire of the variants they contain:

8 Several other parallel readings from Rabbi Meir’s Torah have also been put forward as witnesses to readings from the Severus Scroll. See Appendix 1 below on the significance of these parallels. 9 SevD146, D403. For a description of the manuscript, see D. S. Sassoon, OHEL DAWID: Descriptive Catalogue of the Hebrew and Samaritan Manuscripts in the Sassoon Library, London (2 vols.; Oxford: Oxford University Press, 1932), 1:xi, 6–14; and B. Narkiss, Hebrew Illuminated Manuscripts (Jerusalem: Keter, 1969), esp. 72. 10 SevP. For a description of the manuscript, see G. Sed-Rajna, Les manuscrits hébreux enlu- minés des Bibliothèques des France (Notices codicologiques, relevé des inscriptions par S. Fellous; Corpus of Illuminated Manuscripts 7; Oriental Series 3; Leuven: Peeters, 1994), 80–85. 11 M. Himmelfarb, “R. Moses the Preacher and the Testaments of the Twelve Patriarchs,” AJS Review 9 (1984): 55–78 (73–74); eadem, “Some Echoes of Jubilees in Medieval Hebrew Literature,” in Tracing the Threads: Studies in the Vitality of the Jewish Pseudepigrapha (ed. J. C. Reeves; SBLEJL 6; Atlanta: Scholars Press, 1994), 115–41 (115–18). 12 I. M. Ta-Shma, Rabbi Moses Hadarshan and the Apocryphal Literature (Studies in Jewish History and Literature; Jerusalem: Touro Graduate School of Jewish Studies, 2001) (in Hebrew). 182 Lange

1) The lists exhibit two distinct sequences for variants 9–16:13 SevP, SevD146, and SevD403: 9→10→11→12→13→14→15→16, SevMBR: 11→9→10→13→12→14→16→15. 2) SevP, SevD146, and SevD403 omit variant 17 (Exod 31:13), against SevMBR. 3) SevMBR and SevP include variant 20 (Lev 15:8), against SevD146 and SevD403. 4) The individual entries of SevMBR are often shorter than those in the other lists.

This all means that SevP, SevD146, and SevD403 belong to one textual tradition of the variant list, while SevMBR represents a different one. Since SevMBR agrees with SevP against SevD146 and SevD403 in including variant 20, SevP and SevMBR share the same archetype; SevD146 and SevD403 preserve a more corrupted version of the text of the SevP/D group than SevP. The relationship between SevP and SevMBR is confirmed by variants 5, 8, 12, 15, 16, 23, 24, 29, 31, and 32 where SevP reads with SevMBR against SevD146 and SevD403. The connections among the SevP/D group may be seen, on the other hand, in the prolegomenon and subscript of the variant list; these are almost identical in SevP, SevD146, and SevD403, while SevMBR has in both cases a significantly different text.14 Within the SevP/D group itself, SevD146 and SevD403 read a shorter text than SevP for a number of the variants, which often does not allow one to identify the biblical verse for which a variant reading is noted. Three observations emerge from this comparison. First, while SevP is the latest copy of the Severus Scroll variant list, it represents the best textual witness, although it is not itself free from scribal corruption. Second, the textual differences between the two copies of the Severus Scroll variant list in the Farḥi Bible (SevD146 and SevD403) show that even the same scribe could produce two slightly different versions of the same list. Third, the differences between SevMBR and SevP/D predate the use of the list in Midrash Bereshit Rabbati.15

13 See the discussion in Appendix 2 below, on the system followed here for the numbering of the variants. 14 The prolegomenon and subscript of this copy were most likely rephrased to integrate the list better into the text of Midrash Bereshit Rabbati. 15 This may be seen by looking at the variant for Gen 45:8, which introduces Bereshit Rabbati’s in ,לאב פרעה quotation of the list. SevP, SevD146, and SevD403 cite the variant reading as the ;וישני לאב לפרעה Bereshit Rabbati quotes the variant as .לאב לפרעה place of MT’s must actually reflect an error ,וישני and the erroneous לפרעה form, including the correct in the reproduction of the SevComp reading in the precursor to the list used by Bereshit Rabbati. The confusion in the midrash itself could only have occurred if Bereshit Rabbati’s copy of the list had mistakenly omitted the Severus reading altogether. This means that Severus Scroll Variant List 183

My observations result in the following stemma of the textual tradition of the Severus Scroll variant list.

Severus Scroll Compared Text

Severus Scroll Variant List

Sev D SevMBR Sev D

Sev P

1 The Severus Scroll Variant List: Scholarly Assessments and Critical Challenges

Influenced by Lieberman,16 Kutscher,17 and Siegel,18 most scholars today regard the Severus Scroll as a “vulgar” text, comparable to Qumran scrolls like 1QIsaa which go back to the late Second Temple period.19 For reasons of space

Bereshit Rabbati incorporated a copy of the Severus Scroll variant list which had already been abbreviated. See Appendix 1 for further discussion of the significance of this variant. 16 S. Lieberman, Hellenism in Jewish Palestine: Studies in the Literary Transmission, Beliefs and Manners of Palestine in the I Century B.C.E.–IV Century C.E. (2d ed.; Text and Studies of the Jewish Theological Seminary of America 18; New York: Jewish Theological Seminary of America, 1962), esp. 23–26. 17 E. Y. Kutscher, The Language and Background of the Isaiah Scroll (1QIsaa) (STDJ 6; Leiden: Brill, 1974), esp. 87–89 (originally published in Hebrew: Jerusalem: The Hebrew University Magnes Press, 1959). 18 Siegel, Severus Scroll. 19 See, e.g., M. H. Segal, “The Promulgation of the Authoritative Text of the Hebrew Bible,” JBL 72 (1953): 35–47 (45–47); M. Greenberg, “The Stabilization of the Text of the Hebrew Bible, Reviewed in the Light of the Biblical Materials from the Judean Desert,” JAOS 76 (1956): 157–67, esp. 161; E. Tov, Textual Criticism of the Hebrew Bible (Minneapolis: Fortress, 1992), 120 and 195 (cf. 2d rev. ed. 2001, 120 and 195; in the 3d rev. and expanded ed. [2012], Tov uses instead the phrase “free textual approach” [113 and 184]); J. H. Tigay, “ ‘Archeology’ of the Bible and Judaism in Late Antiquity and the Middle Ages,” in The Archeology of 184 Lange constraints I will cite here only the two representative assessments of the Severus Scroll by Kutscher and Siegel.20 Kutscher in his magisterial work on 1QIsaa evaluates the Severus Scroll as follows:

[T]his text was also of a Vulgar type, or perhaps rather a text influenced by this type (though it is of course possible that not all var. lect. of the copy were listed). The statement concerning this scroll’s origin is also interesting. It was said to have been brought to Rome after the destruction of the Temple. This tradition is written in Aramaic and hence is many centuries earlier than the compilation of the Midraš Berešit Rabbati as a whole.21

Siegel is less cautious than Kutscher:

The Severus Scroll and 1QIsaa were early vulgata of the sort which later Jewish scholars sought to withdraw from circulation because of known divergent readings and/or orthography . . . The three medieval sources for the Severus Scroll preserve truly ancient scribal conventions and variants.22

One of the key problems in the study of the Severus Scroll variant list is the lack of accurate transcriptions of its four extant copies. The early transcriptions by Epstein,23 Harkavy,24 and Neubauer25 were produced without access to proper photographs and suffer regularly from the modern scribal corruption of editorial mistakes. As an example, Harkavy’s transcription of the first variant in SevD146 (Gen 1:31) will suffice. Harkavy reads:26

והנה טוב מאד מות

Jordan and Beyond: Essays in Honor of James A. Sauer (ed. L. E. Stager, J. A. Greene, and M. D. Coogan; Winona Lake: Eisenbrauns, 2000), 490–97, esp. 492. 20 I will provide a detailed history of research in The Severus Scroll. 21 Kutscher, Language and Background, 89. 22 Siegel, Severus Scroll, 87. 23 See Epstein, “Titus,” passim; idem, “Biblische Textkritik bei den Rabbinen,” in Recueil des travaux rediges en memoire du Jubile scientifique de Daniel Chwolson (ed. D. Günzburg; Berlin: Calvary, 1899), 42–56, 49–55. Jerusalem: Karmiel, 1969–1970), 102–3 (abbreviated) חדשים גם ישנים ,A. A. Harkavy 24 reprint of the 1885 edition [pp. 4–5]); cf. also idem, “Things Old and New: Memories from My Trip to Jerusalem,” Hapisgah 1 (1895): 58–59 (in Hebrew). 25 A. Neubauer, “Der Pentateuch der sogenannten Severus-Synagoge,” MGWJ 36 (1887): 508–9. .cf. Epstein, “Biblische Textkritik,” 49 ;102 ,חדשים ,Harkavy 26 Severus Scroll Variant List 185

But the facsimile published by Loewinger leaves no doubt that SevD146 has in fact a very different text:27

וירא אלהים את כל אשר עשה והנה טוב מאד מות היה כתיב

The early and mostly unreliable transcriptions by Epstein, Neubauer, and Harkavy negatively influenced all subsequent studies until Loewinger. The first scholar to publish facsimiles of SevD146, SevD403, and SevP was Loewinger.28 However, he provided transcriptions and critical notes only for SevMBR and SevP (he discussed SevD146 and SevD403 in the critical notes to SevP). The only existing monograph on the Severus Scroll, by J. P. Siegel,29 includes the text of SevMBR, SevP, and SevD146, but not that of SevD403. Siegel’s book suffers not only from typographical errors,30 but also from grave errors in the transcription of SevMBR, SevD146 and SevP—including the accidental omission of one of the variant phrases from his SevD text.31 As a consequence of this problematic editorial history of the variant list, studies of the list up to the present time have been based for the most part on faulty transcriptions of its individual witnesses. Any text-critical analysis of the Severus Scroll variant list thus suffers from a flawed textual basis. As a first step towards remedying this situation, I present here a critical investigation of the list, in which I compare the extant variant readings of the Severus Scroll with all textual witnesses available from the Second Temple period, as well as the Masoretic manuscripts included in the works of Kennicott32 and De Rossi.33 I will then discuss selected variants and draw

ספר תורה שהיה גנוז בבית כנסת סוירוס ברומא: יחסו אל מגילות ישעיהו“ ,S. Loewinger 27 .Beth Mikra 15 (1970): 237–62, plate 2 ”,במדבר יהודה ואל ‘תורתו של רבי מאיר’ 28 S. Loewinger, “Prolegomenon,” in V. Aptowitzer, Das Schriftwort in der rabbinischen Literatur (ed. S. Loewinger; New York: Ktav, 1970; reprint of the edition of Vienna 1906), ”.ספר תורה“ ,vii–xlv, xxv–xxxviii; idem 29 Siegel, Severus Scroll. 30 Cf. the reviews by B. J. Roberts ( JTS 28 [1977]:132–34); L. H. Schiffman ( JBL 96 [1977]: 299– 300), and S. Z. Leiman ( JAAR 46 [1978]: 73–74). Since .את בכרתו ליעקב מכרתו כתי׳ ,Siegel (Severus Scroll, 90) does not list the variant 31 Siegel discusses this variant on pages 23–24 of his book, there can be no doubt that the text is missing due to inattentiveness. 32 B. Kennicott, Vetus Testamentum hebraicum: Cum variis lectionibus ex codicibus manu- scriptis et impressis (2 vols.; Oxford: Clarendon, 1776–1780). 33 G. B. De Rossi, Variae lectiones Veteris Testamenti librorum: Ex immensa manuscriptorum editorumque codicum congerie haustae et ad Samaritanum textum, ad vetustissimas versiones, ad accuratiores sacrae criticae fontes ac leges examinatae (5 vols.; Parma: Bodoni, 1784–1798). 186 Lange some conclusions by comparing the evidence of the Severus list with that of the biblical Dead Sea Scrolls.

2 A Critical Assessment of the Severus Scroll Readings34

As already argued above, none of the extent witnesses of the Severus Scroll variant list is without textual problems. All thirty-three variants are preserved only in SevMBR. But the text of SevMBR is abbreviated to such a degree that some variant readings become incomprehensible, while for others, only the text of SevComp is given. In addition to the different text sequences of SevMBR, SevD146, SevD403 and SevP which I described above, variant 17 (Exod 31:13) is missing in SevD146 and SevD403 as well as in SevP; and variant 20 (Lev 15:8) is missing in SevD146 and SevD403. The variant readings of the Severus Scroll which I list below follow the eclectic text of the Severus Scroll variant list which I have compiled through the critical study of all four witnesses.35

2.1 Ten Orthographic, Phonetic, and Paleographic Variants vs. MT 36

LXX καλὰ λίαν ;טוב מאד MT, SP [טוב מות Gen 1:31 Sev – כיתנת SP ;כתנות MT [כתנוד Gen 3:21 Sev –

34 The Masoretic text (MT) is quoted in the following lists according to the Biblia Hebraica Stuttgartensia (ed. K. Elliger and W. Rudolph; 5th rev. ed.; Stuttgart: Deutsche Bibelgesellschaft, 1997). The text of the Samaritan Pentateuch (SP) is quoted according to the edition of A. Tal and M. Florentin, The Pentateuch: The Samaritan Version and the Masoretic Version (Tel Aviv: Tel Aviv University Press, 2010 [in Hebrew]). The Septuagint is quoted according to the edition of J. W. Wevers in the series Septuaginta Vetus Testamentum Graecum of the Göttingen Academy of Sciences. The text of the LXX is given in the lists below only if relevant. If the Greek text translation could be the result of either the Severus Scroll reading or the MT reading—i.e., if the LXX is inconclusive—its reading is not listed. The same is true for SP readings, which are not included when the paleography of the Samaritan alphabet does not allow for comparison with the square script manuscript of the Masoretic text. This is especially true in the case of final characters, which are absent in the Samaritan script. 35 See Lange, The Severus Scroll; and see Appendix 2 below for a synoptic table of the four witnesses. 36 In the following lists, MTL refers to Codex Leningradensis, MTKen and MTRos, followed by a number, refer to the medieval manuscripts collected by Kennicott (Vetus Testamentum hebraicum) and De Rossi (Variae lectiones) respectively. The sign > indicates that text is lacking in one or more textual witnesses. Biblical Dead Sea Scrolls are labeled according to E. Tov, Revised Lists of the Texts from the Judaean Desert (Leiden: Brill, 2010). Severus Scroll Variant List 187

יום מותי MT, SP [יוממותי Gen 27:2 Sev – השדה SP ;שדה MT [סדה Gen 27:27 Sev – בן עדה MT, SP [בנעדה Gen 36:10 Sev – 4QGenf, LXX < ;שם MT, SP [שמ Gen 48:7 Sev – מדם MT [מדמ Lev 4:34 Sev – אביתם MT [אביתמ Deut 1:26 Sev – הם MT [המ Deut 3:20 Sev – הבנים MT, SP [האבנים Deut 22:6 Sev –

2.2 Twenty-Three Textual Variants vs. MT

and [- מ̊ ה] cf. 4QAges of Creation A [4Q180] 2–4 ii 6) הכצעקתם Gen 18:21 Sev – 4QAges of Creation A (4Q180) 2–4 ;הך צעקתה SP ;הכצעקתה LXX [αὐτῶν])] MT LXX εἰ κατὰ τὴν κραυγὴν αὐτῶν ;ה̊זעקת̊מ̊ה̊ ii 6 ומארץ MT, SP [ומארע Gen 24:7 Sev – LXX τὰ πρωτοτόκια ;בכורתו SP ;בכרתו MT [מכרתו Gen 25:33 Sev – LXX Ἰεοὺς ;יעוש 4QGen–Exoda, MTQere, SP [יעיש Gen 36:5 Sev, MTKetib – LXX Ἰεοὺς ;יעוש MTQere, SP, 1 Chr 1:35 [יעיש Gen 36:14 Sev, MTKetib – ,MTL [מצרימה Gen 43:15 Sev, SP, MTKen 4, 17, 18, 84, 150, 196, 615, MTRos 6, 16, 549 – מצרים MTKen 155, and MTRos 262, 419, 443, 543, 668 לאב לפרעה cf. LXX ὡς πατέρα Φαραὼ] MT, SP ;לאב פרעה Gen 45:8 Sev – מצרימה MT, SP [מצרים Gen 46:8 Sev, MasGen – LXX ἐκ Ῥαμεσσὴ ;מרעמסס 2QExoda, MT, SP [מרעמס Exod 12:37 Sev – LXX τοῖς υἱοῖς Ἰσραήλ ;לבני ישראל MT, SP [לבית ישראל Exod 19:3 Sev – lxx μοχλούς ;בריחים ,Exod 26:27 > Sev] 4QpaleoExodm, MT, SP – היא 4QpaleoExodm, MTQere, SP [הוא Exod 31:13 Sev, MTKetib – LXX ἄμωμον ;תמימה MT, SP [תמימים Lev 14:10 Sev – LXX ὕδατι ;במים MT, SP [במים חיים Lev 15:8 Sev – [(כל הבא :LXX πᾶς ὁ εἰσπορευόμενος (also MTSebir ;כל הבא Num 4:3 Sev, SP – כל בא MT ,[כ]ל בא 2QNuma ,כל[ ב] א̊ 4QLev–Numa לדרתיכם MT, SP [לדריכם Num 15:21 Sev – LXX καὶ ἔσχατον ;ואחר תאסף SP ;אחר תאסף MT [אשר תאסף Num 31:2 Sev – προστεθήσῃ – Num 31:12 Sev, MTL, MTKen 128, 170, MTRos 2, 12, 18, 187, 196, 230, 251, 262, 296, 419, 440, ,MTKen 4, 9, 69, 75, 80, 104, 107, 109 [ואל עדת בני ישראל 851 ,683 ,669 ,668 ,656 ,649 ,529 ,444 129, 132, 150, 176, 181, 184, 196, 199, 226, 228, 615 and MTRos 16, 174, 464, 467, 495, 503, 543, 549, LXX καὶ πρὸς πάντας υἱοὺς Ἰσραὴλ ;ואל כל עדת בני ישראל SP ,853 ,789 ,678 בני MTL, MTRos 16, 592, 766, SP [בן יוסף Num 36:1 Sev, MTKen 107, 168, 247, MTRos 262 – LXX υἱῶν Ἰωσήφ ;יוסף LXX Ἀμορραίων ;האמרי MT, SP [האמר Deut 1:27 Sev – 188 Lange

LXX κατακεκαυμένον ;שרפה MT, SP [שרפת Deut 29:22 Sev – LXX ὥσπερ κατεστράφη ;כמהפכת MT, SP [כמהפת Deut 29:22 Sev – LXX Διασπερῶ αὐτούς ;אפיהם SP ;אפאיהם MT [אף איהם Deut 32:26 Sev –

3 The Character of the Severus Scroll Variant List

For a proper analysis of the Severus Scroll variants, it needs to be understood that the compiler of the list noted variants with respect to only one late ancient manuscript, which came out of the broader textual tradition of the consonantal text of MT. This means that any reconstruction of the relationship of the Severus Scroll to other ancient texts and textual versions must operate from the presupposition that the Severus Scroll read with MT, except when the compiler of the variant list indicated otherwise. This raises the question of how carefully the original variant list of the Severus Scroll was compiled; that is, did it include only a small part of the variant readings attested by the Scroll, or most if not all of them? The answer to this question would not only allow for speculation as to how the Severus Scroll related to other ancient texts and textual versions of the Pentateuch but also for an approximation as to how many variant readings the compiler might have missed when he produced his list in late antiquity.

3.1 The Nature of the Severus Scroll Variants For the most part, the variants recorded by the Severus Scroll variant list do not extend beyond a few characters in a given word. The total deviation of the Severus Scroll from the text with which it was compared amounts to only 53 characters, including missing or added spaces. In four cases, the list tags the use of medial mem in final position (see Gen 48:7; Lev 4:34; Deut 1:26 and 3:20); in Gen 27:2; 36:10; and Deut 32:26, the difference recorded between the list and its comparative text is the lack or presence of a space between words. Three additional cases concern differences in the in Gen 27:27 שדה for סדה orthographic representation of phonemes, such as (cf. the variant readings in Gen 1:31; 3:21), or the use of an Aramaic form instead Three cases of waw–yod .(ומארץ instead of ומארע of a Hebrew one (Gen 24:7 confusion can be found (Gen 36:5, 14; Exod 31:13); in four more cases a single in Deut 29:22; cf. Exod 12:37) or confused כמהפת ,.character was omitted (e.g in Gen 25:33; cf. Deut 29:22). Grammatical disagreements include מכרתו ,.e.g) the use of singular rather than plural forms (Num 36:1) and vice versa (Lev 14:10); the use of different gender endings (Num 15:21; cf. Lev 14:10); the use of Severus Scroll Variant List 189 different suffixes (Gen 18:21); the presence or absence of a he-locale (Gen 43:15; 46:8); and the use of a gentilic noun (Deut 1:27). Differences in wording extend Exod 19:3). Both the short and long texts) לבני and לבית only to the confusion of of the Severus Scroll variant list take note of the presence or absence of both a preposition (Gen 45:8) and an article (Num 4:3). Only in three cases does the in Num 31:12) or כל ;Exod 26:27 בריחים) Severus Scroll list report a missing word .(in Lev 15:8 חיים) an added term The character of these variants has two significant implications. First, the compiler of the Severus Scroll variant list paid close attention to tiny textual and orthographic details and noted tiny variant readings which could easily have eluded him. The Severus Scroll variant list was therefore carefully executed and meticulously compiled. Second, and following from this observation, the Severus Scroll itself seems not to have featured deviations of major significance from the scroll with which it was compared. The possibility may not be excluded, of course, that the compiler missed some variant readings attested in the Severus Scroll. It seems unlikely, though, that someone as metic- ulous as this compiler would have overlooked more than a negligible amount of such evidence. I will come back to this issue below.

3.2 The Nature of the Severus Scroll’s Comparison Text All studies of the Severus Scroll variant list of which I am aware simply presup- pose that the compiler compared the Severus Scroll with a “master” text identical with the consonantal text of MT, which in turn is seen to be represented by the consonantal text of the Leningrad Codex. These, however, are problematic assumptions; given the seeming antiquity of the list itself,37 the Severus Scroll’s comparison text must also date to relatively early in the Amoraic period, that is, well before the Leningrad Codex. Limited textual diversity in rabbinic times seems to be confirmed, in fact, by rabbinic quotations of biblical texts.38 Of

37 See the assessments of Kutscher and Siegel, noted above (p. 184). 38 See, e.g., the studies of Aptowitzer, Das Schriftwort in der rabbinischen Literatur; H.-G. von Mutius, Nichtmasoretische Bibelzitate im Midrasch Ha-Gadol (13./14. Jahrhundert) (Judentum und Umwelt 80; Frankfurt a.M.: Lang, 2010). Especially instructive are the variant readings from rabbinic quotations noted in the extant volumes produced by the Hebrew University Bible Project: M. H. Goshen-Gottstein, ed., The Book of Isaiah (Jerusalem: The Hebrew University Magnes Press, 1995); C. Rabin, S. Talmon, and E. Tov, eds., The Book of Jeremiah (Jerusalem: The Hebrew University Magnes Press, 1997); M. H. Goshen-Gottstein and S. Talmon, eds., The Book of Ezekiel (Jerusalem: The Hebrew University Magnes Press, 2004). 190 Lange course, this kind of textual variation should not be compared with the textual plurality of Second Temple Judaism,39 since it results mostly if not exclusively from comparatively limited scribal corruption. With regard to its comparison scroll, the Severus Scroll variant list points to a similarly limited textual plurality. I have noted above a total of six cases (Gen 36:5, 14; 43:15; Exod 31:13; Num 31:12; 36:1) where a Severus Scroll variant reading agrees with an MT Ketib against a Qere reading, or where different Masoretic manuscripts support the readings of both Sev and its comparison scroll. A good example for this group of readings is Num 31:12, where the The Severus .ואל כל עדת בני ישראל in the phrase כל Severus Scroll did not have is also that of the Leningrad Codex (MTL), while SP ואל עדת בני ישראל ,reading -If the com .כל and several other Masoretic manuscripts read the phrase with piler of the variant list noted variant readings which nevertheless agree with the Leningrad Codex and other important MT codices, then the comparison manuscript was not itself identical with the text of the Leningrad Codex: it represented an alternate stream of the textual transmission of the consonantal text of MT. For lack of a better term I call this alternate text SevComp. The variant readings of SevComp deserve to be studied in their own right for the light they shed on the textual transmission of the Pentateuch in late antiquity. But such an analysis goes beyond the constraints of the present article and will have to be done elsewhere.

3.3 The Orthographic, Phonetic, and Paleographic Variants of the Severus Scroll Among the nontextual readings of the list, two different types of variants may be distinguished in the Severus Scroll: variants of paleographic representation, and variants in the depiction of phonemes. Several of the orthographic variants in the Severus Scroll are of the first type. Four out of the total of ten nontextual variants involve the use of medial mem in final position:

4QGenf, LXX < ;שם MT, SP [שמ Gen 48:7 Sev – מדם MT [מדמ Lev 4:34 Sev – אביתם MT [אביתמ Deut 1:26 Sev – הם MT [המ Deut 3:20 Sev –

39 Contra von Mutius, Nichtmasoretische Bibelzitate, xvii–xxi. Severus Scroll Variant List 191

One example of medial mem in final position can actually be found in the Since the publication of 1QIsaa, it 40.(המ פרוצים) Masoretic text of Neh 2:13 has become the scholarly consensus that occurrences of medial mem in final position as attested by the Severus Scroll variant list and MT Neh 2:13 reflect an earlier scribal practice, also attested, for example, in early Hasmonean manuscripts.41 Based on the evidence of the Severus Scroll and 1QIsaa, Siegel wanted to reconstruct an ancient orthographic rule that monosyllabic words were to be written with medial mem in final position.42 But following the publication of all the Dead Sea Scrolls it has become evident that medial characters are used in final position with monosyllabic words (4QPsx 2 [Ps 89:20]; 4QPsx 8 [Ps 89:31]; 4QToh A [4Q274] 2 i 2, 8); bisyllabic words (4QpapSa [4Q255] 2 5; 4QHa [4Q427] 7 ii 10, 17; 4QApocryphal Psalm and Prayer [4Q448] 2:2, 3, 9); and even longer words (4QPsx 1 [Ps 89:20]; 4QTa? [4Q365a] 2 ii 6; 4QApocryphal Psalm and Prayer [4Q448] 2:8; 3:1).43 Furthermore, the Severus Scroll variant list itself attests to a use of a medial mem in final position in a trisyllabic word: .אביתמ Deut 1:26 More paleographic reference material from the Dead Sea Scrolls and elsewhere also allows the conclusion that the use of final characters became consistent in Hebrew writing late in the Second Temple period. Ada Yardeni44 observes an inconsistent usage of final characters in the various manuscripts from the Dead Sea, until the Herodian book hand developed. Describing the early Herodian book hand, she notes: “A regular distinction between medial and final forms already exists in the early Herodian book-hand, and the medial letter-signs become somewhat shorter.”45 With all probability the four uses of medial mem in final position in the Severus Scroll thus go back to a time

40 Cf. J. Leeven, “The Orthography of the Hebrew Scroll of Isaiah A,” in Proceedings of the Twenty-Second Congress of Orientalists Held in Istanbul September 15th to 22nd, 1951 (ed. Z. V. Togan; 2 vols.; Leiden: Brill, 1953–1957), 2:577–83, 578. Segal, “Promulgation,” 47, also points to the use of medial nun in final position in the Ketib of Job 38:1 and 40:6. 41 See Leeven, “Orthography,” 578–79; Kutscher, Language and Background, 88; Loewinger, .250 ”,ספר תורה“ ,Prolegomenon,” xxxi; Loewinger“ 42 Siegel, Severus Scroll, 1–14. 43 Cf. Emanuel Tov, Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (STDJ 54; Leiden: Brill, 2004), 230–32; the list is taken from page 232. 44 A. Yardeni, “A Deed of Sale from the Judaean Desert: Naḥal Ṣeʾelim 9,” Tarbiz 53 (1994): 299–320 + xix, 308 (in Hebrew); eadem, The Book of Hebrew Script: History, Paleography, Script Styles, Calligraphy & Design (3d ed.; Jerusalem: Carta, 2010), 164–76. 45 Yardeni, Hebrew Script, 174. 192 Lange before the early Herodian book hand. But the fact that the Severus Scroll variant list notes only four cases of this praxis shows that the use of medial characters in final position was not a systematic feature of the Severus Scroll. The four variants in question should be understood as overlooked instances of an earlier orthographic convention in the manuscript tradition of the Torah that survived an orthographic revision in the time of the early Herodian book hand. Another difference in paleographic representation between the Severus Scroll and MT is seen in two cases where two words were written together, rather than separately as in the MT:

יום מותי MT, SP [יוממותי Gen 27:2 Sev – בן עדה MT, SP [בנעדה Gen 36:10 Sev –

Both cases involve relatively short words which are closely connected syntacti- MTL as well as other Masoretic codices have a maqqep ,בנעדה cally. In the case of Joining together two or three syntactically .בן־עדה ,.i.e ,עדה and בן between related words reflects a practice attested in Dead Sea Scrolls from the second century BCE through the first century CE. One case taken from Mur 88 (Minor instead ואימזה Prophets Scroll) may serve as an example. Mur 88 10:14 reads -in Jonah 1:8.46 It should be noted that not only ancient manu וְאֵי־מִ זֶ ה of MT’s scripts but also medieval Masoretic codices disagree among themselves as to whether some words should be connected with a maqqep or written together. Two examples can be found in the ninth-century Karasu–Bazar codex of the -respec , חֶ פְ י צִ ֔ בָ ּה and ּפְקַ חקֽוחַ ,Latter Prophets, which reads in Isa 61:1 and 62:4 47. חֶ פְ י צִ ־ ֔ בָ ּה and ּפְקַ ח־קֽוחַ tively, rather than It should further be noted that the proto-Masoretic biblical manuscripts from Masada, Naḥal Ḥever and Wadi Murabbaʾat‌ (MasPsa, 5/6ḤevPs, MurGen, MurExod, MurNum, MurIsa) leave almost no spaces between their words.48 All of these scrolls were produced in the first century CE or the early second cen- in the Severus בנעדה and יוממותי tury CE. The omitted space between the words Scroll could therefore easily be the result of a scribal error which occurred in copying such a Vorlage, with almost no spaces. The omission could have like-

46 For more examples, see Tov, Scribal Practices, 133–35. 47 See the list of orthographic variants in F. I. Andersen, “The Orthography of D62,” in Studies in Hebrew and Aramaic Orthography (ed. D. N. Freedman, A. D. Forbes, and F. I. Andersen; Biblical and Judaic Studies 2; Winona Lake: Eisenbrauns, 1992), 253–93 (278). For the date of the Karasu–Bazar codex, see ibid., 256–58. 48 Cf. Tov, Scribal Practices, 132, 133. Severus Scroll Variant List 193 wise occurred, however, any time later in the (proto-)Masoretic textual tradi- the Severus Scroll is thus well ,בנעדה and יוממותי tion. In its writing together of within the margin of textual differences among proto-Masoretic and medieval Masoretic manuscripts. The remaining orthographic variants of the Severus Scroll involve the phonetic depiction of specific words. In Gen 27:7 the Severus Scroll used a samekh Samekh .(השדה SP ,שדה MT [סדה to denote the phonetic value of a śin (Sev is used to represent the character śin in several Dead Sea Scrolls, as well as in Hebrew inscriptions up to the first century CE. Examples include a phylac- is spelled with samekh in 4QPhyl שדה :tery and a manuscript of the War Scroll 49.(ספות) Deut 5:21); and samekh represents śin in 1QM 10:12 סדהו) G (4Q134) 1:26 Two further examples of orthographic variants in the Severus Scroll include the representation of the voiced and voiceless alveolar plosives (as in the the Hebrew character dalet , מְ ֹא ד English words debt and tap). In the case of normally represents the voiced alveolar plosive, but the Severus Scroll in A similar confusion of characters can .(טוב מות) Gen 1:31 used a taw instead be observed in the transcription of dalet by θ in Septuagint manuscripts; as ζαβουθ or ζαββουθ in 1 Kgs 4:5 זבוד an example is the transliteration of (LXXA, B).50 In Gen 3:21, the voiceless alveolar plosive is represented by dalet instead כתנוד) instead of the taw which is normally employed for this purpose Loewinger points to a parallel for this phenomenon in 1QIsaa 50:28 .(כתנות of The interchangeability of dalet 51.גת instead of MT’s גד Isa 63:2), which reads) and taw makes it unlikely, to my mind, that the difference in consonants indicates a difference in pronunciation; i.e., pronunciation as fricatives (as in English this/then) rather than plosives. Note, too, that the Severus Scroll, like many Dead Sea Scrolls, may employ waw rather than aleph as a mater lectionis to represent ō. Kutscher showed -might have been pronounced mōd, and might thus occa מְ ֹא ד that the adverb [4QPsf [4Q88) מוד see Isa 16:6 in 1QIsaa 13:23) or) מואד sionally be spelled

49 Cf. E. Qimron, The Hebrew of the Dead Sea Scrolls (HSS 24; Atlanta: Scholars Press, 1986), 24; for further examples of this phenomenon, see Kutscher, Language and Background, 185; and Siegel, Severus Scroll, 26. 50 See A. Sperber, “Hebrew Based upon Greek and Latin Transliterations,” HUCA 12–13 (1937– 1938): 103–274 (128–29, 146–47; further examples can be found on pp. 128–29); idem, A Historical Grammar of Biblical Hebrew: A Presentation of Problems with Suggestions for Their Solution (Leiden: Brill, 1966), 226, 519. Cf. also Loewinger, “Prolegomenon,” xxviii; .247 ”,ספר תורה“ ,idem .48–247 ”,ספר תורה“ ,Loewinger, “Prolegomenon,” xxix; idem 51 194 Lange

8:10 [Apostrophe to Zion]; 11QPsa [11Q5] 22:1 [Apostrophe to Zion]; 4QTohorot A [4Q274] 3 ii 9).52 In the Severus Scroll at Gen 1:31, both orthographic pecu- liarities—i.e., the representation of the voiced alveolar plosive by taw instead of dalet and the use of waw instead of aleph as a mater lectionis—coincided, It seems likely that the Vorlage of the Severus Scroll (or .מות became מאד and a form that was then ,מוד an even earlier copy in its textual tradition) read either by the Severus Scroll or its Vorlage. 53 ,מות changed to Another phonetic variant listed for the Severus Scroll is an unusual plene spelling in Deut 22:6, where aleph expresses the patach of the determinative ha The use of aleph for this purpose is rare even within .(הבנים MT, SP [האבנים Sev) so-called Qumran orthography, but it is not unheard of. Kutscher points to the in 1QIsaa 47:16 (Isa 59:5).54 וְהַ זּורֶ ֖ ה for והאזורה form The survival of orthographic irregularities in the phonetic depiction of words in the Severus Scroll should not be overemphasized, because the same phenomenon can be observed in the Leningrad Codex itself. As an example, I note that in forty-three cases, the Leningrad Codex spells the suffix of the sec- rather than ,-כה ond person masculine singular with the consonants kaph he and ( ַ ו אֲ בָ רֶ ְ כ כָ ֛ ה) Examples include Gen 27:7 .ך with the customary final kaph In my opinion, the remnants of divergent orthographic .(עַ ל־יָ �֣דְ כָ֔ ה) Exod 13:16 systems in the text of MT itself as well as in the Severus Scroll can best be explained as the result of an orthographic revision that overlooked isolated readings.55 To summarize thus far: None of the orthographic variants recorded for the Severus Scroll differs morphologically from the MT. These variants are mostly paleographic in nature, and otherwise involve diversity in phonetic depictions, including a discrepancy in the use of a mater lectionis. All these variants may be understood as traces of earlier orthographic features which were overlooked in an orthographic revision of the text of the Torah.

52 See Epstein, “Titus,” 342–43; idem, “Biblische Textkritik,” 51–52, writing even before the Dead Sea Scrolls were found; cf. esp. Kutscher, Language and Background, 87. 53 Siegel, Severus Scroll, 17, argued similarly, but without reference to 4QPsf (4Q88) 8:10 (Apostrophe to Zion); 11QPsa (1Q5) 22:1 (Apostrophe to Zion); 4QTohorot A (4Q274) 3 ii 9. 54 Kutscher, Language and Background, 88, 161; cf. Siegel, Severus Scroll, 40, with further examples. 55 On this phenomenon, see A. Lange, “The Question of the So-Called Qumran Orthography, the Severus Scroll, and the Masoretic Text,” Hebrew Bible and Ancient Israel 3 (2014): 424–75. Severus Scroll Variant List 195

3.4 The Textual Variants of the Severus Scroll Assessing the textual character of the Severus Scroll in comparison with other textual witnesses of the Torah from the Second Temple period meets several obstacles:

1) The ancient compiler of the Severus Scroll variant list collected variant readings against only a single proto-Masoretic manuscript. This means we must presume that the Severus Scroll read with MT against the LXX, or the SP, or a given Qumran Scroll, only in those cases where these textual witnesses differ from MT. 2) As noted above (pp. 189–90), the manuscript with which the compiler of the variant list compared the Severus Scroll was at variance with the medieval Ben Asher master copy, MTL. 3) As noted above, it cannot be excluded that the compiler of the Severus Scroll variant list overlooked some variant readings in the Severus Scroll. The number of these overlooked variants cannot have been very high, however, given his general meticulousness.

Even taking these obstacles into consideration, however, the comparison of the textual variants of the Severus Scroll variant list with other witnesses from the Second Temple period yields some significant results. In general, three different kinds of textual variants can be distinguished in the Severus Scroll list, 1) cases of scribal error, 2) cases of textual harmonization, and 3) rare cases in which the Severus Scroll preserves a more original reading against its comparison text. Against Siegel’s claims,56 I have found no variants in the list which go back to scribal interpretative efforts, beyond harmonization. For reasons of space, I will not discuss here all the textual variants of the Severus Scroll. Instead, I will present some sample cases for each type of textual variant. Examples of scribal error are listed for Gen 24:7 and Exod 12:37. At Gen 24:7, as attested in ,ומארץ instead of ומארע the Severus Scroll reads the Aramaism MT and SP. This variant goes back either to a confusion of ʿayin and ṣade,57 or for ומארע to the Aramaic mindset of a bilingual scribe who wrote the Aramaic

56 Siegel, Severus Scroll, 15–42. .Siegel, Severus Scroll, 21–22 ;248 ”,ספר תורה“ ,Loewinger 57 196 Lange

was מרעמסס At Exod 12:37, the second samekh of 58.ומארץ the Hebrew word cf. LXX ἐκ Ῥαμεσσὴ).59 ;מרעמס lost due to haplography (Sev A good example of textual harmonization is found at Lev 15:8. In Lev LXX ὕδατι) was clearly ;במים compare with MT, SP) במים חיים the phrase ,15:8 created by the Severus Scroll or its Vorlage in harmonization with Lev 15:13 60. וְרָחַץּבְׂשָ רו ּבְמַ יִם חַ ּיִים An example where the Severus Scroll has a more original reading can be found in Gen 25:33:

LXX τὰ πρωτοτόκια ;בכורתו SP ;בכרתו MT [מכרתו Gen 25:33 Sev –

the privilege of the firstborn,” MT, SP, LXX), the Severus“) בְ כ ֺרָ ה Instead of sword”);61 otherwise attested“) מְ ֵ כ רָ ה Scroll employs in Gen 25:33 the rare word occurs also Gen בְ כ ֺרָ ה only in Gen 49:5, where its meaning is debated.62 Since as Siegel מְ ֵ כ רָ ה it is unlikely that a scribe changed it intentionally to ,34 ,32 ,25:31 argues: “The Severus Scroll variant suggests that Rome was an illegitimate master of Israel, with only tenuous and temporary claims to the Land.”63 To argue for such an exegetical variant presupposes not only ongoing Roman rule over Judea but also a typological interpretation of Esau as Rome. Resistance to Roman rule is most likely in the first century CE. But Siegel envisions that the Severus Scroll was copied in Hasmonean times! Furthermore, the typological

:in idem, Ketav Lashon Wa-Sefer ”,ספר אורייתא דאשתכח ברומא“ ,Cf. A. M. Habermann 58 Reflections on Books, Dead Sea Scrolls, Language and Folklore (Jerusalem: Rubin Mass, 1973), 166–75 (170) (first published in Sinai 32 [1953]: 161–67). 59 I see no reason to regard this reading as an abbreviation (thus originally C. D. Ginsburg, Introduction to the Massoretico-Critical Edition of the Hebrew Bible [London: Trinitarian Bible Society, 1897], 416); the Dead Sea Scrolls have demonstrated that abbreviations are nonexistent in biblical manuscripts from the Second Temple period (Tov, Scribal Practices, 235; cf. Tov, Textual Criticism [3d ed.], 238–39). Haplography is much more common and would better explain the Severus Scroll reading. 60 Cf. Ginsburg, Introduction, 418; Segal, “Promulgation,” 46. with the וימכר את מכרתו It would of course be simpler to assume a figura etymologica 61 meaning “he sold him that which he sold” (thus the suggestion of Michael Segal). But merchandise”) nor a feminine“) מֶ כֶ ר in prerabbinic Hebrew neither a feminine form of business, sale”) is attested. Such a feminine form is required, though, by the“) מַ ּכָ ר form of sales”) is attested only in rabbinic“) מכרות The plural noun .מכרתו of ת– feminine ending literature. see D. J. A. Clines, ed., The Dictionary of , מְ ֵ כ רָ ה For the range of possible significations of 62 Classical Hebrew (8 vols.; Sheffield: Sheffield Academic Press and Sheffield Phoenix Press, 1993–2012), 5:274. 63 Siegel, Severus Scroll, 24; cf. ibid., 23–24. Severus Scroll Variant List 197 correlation between Esau and Rome is absent from prerabbinic Jewish literature, as far as we know, and became prominent only later, in rabbinic litera- goes back to a מכרתו ture.64 M. H. Segal therefore considers that the reading scribal confusion of bet and mem.65 While a bet–mem confusion66 is attested in some Septuagint readings, it is not as widespread as one would expect in the biblical Dead Sea Scrolls. It seems hence more likely to me that the Severus :a more original reading than the extant witnesses מכרתו Scroll preserves with Together with his birthright, Esau sold his sword, the latter being a symbol of is a rare word which is difficult to understand in the מְ ֵ כ רָ ה his power.67 Because context of Gen 25:33, the Severus Scroll attests to the lectio difficilior. Early on in the textual transmission of the book of Genesis a scribe corrected the diffi- in Gen 25:33 to harmonize it with the occurrences of מכרתו cult-to-understand in Gen 25:31, 32, 34. In addition to the Severus Scroll, a reflection of this בְ כ ֺרָ ה more original reading might have been preserved in Gen. Rab. 98:5 which correlates Gen 49:5 and Gen 25:33 explicitly.68 Disregarding questions of scribal error, scribal harmonization, or textual originality, it is also of interest to note that five variant readings in the Severus Scroll list are attested by other textual witnesses from the Second Temple period outside of the proto-Masoretic textual tradition; namely, the Old Greek, the Samaritan Pentateuch, and MasGen. In Gen 18:21, the third person masculine plural suffix in Sev corresponds to the Septuagint’s reading αὐτῶν In Gen 43:15, the Severus Scroll .([- מ̊ ה] cf. 4QAges of Creation A [4Q180] 2–4 ii 6) ,Conversely, in Gen 46:8 .מצרים against MT’s מצרימה agrees with SP in reading attested מצרימה with MasGen, instead of the מצרים the Severus Scroll reads פרעה by MT and SP. Furthermore, while in Gen 45:8 the Severus Scroll reads of MT and SP, its reading is remarkably close to the לאב לפרעה against the לאב הבא Septuagint’s ὡς πατέρα Φαραὼ. Finally, in Num 4:3, the Severus Scroll reads .of MT, 4QLev–Numa, and 2QNuma בא with SP and LXX, against the These five agreements with other textual witnesses from the Second Temple Period outside the proto-Masoretic textual tradition provide comparative evidence beyond that of 1QIsaa to corroborate the antiquity of the Severus Scroll variants, and hence of the Severus Scroll itself.

64 For the interpretation of Esau in Jewish literature see the volume of essays, Esau—Bruder und Feind (ed. G. Langer; Göttingen: Vandenhoeck & Ruprecht, 2009). 65 Segal, “Promulgation,” 47; cf. A. Sperber, Historical Grammar, 479–80. 66 For the confusion of bet and mem, see Tov, Textual Criticism (3d ed.), 230–31. ,Loewinger ;170 ”,ספר אורייתא“ ,Cf. Epstein, “Biblische Textkritik,” 52; contrast Habermann 67 .248 ”,ספר תורה“ 68 Cf. Siegel, Severus Scroll, 24. 198 Lange

3.5 Severus Scroll “Variants” in Agreement with the Masoretic Tradition Severus Scroll readings are not attested solely in the biblical Dead Sea Scrolls, the Old Greek, and the Samaritan Pentateuch. As discussed above (3.2) the Severus Scroll agrees six times with parts of the MT textual tradition as well. In three cases, i.e., Gen 36:5, 14; and Exod 31:13, the Severus Scroll reads with the MT-Ketib against the MT-Qere reading. In three additional cases—Gen 43:15; Num 31:12; and 36:1—a Severus Scroll reading is aligned with a group of medieval Masoretic manuscripts; in the case of Num 31:12 this includes MTL, as well. In addition to these six passages, there are two more cases in which the Severus Scroll reads with the consonantal text of MT against other texts from the Second Temple period.

השדה LXX ἀγροῦ] SP ;שדה MT ;סדה Gen 27:27 Sev – 4QGenf, LXX < [שם MT, SP ,שמ Gen 48:7 Sev –

The compiler included these two readings in the list because of their ortho- Gen 48:7 medial ;ש instead of ס graphic differences from SevComp (Gen 27:27 in final position; see discussion above, pp. 188, 190–193). Apart from these מ orthographic characteristics, however, the Severus Scroll reads in both cases with MT against other texts from the Second Temple period. That Sev agrees several times with MT against the Old Greek, the Samaritan Pentateuch, and even a biblical Dead Sea Scroll raises the issue of its textual affiliation. Was Sev really a vulgar manuscript69 or was it more closely related to the proto- Masoretic textual tradition than previously thought?

4 The Severus Scroll Variants in Context

Do the orthographic and textual characteristics of the Severus Scroll described above justify its categorization as a “vulgar” text? To answer this question, a look at the orthographic and textual evidence of the proto-Masoretic scrolls from the Judean Desert provides a comparative context:70

69 See discussion above, pp. 183–84. 70 For the orthographic and textual classification of the individual manuscripts in the following table as proto-Masoretic, see A. Lange, Handbuch der Textfunde vom Toten Meer, Vol. 1: Die Handschriften biblischer Bücher von Qumran und den anderen Fundorten (Tübingen: Mohr Siebeck, 2009), ad loc. Severus Scroll Variant List 199

Manuscript Number of Number of Percent Number of Percent sigla preserved Orthographic Orthographic textual Textual words and variants deviation variants deviation remnants vs MTL from MTL vs MTL from MTL of words

4QGenb (4Q3) 358 1 0.28% 0 0% SdeirGen 117 0 0% 0 0% (Sdeir 1) MurGen, 270 0 0% 0 0% Exod,Numb (Mur 1) MasLevb 457 071 0% 0 0% (Mas1b) XLevc (Ms. 107 0 0% 0 0% Schøyen 4611) 4QDeute 126 1 0.8% 1 0.8% (4Q32) 4QDeutg 151 0 0% 0 0% (4Q35) MurXII 3803 2373 0.6% 28 0.74% (Mur88)72

As indicated by the foregoing chart, the orthographic and textual variation between proto-Masoretic Torah manuscripts seems to have been negligible at the end of the Second Temple period. Because the orthographic variants of the Severus Scroll list are typical for manuscripts from the Second Temple period, and because the list notes in particular tiny textual and orthographic variants, the type of orthographic variants and the small amount of variants listed makes it likely that, orthographically speaking, the Severus Scroll may be located among the proto-Masoretic manuscripts of the late Second Temple period.

.in Lev 11:28 המה instead of הםה But note 71 72 MurXII (Mur88), the Minor Prophets Scroll from Wadi Murabba⁠ʾat, is one of the most often-cited examples of a proto-Masoretic manuscript from the Second Temple period. 73 J. T. Milik, “Rouleau des Douze Prophètes,” in Les grottes de Murabba’at (ed. P. Benoit, J. T. Milik, and R. de Vaux; 2 vols.; DJD 2; Oxford: Clarendon, 1961), 1:181–205 (183–84, 205). 200 Lange

A statistical comparison with the manuscripts listed above bears out this assessment. The Severus Scroll list reports ten orthographic variant readings out of 79,977 words of Torah text;74 this equals an orthographic variation against MT of 0.01%. Even if the compiler of the Severus Scroll variant list missed twice as many orthographic variant readings as he listed, orthographically speaking the Severus Scroll would be well within the range of orthographic variation manifested by other proto-Masoretic manuscripts from the Second Temple period. Against the claims of Lieberman, Kutscher, and Siegel,75 the Severus Scroll variant list provides therefore no orthographic evidence that the Severus Scroll was a vulgar manuscript. The limited orthographic similarities between 1QIsaa and the Severus Scroll are by far not extensive enough to justify such a claim. A similar situation obtains in regard to the Severus Scroll textual variants. Of the twenty-three textual variants reported above, fifteen do not align with any of the three medieval versions of the Hebrew Bible. These readings might thus be characterized as nonaligned in character. Two additional readings align with SP and/or the LXX, a fact which brings the total number of readings against MT to seventeen. In six cases, the Severus Scroll reads with at least a part of the medieval MT textual tradition (in one of these cases also with SP). The antiquity of the Severus Scroll variants, already noted by Kutscher and Siegel (see above, p. 184), is underscored by the fact that several of these readings are attested in other textual witnesses of the Pentateuch that go back to the Second Temple period (Gen 36:5 [MTKetib]; 36:14 [MTKetib]; 43:15 [SP]; 45:8 [LXX]; 46:8 [MasGen]; Exod 31:13 [MTKetib]; Num 4:3 [SP, LXX]). At first glance, these twenty-three textual variants against MT might seem to confirm an assessment of the Severus Scroll as a vulgar text. But a simple count of variant readings is misleading; a statistical comparison with the biblical Dead Sea Scrolls listed in the table above gives a different picture of the significance of the evidence. Today we know of seventeen cases in which the Severus Scroll disagreed with the whole of the MT textual tradition. This equals a textual deviation from MT of ca. 0.02%. Even if the compiler of our variant list missed some variants and we double or triple this number of textual variants, the Severus Scroll would still display surprisingly little textual deviation

74 This number derives from a computerized count of the number of words in the Pentateuch text of codex MTL by G. E. Weil, “Les décomptes de versets, mots et lettres du Pentateuque selon le manuscrit B 19a de Leningrad: Un essai d’arithmétique des scribes et des mas- sorètes,” in Mélanges Dominique Barthélemy: Etudes bibliques offertes à l’occasion de son 60e anniversaire (ed. P. Casetti, O. Keel, and A. Schenker; OBO 38; Fribourg: University Press, 1981), 651–703, esp. 670. 75 See nn. 16–22. Severus Scroll Variant List 201 from the consonantal MT text of the Torah; i.e., not more than 0.04 or 0.06%. The meticulousness with which the list was compiled makes it unlikely in any event that the Severus Scroll differed more than 1% from the consonantal text of MT. The table above puts these numbers into further perspective. While several manuscripts noted above do not attest to any variant readings against MT, this putative 100% agreement with MT should not be overemphasized. All these manuscripts are damaged, and only a relatively small amount text survives from them. The comparison of the Severus Scroll with one of the better preserved proto-Masoretic manuscripts from the Judean Desert, MurXII, is particularly telling. MurXII itself exhibits 0.74% deviation from MT among its 3803 preserved words and remnants of words. This means that MurXII differs almost thirty-seven times more from the MT text of the Minor Prophets than the recorded Severus Scroll variants differ from the Pentateuch’s MT text. In any case, the Severus Scroll’s deviation from MTL is well within the margins of the textual deviation exhibited by other proto-Masoretic manuscripts from the Second Temple period against this medieval Masoretic master copy.

5 Conclusions: The Textual Affiliation of the Severus Scroll

On the basis of the above comparisons, it is very likely that the Severus Scroll represents an average proto-Masoretic Torah scroll from the Second Temple period, and not a vulgar text. Its general closeness to the proto-Masoretic textual tradition of the Pentateuch is underscored, not only by the six readings on the list that align with a part of the Masoretic textual tradition, but also by the two variant readings which differ orthographically from MT but agree with it textually. The classification of the Severus Scroll as a “vulgar” text, by Lieberman, Kutscher, Siegel, and others,76 focused on the nature of the readings listed in the Severus Scroll list but neglected to interpret them in comparison with proto-Masoretic manuscripts from the Second Temple period. But if the Severus Scroll was an average representative of a standard proto- Masoretic text of the late Second Temple period, why were the variant readings of this scroll recorded as something out of the ordinary? The answer to this question must remain speculative because there are only a very few Hebrew Torah manuscripts preserved from the third through the fifth century CE.77

76 See nn. 16–22. 77 Textual evidence for the Late Ancient Hebrew text of the Pentateuch is preserved through its use in inscriptions and rabbinic literature (see the literature cited in n. 38 above), as well as in Masoretic annotations about readings of lost master copies (see, e.g., Ginsburg, Introduction, 430–37). Manuscripts which can be attributed to this period are exceed- 202 Lange

It seems likely, though, that the textual standardization of the proto-Masoretic text, evident already in the Qumran manuscripts, continued in late antiquity. This would mean that Torah scrolls from late antiquity were textually more advanced and more standardized than the Torah scrolls from the late Second Temple period. A more advanced textual standardization would explain why a scroll that represented the proto-Masoretic standard text at the end of the Second Temple period became known for its unusual readings later on. So in the end: Was the Severus Scroll a vulgar text? According to the standards of the late Second Temple period the Severus Scroll was most probably, both orthographically and textually, a high quality Torah scroll, which attested to the evolving proto-Masoretic standard text of the Pentateuch. When the Severus Scroll variant list was compiled, several hundred years later, the few disagreements of a hallowed Torah scroll from the Jerusalem Temple with the developing Masoretic standard text were carefully documented and passed on among those responsible for the ongoing textual standardization of the Hebrew Bible.

Appendix 1: The Severus Scroll and Rabbi Meir’s Torah

R. Moshe ha-Darshan was the first to make a connection between the Severus Scroll and Rabbi Meir’s Torah, noting that the two a shared variant reading (Midrash Bereshit

ingly rare. Only the paleographic dating of Ms. Heb. d. 89 in Oxford’s Bodleian Library, a small fragment with remnants of Exod 2:23–25, to the second or third century CE is undisputed (see, e.g., A. Yardeni, Hebrew Script, 73). Two manuscripts, from the Staatliche Museen in Berlin (P 10598) and the Cambridge Genizah Collection (T.–S. NS 3.21), preserve parts of Numbers and Genesis respectively. Their paleographic dating is debated, however. That Sirat discusses P 10598 together with Ms. Heb. d. 89 implies a similar date for both manuscripts (see C. Sirat, Les papyrus en caractères hébraïques trouvés en Egypte [with contributions by M. Beit-Arié, M. Dukan, F. Klein-Franke and H. Harrauer; calligraphy and illustrations by A. Yardeni; Paris: Centre national de la recherche scientifique, 1985], 35); but the Trismegistos database proposes a dating range of 700–899 CE (http:// www.trismegistos.org/tm/detail.php?tm=113848; last accessed November 2014). For T.–S. NS 3.21, Sirat suggests a date in the fifth or sixth century CE (idem, “Genesis Discovery,” Genizah Fragments 23 [1992]: 2; idem, “Earliest Known Sefer Torah,” Genizah Fragments 24 [1992]: 3; idem, “Rouleaux de la Tora antérieurs à l’an mille,” Comptes rendus des séances de l’Académie des Inscriptions et Belles-Lettres [1994]: 861–87 [with the collaboration of שעטנ"ז“) M. Dukan and A. Yardeni]); while A. Yardeni thinks of the eighth or ninth century in Proceedings of ”,ג"ץ ופרשות פתוחות וסתומות בקטע חדש של ספר בראשית מן הגניזה the Tenth World Congress of Jewish Studies, Div. D., vol. 1: The Hebrew Language, Jewish Languages [Jerusalem: World Union of Jewish Studies, 1990], 173–80). Severus Scroll Variant List 203

Rabbati 209:10–15).78 In modern times, several authors have emphasized the seemingly close relationship of the two texts;79 some have even claimed that they are one and the same.80 Most publications, however, argue for a somewhat looser connection between Rabbi Meir’s Torah and the Severus Scroll.81 Is Rabbi Meir’s Torah scroll therefore to be identified with the Severus Scroll or at least to be seen as closely related to it? In rabbinic literature, six variant readings are attributed to Rabbi Meir. Two are from the book of Isaiah and four come from the book of Genesis:82

והנה טוב מאד והנה טוב מות :([Gen 1:31 (Gen. Rab. 9:5 [1:70 – כתונות אור :[Gen 3:21 (Gen. Rab. 20:12 [1:196 – וישני לאב :(Gen 45:8 (MBR 209:13 – ובן דן חושים :([Gen 46:23 (Gen. Rab. 94:9 [3:1181–82 –

Three out of these four readings also occur in the Severus Scroll variant list; i.e., Gen 1:31; 3:21; and 45:8. But only in the case of Gen 1:31 do the Severus Scroll and Rabbi Meir’s This is .מאד instead of MT’s מות Torah agree in a variant reading against MT. Both read an agreement in orthographic convention, however, rather than in textual difference. In the other two cases, the deviant reading found in R. Meir’s Torah is not reflected in the variant list. For Gen 3:21, the Severus Scroll deviated from the list’s comparison Rabbi Meir’s Torah, however, deviated from .(כתנוד Sev) כתנות scroll in the writing of -There is no indica .עור rather than אור the MT tradition in a different way, by writing tion in any copy of the Severus Scroll variant list that either SevComp or the Severus

78 Albeck, Midraš Berešit Rabbati, 209. 79 E.g., Epstein, “Titus,” 346; Lieberman, “Hellenism,” 25; Tov, Textual Criticism (3d ed.), 113; T. Arndt, “Zur Tora des Rabbi Meʾir: Bemerkungen zu Uwe Glessmer,” Mitteilungen und Beiträge der Forschungsstelle Judentum 12–13 (1997): 87–91 (87). 80 E.g., E. Tov, “The Text of the Old Testament,” in The World of the Bible (ed. A. S. van der Woude; Bible Handbook 1; Grand Rapids: Eerdmans, 1986), 156–90 (159); P. D. Wegner, A Student’s Guide to Textual Criticism of the Bible: Its History, Methods and Results (Downers Grove: InterVarsity, 2006), 124; R. Price, Searching for the Original Bible (Eugene, Ore.: Harvest House, 2007), 56; J. D. H. Norton, Contours in the Text: Textual Variation in the Writings of Paul, Josephus, and the Yaḥad (Library of New Testament Studies 430; London: T&T Clark, 2011), 113. 81 E.g., Loewinger, “Prolegomenon,” xxxi; Siegel, Severus Scroll, 43. 82 The two Isaiah readings are from Isa 21:11 and 34:7. For the complete list, see Loewinger, (Siegel’s list (Severus Scroll, 42–48 ;258 ”,ספר תורה“ ,Prolegomenon,” xxxii; and Loewinger“ does not include the Isaiah variants. The numbers given in parentheses in the list that follows refer to the critical editions of MBR (see n. 1) and Genesis Rabbah respectively (J. Theodor and C. Albeck, eds., Bereschit Rabba mit kritischem Apparat und Kommentar: Parascha XLVIII–LXXXVI [2d prtg.; Jerusalem: Wahrmann, 1965]). 204 Lange

Scroll itself differed from the MT tradition on this word.83 For Gen 45:8, SevMBR quotes ,(with the MT tradition) וישימני while SevP reads for SevComp ,וישני SevComp as reading -in SevMBR thus must go back to a scribal error within the tex וישני instead. The reading tual tradition of the Severus Scroll variant list which was already evident in the copy of the list employed by Moshe ha-Darshan.84 Rabbi Meir copied this reading from a Torah scroll which suffered from the same copyist error as did Moshe ha-Darshan’s copy of the Severus Scroll variant list. In view of the fact that it shares only one agreement with the Severus Scroll, and one additional possible but unlikely reading with SevComp, Rabbi Meir’s Torah may neither be identified with the Severus Scroll nor be seen as closely related to it. This is all the more the case since the single agreement between R. Meir’s Torah and the Severus Scroll reflects a shared spelling convention. A relationship between the two becomes even less likely when the nature of Rabbi Meir’s Torah scroll is recognized. Next to its running text, Rabbi Meir made marginal notes in this manuscript about divergent readings that he had encountered in his professional career as a scribe.85 The variant readings recorded in Rabbi Meir’s Torah derived therefore most likely from a wider range of different manuscripts. Rabbi Meir’s Torah hence can be regarded neither as another version of the Severus Scroll, nor as a manuscript closely related to it.

Appendix 2: Synopsis of the Textual Witnesses of the Severus Scroll Variant List

The transcriptions given below of the four textual witnesses of the Severus Scroll variant list are newly done, prepared by myself through fresh examinations of the manuscripts. All are preliminary publications; full critical editions of each, along with an eclectic text constructed from the best readings across the four witnesses will appear in my forthcoming monograph, where I will discuss my text-critical decisions in detail.86 Each version of the list differs in its ordering of the variants. The master sequence followed here is based primarily on that of SevP.87 Within each column, the individual variants are numbered according to their places in the individual copy.

,Epstein, “Biblische Textkritik,” 49; U. Glessmer ;102 ,חדשים ,Contra Harkavy 83 “Mehrdeutigkeit des Bibeltextes und theologische Sprachfindung: ‘Kleider der Haut’ oder ‘Kleider des Lichts’ in Gen 3,21?” Mitteilungen und Beiträge der Forschungsstelle Judentum 12–13 (1997): 65–86 (82); and Arndt, “Zur Tora des Rabbi Meʾir,” 87. 84 See n. 15 above, and cf. Segal, “Promulgation,” 46. 85 Cf. Epstein, “Biblische Textkritik,” 48–49; Segal, “Promulgation,” 45; Lieberman, ”,ספר תורה“ ,Hellenism,” 24–25; Loewinger, “Prolegomenon,” xxxii–xxxiv; and Loewinger“ 259–60. 86 Lange, The Severus Scroll. 87 SevP diverges from the master sequence in one element only: the omission of variant 17, which, as discussed in the main body of the paper, is found only in SevMBR. Severus Scroll Variant List 205

SevMBR SevD403 SevD146 SevP Master sequence

Prologue אלין פסוקיא דהוו אלין פסוקיא דהוו פסוקיא דהוו דין הוא מן מליא כתיבין בספר כתיבין בספר כתבין בספר דכתיבן באורייתא אוריתא דאישתכח אוריתא דאשתכח אוריתא דאשכח דנפקת מן ירושלם ברומי והיא גנוזה ברומא והי גנוזה ברומא והיא גנוזה בשביתא וסלקת וסתומא בכנשתא וסתומה בכנשתא וסתימה בכנשתא לרומי והות גניזא דסירוס בשנוי דסוירוס בשנוי דסוירוס בשנוי בכנישתא דאסוירוס אותיות ותיבות אותיות ותיבות אותיות ותיבות 1 )1( וירא ﭏהים את )1( וירא אלהים את )1( וירא אלהים את )1( וירא אלהים את Gen 1:31 כל אשר עשה והנה כל אשר עשה והנה כל אשר עשה והנה כל אשר עשה והנה טוב מאד מות היה טוב מאד מות היה טוב מאד מות היה טוב מאד כתוב כתיב כתי 2 )2( כתנות עור )2( כתנות עור )2( כתנות עור )2( לאדם ולאשתו Gen 3:21 וילבשם כתנוד היה כתנוד כתי כתנוד כתי כתנות עור כתוב 3 )3( הכצעקתה )3( הכצעקתה )3( הכצעקתה )3( ארדה נא Gen 18:21 הבאה אלי עשו הכצעקתם כתי הקצעקתם כתי ואראה הכצעקתם כלה הכצעקתם היה כתוב 4 )4( ואמר יי אלהי )4( ואמר ײי אלהי )4( ואמר ײי אלהי )4( ה׳ אשר לקחני Gen 24:7 אדני אברהם ומארע אדני אברהם ומארע אדני אברהם ומארע מביתי ומארצי היה כתוב כתי כתי 5 )5( וימכר את )5( את בכרתו )5( את בכורתו )5( וימכור את Gen 25:33 בכורתו ליעקב ליעקב ליעקב מכירתו מכרתו היה כת מכרתו כתי מכרתו כתי 6 )6( הנה נא זקנתי )6( הנה נא זקנתי )6( הנה נא זקנתי )6( הנה נא זקנתי Gen 27:2 יוממתי היה כתוב יוממתי כתי יוממת כתי לא ידעתי יום מותי 7 )7( ראה ריח בני )7( ריח בני )7( ריח בני )7( כריח שדה אשר Gen 27:27 כריח שדה כריח שדה כריח שדה ברכו ה׳ סדה היה כתוב סדה כתי סדה כתי 8 )8( יעוש )8( יעוש )8( יעוש )8( ואהליבה ילדה Gen 36:5 דואהליבמה ילדה דואהליבמה יעיש דואהליבמה יעיש את יעוש יעיש היה כתוב כתי כתי 9 )9( וכן דואלה היו )9( וכן דואלה היו )9( וכן דואלה היו )11( ואלה היו בני Gen 36:14 בני יעוש היה כתוב בני יעוש יעיש כתי בני יעוש יעיש כתי אהליבמה יעוש 10 )10( ויקמו וירדו )10( ויקומו וירדו )10( ויקומו וירדו )9( ויקומו וירדו Gen 43:15 מצרימה מצרים מצרימה מצרים מצרימה מצרים מצרימה היה כתוב כתי כתי 11 )11( אליפז בן עדה )11( אליפז בן עדה )11( אליפז בן עדה )10( אליפז בן ענה Gen 36:10 בנעדה היה כתוב בנעדה כתי בנעדה כתי 206 Lange

(cont.)

SevMBR SevD403 SevD146 SevP Master sequence

12 )12( וישימני לאב )12( לאב לפרעה )12( לאב לפרעה )13( וישני לאב Gen 45:8 לפרעה פרעה היה פרעה כתי פרעה כתי לפרעה כתוב 13 )13( ואקברה שם )13( ואקברה שם )13( ואקברה שם )12( ואקברה שם Gen 48:7 שמ היה כתוב שמ כתי שם כתי בדרך 14 )14( ואלה שמות )14( ואלה שמות )14( ואלה שמות )14( ואלה שמות Gen 46:8 בני ישראל הבאים בני ישראל הבאים בני ישראל הבאים הבאים מצרימה מצרימה מצרים היה מצרימה מצרים כתי מצרימה מצרים כתי כתוב 15 )15( ויסעו בני )15( מרעמסס )15( מרעמסס )16( ויסעו בני Exod 12:37 ישראל מרעמסס מרעמס כתי מרעמס כתי ישראל מרעמסס מרעמס היה כתוב 16 )16( כה תאמר )16( לבית יעקב )16( לבית יעקב )15( כה תאמר לבני Exod 19:3 לבית יעקב ותגיד ותגיד לבני ישראל ותגד לבנ֯י֯ י֯ש֯ ר֯ אל יעקב ותגד לבני ישראל לבית לבית כתי תרויהו לבית ישראל כתי לבני ישראל היה כתו תרויהו תר֯ ו֯ יהו 17 )17( כי אות היא Exod 31:13 ביני 18 )17( וחמשה )17( וחמשה )17( ֯ ו ח֯ מ֯ ש֯ ה֯ )18( חמש לקרש Exod 26:27 בריחים לקרשי לא בריחים לקרשי לא ב֯ר֯ יחים לקרשי לא צלע תנינא לית בה היה כתוב בריחים היו כתי בריחים היה כתו בו בריחים ברחים 19 )18( ולקח הכהן )18( ולקח הכהן )18( ולקח הכהן )19( ולקח הכהן Lev 4:34 מדם מדמ היה כתוב מפר מדמ כתי מדם מדמ כתי מדמה דאם כבש 20 )19( וכי ירק הזב )20( וכי יטהר הזב Lev 15:8 במים חיים היה במים חיים כתוב 21 )20( וכבשה אחת )19( וכבשה אחת )19( וכבשה אחת )21( וכבשה אחת Lev 14:10 בת שנתה תמימה בת שנתה תמימה בת שנתה תמימה בת תמימים תמימים היה כתוב תמימים כתי תמימים כתי 22 )21( כל בא לצבא )20( כל בא לצבא )20( כל בא לצבא )22( וכל הבא לצבא Num 4:3 דקהת הבא היה דקהת הבא כתי דקהת הבא כתי בקרות בה כתוב 23 )22( מראשית )21( עריסותיכם )21( עריסותיכם )23( מראשית עריס Num 15:21 עריסותיכם לדרתיכם לדרותיכם לדרותיכם תתנו לדריכם היה כתוב לדריכם כתי לדריכם כתי לדריכם Severus Scroll Variant List 207

(cont.)

SevMBR SevD403 SevD146 SevP Master sequence

24 )23( נקם נקמת )22( מאת המדינים )22( מאת המדינים )24( נקום נקמת Num 31:2 בני ישראל מאת אחר תאסף אשר אחר תאסף אשר אשר המדינים אחר תאסף כתי כתי אשר היה כתוב 25 )24( ויבאו אל משה )23( ויבאו אל משה )23( ויביאו אל משה )25( ויבאו אל משה Num 31:12 ואל כל עדת ואל כל עדת ואל כל עדת ואל עדת בני ישראל לא היה כתו בו כל לא היה בו כל לא היה בו כל 26 )25( ויקרבו ראשי )24( ויקרבו ראשי )24( ויקרבו ראשי )26( ויקרבו ראש Num 36:1 בני יוסף בן יוסף בני יוסף בן יוסף בני יוסף בן יוסף ממשפחת בית יוסף היה כתוב כתיב כתי 27 )26( ולא אביתם )25( ולא אביתם )25( ולא אביתם )27( ולא אבית Deut 1:26 לעלות אביתמ היה אביתמ כתי אביתמ כתי לעלות כתוב 28 )27( ויעשו גם הם )26( ויעשו גם הם )26( ויעשו גם הם )28( וירשו גם הם Deut 3:20 המ היה כתוב גם המ כתי גם המ כתי את הארץ 29 )28( לתת אותנו ביד )27( אותנו ביד )27( אותנו ביד )29( לתת אותנו ביד Deut 1:27 האמרי האמור היה האמרי האמר כתי האמרי האמר כתי האמורים כתוב 30 )29( לא תקח האם )28( לא תקח האם )28( לא תקח האם )30( ולא תקח האם Deut 22:6 על הבנים האבנים על הבנים האבנים על הב֯נ֯ ים האבנים על האבנים היה כתוב כתי כתי 31 )30( גפרית ומלח )29( ומלח שרפה )29( ומלח שרפה )31( גפרית ומלח Deut שרפה שרפת היה שרפת כתי שרפת כתי שרפת 29:22a כתוב 32 )31( כמהפכת )30( כמהפכת סדם )30( כמהפכת סדם )32( כמהפכת סדום Deut אלהים את סדום כמהפכת אלהים כמהפכת אלהים ועמורה 29:22b כמהפת היה כתו כתי כתי 33 )32( אמרתי )31( אמרתי )31( אמרתי )33( אמרתי Deut 32:26 אפאיהם אף אי הם אפאיהם אף איהם אפאיהם אף איהם אף איהם היה כתוב כתי כתי Postscript ויבא מורה צדק יבא מורה צדק יבא מורה צדק כן הוו כתיבין במהרה בימינו ויאמר אלינו בימים במהרה בימי֯נ֯ ו אמן באורייתא דנפקת ויאמר לנו מירושלם Where is the Lost Ark of the Covenant? The True History (of the Ancient Traditions)

Chaim Milikowsky

In the course of this paper I will not reveal the present location of the Ark of the Covenant. Nor will those who persevere and read the paper until the end know what happened to the Ark at the time that the First Temple was destroyed.1 For better or for worse, my objective is run-of-the-mill and scholarly: namely, to present the fascinatingly divergent traditions found in rabbinic literature and in other Jewish literatures of the Hellenistic–Roman period pertaining to the “lost Ark,” and to try to understand their origins and their lines of development. This paper divides naturally into three sections: the first section focuses upon nonrabbinic texts; the second focuses upon rabbinic texts; and the third attempts to place both corpora together within one larger framework. This larger framework will allow me, on the one hand, to make some general comments about the treatment of the Ark across the board in this period, high- lighting the main topics of interest and concern; and on the other hand, to delineate some differences among the members of each group of texts, as well as between the groups themselves.

Many texts tell us that the Ark of the Covenant was hidden away in the period leading up to the destruction of the First Temple. In some of these texts, the Ark is the sole item hidden away, and in others the Ark is one among a variety of different items, some of them Temple vessels and others unrelated to the Temple.2 It must, however, be emphasized that in all these texts, the hiding of the Ark is a central focus. It seems reasonable to assume that the traditions about the hiding of the Ark in the period leading up to the destruction of the First Temple were generated as a response to the absence of the Ark from the Second Temple. Indeed, a passage found in several places in rabbinic literature reports explicitly that

1 Contrast, e.g., M. Haran, “The Disappearance of the Ark,” IEJ 13 (1963): 46–58; J. Day, “Whatever Happened to the Ark of the Covenant?” in Temple and Worship in Biblical Israel (ed. J. Day; London: T&T Clark, 2005), 250–70. 2 These matters will be discussed below.

חמשה דברים היה המקדש האחרון חסר מן הראשון, ואילו הן: אש וארון ואורים ותומים ושמן המשחה ורוח הקודש.

The last Temple was lacking five things from the first Temple; and they are the fire, the Ark, the Urim ve-Tummim, the anointing oil, and the Spirit of Holiness.3

אלו חמשה דברים שהיו בין מקדש ראשון למקדש שני ואלו הן: ארון וכפורת וכרובים אש ושכינה ורוח הקודש ואורים ותומים.

These are five things in which the First Temple differed from the Second Temple: the Ark, the Ark-cover, the Cherubim, the fire, the Shekhinah, the Spirit of Holiness, and the Urim ve-Tummim.4

3 The quote is from y. Taʿan. 2:1 (65a); very similar are y. Mak. 2:7 (32a); y. Hor. 3:3 (47c) (which and Song Rab. 8:9.3 (which has a slightly different order). Please ;(משחה ורוח lacks the items note: Translations of biblical verses generally follow either the RSV or the NJPS or are an amalgamation of both, with occasional alterations. Translations of rabbinic texts are based upon the Soncino translations, with occasional alterations. Similarly, translations of works included in OTP are taken from those volumes, with occasional modifications. 4 So b. Yoma 21a, following the majority of manuscripts. Note the differences from the previously cited list ( y. Taʿan. 2:1 [65a]): the anointing oil is missing here; additional elements are the Ark-cover, the Cherubim, and the Shekhinah. Note also that although this passage, like y. Taʿan. 2:1 (65a), purports to list five things, the list actually comprises seven items; see Rashi, ad loc., s.v. ʾaron. Ms. London, Harl. 5508 of b. Yoma lacks the Shekhinah and the Spirit of Holiness; according to R. N. Rabbinovicz in his Diqduqei Soferim (15 vols. in 8; Munich: H. Roesl, E. Huber, 1867–1886 [1872]), ad loc., this is the original reading of b. Yoma here. I find it difficult, however, to envision a textual scenario which would explain how the original reading was retained only in this manuscript. Ramban, in his commentary to Song of Songs, reads “the Ark, the Urim ve-Tummim, the anointing oil, the fire, and the Shekhinah” (Kitvei Rabbenu Moshe ben Naḥman [ed. C. B. Chavel; 2 vols.; Jerusalem: Mosad ha-Rav Kook, 1963], 2:500); but from his phrasing here, it appears that this is his own emendation to the text of the Talmud. See also the parallel to this text in Midrash Ḥaserot ve-Yeterot, in Batei Midrashot (ed. S. A. Wertheimer; rev. ed. A. J. Wertheimer; 2 vols.; Jerusalem: Ktav va-Sefer, 1968), 2:252, especially the readings cited by Wertheimer in his apparatus. In a relatively late midrashic source we find a strange fusion of the tradition concerning “hiding” items from the First Temple and the tradition concerning those items “present” in the First Temple and “lacking” in the Second Temple; see Num. Rab. 15:10 (= Tanḥ. Behaʿalotekha 6:11 [Midrash Tanḥuma (ed. S. Buber; Vilna: Romm, 1885), 50]). Of the five items said to have been hidden away, one 210 Milikowsky

Regarding the period towards the end of the Second Temple, the lack of the Ark is made more explicit. Indeed, in one place the Mishnah refers to the time “after the Ark was taken . . .” (m. Yoma 5:2);5 and in another place the Mishnah cites a tradition “that there [= in the storehouse for wood] the Ark was hidden” (m. Šeqal. 6:1).6 In a similar manner, Josephus speaks of “the innermost cham- ber . . . and nothing is found there. . . . [I]t is called by the name ‘holy of holies’ ” ( J.W. 5.219). Yet not all traditions attest straightforwardly to the absence of the Ark. Louis Ginzberg, in his Legends of the Jews,7 points to a mishnah in tractate Yoma (m. Yoma 5:1) which relates that the High Priest “walks to his left along the curtain until he reaches the Ark . . .”;8 Ginzberg argues that it should be inferred from this mishnah that the Ark was present in the period of the Second Temple. Presumably he would understand the next mishnah (5:2, cited above), “after the Ark was taken . . . ,” as referring to an occurrence in the days of the Second Temple. In truth, one cannot conclude from these passages that the Ark was

of them is the Spirit of Holiness, and surely such a notion is incongruous at best. Regarding the hiding away of the fire, also mentioned there, see 2 Macc 1:19–22; and Sefer Yosippon, Chapter 7 (ed. D. Flusser, Jerusalem: Bialik Institute, 1979), 1:44–45). On this latter text see further L. Ginzberg, The Legends of the Jews (trans. H. Szold; 7 vols.; Philadelphia: Jewish Publication Society, 1909–1938), 6:440; and Flusser’s comment in Sefer Yosippon, 2:132. משניטל ארון, אבן היתה שם מימות נביאים ראשונים ושתייה היתה נקראת, גבוה :M. Yoma 5:2 5 After the Ark was taken, there was“) מן הארץ שלש אצבעות, ועליה היה נותן. נטל את הדם . . . a stone from the days of the earlier prophets, called the Shetiyah, three fingers above the ground, on which he would place [the pan of burning coals]. He would take the blood . . .”). שלשה עשר שופרות, שלשה עשר שולחנות, שלש עשרה השתחויות היו במקדש. :M. Šeqal. 6:1 6 של בית רבן גמליאל ושל בית רבי חנניה סגן הכהנים היו משתחוין ארבע עשרה, והיכן היתה There were in the“) יתרה? כנגד דיר העצים, שכן מסורת בידם מאבותיהם ששם הארון נגנז Temple thirteen chests, thirteen tables and thirteen prostrations. [Members] of the household of Rabban Gamaliel and of R. Ḥananiah the segan of the priests used to prostrate themselves fourteen [times]. Where was the additional [prostration]? In front of the storehouse for wood, for they had a tradition from their fathers that there the Ark was hidden”). 7 Ginzberg, Legends of the Jews, 6:378. החיצונה היתה פרופה מן הדרום ופנימית מן הצפון. מהלך ביניהן עד שמגיע לצפון. :M. Yoma 5:1 8 הגיע לצפון הופך פניו לדרום מהלך לשמאלו עם הפרוכת, עד שהוא מגיע לארון. הגיע לארון נותן את המחתה בין שני הבדים. צבר את הקטרת על גבי גחלים ונתמלא כל הבית כולו עשן (“The outer curtain was held back by a clasp on the south side and the inner curtain on the north side. He walks along between them until he reaches the north side. When he reaches the north side he turns his face to the south and walks to his left along the curtain until he reaches the Ark. When he reaches the Ark he places the pan of burning coals between the two bars. He heaped up the incense upon the coals and the whole house became full of smoke”). the Lost Ark of the Covenant 211 absent from the Temple during the entire Second Temple period. But, on the other hand, one should not accept Ginzberg’s claim that this mishnah testifies to the presence of the Ark in the Second Temple: very plausibly, m. Yoma 5:1 is a theoretical description of the Yom Kippur ritual and not an historical one. In that case, mishnah 5:2 is using the ambiguous word “taken” because its formu- lator knew of the argument concerning whether the Ark was hidden away or exiled and did not wish to take a position on the question.9 Thus far, I have highlighted the theme of the fate of the Ark in the texts under discussion. Connected to but distinct from the question of its fate is that of at whose hands the Ark was hidden or exiled. The two mishnaic passages cited above speak of the taking and hiding of the Ark without mentioning who took it or hid it. Other rabbinic texts, however, state explicitly that the Ark was hidden by King Josiah, and there is no rabbinic text that claims otherwise. I have not, however, found this tradition naming Josiah outside of rabbinic literature.10 As noted above, many nonrabbinic as well as early Christian sources discuss the hiding of the Ark and other Temple vessels. All those texts that tell us who hid the vessels, name the prophet Jeremiah. Outside of rabbinic literature I have found no text claiming that the Ark was hidden by any biblical figure other than Jeremiah. To recapitulate this point: notwithstanding the disparities found within rabbinic literature concerning the hiding of the Ark, there is a consensus that it was hidden by Josiah. Similarly, notwithstanding the disparities in nonrabbinic literature concerning the hiding of the Ark, the consensus is that it was hidden by Jeremiah. Later in this paper (section 3) I shall argue that this difference of opinion between nonrabbinic and rabbinic texts has something quite important to teach us about differences between these texts themselves (and possibly between the various communities who created them), in terms of their modes of exegesis of and expansion upon the Bible.

9 See discussion below. B. Z. Wacholder is surely right in pointing out that ancient Jewish texts evince a special interest in the Ark, much more intense than their interest in any other utensil of the Temple (Eupolemus: A Study of Judaeo-Greek Literature [HUCM 3; Cincinnati: Hebrew Union College, 1974], 242); but it did not occur to him that this interest might have been due to the absence of the Ark. See also V. Aptowitzer, Parteipolitik der Hasmonäerzeit im rabbinischen und pseudo-epigraphischen Schrifttum (Veröffentlichungen der Alexander Kohut Memorial Foundation 5; Vienna: Kohut Foundation, 1927), 1, 192 (n. 2). 10 All these texts will be quoted and discussed below. 212 Milikowsky

1 The Concealment of the Ark in Nonrabbinic Sources

When we turn to nonrabbinic Jewish sources and early Christian sources that deal with the fate of the Ark, we find that although many state explicitly that Jeremiah hid the Ark, there are wide divergences in regard to the details surrounding his actions.11 Second Maccabees 2:1–8 reports that Jeremiah hid the Tabernacle,12 the Ark, and the incense altar13 on the mountain upon which

11 The arrangement of texts within the next few paragraphs is conceptual and not chronological; I begin with those that specify Jeremiah as concealer of the Ark; move to those that propose a different concealer; and then discuss those texts that refer to concealment more generally. For several of the works to which I refer there are major differences of opinion among scholars regarding the date of their composition as well the religious identities of their creators; these questions, while important, are not at all crucial for me here, especially given the methodological presuppositions I present below concerning the fluidity of these traditions. 2 Maccabees of course is part of what is generally termed the Apocrypha; translations and discussions of all the other texts for which no specific reference is given can be found in OTP. 12 2 Baruch mentions the concealment of the vessels of the Tabernacle (see below, n. 16), and it seems probable that these two traditions are somehow connected, for both introduce the motif of concealing the Tabernacle or its vessels at the time of the destruction of the Temple. T. Soṭah 13:1 (The Tosefta According to Codex Vienna [ed. S. Lieberman; 4 vols.; 2d ed.; Jerusalem: Jewish Theological Seminary of America Press, 1992–1995], 3:229) states that the Tabernacle was hidden away when the First Temple was built. See also the midrashic Genizah fragment published in J. Mann and I. Sonne, The Bible as Read and Preached in the Old Synagogue (2 vols.; New York: ‎Ktav,‎ 1966–1971), Hebrew section, 2:189; and the comments of S. Lieberman, Tosefta ki-feshuṭah: A Comprehensive Commentary on the Tosefta (10 vols.; New York: Jewish Theological Seminary of America, 1955–1988), 8:733–34. 13 It is not at all apparent why the incense altar is included in this list. It seems that the other items listed are those which, in the opinion of the author of 2 Maccabees or his source, were extant in the First Temple but not in the second. It seems far-fetched, however, to suggest that either at the time of the composition of 2 Maccabees, or at the time of the composition of this letter which is included in 2 Maccabees, there was no incense altar in the Temple. Note the peculiar supposition of E. Schürer (The History of the Jewish People in the Age of Jesus Christ [trans. T. A. Burkill et al.; rev. and ed. G. Vermes, F. Millar, and M. Goodman; 3 vols. in 4; Edinburgh: T&T Clark, 1973–1987], 2:279), that indeed during part of the time that the Second Temple stood it contained no incense altar. This passage in 2 Maccabees, however, is not relevant to his argument, inasmuch as he limits his supposition to the period before the Hasmonean wars; Schürer agrees that regarding the period following these wars, the evidence that there was an incense altar in the Temple is unas- sailable. J. A. Goldstein has suggested (1 Maccabees [AB 41; Garden City, N.Y.: Doubleday, 1976], 547 n. 1) that the author of this passage in 2 Maccabees saw a special value in the the Lost Ark of the Covenant 213

Moses had ascended to see the land of Israel; the text states that the place of its hiding will remain concealed until God gathers together his nation and the glory of God is revealed. Similarly, the Lives of the Prophets, Jeremiah 2:11, writes that Jeremiah caused the Ark and its contents to be swallowed by the rock in the desert next to the mountains where Moses and Aaron lay, a reference to their graves. In 4 Baruch (The Rest of the Words of Jeremiah) 3:9–20, Jeremiah and Baruch follow God’s command and tell the earth to open up and swallow the Temple vessels.14 The Hellenistic Jewish historian Eupolemos likewise links Jeremiah and the Ark: Nebuchadnezzar took to Babylon the gold, the silver, and the

original incense altar of the First Temple, which was not returned to the Second Temple, and that this lack negatively affected the ideal status of the later Temple. D. R. Schwartz (2 Maccabees [Jerusalem: Yad Izhak Ben-Zvi, 2004], 89 [in Hebrew]) suggests that the absence of the incense altar from the list of Temple vessels plundered at the time of the Destruction (2 Kgs 25:13–17) led to the creation of the tradition that it was among the items hidden by Jeremiah. This suggestion does not seem convincing, inasmuch as the table and the menorah are likewise missing from the list of plundered Temple vessels, yet no text claims that Jeremiah concealed these items. 2 Baruch likewise mentions the hiding of the incense altar together with the Ark of the Covenant; see below, n. 16. See also I. Kalimi and J. D. Purvis, “The Hiding of the Temple Vessels in Jewish and Samaritan Literature,” CBQ 56 (1995): 680. It is noteworthy that in rabbinic literature the table and menorah that Moses made are cited as vessels that were in use in the First Temple (see t. Soṭah 13:1 [Lieberman, Tosefta, 3:229], and the list of parallels cited by Lieberman, Tosefta ki-feshuṭah, 8:734); but no mention is made of the incense altar that Moses made. 14 This passage speaks of concealing all the Temple vessels; but this notion seems to contradict the biblical account, inasmuch as several passages in the Bible state that Temple vessels were taken to Babylon: 2 Chr 36:7 and Dan 1:2 (both referring to the defeat of Jehoiakim); 2 Chr 36:10 (the defeat of Jehoiachin); 2 Chr 36:18 (the defeat of Zedekiah; and compare to 2 Kgs 25:13–17 and Jer 52:17–22); Dan 5:2 (the feast of Belshazzar); and Ezra 1:7–10 (the return of the Temple vessels at Cyrus’s command). See also Jer 27:19–22 and Bar 1:8. It is worthy of emphasis that of all the pseudepigraphal works extant, only 4 Baruch professes that all the Temple vessels were concealed, pace G. W. E. Nickelsburg, Jr. (“Narrative Traditions in the Paralipomena of Jeremiah and 2 Baruch,” CBQ 35 [1973]: 60–68 [67]). Nickelsburg’s argument that the hiding of the vessels is motivated by the need to prevent them from becoming impure by falling into the hands of the Gentile conquerors is supported only by 2 Bar. 80:2—which he does not, however, cite. Since all the other sources generally mention only the Ark and a few other isolated items, it seems much more probable that their absence was the crucial catalyst in the development of the tradition, and that the idea of preventing impurity was subsequently attached to the earlier stratum. 214 Milikowsky copper from the Temple, “apart from the Ark and the tablets that were in it; these Jeremiah guarded.”15 Second Baruch 6:7–9 relates that the earth swallowed up a number of Temple vessels, possibly including the Ark, but explicitly mentioning the two Tablets contained therein; this text asserts, in contrast to 4 Baruch and Eupolemos, that the agent of concealment was an angel.16 Two other texts mention concealment but do not identify the concealer: L.A.B. 26:13 states simply that the tablets and the stones of the breastplate were hidden after the Temple was destroyed; and according to the Coptic Jeremiah Apocryphon, Nebuchadnezzar was told at the time of the conquest of Jerusalem that the Ark was covered in dirt on the mountain of Jericho.17 In light of this accumulation of sources, two of which—2 Maccabees and Eupolemos—were composed long before the beginning of the Common Era,

15 According to Eusebius, Praeparatio evangelica 9.39.5 (454d) (my translation). This fragment is not explicitly attributed to Eupolemos, but there are excellent reasons to consent to this commonly accepted attribution; see Wacholder, Eupolemus, 227 n. 1. 16 Most of the items in the list can be easily identified, but not all. Among the items explicitly mentioned are the curtain, the Ark-cover, the incense altar, the priestly garments, the vessels of the Tabernacle, and the holy ephod, and it has been suggested that this last item actually refers to the Ark; see R. H. Charles, “II Baruch: I: The Syriac Apocalypse of Baruch,” in idem, The Apocrypha and Pseudepigrapha of the Old Testament (Oxford: Clarendon, 1913), 2:484; P. Bogaert, L’Apocalypse syriaque de Baruch: Introduction, Traduction du Syriaque et Commentaire (2 vols.; SC 144, 145; Paris: Cerf, 1969), 2:22. Worthy of note is the fact that the angel is said to have entered the Holy of Holies and to have taken these items from there, but a number of them are not generally connected to that specific sacred space. 17 See K. H. Kuhn, “A Coptic Jeremiah Apocryphon,” Le Muséon 83 (1970): 95–135, 291–350 (298). I do not know if “the mountain of Jericho” mentioned here denotes the same place as the Mountain of Moses mentioned by 2 Maccabees (cited above). On the face of it, this does not appear probable, but I fail to see how else to understand this tradition. The hiding of the Ark in the mountain of Jericho appears to contradict a passage found earlier in the Coptic Jeremiah Apocryphon, which states that Zedekiah the king commanded that he be brought the Ark and the Tablets, and that he used them as part of his idolatrous rites (124). This apocryphon also mentions the hiding of vessels in the Temple before the Destruction (302), but in this context describes only the hiding of the High Priest’s vestments in the cornerstone of the Temple (presumably a Christian motif); the priestly frontlet was sent up to the sun to be hidden there. In an Arabic version of this work there follows the further statement that at that same time Jeremiah hid the remaining Temple vessels (see A. Mingana, ed. and trans., “A Jeremiah Apocryphon,” Woodbrooke Studies [Cambridge: W. Heffer, 1927], 1:125–38, 148–233 [173] [= BJRL 11 (1927): 377]); but the Coptic version published by Kuhn states explicitly that the remaining vessels were taken to Babylon by Nebuchadnezzar. the Lost Ark of the Covenant 215 it is clear that the tradition connecting Jeremiah with the hiding of the Ark is very early.18 Christian Wolff has concluded on the basis of these texts that in the earliest formulation of this tradition, Jeremiah hid the vessels in the Mountain of Moses.19 This conclusion takes the earliest text, 2 Maccabees, as witness to the earliest expression of this tradition. In Wolff’s reconstruction, there developed an independent tradition that the vessels were hidden in the earth below the Temple. He further concludes that, since the only textual evidence for the tradition that Jeremiah hid the vessels in the earth beneath the Temple is the relatively late composition 4 Baruch, this tradition must have developed very late, from the amalgamation of the two early traditions: “Jeremiah hid the items in the Mountain of Moses”; and “the items were concealed in the ground beneath the Temple.” From a methodological point of view, I find it hard to accept Wolff’s conclusions. It is immediately obvious that during the days of the Second Temple many variants of these traditions were extant, of which only a small fraction have reached us. We should not necessarily reconstruct developments within the histories of these traditions, which occurred over the course of hundreds of years, on the basis of a few surviving witnesses. And we should certainly not assume, as Wolff does, that the dates of composition of the surviving documents necessarily fix the dates of the traditions themselves. There are two basic points of tension among these texts: the identity of the concealer (was it Jeremiah or an angel?); and the place of concealment (was it in the earth beneath the Temple or at the Mountain of Moses?). In the two earliest sources, 2 Maccabees and Eupolemos, the concealer is Jeremiah; assur- edly, then, this is an early tradition. With regard to the place of concealment, even though 2 Maccabees asserts that the location was the Mountain of Moses, I would argue that this assertion reflects a later stage in the development of the tradition. The Mountain of Moses—that is the place where Moses saw the land—is also, of course, Moses’s burial place; see Deut 32:48–50:

18 For a sampling of later traditions about the hiding of Temple vessels see J. R. Davila, “Scriptural Exegesis in the Treatise of the Vessels: A Legendary Account of the Hiding of the Temple Treasures,” in With Letters of Light: Studies in the Dead Sea Scrolls, Early Jewish Apocalypticism, Magic, and Mysticism in Honor of Rachel Elior (ed. D. V. Arbel and A. A. Orlov; New York: de Gruyter, 2011), 45–61; H. Newman, “At Cross Purposes: The Ritual Execution of Haman in Late Antiquity,” in Between Personal and Institutional Religion: Self, Doctrine, and Practice in Late Antique Eastern Christianity (ed. B. Bitton-Ashkelony and L. Perrone; CELAMA 15; Turnhout: Brepols, 2013), 311–36 (329–31). 19 C. Wolff, Jeremia im Frühjudentum und Urchristentum (TUGAL 118; Berlin: Akademie- Verlag, 1976), 68. 216 Milikowsky

That very day, the Lord spoke to Moses: “Ascend this mountain of Abarim to Mount Nebo, which is in the land of Moab facing Jericho, and view the land of Canaan which I am giving the Israelites as their holding. You shall die on the mountain that you are about to ascend and be gathered there to your kin. . . .”

I suspect that the tradition that connects the place of the concealment with the site of Moses’s burial is dependent upon the fact that the location of Moses’s grave was itself unknown. In other words, the tradition that the Ark was concealed in the Mountain of Moses seeks to explain why the Ark, which had presumably been hidden towards the end of the days of the First Temple, was not returned in the days of the Second Temple, a particularly difficult question if we assume that the Ark was hidden in the ground beneath the Temple. This line of reasoning leads us to the conclusion that the notion of the concealment of the Ark in the Mountain of Moses was a secondary development. There is good reason to assume that Jeremiah was originally said to have concealed the Ark beneath the Temple (albeit no such early tradition exists); and that, because of the difficulty just delineated, the place of concealment was transposed to the Mountain of Moses. That is, once it was clear that the location of the Ark could not be identified, it was natural to link it to another location that could not be identified, a location which the Bible itself explicitly says is unknown:

So Moses the servant of the Lord died there, in the land of Moab, at the command of the Lord; He buried him in the valley in the land of Moab, near Beth-Peor; and no one knows his burial place to this day. (Deut 34:5–6)

The linking of the hiding of the Ark with Moses’s grave adds the suggestion that the Ark’s location was concealed deliberately, “to this day.” To my mind, there is no reason to assume that these variant traditions necessarily stem from a single Ur-tradition. Nonetheless, the arguments proposed lead me to conclude that of the variants that have survived, the earliest is the tradition that Jeremiah hid the Ark in the ground beneath the Temple.20

20 It may perhaps be relevant to point out that Jeremiah was a priest. the Lost Ark of the Covenant 217

A few years ago, Steven Weitzman suggested that a specific biblical verse had “inspired” (to use his term) the tradition that Jeremiah was the concealer of the Ark.21 The verse in question is Jer 3:16:

והיה כי תרבו ופריתם בארץ בימים ההמה, נאם ה', לא יאמרו עוד ארון ברית ה' ולא יעלה על לב ולא יזכרו בו ולא יפקדו ולא יעשה עוד

And when you increase and are fertile in the land, in those days—declares the Lord—men shall no longer speak of the Ark of the Covenant of the Lord, nor shall it come to mind. They shall not mention it, or miss it, or make another.22

Weitzman points to 2 Macc 2:4, which relates that it was “in the writing” (ἐν τῇ γραφῇ) that the prophet, i.e., Jeremiah, received an oracle (χρηματισμοῦ) telling him to take the Ark and the tent with him: Weitzman suggests that Jer 3:16 must be that oracle. By the nature of the beast, it is extremely difficult to prove that a specific biblical verse is not the jumping-off point for a later tradition. Nonetheless, Weitzman’s suggestion seems to me to be extremely improbable, for several reasons: 1) According to 2 Macc 2:4, this same oracle told Jeremiah to take the tent as well, an item not found in Jer 3:16 or its surrounding verses. 2) Similarly, 2 Macc 2:1 cites “the writings” as reporting that Jeremiah told the exiles to take with them some of the fire; clearly, then, we must postulate some unknown book as the source of the traditions noted in both 2:1 and 2:4. That is, there is good to reason to think that 2 Maccabees is engaged here in non- exegetical narrative expansion, rather than exegetical activity. 3) From the narrative flow in 2 Maccabees, we may surmise that the oracle about the Ark and the tent was given to Jeremiah after he told the exiles to take the fire and not to forget the commandments; that is, after the exile was already underway, Jeremiah received an oracle to take the tent and the Ark. This excludes Jer 3:16

21 Weitzman first offered this suggestion in a festschrift dedicated to James Kugel (S. Weitzman, “Myth, History, and Mystery in the Copper Scroll,” in The Idea of Biblical Interpretation: Essays in Honor of James L. Kugel [ed. H. Najman and J. H. Newman; JSJSup 83; Leiden: Brill, 2004], 239–55 [243]), and though a bit hesitant, was quite taken with the idea. The argument reappears in Weitzman’s book, Surviving Sacrilege: Cultural Persistence in Jewish Antiquity (Cambridge, Mass.: Harvard University Press, 2005), 25–28. -Meir Ish-Shalom (Friedmann) already indi :אין חדש תחת השמש As an aside, I note that 22 cated a potential connection with this verse in his commentary to Baraita d’Melekhet ha-Mishkan, which was published more than a century ago (Vienna: Israel.-Theol. Lehranstalt, 1908), 56. 218 Milikowsky as the oracle in question. 4) As Weitzman was of course aware,23 the entire context of Jer 3:16 is a prediction of future bliss; it seems improbable that this verse could be taken as a command to Jeremiah to hide the Ark at the moment of the destruction of the Temple. In the final analysis, therefore, although the meaning of Jer 3:16 remains unclear,24 I cannot accept Weitzman’s suggestion. Menahem Kister has noted an exemplary intertextual link between Jer 22:29 and 2 Bar. 6:7–9, a link which might imply that the former verse is the biblical source for the notion that Jeremiah was the concealer of the Ark.25 Second Baruch 6:8 states that before the earth swallowed the Temple vessels, an angel spoke to the earth and said “Earth, earth, earth, hear the word of . . . God [and receive what I commit to you].” The first part of this speech is an obvious appropriation of Jer 22:29. Kister suggested two possibilities for understanding this intertextual connection: 1) that it represents the felicitous reuse of an apt biblical phrase in a new context; or 2) that the use of the quotation here points to an exegetical understanding of the verse as implying that Jeremiah hid the Temple vessels.26 The context of Jer 22:29 is the complete condemnation and punishment of Jehoiachin. As Kister himself notes, the chronological context here does not fit that of the hiding of the vessels: all the nonrabbinic traditions about the concealment place this critical incident among the events immediately surrounding the destruction of the Temple, while the exile of Jehoiachin took place eleven years previously.27 It is also worth emphasizing that 2 Baruch itself, though it alludes to this verse in Jeremiah, does not know that Jeremiah supposedly hid the vessels, but rather suggests that an angel was the agent.28

23 Cf. Weitzman, Surviving Sacrilege, 27. 24 See M. Haran, “The Removal of the Ark of Covenant,” in idem, The Bible and Its World: Selected Literary and Historical Studies (Jerusalem: The Hebrew University Magnes Press, 2009), 146–54 (149–50) (in Hebrew). in Higayon ”,אחור וקדם: אגדות ודרכי מדרש בספרות החיצונית ובספרות חז"ל“ ,See M. Kister 25 l’Yona: New Aspects in the Study of Midrash, Aggadah and Piyut in Honor of Professor Yona Fraenkel (ed. J. Levinson et al.; Jerusalem: The Hebrew University Magnes Press, 2006), 231–59 (240–41 n. 38). Kister does not connect the verse to the idea that Jeremiah hid the vessels, but does suggest that this verse was an impetus to the creation of the motif of concealment. The allusion itself is already noted by Charles, “2 Baruch,” 2:484. 26 Some of the arguments offered below would also apply to the hypothesis that this verse generated the tradition that an angel concealed the Temple vessels. 27 Kister points to a rabbinic tradition about Jehoiachin throwing the Temple keys towards the heavens, and to a similar motif connected with the actual destruction of the Temple. I do not think the parallel is convincing. 28 It is true these traditions are quite fluid; nonetheless, it seems far-fetched to postulate that the hypothetical concealment motif originally generated from this verse could have the Lost Ark of the Covenant 219

All these considerations lead me to conclude that Kister’s first suggestion is the correct one: 2 Baruch 6:8 does not testify to an exegetical tradition grounded in Jer 22:29, but simply to an exquisite secondary usage of a poignant biblical phrase. Most probably, Jeremiah was chosen to be the concealer for the simple reason that he was the most prominent prophetic figure in Jerusalem at the time of the Destruction, the primary prophet of the final Babylonian conquest and exile.

2 Rabbinic Traditions about the Concealment of the Ark

Traditions about the concealment of the Ark abound in rabbinic literature. Among these traditions, as noted above, there is no trace of the notion that the Ark was hidden by Jeremiah. All rabbinic texts that deal with the question of who hid the Ark identify the concealer as Josiah.29 Among the earliest witnesses to this tradition (late first century–early second century CE) is Seder Olam, chapter 24, lines 16–18:30

metamorphosed so radically. The original form of the tradition would presumably have related that at the time of Jehoiachin’s exile, Jeremiah talked to the earth to hide something specific—the Temple continued to function at this time, and so the concealment cannot have included the Temple vessels in general. To get from such an interpretation to the tradition found in 2 Baruch would require that Jehoiachin’s exile become the destruction of the Temple; that the concealment of a specific item become the more general concealment of the Temple vessels; and that Jeremiah himself become an angel—all changes gainsaid by the biblical context. And after all these metamorphoses the original biblical verse is still retained? On a more basic level, I find it difficult to see how an exegete could have used the verse in Jeremiah to conclude anything about any concealment. 29 Many other passages simply comment on its concealment. See, e.g., m. Šeqal 6:1; t. Šeqal 2:18 (Lieberman, Tosefta, 2:212); t. Yoma 2:15 (Lieberman, Tosefta, 2:832); b. Yoma 53b; ʾAbot R. Nat. A, 41 (ʾAbot de Rabbi Nathan [ed. S. Schechter; Vienna: Knopflmacher, 1887], 331). Some of these texts are cited in the synoptic table of parallels in the Appendix to this paper. 30 See C. Milikowsky, ed., Seder Olam: Critical Edition, Introduction and Commentary (2 vols.; Jerusalem: Yad Izhak Ben-Zvi, 2013), 1:299 (in Hebrew). This identification is also found in an additional seven rabbinic texts: t. Soṭah 13:1 (Lieberman, Tosefta, 3:229); y. Šeqal. 6:1 (49c); y. Soṭah 8:3 (22c); b. Yoma 52b; b. Hor. 12a; b. Ker. 5b; Baraita d’Melekhet ha-Mishkan, chapter 7 (Ish-Shalom, Baraita, 49; see also Baraita d’Melekhet ha-Mishkan [ed. R. Kirschner; Cincinnati:‪ ‎Hebrew Union College Press,‎‪ 1992], 184–85); L. Ginzberg, Genizah Studies in Memory of Doctor Solomon Schechter [3 vols.; New York: Jewish Theological Seminary of America, 1928–1929], 1:379). Six of these texts are included in the 220 Milikowsky

יאשיהו גנז את הארון ואמר ללוים המבינים לכל ישראל הקדשים לה': “תנו את ארון הקדש בבית אשר בנה שלמה בן דוד מלך ישראל אין לכם משא בכתף ועתה עבדו את ה' אלהיכם ואת עמו ישראל.”

Josiah hid the Ark and said to the Levites, teachers of all Israel, who were holy to the Lord: “Put the holy Ark in the House that Solomon son of David, king of Israel, built; you cannot carry it on your shoulders. See now to the service of the Lord your God and His people Israel.” (2 Chr 35:3)

From the words, “Put the holy Ark in the House that Solomon son of David, king of Israel, built,” Seder Olam concludes that Josiah hid the Ark. Seder Olam does not make explicit its exegesis; but it seems that in the author’s understanding, the command, “Put the holy Ark . . .” refers to a hidden place within the Temple. Inasmuch as the Ark was in the Temple when Josiah spoke to the Levites, he must have been ordering the Levites to move the Ark from one place to another within the Temple itself.31 The phrase, “You cannot carry it on your shoulders,” Seder Olam understands to mean, “[the Ark] will not go down with you to Babylon, that you should [in future days] bring it up [from there] on your shoulders.” A viewpoint about the fate of the Ark similar to that of Seder Olam is cited in Baraita d’Melekhet Ha-Mishkan, chapter 7, as the last of four opinions:32

synopsis which appears as the appendix to this article; Baraita d’Melekhet ha-Mishkan is not included in the synopsis but is cited immediately below. 31 Louis Ginzberg (Legends of the Jews, 6:378) mentions a Jewish tradition cited by Ps.- Jerome on 2 Chr 35:3, which suggests that Ahaz removed the Ark from the Temple and that it remained outside of the Temple precinct until the days of Josiah. (Ginzberg was of the opinion that this commentary, clearly not authored by Jerome, was nonetheless the work of an early Christian exegete, and even devoted an entire short monograph to the rabbinic parallels contained within it [L. Ginzberg, Die Haggada bei den Kirchenvätern, I: Die Haggada in den pseudo-hieronymianischen “Quaestiones” (Amsterdam: [s.n.], 1899)]. Today, however, a more accepted view is that the work was composed by a ninth-century Jewish convert to Christianity and therefore has no relevance to early biblical exegesis; see A. Saltman, Pseudo-Jerome: Quaestiones on the Book of Samuel [StPB 26; Leiden: Brill, 1975], 12). See also the commentaries of Ps.-Rashi and R. David Kimchi to 2 Chr 35:3; both independently suggest that Manasseh or Amon had removed the Ark from the Temple, and therefore Josiah had to return it. 32 For edition information, see n. 30 above. The angle brackets are used by Kirschner, Baraita, to indicate additions to the diplomatic transcription of his base text, Ms. Oxford Opp. 726 (identified in Kirschner’s edition as Ms. Oxford Bodleian 370.11, which is its catalog number, not a shelf mark). Baraita d’Melekhet Ha-Mishkan is, like Seder Olam, a tannaitic composition, though it was composed, to my mind, a century or so later, in the first half of the Lost Ark of the Covenant 221

היכן ארון נתון? >ר' יהודה בן לקיש אומר< במקומו בבית קדשי הקדשים הוא שנ' “ויהי שם עד היום הזה.” וחכמי' או': בלשכת דיר העצים. ר' אל>י<עז' >אומר< לבבל ירד שנ' “ולא יוותר דבר אמר יי”; ואין דבר אלא דברות, הארון. ר' או': יאשיהו גנזו שנ' “ויאמר ללוים המבינים את העם וגו'”; אמ' לא ירד עמכם לבבל שתעלו אותו בכתף.

Where is the Ark found? [R. Yehudah b. Laqish says:] It is in its place in the House of the Holy of Holies, as it says, “[The poles projected beyond the Ark and the ends of the poles were visible from the front of the inner sanctuary, but they could not be seen from the outside]; and there they remain to this day” (1 Kgs 8:8; 2 Chr 5:9). The Sages say: It is in the Office of the Storing of the Wood. R. Eliezer says: It went down to Babylon, as it says, “[A time is coming when everything in your palace which your ancestors have stored up to this day will be carried off to Babylon;] not a thing [davar] will remain behind” (2 Kgs 20:17; Isa 39:6); and davar is the commandments [dibrot]—the Ark. Rabbi say: Josiah hid it, as it says, “And he said to the Levites, teachers of all Israel, holy to the Lord, ‘Put the Holy Ark in the House that Solomon son of David, king of Israel, built; you cannot carry it on your shoulders’ ” (2 Chr 35:3); he said: “It shall not go down with you to Babylon, that you shall bring it up on your shoulders.”

Baraita d’Melekhet Ha-Mishkan does not explain, however, what underlies Josiah’s statement that the Ark is not to go to Babylon. It is likewise difficult to determine Josiah’s motivation from the language of t. Soṭah 13:1:33

אמר להם: גנזו אותו שלא יגלה לבבל כשאר כל הכלים שתחזירהו למקומן.

He said to them: “Conceal it, that it should not be exiled to Babylon like all the other vessels, that you will return it to its place.”

These last two texts seem to be making the following claim: If the Ark is taken to Babylon, the exiles will not be able to bring it back with them when they return to Israel, since they have been forbidden to carry it on their shoulders.

the third century CE or perhaps a bit earlier; see M. Kahana, “Initial Observations regarding the Baraita deMelekhet haMishkan: Text, Redaction and Publication,” in Melekhet Maḥshevet: Studies in the Redaction and Development of Talmudic Literature (ed. A. Amit and A. Shemesh; Ramat-Gan: Bar-Ilan University Press, 2011), 55–67 (in Hebrew). 33 See the full citation in the synopsis at the end of the article. 222 Milikowsky

In a close parallel to t. Soṭah 13:1, found in y. Šeqal. and y. Soṭah,34 the words attributed to Josiah are different:

אמר להם: אם גולה הוא עמכם לבבל אין אתם מחזירין אותו עוד למקומו.

If it is exiled with you to Babylon, you will not return it to its place.

The meaning of the Yerushalmi passage may be identical to that of the Tosefta passage (as I explained it above); or, it may instead represent an accusation on Josiah’s part: if the Ark is exiled with the people of Israel, they will not bring it back with them when they return. From the language of all the texts just cited, it can be inferred that Josiah concealed the Ark because of the imminent exile. As is related in 2 Kgs 22:11–20 and 2 Chr 34:19–28, Josiah understood from the reading of the Torah discovered by Hilkiah that a catastrophe was about to occur. Huldah the prophetess had confirmed that this was true, but had told him that it would occur only after his death. Rabbinic literature identified Deut 28:36, which refers explicitly to exile, as the specific verse that Josiah heard, and which caused him to hide the Ark:

יולך ה' אתך ואת מלכך אשר תקים עליך אל גוי אשר לא ידעת אתה ואבתיך ועבדת שם אלהים אחרים עץ ואבן.

The Lord will drive you, and the king you have set over you, to a nation unknown to you or your fathers, where you shall serve other gods, of wood and stone.35

Although the thrust of these passages seems to be that Josiah’s action saved the Ark from exile, five rabbinic passages—Baraita d’Melekhet Ha-Mishkan chapter 7; t. Šeqal 2:18; t. Soṭah 13:1; y. Šeqal 6:1 (49c); and b. Yoma 53b—offer an alternative opinion; that is, that the Ark was indeed exiled to Babylon. All five passages attribute this opinion to Rabbi Eliezer, and two of them attribute it in addition to Rabbi Shimon. A nonrabbinic work, 4 Ezra, which deals with the events surrounding the destruction of the First Temple, also states that the Ark of the Covenant was captured (10:22).36

34 See n. 30. 35 Among our parallel versions, see the passages in t. Soṭah, y. Šeqal, y. Soṭah, b. Yoma, and b. Horayot, all cited above, n. 30. 36 M. E. Stone, Fourth Ezra: A Commentary on the Book of Fourth Ezra (Hermeneia; Minneapolis: Fortress, 1990), 316. the Lost Ark of the Covenant 223

All the passages in the Babylonian and Palestinian Talmuds which affirm that Josiah hid the Ark are more or less parallel to Toseftaֹ Soṭah. We are dealing therefore with two independent sources, t. Soṭah and its parallels, on the one hand, and Baraita d’Melekhet Ha-Mishkan, on the other. For the purposes of this analysis I shall focus on t. Soṭah as the paradigmatic representative of the first group.37 The Tosefta establishes the identity of the concealer with the words: “And who hid it? Josiah the king hid it.” These words appear as an addendum to a list of items hidden along with the Ark: “When the Ark was hidden, hidden with it was the jar of manna. . . .” Had these two statements been formulated at the same time, much more probable would have been the following construction: “When Josiah hid the Ark, hidden with it was the jar of manna. . . .”—or something similar. This leads to the conclusion that this passage was formulated in two stages, and that the name of the concealer was added at the second stage. Furthermore, it also seems logical to infer that the second statement, identifying Josiah as the concealer, was originally formulated independently of the first statement; otherwise, we would expect all the parallel texts to have the formulation found in y. Soṭah: “And who hid them?” not, “And who hid it?” Quite plausibly, the question, “And who hid it?” was originally asked in response to the ending of m. Šeqal. 6:1: “. . . for they had a tradition from their fathers that there the Ark was hidden.”38 The Baraita d’Melekhet Ha-Mishkan cites several opinions regarding the location of the Ark during the period of the Second Temple; four parallels to this passage are found elsewhere in rabbinic literature.39 Not one of them includes the information that Josiah was the concealer. This in itself would seem to confirm that this statement was lacking in the original formulation of the passage and was added only in its later formulation in the Baraita d’Melekhet Ha-Mishkan. Furthermore, it can be shown that the statement asserting that Josiah was the concealer fits imperfectly into the context of Baraita d’Melekhet Ha-Mishkan. The passage begins with the question, “Where is the Ark found?” and three opinions are cited. R. Yehudah ben Lakish maintains that it is found “in its

37 This should not be taken to imply that the formulation of t. Soṭah is more original than that of its parallels. Rather, for the sake of convenience, in this discussion, “t. Soṭah” designates “t. Soṭah and its parallels.” 38 See n. 6 above for text and translation. 39 The four parallel texts are t. Šeqal. 2:18; t. Soṭah 13:1; y. Šeqal. 6:1 (49c); and b. Yoma 53b; see nn. 29 and 30 above, and see the synoptic table of parallels at the end of the paper for the texts themselves. None of these parallels are very close to Baraita d’Melekhet Ha-Mishkan. 224 Milikowsky place in the House of the Holy of Holies”; the sages hold that it is “in the Office for the Storing of the Wood”; R. Eliezer says, “It went down to Babylon.” Finally, Rabbi states that “Josiah hid it.” Note that the first three opinions name alternate locations and thus conflict with one another; but the opinion of Rabbi, dealing not with the Ark’s location but with the identity of its concealer, is consistent with both the view of R. Yehudah ben Lakish and the view of the sages. Thus, in all the rabbinic texts which state that Josiah hid the Ark, aside from Seder Olam, this identification appears at a late stage in the formulation of the passage in question. The hypothetical earlier formulations of these passages, as I have reconstructed them, do not state that Josiah hid the Ark; only as a result of the influence of Seder Olam, or a similar source, did the opinion prevail that the concealer was Josiah.

3 Concluding Observations

As we have seen above, the tradition identifying Jeremiah as the concealer of the Ark, a tradition found only in nonrabbinic texts, took shape very early; the assertions that Josiah was the concealer, however, occur in secondary formulations of all the passages within which this tradition is found, apart from Seder Olam. As I have shown, the identification of Josiah as the concealer in Seder Olam derives from the interpretation of Josiah’s directive to the Levites in 2 Chr 35:3. That is, the opinion that Josiah concealed the Ark originated either in the interpretative activity of Seder Olam itself or in that of a source upon which it drew. It is important to stress that the opinion that Josiah was the concealer is grounded in the exegesis of the biblical text, but the opinion that Jeremiah hid the Ark is not textually grounded. In other words, the shift from Jeremiah to Josiah is in reality a shift from one view of biblical interpretation to another.40 The nonrabbinic texts discussed in this paper espouse a relatively restraint- free approach to biblical interpretation—or perhaps better, to biblical narra-

40 I find convincing neither of the suggestions made by M. F. Collins (“The Hidden Vessels in Samaritan Traditions,” JSJ 3 [1972]: 97–116 [105]), in her attempt to explain the shift from Jeremiah to Josiah. She writes that “[t]his change may be due partially to a post-70 Javnean reaction against Jewish and Christian apocalypticism, and partially to a struggle between the royal and the charismatic models in the Rabbinic Academies.” I fail to see why Josiah is more amenable to an anti-apocalyptic model, and I know of no “royal model” in the rabbinic academies. the Lost Ark of the Covenant 225 tive expansion—which allows the addition of specific details to the biblical plotline even when these details lack a foundation in the biblical text. This is replaced in the rabbinic documents by an approach which aspires to present a biblical source text for each added detail. All the traditions we have looked at share common concerns: where did the Ark end up, and who was responsible? That they generate different answers— beneath the Temple, at Moses’s grave, or perhaps in Babylon; Jeremiah versus Josiah—might be thought to result from ideological or theological differences between them. I have suggested, however, that at least the shift from Jeremiah to Josiah stems from a difference in exegetical methodology. I am not claiming, of course, that nonrabbinic narrative expansion never had a textual basis; indeed, James Kugel has without a doubt shown just how often exegesis underlies some piece of narrative tradition that contains not a trace of an obvious exegetical foundation.41 To my mind, however, it is nonetheless true that early rabbinic exegetical practice much preferred narrative expansion with a textual basis, and indeed was most emphatic on this point. This subject, I believe, is worthy of further analysis, and I hope to return to it in the future.

41 See especially J. L. Kugel, Traditions of the Bible: A Guide to the Bible As It Was at the Start of the Common Era (Cambridge, Mass.: Harvard University Press, 1998), passim. 226 Milikowsky

Appendix: Rabbinic Texts on the Hiding of the Ark

תוספתא סוטה יג:א תוספתא יומא ב:טו תוספתא שקלים ירושלמי שקלים ו:א ירושלמי סוטה ח:ג בבלי יומא נב ע״ב בבלי הוריות יב ע״א בבלי כריתות ה ע״ב בבלי יומא נג ע״ב ב:יח מט ע״ג )לא ברצף( כב ע״ג

משנגנז ארון נגנז צנצנת המן וצלוחית משנגנז הארון משנגנז הארון נגנז והתניא משנגנז ארון נגנזה והתניא משנגנז ארון נגנז והתניא משנגנז ארון נגנז עמו צנצנת המן של שמן המשחה נגנז עמו צנצנת עמו צנצנת המן עמו צנצנת המן וצלוחית שמן המשחה וצנצנת המן צנצנת המן וצלוחית שמן וצלוחית של שמן ומקלו של אהרן המן וצלוחית שמן וצלוחית שמן המשחה שמן המשחה ומקלו של ומקלו של אהרן שקדיה המשחה ומקלו של אהרן המשחה ומקלו של שקדיה ופרחיה המשחה ומקלו של ומקלו של אהרן אהרן ושקדיה ופרחיה ופרחיה וארגז ששלחו שקדים ופרחים וארגז אהרן שקדיה וארגז דרון שהושיבו אהרן ופרחיו ופרחיו ושקידיו וארגז וארגז ששגרו פלשתים פלשתים דורון לישראל ששגרו פלשתים דורון ופרחיה וארגז פלשתים כבוד לאלי ושקידיו וארגז שהשיבו פלשתים דורון לאלהי ישראל שנאמר )שמ״א ו:ח( ואת לאלהי ישראל שנאמר שהשיבו פלשתים ישראל היו כולם שהשיבו פלשתים אשם לאלהי ישראל שנאמר )שמ״א ו:ח( וכלי כלי הזהב אשר השבותם )שמ״א ו:ח( ואת כלי דרין לאלי ישראל בבית קדש הקדשים אשם לאלהי ישראל הזהב אשר השבתם לו לו אשם תשימו בארגז הזהב אשר השיבותם לו כולם היו בבית קדש משנגנז ארון נגנזו אשם תשימו בארגז מצדו מצדו ושלחתם אותו אשם תשימו בארגז מצדו הקדשים ומשנגנז עמו ושלחתם אתו והלך והלך ארון נגנזו עמו דברים ומי גנזו יאשיהו מי גנזו יאשיהו גנזו ומי גנזן יאשיהו וכיון ומי גנזו יאשיהו גנזו מה ומי גנזו יאשיהו מלך ומי גנזו יאשיה מלך המלך גנזו מה ראה כיון שראה שכתוב שראה שכתוב יולך ראה שגנזו ראה שכתוב יהודה גנזו שראה שכתוב יהודה גנזו שנאמר ויאמר כיון שראה כתו׳ יולך יי׳ אותך ואת יי׳ אותך ואת מלכך )דברים כח:לו( יולך ה׳ בתורה )דברים כח:לו( המלך אל הכהנים תנו בתורה יולך ה׳ מלכך אשר תקים הדא היא דכתיב אתך ואת מלכך אשר תקים יולך ה׳ אותך ואת מלכך את ארון הקדש אותך ואת מלכך עליך אל גוי אשר ויאמר ללוים המבינים עליך עמד וגנזו שנאמר וגו׳ צוה וגנזום שנאמר וגו׳ פיקד ללוים לא ידעת אתה לכל ישראל הקדושים )דה״ב לה:ג( ויאמר ללוים )דה״ב לה:ג( ויאמר וגנזוהו שנ׳ ויאמר ואבתיך עמד וגנזו ליי׳ תנו את ארון המבינים לכל ישראל ללוים המבינים לכל ללוים המבינים לכל הדא הוא דכתיב הקודש בבי׳ אשר בנה הקדושים לה׳ תנו את ארון ישראל הקדושים לה׳ תנו ישראל הקדשים ויאמר ללוים שלמה בן דוד מלך הקדש בבית אשר בנה את ארון הקדש בבית לה׳ תנו את ארון המבינים ולכל ישראל אין לכם משא שלמה בן דויד מלך ישראל אשר בנה שלמה בן דוד הקדש בבית אשר ישראל הקדושים בכסף )דה״ב לה:ג( אין לכם משא בכתף עתה מלך ישראל אין לכם בנה שלמה מלך ליי׳ תנו את ארון אמר אם גולה הוא עבדו את ה׳ אלהיכם ואת משא בכתף עתה עבדו ישראל אין להם הקודש בבית אשר עמכם לבבל אין אתם עמו ישראל את ה׳ אלהיכם ואת עמו משא בכתף )דה״ב בנה שלמה בן מחזירין אותו למקומו ישראל לה:ג( אמ׳ להם גנזו דוד מלך ישר׳ אין אלא עתה עבדו את אותו שלא יגלה לכם משא בכתף יי׳ אלהיכם ואת עמו לבבל כשאר כל )דה״ב לה:ג( אמר ישראל )שם( הכלים שתחזירוהו להן אם גולה הוא למקומו שנ׳ עתה עמכם לבבל עוד עבדו את ה׳ אין אתם מחזירין אלהיכם ואת עמו אותו למקומו אלא ישר׳ )שם( מיד גנזו עתה עבדו את יי׳ אותו הלוים אלהיכם ואת עמו ישראל )שם( the Lost Ark of the Covenant 227

Appendix: Rabbinic Texts on the Hiding of the Ark

(cont.)

ר׳ ליעזר אומ׳ ארון ר׳ ליעזר אומ׳ ארון תני בשם רבי דתניא רבי אליעזר גלה לבבל שנ׳ לא גלה לבבל שנא׳ אליעזר הארון גלה אומר ארון גלה לבבל יותר דבר אמר ה׳ לא יותר דבר אמר עמהן לבבל מה שנאמר )דה״ב לו:י( )מל״ב כ:יז; יש׳ ה׳ ואין דבר אלא טעמא לא יוותר ולתשובת השנה שלח לט:ו( ואין דבר אלא דברות שבו דבר אמר יי׳ )מל״ב המלך נבוכדנאצר דברות שבו כ:יז; יש׳ לט:ו( אין ויבאהו בבלה עם כלי דבר אלא חמדת בית ה׳ שהדיברות לתוכו

ר׳ שמעון אומ׳ הרי ר׳ שמעון אומ׳ הרי וכן הוא אומר רבי שמעון בן יוחאי הוא אומ׳ ולתשובות הוא אומ׳ ולתשובת ולתשובת השנה אומר ארון גלה לבבל השנה שלח המלך השנה שלח המלך שלח המלך שנאמר )יש׳ לט:ו( לא נבוכדנצר ויביאהו נבוכדנצר ויביאהו נבוכדנצר ויביאהו יותר דבר אמר ה׳ אלו בבלה עם כלי בבלה עם כלי בבלה עם כלי עשרת הדברות שבו חמדת בית ה׳ חמדת בית ה׳ אילו חמדת בית יי׳ אי )דה״ב לו:י( זה ארון הן כלי חמדת בית זהו חמדת בית יי׳ ה׳ זה ארון )דה״ב לו:י( זה הארון ר׳ יהודה בן לקיש ר׳ יהודה בן לקיש רבי שמעון בן לקיש רבי יהודה )בן לקיש( או׳ ארון נגנז או׳ ארון במקומו אמר במקומו היה אומר ארון במקומו במקומו שנ׳ ויאריכו ניגנז שנ׳ ויאריכו הארון גנוז הדא נגנז שנאמר )מל״א הבדים ויראו וגו׳ הבדים ויראו ראשי הוא דכתיב ויאריכו ח:ח( ויראו ראשי ויהיו שם עד היום הבדים מן הקודש הבדים ויראו ראשי הבדים מן הקדש על הזה )מל״א ח:ח( אל פני הדביר ולא הבדים אל הקדש פני הדביר ולא יראו יראו החוצה ויהיו אל פני הדביר ולא החוצה ויהיו שם עד שם עד היום הזה יראו החוצה )מל״א היום הזה ח:ח( כתיב ויראו ואת אמר ולא יראו אלא נראין ולא נראין בולטין ויוצאין כשני דדי האשה ורבנן אמרין בלישכת דיר העצים היה הארון גנוז the Lost Ark of the Covenant 229

Sergey Minov

The sinister figure of Satan, about whom the canonical writings of the Bible provide so little information, puzzled many generations of Jewish and Christian theologians and exegetes from antiquity through the middle ages.1 Perhaps the most significant challenge posed by this mythological figure for the thinkers of the nondualist mainstream in both Judaism and Christianity was the need to explain why and how Satan, created originally as good, became quite the opposite—a quintessential and paradigmatic enemy of the omnipo- tent and good deity, and of humanity. A number of different explanations have been suggested to account for the fall of Satan. Generally speaking, they may be divided into two major groups: 1) stories in which Satan forfeits his original quality of goodness because he tries to challenge God himself and usurp his place in heaven;2 and 2) stories in which he loses his exalted status because of his enmity towards the primeval humans, God’s creatures.3 The purpose of the present study is to investigate one particular version of the myth of the fall of Satan, which belongs to the second group. Its distin- guishing feature is the combination of the two closely related but nevertheless distinctive submotifs: 1) the veneration of the newly created Adam by the

* I am most grateful to Prof. Menahem Kister for his generous help and insightful comments on an earlier version of this paper, which helped to improve it significantly. Needless to say, the responsibility for any shortcomings that remain is mine alone. 1 On the origins of this mythological figure and the earliest stages in its development in ancient Jewish and Christian traditions, see J. B. Russell, Satan: The Early Christian Tradition (Ithaca: Cornell University Press, 1981); N. Forsyth, The Old Enemy: Satan and the Combat Myth (Princeton: Princeton University Press, 1987); E. H. Pagels, The Origin of Satan (New York: Vintage, 1996); H. A. Kelly, Satan: A Biography (Cambridge: Cambridge University Press, 2006); A. A. Orlov, Dark Mirrors: Azazel and Satanael in Early Jewish Demonology (Albany, N.Y.: State University of New York Press, 2011). 2 This explanation often invokes the motif of Satan’s pride vis-à-vis God. 3 This explanation often invokes the motif of Satan’s “envy” towards Adam. For a concise presentation of these two positions, see J.-M. Rosenstiehl, “La chute de l’Ange: Origines et développement d’une légende; ses attestations dans la littérature copte,” in Écritures et traditions dans la littérature copte: Journée d’études coptes, Strasbourg 28 mai 1982 (Cahiers de la Bibliothèque copte 1; Louvain: Peeters, 1983), 37–60, esp. 37–53.

© koninklijke brill nv, leiden, ��5 | doi ��.��63/9789004299139_011 Satan’s Refusal to Worship Adam 231 angelic forces; and 2) the refusal of Satan to participate in this act, which results in his rejection by God. One of the earliest attestations of this interpretation of Satan’s demotion comes from the apocryphal composition known as the Life of Adam and Eve. In the first part of my investigation, I discuss the account of the fall of Satan as it is presented in the Life. The main thrust of this discussion is that this interpretation of Satan’s fall is deeply rooted in the context of ancient Jewish speculation on the figure of Adam; I argue that this account took its point of departure from a Jewish tradition about the veneration of Adam by the angels. This latter tradition is itself attested in such diverse sources as the Slavonic apocryphon 2 Enoch and some rabbinic texts. In the second section I offer an overview of the reception history of this originally Jewish tradition in the Syriac Christian milieu, from its earliest appearance during late antiquity, in the sixth-century composition known as the Cave of Treasures, until the early modern period. In the process, I explore how this tradition was adapted to and functioned within a wide range of literary genres and rhetorical settings. I place particular emphasis on how this tradition became an important topic of the Christian dialogue with Islam, in the context of the complex cross-cultural exchange that characterized societies of the medieval Near East. Because this explanation of Satan’s fall gained canonical status in the Muslim tradition, where it appears already in the Qurʾān, some later Syriac-speaking Christians began to perceive it as problematic and tried to marginalize it; those who continued to use this tradition also mobilized it for the purpose of polemic against Islam. I connect the diversity among Syriac- speaking Christians in the usage of this account with its popularity as an element of the mythological discourse that was shared by many groups across the Islamicate world: a world which was shaped by the tradition of the dominant Muslim majority, but was open to a certain degree to the participation of various religious minorities.

1 The Fall of Satan in the Life of Adam and Eve

One of the earliest attestations of the explanation of Satan’s fall as a result of the conflict with Adam comes from the Life of Adam and Eve, a retelling of the life story of the primeval couple written most probably during the period 100–300 CE.4 This work, the original language of which was apparently Greek,

4 For general information on this work, see M. E. Stone, A History of the Literature of Adam and Eve (SBLEJL 3; Atlanta: Scholars Press, 1992); and M. de Jonge and J. Tromp, The Life of Adam 232 Minov is preserved in several recensions—Greek, Latin, Armenian, Georgian, and Old Slavonic.5 The account of Satan’s fall that interests us appears in paragraphs 11–17 of the Latin, Armenian, and Georgian versions of the Life.6 There, Satan succeeds in tricking Eve for a second time by obstructing her efforts to do penance in the river Tigris. Adam then asks him, in despair, for the reason of his persistent hostility towards the two humans. In his reply, Satan explains that because of Adam, he has lost his exalted place in the heavenly hierarchy of angels. Satan further relates how, when Adam was first formed, animated, and endowed with the image and likeness of his creator, the archangel Michael brought him to bow down to God. This act of worship was followed by God’s public recognition that Adam had indeed been created in his image and likeness. Then, Michael summoned the rest of angels, ordered them to “worship” Adam,7 and set the example by bowing down (or prostrating himself) before the first man. When the archangel ordered Satan to comply the latter refused, on the grounds that having been created first, he could not worship Adam, who had been created last. The angels under Satan’s command followed their leader and also refused to worship Adam. Displeased by this act of disobedience, God expelled Satan and his angels from the heavens.

and Eve and Related Literature (Guides to the Apocrypha and Pseudepigrapha 4; Sheffield: Sheffield Academic Press, 1997). 5 These versions have been conveniently published in synoptic format, together with English translations, in G. A. Anderson and M. E. Stone, A Synopsis of the Books of Adam and Eve (2d rev. ed.; SBLEJL 17; Atlanta: Scholars Press, 1999). There are also recent critical editions of the Greek and Latin versions of the Life: J. Tromp, The Life of Adam and Eve in Greek: A Critical Edition (PVTG 6; Leiden: Brill, 2005); and J.-P. Pettorelli, J.-D. Kaestli, A. Frey, and B. Outtier, Vita latina Adae et Evae: Synopsis vitae Adae et Evae, Latine, Graece, Armeniace et Iberice (2 vols.; CCSA 18–19; Turnhout: Brepols, 2012). 6 For the texts and an English translation, see Anderson and Stone, Synopsis, 13–18e. For the Latin recension, see also Pettorelli et al., Vita latina, 1:300–315 (Latin text and French translation); 2:778–89 (synopsis of the Latin, Armenian, and Georgian versions). For an alternative English translation of the Latin version by M. D. Johnson, see OTP 2:260–64. 7 The Latin text describes this action using the verb adorare, which undoubtedly renders the verb προσκυνεῖν of the original Greek prototype of the Life. The verb προσκυνεῖν originally designated a hand-kissing gesture of devotion towards the gods; later on it came to describe the act of kneeling or prostrating oneself in obeisance, whether before gods or kings. Thus, during the period of the Life’s composition, i.e., the Roman Imperial era, both the Greek προσκυνεῖν and the Latin adorare were used by classical authors to describe obeisance before earthly rulers as well as the worship of the gods. See B. M. Marti, “Proskynesis and Adorare,” Language 12/4 (1936): 272–82. Satan’s Refusal to Worship Adam 233

It should be noted here that although the story of Satan’s fall is not attested in the existing Greek witnesses of the Life, there are compelling reasons to think that it formed an integral part of the Greek Vorlage of this composition, from which the Latin and other versions derive. This has been convincingly argued by Michael Stone on the basis of evidence provided by the Armenian and Georgian versions.8 There is continuing debate among students of apocryphal literature as to whether the Life should be regarded as a Jewish or a Christian work. While previous scholarship primarily considered the book to be a Jewish composition, more and more scholars during the last decades have begun to challenge this consensus and to advance the argument that the Life is a Christian text.9 Nevertheless, the question of the Life’s confessional milieu is still sub iudice. Accordingly, at the moment it seems preferable to follow the cautious assessment of George Nickelsburg, who in a recent survey of the status quaestionis with regard to this issue concludes that, “in the present state of the discussion the provenance of the versions of the Life of Adam and Eve is uncertain, but seems to tip in favor of Christian authorship of the Life of Adam and Eve in the versions in which it is now extant.” At the same time, Nickelsburg warns scholars of the dangers of using the various versions of the Life uncritically, either “as attestations of first-century Jewish religious thought or as certain testimonies to an as yet undefined sector of the second- or third-century church.”10 Since Nickelsburg’s observation that the presence of Jewish traditions in the versions of the Life “by no means excludes Christian authorship” works both ways,11 I will not address here the general issue of the origins of the Life as a whole composition. Instead, I will focus on the question of whether the particular tradition concerning the fall of Satan that appears in this work might be better understood as Jewish or Christian.

8 See M. E. Stone, “The Fall of Satan and Adam’s Penance: Three Notes on The Books of Adam and Eve,” JTS n.s. 44/1 (1993): 143–56 (153–56). 9 See M. de Jonge, “The Christian Origin of the Greek Life of Adam and Eve,” in Literature on Adam and Eve: Collected Essays (ed. G. A. Anderson, M. E. Stone and J. Tromp; SVTP 15; Leiden: Brill, 2000), 347–63; R. Nir, “The Aromatic Fragrances of Paradise in the Greek Life of Adam and Eve and the Christian Origin of the Composition,” NovT 46/1 (2004): 20–45; J. R. C. Cousland, “The Latin Vita—A ‘Gospel’ of Adam and Eve?” in . . . And So They Went Out: The Lives of Adam and Eve as Cultural Transformative Story (ed. D. V. Arbel, J. R. C. Cousland, and D. Neufeld; London: T&T Clark, 2010), 121–42, 157–59. 10 G. W. E. Nickelsburg, Jewish Literature between the Bible and the Mishnah: A Historical and Literary Introduction (2d ed.; Minneapolis: Fortress, 2005), 332. 11 Nickelsburg, Jewish Literature, 332. 234 Minov

For most of the past century, it has been commonly assumed by students of ancient Judaism that the myth of Satan’s fall in the Life was Jewish in origin.12 This consensus, however, was not unanimous. Among the first scholars to clas- sify the myth of Satan’s refusal to worship Adam as Christian were Jewish students of Islam. Thus, Abraham Geiger, in his pioneering work on the Jewish background of the Qurʾān, commented in relation to the Qurʾānic story of the devil’s refusal to worship Adam that it “bears unmistakable marks of Christian development, in that Adam is represented in the beginning as the God-man, worthy of adoration, which the Jews are far from asserting.”13 Recently, Jean-Daniel Kaestli has made perhaps the most sustained attempt to demonstrate the Christian origin of this myth.14 The main arguments presented by Kaestli are: (a) the interpolated and Christian character of the myth of Satan’s fall in the long recension of 2 En. 29:4–5, used by some scholars to support the theory of the Jewish origin of the story of Satan’s fall in the Life; (b) the irrelevance of the story of Satan’s fall in the medieval midrash, Bereshit Rabbati,15 for the reconstruction of the earliest stages in the development of this tradition; (c) the difficulty involved in proving that the notion of the angelic veneration of Christ in the New Testament, i.e., Heb 1:6, is dependent

12 Cf. L. Ginzberg, The Legends of the Jews (trans. H. Szold; 7 vols.; Philadelphia: Jewish Publication Society, 1909–1938), 5:85; C. H. T. Fletcher-Louis, “The Worship of Divine Humanity as God’s Image and the Worship of Jesus,” in The Jewish Roots of Christological Monotheism: Papers from the St. Andrews Conference on the Historical Origins of the Worship of Jesus (ed. C. C. Newman, J. R. Davila, and G. S. Lewis; JSJSup 63; Leiden: Brill, 1999), 112–28 (127–28); and most recently, J. Dochhorn, “The Motif of the Angels’ Fall in Early Judaism,” in Angels: The Concept of Celestial Beings—Origins, Development and Reception (ed. F. V. Reiterer, T. Nicklas, and K. Schöpflin; Deuterocanonical and Cognate Literature Yearbook 2007; Berlin: de Gruyter, 2007), 477–95 (486–91). 13 A. Geiger, Judaism and Islam: A Prize Essay (trans. F. M. Young; Madras: M.D.C.S.P.C.K. Press, 1898), 77. A similar opinion was expressed by another Jewish scholar, Leo Jung, who in his comprehensive overview of Jewish, Christian, and Muslim speculations concerning the fall of the angels, characterizes this tradition as “extra-talmudic and Christian”; see L. Jung, “Fallen Angels in Jewish, Christian, and Mohammedan Literature: A Study in Comparative Folk-Lore,” JQR n.s. 15/4 (1925): 467–502; 16/1 (1925): 45–88; 16/2 (1925): 171–205; 16/3 (1926): 287–336; the quotation is from 16/1:61. 14 See J.-D. Kaestli, “Le mythe de la chute de Satan et la question du milieu d’origine de la Vie d’Adam et Ève,” in Early Christian Voices in Texts, Traditions and Symbols: Essays in Honor of François Bovon (ed. D. H. Warren, A. G. Brock, and D. W. Pao; Biblical Interpretation Series 66; Leiden: Brill, 2003), 341–54. 15 Where the narrative contours and the general character of the composition suggest Muslim influence (see further below, n. 129). Satan’s Refusal to Worship Adam 235 on the account of the veneration of Adam by angels in the Life.16 In Kaestli’s opinion, the myth of Satan’s fall in the Life originated as a result of Christian interpretation of Wis 2:23–24 (“through envy of the devil death came into the world”), in light of Rom 5:12 (“sin entered the world through one man, and death through sin”).17 While Kaestli’s critique of some of the arguments for a Jewish origin of the Life’s account is legitimate, it is not sufficient in and of itself to prove the Christian origin of this tradition. Moreover, his suggestion that the Christian reading of Wis 2:24 through the lens of Rom 5:12 lies at the basis of the tradition is too sketchy. Even more important, it misses the point of the narrative, since according to the Life, Satan is driven not so much by the feeling of envy as by that of pride, which finds expression in his claim that he is superior to Adam. Kaestli’s hypothesis also does not adequately explain the connection between Satan’s demotion and the angelic veneration of Adam. In fact, there are several considerations which render the theory of the Jewish origin of this tradition preferable to the theory of a Christian origin. One of the most important concerns the antiquity of the motif of angelic sub- jugation to “deified” humans in Jewish sources. We have been able to trace the prehistory of this motif because of the evidence provided by 2 Enoch. Chapters 21–22 of this work relate the story of Enoch’s ascent to heaven, during which the patriarch joins the angelic ranks.18 There is a particular episode of this narrative, found in 2 En. 22:6–7, that echoes features of the story of Satan in the Life. In both recensions of 2 En. 22:6, it is said that God has brought Enoch to heaven in order to “test” his angelic retinue.19 Furthermore, according to both recensions of 2 En. 22:7, the angels pass this test successfully by acknowledging the patriarch’s new exalted status and performing an act of “obeisance.”20 Michael Stone was the first to notice a basic similarity between this story and

16 In which case the latter would predate the former. 17 Kaestli, “Le mythe de la chute,” 353–54. 18 For the Slavonic text, see A. Vaillant, Le Livre des secrets d’Hénoch: Texte slave et traduction française (Textes publiés par l’Institut d’Études slaves 4; Paris: Institut d’Études slaves, 1952), 22–26; or the more recent edition by G. Macaskill, The Slavonic Texts of 2 Enoch (Studia Judaeoslavica 6; Leiden: Brill, 2013), 96–103. For an English translation by F. I. Andersen, see OTP 1:134–39. 19 According to the long recension: “and the Lord said to his servants, while testing them” (и реч̑ г͠ь слꙋгам своим искѹшаѫ их); contrast the short recension: “and the Lord tested his servants, while speaking to them” (и искyси г͠ь слyги своѧ гл͠а к ним); ed. Macaskill, Slavonic Texts, 100–101. 20 While the long recension states explicitly that the target of the angelic obeisance was God himself—“and the glorious ones worshipped the Lord” (и поклонишѫс҄ славнїи г͠ви)—the 236 Minov that of Satan’s fall in the Life.21 As he points out, the fact that these two apocryphal narratives share this motif of God’s test of the angels’ obedience through the elevation of a human being leads to the conclusion that the author of 2 Enoch 21–22 “knew a story of the rebellion of Satan that strongly resembled that which is found in chapters 11–17 of the primary Adam book.”22 These insights of Stone have been adopted and further developed by Gary Anderson, who arrives at a similar conclusion: “One cannot imagine that the tradition in the Enoch materials was created independently from the tradition found in the Vita.”23 Moreover, as has rightly been pointed out by Anderson, the story of Satan’s fall in the Life is based on the biblical doctrine of election.24 In fact, it might be regarded as an attempt to make sense of the figure of Satan, underdeveloped in Scripture, by looking at this figure through the lenses of the motif of election and elevation of the beloved son. In Anderson’s opinion, the motif of election allows us to anchor this narrative in Jewish tradition. Stone’s hypothesis of a close connection between 2 En. 22:6–7 and Life 11–17 has recently been supported by Andrei Orlov as well. Orlov considers the former composition to be engaging in intertextual polemics with the Adamic traditions attested in the latter.25 In addition, Orlov has convincingly argued that the Adamic tradition underlying 2 En. 22:6–7 “is not an interpolation, but belongs to the original core of the Slavonic apocalypse.”26 According to Orlov, “the tradition of angelic veneration [of Adam] is interwoven into the original fabric” of 2 Enoch’s text. This may be seen in such passages of the long recension as 7:3–4, where during the patriarch’s ascent through the second heaven, the fallen angels that are imprisoned there “bow down” to him and implore him to intercede on their behalf before God; and 18:3, where these same angels are identified as companions of Satanael.27 The presence in 2 Enoch of the motif of God testing the angels by elevating a human being provides us with strong testimony in favor of the antiquity of

short recension is more ambiguous—“and the glorious ones worshipped” (и поклонишас҄ славнии); ed. Macaskill, Slavonic Texts, 100–101. 21 See Stone, “The Fall of Satan,” 144–48. 22 Stone, “The Fall of Satan,” 148. 23 G. A. Anderson, “The Exaltation of Adam and the Fall of Satan,” Journal of Jewish Thought and Philosophy 6 (1997): 105–34 (125). 24 Anderson, “Exaltation of Adam,” 133–34. 25 A. A. Orlov, The Enoch–Metatron Tradition (TSAJ 107; Tübingen: Mohr Siebeck, 2005), 219–22. 26 Orlov, Enoch–Metatron, 221. 27 Orlov, Enoch–Metatron, 221. For the texts, see Macaskill, Slavonic Texts, 54–55, 86 [Slav. text]; Andersen, OTP 1:114, 130 [Engl. trans.]. Satan’s Refusal to Worship Adam 237 the Adam-related version of the motif found in the Life. According to Orlov, the tradition of the angelic veneration of Adam must have been in circulation by the first century ce, when it was appropriated into the Enochic text.28 Such an early dating for this motif appears quite plausible, especially in light of recent developments in research on 2 Enoch that locate this apocryphal composition firmly within the context of Second Temple Judaism.29 Furthermore, some scholars argue that the motif of the angelic worship of Adam stands behind the christological description in the first chapter of the Epistle to the Hebrews (1:3–6), which depicts the exalted status of Christ as the “firstborn” (πρωτότοκος) of God, whom “all God’s angels worship.”30 One also discovers this motif reverberating through the corpus of Jewish writings from late antiquity. Thus, several rabbinic works explore the connection between the biblical notion of Adam as the “image of God” and the motif of his worship by the angels. One of the earliest examples of this kind is found in Genesis Rabbah, a fifth-century midrashic collection of Palestinian provenance. In the section dealing with Gen 1:26–27, this work transmits a midrashic tradition, attributed to R. Hoshaya: When God created Adam “in his image,” the -and wanted to honor him by exclaim (טעו בו) ”ministering angels “erred in him i.e., by acclaiming him using the first word of the angelic ;(קדוש) ”!ing “Holy prayer before God given in Isa 6:3. Only God’s direct intervention, by putting Adam to sleep in order to demonstrate his mortal nature, prevents the angels from compromising themselves in the eyes of their Creator, who ought to have been the sole object of their worship and adoration. To illustrate this explanation, the compiler of the midrash supplements it with a parable about a king who rode in a chariot, together with his governor. The king’s subjects wanted to -from Lat. Domine), but became con ,דומיני) ”!hail him by exclaiming “Sovereign fused as to which of the two riders they should address. To dispel their doubts, the king pushed the governor out of the chariot, thereby demonstrating his own superiority.31

28 Orlov, Enoch–Metatron, 219 n. 32. 29 See New Perspectives on 2 Enoch: No Longer Slavonic Only (ed. A. A. Orlov and G. Boccaccini; Studia Judaeoslavica 4; Leiden: Brill, 2012), especially the contributions by C. Böttrich, A. A. Orlov, and D. Stökl Ben-Ezra. 30 See on this J. Dochhorn, “Die Christologie in Hebr 1,1‒2,9 und die Weltherrschaft Adams in Vit Ad 11‒17,” in Biblical Figures in Deuterocanonical and Cognate Literature (ed. H. Lichtenberger and U. Mittmann-Richert; Deuterocanonical and Cognate Literature Yearbook 2008; Berlin: de Gruyter, 2009), 281–302. 31 Gen. Rab. 8:10; ed. J. Theodor and C. Albeck, Midrash Bereschit Rabba: Critical Edition with Notes and Commentary (3 vols.; 2d rev. ed.; Jerusalem: Wahrmann, 1965), 1:63–64. For an English translation, see H. Freedman, Midrash Rabbah: Genesis (2 vols.; London: Soncino, 238 Minov

This midrashic passage establishes an intertextual connection between Gen 1:26–27, Gen 2:21, and Isa 2:22 (which is quoted at the conclusion), bringing them together on the basis of the extracanonical motif of the angelic veneration of the newly created Adam. Remarkably, however, the midrash employs this motif in a manner rather different from both the Life and 2 Enoch. In a striking contrast to these apocryphal works, the rabbinic text completely reverses the religious significance of the veneration of Adam, presenting the angels’ attempt to worship the first man as an example of misdirected and illegitimate religious activity, incompatible with the worship of God alone. An additional rabbinic example of an inversion of the motif of angelic veneration of Adam is found in the relatively late midrash, Pirqe de-Rabbi Eliezer.32 (כל הבריות) ”Chapter 11 of this composition relates that when “all the creatures saw the glorious appearance of the newly created Adam, who was “adorned ,they “became afraid of him ,(מתואר כדמות אלהים) ”with the Divine Image and they came to do obeisance to ,(בוראן) thinking that he was their Creator -In distinction from Genesis Rabbah, here it is Adam him 33”.(להשתחוות לו) him self who deflects the creatures from the crime of lèse-majesté and directs them to the worship of their true creator. Another rabbinic variation on the theme of the angelic veneration of Adam comes from the fragments of the midrash Tanḥuma–Yelamdenu, published by Ephraim Urbach. According to this midrash, the first reaction of the angels and gigantic proportions of (דמות חדשה) ”when they saw the “new appearance It is only the intervention .(להשתחוות לו) ”Adam, was to “do obeisance to him of the archangel Michael, who quotes Isa 2:22, that prevents them from committing this mistake.34 There can be little doubt that these rabbinic stories stand in a close relation to the tradition of the angelic worship of Adam. It is a more difficult task, however, to establish with precision what motivated the rabbinic inversion

1939), 1:61. This parable is also attested, although without attribution to R. Hoshaya, in Eccl. Rab. 6:9. 32 On the question of earlier apocryphal material in this midrash, see A. Urowitz- Freudenstein, “Pseudepigraphic Support of Pseudepigraphical Sources: The Case of Pirqe de Rabbi Eliezer,” in Tracing the Threads: Studies in the Vitality of Jewish Pseudepigrapha (ed. J. C. Reeves; SBLEJL 6; Atlanta: Scholars Press, 1994), 35–53. 33 Ed. D. Börner-Klein, Pirke de-Rabbi Elieser: Nach der Edition Venedig 1544 unter Berücksichtigung der Edition Warschau 1852 (SJ 26; Berlin: de Gruyter, 2004), 113–15. For an English translation, see G. Friedlander, Pirḳê de Rabbi Eliezer: The Chapters of Rabbi Eliezer the Great (London: Kegan Paul, Trench, Trübner, 1916), 79–80. 34 E. E. Urbach, “Fragments of Tanḥuma–Yelamdenu,” Kobez al Yad 6/1 [16] (1966): 1–54 (24–25) (in Hebrew). Satan’s Refusal to Worship Adam 239 of this motif. Students of rabbinic literature have suggested that the peculiar treatment of this theme in Genesis Rabbah and Pirqe de-Rabbi Eliezer exhibits distinctively polemical overtones; a number of attempts have been made to reconstruct the possible adversary against whom this polemic was aimed. One of the first scholars to make this point was Alexander Altmann, who argued that the motif of the adoration of Adam by the angels was not originally Jewish, but developed in Gnostic circles, against which the rabbis waged a polemic by inverting this heterodox mythologumenon.35 Although Altmann’s intuitions about the polemical tendency behind the censure of the motif in these rabbinic works do certainly deserve scholarly attention, his attempt to connect the motif itself with Gnosticism can hardly be accepted as plausible, especially in light of what we now know about this late ancient form of religi- osity and its mythology, half a century after the discovery of the Nag Hammadi library.36 Another solution to the problem of the polemical context of these midrashim has been offered recently by Peter Schäfer.37 In his analysis of several passages in the rabbinic corpus that deal with the figure of Adam, including those aimed against the angelic veneration of the first man, Schäfer comes to the conclusion that “the rabbinic polemic against Adam as a supernatural and (semi)divine being is aimed at possible christological interpretations of the Adam myth.”38 This hypothesis deserves to be taken seriously in light of a growing awareness, among students of rabbinic Judaism, of Christianity as an important constitutive factor in development of the new form of Jewish identity promulgated by the rabbis.39

35 See A. Altmann, “The Gnostic Background of the Rabbinic Adam Legends,” JQR n.s. 35/4 (1945): 371–91 (379–87). 36 For general surveys of Gnosticism, see M. A. Williams, Rethinking “Gnosticism”: An Argument for Dismantling a Dubious Category (Princeton: Princeton University Press, 1996); K. L. King, What is Gnosticism? (Cambridge, Mass.: Harvard University Press, 2003); R. van den Broek, Gnostic Religion in Antiquity (Cambridge: Cambridge University Press, 2013). 37 See P. Schäfer, The Jewish Jesus: How Judaism and Christianity Shaped Each Other (Princeton: Princeton University Press, 2012), 203–13. 38 Schäfer, Jewish Jesus, 212. 39 See, for example, D. Boyarin, Border Lines: The Partition of Judaeo–Christianity (Divinations: Rereading Late Ancient Religion; Philadelphia: University of Pennsylvania Press, 2004); I. J. Yuval, Two Nations in Your Womb: Perceptions of Jews and Christians in Late Antiquity and the Middle Ages (trans. B. Harshav and J. Chipman; Berkeley: University of California Press, 2006); M. R. Niehoff, “Creatio ex Nihilo Theology in Genesis Rabbah in Light of Christian Exegesis,” HTR 99/1 (2006): 37–64; D. Milson, Art and Architecture of 240 Minov

While not denying the general plausibility of Schäfer’s hypothesis, I would like nevertheless to modify it along more minimalistic lines, present implicitly already in Schäfer’s own analysis of the motif of the angelic veneration of Adam in rabbinic sources. In his discussion of the genesis of this motif, Schäfer draws attention to the concept of the divine Word–Logos developed by Philo of Alexandria.40 At one point, Philo introduces the Logos as “God’s Firstborn, the Word, who holds the eldership among the angels; their ruler, as it were.”41 This description of the Logos is followed by a list of his names, which includes among others “the Man after His image” (ὁ κατ’ εἰκόνα ἄνθρωπος).42 In addition, Philo characterizes the Logos as “the eldest-born image of God” (θεοῦ γὰρ εἰκὼν λόγος ὁ πρεσβύτατος).43 As has been pointed out by Schäfer, this portrayal of the Logos is rooted in Philo’s conception of the two distinct Adams represented in the book of Genesis; i.e., the (heavenly) Adam of Gen 1:27 and the (earthly) Adam of Gen 2:7.44 Whereas the latter stands for humanity, the former represents (for Philo) the hypostasized divine Word. One can see from the passages quoted above that the notion of the angelic veneration of Adam might be developed very easily within the framework of Philo’s conception of the Logos; who, on the one hand, is identified with the biblical figure of Adam, and, on the other, is said to be superior to the angels. According to Schäfer, an elevated image of the Logos–Adam, similar to that of Philo, had a deep impact on the development of New Testament christology, which is reflected in such passages as 1 Cor 15:45–49 and Col 1:15–18; it later became a target for the rabbinic midrashim discussed above. However, the midrashic inversion of the angelic veneration motif could also legitimately be interpreted as reflecting a more general rabbinic tendency to marginalize the wide range of mediatorial figures, including the Logos, Enoch, Melchizedek, Metatron, Moses, and some others, who figure prominently in

the Synagogue in Late Antique Palestine: In the Shadow of the Church (AJEC 65; Leiden: Brill, 2007); M. Bar-Asher Siegal, Early Christian Monastic Literature and the Babylonian Talmud (Cambridge: Cambridge University Press, 2013). 40 See Schäfer, Jewish Jesus, 207–9. 41 De conf. ling. 146: τὸν πρωτόγονον αὐτοῦ λόγον, τὸν ἀγγέλων πρεσβύτατον, ὡς ἂν ἀρχάγγελον, πολυώνυμον ὑπάρχοντα; text and translation from F. H. Colson and G. H. Whitaker, Philo, with an English Translation (10 vols.; LCL; London: Heinemann; Cambridge, Mass.: Harvard University Press, 1929–1962), 4:88–89. 42 De conf. ling. 146; Colson and Whitaker, Philo, 4:90–91. 43 De conf. ling. 147; Colson and Whitaker, Philo, 4:90–91. 44 See Schäfer, Jewish Jesus, 207–9. Satan’s Refusal to Worship Adam 241 a number of pre- and nonrabbinic Jewish sources.45 A similar explanation of the rabbinic polemic against the veneration of Adam has been offered by Jarl Fossum, who suggests that these midrashic passages constitute examples of inner-Jewish polemical discourse, directed against “early Jewish mystical teaching about divine Glory, the heavenly Man, or against notions showing a rapprochement, even a confusion, between the heavenly Man and the earthly Adam.”46 As an example of such a Jewish tradition, one may bring forward the account of the elevation of Enoch found in Sefer Hekhalot, also known as 3 Enoch—an early Jewish mystical tractate written in Hebrew.47 In chapter 4 of this work, R. Ishmael asks the angel Metatron, identified also as Enoch, about that he bears. The answer given to the rabbi ,(נער) ”the peculiar name, “Youth is presented in the form of a narrative about Enoch’s ascension to the heavens, ;over the angels (לשר ולנגיד) ”during which God made him “a prince and a ruler although the angels themselves initially opposed this decision, they eventually ויצאו) accepted it, “went to meet” him, and “prostrated themselves before” him And because he, i.e., Enoch–Metatron, was younger than .(לקראתי והשתחוו לפני the angels, “a mere youth among them in days and months and years,” they would call him “Youth.”48 As has been noted by Anderson, the basic plot of Enoch’s elevation in Sefer Hekhalot is “almost identical” to the narratives under consideration from the Life and 2 Enoch.49 Without going into a discussion of the nature of the relationship between this narrative and these other two

45 See on this M. J. H. M. Poorthuis, “Enoch and Melchizedek in Judaism and Christianity: A Study in Intermediaries,” in Saints and Role Models in Judaism and Christianity (ed. J. Schwartz and M. J. H. M. Poorthuis; JCP 7; Leiden: Brill, 2004), 97–120; D. Boyarin, “The Parables of Enoch and the Foundation of the Rabbinic Sect: A Hypothesis,” in “The Words of a Wise Man’s Mouth are Gracious” (Qoh 10,12): Festschrift for Günter Stemberger on the Occasion of his 65th Birthday (ed. M. Perani; SJ 32; Berlin: de Gruyter, 2005), 53–72. 46 J. E. Fossum, “The Adorable Adam of the Mystics and the Rebuttals of the Rabbis,” in Geschichte–Tradition–Reflexion: Festschrift für Martin Hengel zum 70. Geburtstag. Band 1: Judentum (ed. H. Cancik, H. Lichtenberger and P. Schäfer; Tübingen: Mohr Siebeck, 1996), 529–39 (534). 47 For a recent discussion of this composition, see K. Herrmann, “Jewish Mysticism in Byzantium: The Transformation of Merkavah Mysticism in 3 Enoch,” in Hekhalot Literature in Context: Between Byzantium and Babylonia (ed. R. S. Boustan, M. Himmelfarb and P. Schäfer; TSAJ 153; Tübingen: Mohr Siebeck, 2013), 85–116. 48 Ed. P. Schäfer, Synopse zur Hekhalot-Literatur (TSAJ 2; Tübingen: Mohr Siebeck, 1981), 5–7, §§5–6; trans. P. S. Alexander, “3 (Hebrew Apocalypse of) Enoch,” OTP 1:223–315 (258–59). For a discussion of this story, see Anderson, “Exaltation of Adam,” 127–32; Orlov, Enoch–Metatron, 133–36. 49 Anderson, “Exaltation of Adam,” 131. 242 Minov texts, I would only point out that the reaction of the angels toward the elevated human being in Sefer Hekhalot is exactly of the kind against which the rabbinic midrashim wage their polemic. At this point I would like to emphasize, however, that although attempts to solve the problem of the possible polemical background of these midrashim are certainly legitimate, the whole question of their genesis and Sitz im Leben is in need of a more detailed investigation that also takes into consideration the inner dynamics of rabbinic anthropological thinking, especially regarding the place occupied by humankind in the universal hierarchy of created beings.50 Having established the antiquity and Jewish origin of the motif of the angelic veneration of Adam, I also want to note that other narrative elements found in the Life’s account of Satan’s fall are attested in ancient Jewish sources as separate motifs. For instance, Menahem Kister has highlighted several elements common to the account of Satan’s fall in the Life and the picture of Adam in Philo’s On the Creation of the World (77–78); among these are the discussion of the temporal posteriority of Adam and the problematization of the idea of Adam’s authority over all creatures versus his status vis-à-vis the angels.51 Moreover, the first notion is also attested in an early Byzantine Aramaic piyyuṭ for Passover, published by Sokoloff and Yahalom. The second part of this poem relates a dramatic dialogue between Moses and the personified figure of the Red Sea, who refuses to let the people of Israel pass through. Remarkable in this dialogue is that in order to prove his superiority over Moses, the Sea claims that he is “three days older” than Moses himself.52 This claim is based on the biblical cosmogonic account in the first chapter of Genesis, where the creation of the seas takes place on the third day, and the creation of Adam on the sixth day. As has been observed by Kister, in its structure this argument very much resembles Satan’s claim of superiority in the Life, also based on temporal reasons.53 Finally, in his contribution to this volume, Kister convincingly demonstrates, relying on the evidence of the Roman philosopher Celsus and the payyeṭan Yose ben Yose, that in ancient Judaism there existed a tradition

50 See, for instance, M. Kister, “Hellenistic Jewish Writers and Palestinian Traditions: Early and Late,” pp. 150–78 in this volume, at 156–63. 51 See M. Kister, “Jewish Aramaic Poems from Byzantine Palestine and Their Setting,” Tarbiz 76/1–2 (2006–2007): 105–84 (150–52) (in Hebrew). 52 ʾEzel Moshe, in M. Sokoloff and Y. Yahalom, eds., Jewish Palestinian Aramaic Poetry from Late Antiquity (Publications of the Israel Academy of Sciences and Humanities, Section of Humanities; Jerusalem: The Israel Academy of Sciences and Humanities, 1999), 84. This tradition is attested also in such medieval rabbinic works as Midrash Hallel and Midrash Wayosha; see on this point Kister, “Jewish Aramaic Poems,” 146. 53 Kister, “Jewish Aramaic Poems,” 151. Satan’s Refusal to Worship Adam 243 that God had granted to Adam dominion not only over material and mortal creatures, but over angelic forces as well.54 In light of all these considerations, I see no compelling reason to disagree with Jan Dochhorn, who has reexamined the story of Satan’s fall in the Life in the context of early Jewish tradition, and who has come to the conclusion that the Life’s account is “conceivable as originating in a Jewish milieu.”55 As we have seen, the core element around which the Life’s account crystallized, i.e., the motif of the angelic veneration of Adam, may be traced back to the prerabbinic stage in the development of Jewish speculation on the first man. Regardless of whether the author of the Life himself took the step that con- verted this originally Jewish motif into the key to the mystery of Satan’s fall, or whether he relied upon an already existing Jewish tradition, there is nothing in this development that compels us to regard it as distinctively Christian. In my opinion, the only element in the Life’s narrative that might be suspected as Christian is verse 15:3 of the Latin recension, where Satan responds to Michael’s second request to worship Adam by paraphrasing Isa 14:13–14.56 Jewish sources from the Second Temple period do not make an explicit connection between Satan and the figure of Helel ben Shachar in this verse; but early Christian sources take the connection for granted from the end of the second century on.57 Moreover, this paraphrastic response is absent not only from the Armenian and Georgian versions, but from several Latin witnesses of the Life, as well. Taking all these factors into account, it seems justified to regard this verse as a later Christian addition to the originally Jewish story of Satan’s fall in the Life.58

2 The Motif of Satan’s Fall in Syriac Christian Tradition

As has been noted by scholars, accounts of the fall of Satan similar to that in the Life are attested across a wide range of Christian literatures from late

54 Kister, “Hellenistic Jewish Writers and Palestinian Traditions,” 159–61. 55 Dochhorn, “The Motif of the Angels’ Fall,” 491. 56 See Pettorelli et al., Vita latina, 1:310–11; 2:784–85. 57 And perhaps earlier; note the singular reference to Satan’s fall “like lightning” in Luke 10:18. Cf. Tertullian, Adv. Marc. 5.11.11; 5.17.8; Hippolytus, Antichr. 17, 53; Comm. Dan. 4.12; Origen, De. princ. 1.5.4–5; C. Cels. 6.43–44; Hom. Num. 12.3. See also Kaestli, “Le mythe de la chute,” 344. 58 We find Isa 14:13–14 introduced into the story of Satan’s refusal to worship Adam in such early Christian apocrypha as the Questions of Bartholomew 4.53–55 (see below). 244 Minov antiquity and the Middle Ages, encompassing writings in Greek, Syriac, Coptic, Armenian, Slavonic, and several other languages. Due to the limitations of time and the large amount of material, it is impossible to offer here even a per- functory overview of this rich corpus.59 In what follows I shall limit myself to discussing the reception history of this tale among Syriac-speaking Christians. The reason for this choice is that the Syriac material illustrates well the main trends in the appropriation and transmission of this Jewish tradition among Christians more generally. The earliest unambiguous attestation in the Syriac language of the motif of Satan’s refusal to worship Adam comes from the composition known as the Cave of Treasures. Transmitted under the name of Ephrem the Syrian (fourth century),60 this work belongs to the category of “rewritten Bible”: it offers a retelling of sacred history from the first day of creation until Pentecost, based on the Hebrew Bible and the New Testament, but also drawing on patristic and apocryphal sources. It was most likely composed not earlier than the first decades of the sixth century, in the area of northern Mesopotamia that was under the control of the Sasanian Empire.61 The editor of the Syriac text of the Cave, Su-Min Ri, has divided all its textual witnesses into two recensions, the East Syrian and the West Syrian. The story of Satan’s fall is found in both recensions without significant differences, and thus should be considered an integral part of the original composition. In the following analysis, I will rely upon the British Museum Add. Ms. 25875, of the Eastern recension, which so far appears to be the best textual witness of the Cave.62

59 For the Coptic sources, see the good overview in Rosenstiehl, “La chute de l’Ange.” For the Slavonic sources, see A. Kulik and S. Minov, Biblical Pseudepigrapha in Slavonic Traditions (Oxford: Oxford University Press, [forthcoming]), ch. 1. 60 See the discussion below, pp. 250–51, and especially n. 78, on Ephrem’s potential knowledge of aspects of this tradition. 61 For the Syriac text and French translation, see S.-M. Ri, La Caverne des Trésors: Les deux recensions syriaques (2 vols.; CSCO 486–487; Scriptores Syri 207–208; Louvain: Peeters, 1987). For general information, see C. Leonhard, “Observations on the Date of the Syriac Cave of Treasures,” in The World of the Aramaeans III: Studies in Language and Literature in Honour of Paul-Eugène Dion (ed. P. M. M. Daviau, J. W. Wevers, and M. Weigl; JSOTSup 326; Sheffield: Sheffield Academic Press, 2001), 255–93. For a recent reexamination of the date and milieu of this work, see S. Minov, “Syriac Christian Identity in Late Sasanian Mesopotamia: The Cave of Treasures in Context” (Ph.D. diss., The Hebrew University of Jerusalem, 2013), 21–86. 62 I cite the text of this manuscript as it is appears in the critical apparatus of Ri’s edition. For a discussion of the textual tradition of the Cave, see Minov, “Syriac Christian Identity,” 21–31. Satan’s Refusal to Worship Adam 245

In the second chapter, the author of the Cave offers a long and vivid description of the creation of Adam. In this narrative he emphasizes that God had entrusted the first man with authority over all created beings, including angels. Thus, in Cav. Tr. 2.8–11, which relates the creation of Adam from the four elements, the author states that this was done in order that “through them (i.e., the elements) everything which is in the world should be subjugated to him” (ܐܡܠܥܒ ܬܝܐܕ ܡܕܡ ܠܟ ܗܠ ܕܒܥܬܫܢ ܢܘܗܒܕ). More explicitly, God included among the elements “the heat of fire” so “that all the fiery beings and the celestial hosts might be at his (i.e., Adam’s) service.”63 The narrative continues with a description of the glorious appearance of Adam, where the biblical notion of “image and likeness” is interpreted in terms of almost physical resemblance (Cav. Tr. 2.12–14). After he creates Adam and endows him with the triple authority of “king, priest and prophet,” God sets him on his own throne ̈ and gives him “dominion over all creatures” (ܐܬܝܖܒ ܢܝܗܠܟ ܠܥ ܐܗܠܐ ܗܛܠܫܐ) (Cav. Tr. 2.15–19).64 The right of Adam to rule the world is demonstrated by the ensuing episode, in which he gives names to animals and other living creatures, so that they will acknowledge his dominion and worship him. The angelic forces, who hear a heavenly voice publicly confirming the authority of Adam over everything created by God, worship Adam as well:

And all the wild beasts, and all the cattle, and the birds were gathered together before Adam. And they passed in front of Adam and he assigned names to them, and they bowed their heads before him. And everything in nature worshipped him, and submitted themselves to him. And the angels and the heavenly hosts heard the voice of God saying to him: “Adam, behold, I have made you king, and priest, and prophet, and lord, and head, and governor of everything which has been made and created. And they shall be in subjection to you, and they shall be yours and only yours. And I have given you authority over everything which I have created.” And when the angels heard this heavenly voice they all bowed the knee and worshipped him.65

̈ ̈ 63 ܢܘܘܗܢ ܗܢܪܕܘܥܒ ܐܬܘܠܝܚܘ ܐܢܖܘܢ ܢܘܗܠܟܕ :ܐܪܘܢ ܢܡ ܐܚܬܪܘ; ed. Ri, La Caverne des Trésors, 14–16; trans. E. A. W. Budge, The Book of the Cave of Treasures (London: The Religious Tract Society, 1927), 52 (here and elsewhere I have slightly modified Budge’s translation). 64 Ed. Ri, La Caverne des Trésors, 18; trans. Budge, The Book of the Cave, 53. ̈ ̈ ̈ 65 ܢܝܗܠ ܡܣܘ ܡܕܐ ܡܕܩ ܢܘܗܠܟ ܝܖܒܥܘ .ܡܕܐ ܡܕܩ ܐܬܚܖܦܘ ܐܪܝܥܒܘ ܐܬܘܝܚ ܢܝܗܠܟ ܝܫܢܟܬܐܘ ܐܟܐܠܡ̈ ܘܥܡܫܘ .ܗܠ ܘܕܒܥܬܫܐܘ ܐܢܝܟ̈ ܢܘܗܠܟ ܗܠ ܘܕܓܣܘ .ܢܝܗܝܫܖ̈ ܗܠ ܝܢܟܖܐܘ̈ .ܐܗܡܫ̈ ܐܫܪܘ ܐܪܡܘ ܐܝܒܢܘ ܐܢܗܟܘ ܐܟܠܡ ܟܬܕܒܥ ܐܗ ܡܕܐ .ܗܠ ܪܡܐ ܕܟ ܐܗܠܐܕ ܗܠܩ ܐܬܘܠܝܚܘ̈ 246 Minov

Immediately following, at the beginning of the next chapter (Cav. Tr. 3.1–4), the story of Satan’s fall is presented. According to this account, Satan is the only heavenly being who is not content with the exaltation of Adam and resents God’s decision. Driven by envy, he persuades the angels under his command not to bow to Adam:

And when the prince of the lower order of angels saw what greatness had been given to Adam, he became envious of him from that day, and he did not wish to worship him. And he said to his hosts: “You shall not worship him, and you shall not praise him with the angels. Rather, he ought to worship me, because I am fire and spirit; and not that I should worship a thing of dust, which has been fashioned of fine dust.” And as the rebel was nurturing these thoughts, he would not render obedience to God, and out of his own desire and free will he separated himself from God. And he and all his company were cast down and fell—their fall from heaven happened on the sixth day, at the second hour of the day.66

It is noteworthy that in his version of the myth of the fall of Satan, the author of the Cave combines two different explanations of the protagonist’s motivation; i.e., one based on envy and one based on pride. Thus, on the one hand the author points out that the reason for Satan’s refusal to bow down to Adam was his envy of Adam’s exalted status. On the other hand, the explanation that Satan himself gives to the fellow angels is based on the motif of pride, evok- ing Satan’s ontological superiority to Adam—since, as an angel, he is a creature of fire and spirit, he outclasses the first human, who is a mere creature of dust. By making the latter claim, Satan apparently misses the point made in Cav. Tr. 2.8–11, according to which the nature of the human being, created from the four elements, includes “fire” as well. To the best of my knowledge, no Syriac composition predating the Cave contains a similar explanation of the fall of Satan. In search of a possible source of

ܐܢܛܠܘܫ ܟܠ ܬܒܗܝܘ .ܟܝܕܘܚܠܒ ܢܘܘܗܢ ܟܠܝܕܘ ܇ܢܘܕܒܥܬܫܢ ܟܠܘ ܐܝܖܒܘ̈ ܐܕܝܒܥ̈ ܢܘܗܠܟܕ ܐܢܪܒܕܡܘ ̈ ܗܠ ܘܕܓܣܘ ܢܘܗܠܟ ܘܟܪܒ .ܐܠܩ ܬܪܒ ܐܕܗ ܐܟܐܠܡ ܘܥܡܫ ܕܟܘ .ܬܝܪܒܕ ܡܕܡ ܠܟ ܠܥ; ed. Ri, La Caverne des Trésors, 18–20; trans. Budge, The Book of the Cave, 53–54. ̇ ̇ 66 ܘܗ ܢܡ ܗܒ ܡܣܚ ܡܕܐܠ ܗܠ ܬܒܗܝܬܐ ܐܬܘܒܪ ܐܕܝܐܕ .ܐܝܬܚܬ ܐܢܗ ܐܡܓܬܕ ܐܫܝܪ ܐܙܚ ܕܟܘ ̈ .ܐܟܐܠܡ ܡܥ ܢܘܚܒܫܬ ܐܠܘ ܗܠ ܢܘܕܓܣܬ ܐܠ ܗܬܘܠܝܚܠ ܪܡܐܘ̣ .ܗܠ ܕܘܓܣܢܕ ܐܒܨ̣ ܐܠܘ .ܐܡܘܝ ܕܟܘ .ܐܚܝܚܕ ܢܡ ܠܒܓܬܐܕ ܐܪܦܥܠ ܕܘܓܣܐܕ ܐܢܐ ܘܠܘ ܐܚܘܪܘ ܐܪܘܢ ܝܬܝܐܕ ܝܠ ܕܘܓܣܢܕ ܩܕܙ ܘܗ .ܐܗܠܐ ܢܡ ܗܫܦܢ ܫܪܦ ܗܬܘܪܐܚܘ ܢܝܒܨܒ ܘܗܘ̣ .ܐܢܥܡܬܫܡ ܐܘܗ ܐܠܘ ܐܕܘܪܡ ܝܥܪܬܐ ܢܝܠܗ ̈ ̈ ܐܝܡܫ ܢܡ ܐܬܠܘܦܡ ܐܘܗ ܢܝܥܫ ܢܝܬܖܬܒ .ܐܬܫܕ ܐܡܘܝܒ ܗܡܓܬ ܗܠܟܘ ܘܗ̣ ܠܦܢܘ ܦܚܬܣܐܘ; ed. Ri, La Caverne des Trésors, 20–22; trans. Budge, The Book of the Cave, 55–57. Satan’s Refusal to Worship Adam 247 inspiration for the author of the Cave, let us turn, first, to the narrative of the Life. As we compare these two versions of Satan’s demotion, we will discover that, although the general plot underlying them is basically the same, there are several serious differences in the details. First of all, as has been noted by Gary Anderson, the Cave diverges from the Life in that the former relegates the whole story of the angels worshipping Adam, and Satan’s refusal, to the time when the animals are brought before Adam (Gen 2:19–20), while in the latter the episode is placed immediately after God animates Adam with his spirit (Gen 2:7).67 In addition, the Life features the archangel Michael as God’s messenger, who demands that Satan worship Adam, while in the Cave he is not mentioned at all. Moreover, in distinction from the Life, the author of the Cave puts great emphasis on the royal status of the newly created Adam. And, finally, whereas in the Life, Satan’s argument against worshipping Adam is based upon his own temporal priority, in the Cave’s account, he appeals to the difference between the materials from which they were each created; i.e., “fire” and “dust,” respectively.68 All these observations lead us to the conclusion that these two narratives of Satan’s fall are not related to one another through a straightforward process of textual borrowing. This conclusion goes against the suggestion made by Michael Eldridge that the Cave is dependent upon the actual text of the Life;69 it rather supports the verdict of Marinus de Jonge and Johannes Tromp that “the claim that the Cave of Treasures depends directly on the Life of Adam and Eve cannot be substantiated.”70 A more plausible explanation for the origin of the story of Satan’s fall in the Cave would be that the Syriac-speaking author came to know of this myth through one of its earlier Christian reworkings. There are several other Christian writings that might predate the Cave, where we find versions of the story of Satan’s downfall that includes his refusal to worship Adam. Among these may be mentioned the Greek Questions of Bartholomew and Apocalypse

67 Anderson, “The Exaltation of Adam,” 110–11. 68 In the formulation of Satan’s claim that he is created from “fire and spirit,” the author ̈ ̈ of the Cave could be relying upon Ps 104:4 (Peshitta: ܝܗܘܢܫܡܫܡܘ .ܚܘܪ ܝܗܘܟܐܠܡ ܕܒܥ ̈ ܐܕܩܝ ܪܘܢ). Cf. also the mention of “the Watchers of fire and spirit” (ܚܘܪܕܘ ܪܘܢܕ ܐܖܝܥ) in Ephrem’s Hymns on Paradise 6.24; ed. E. Beck, Des heiligen Ephraem des Syrers Hymnen de Paradiso und contra Julianum (CSCO 17; Scriptores Syri 78; Louvain: Peeters, 1957), 24. See also Yaakov Kaduri’s essay in this volume for a discussion of the exegesis of Ps 104:4 and its influence upon rabbinic and earlier notions concerning the nature of the angels. 69 See M. D. Eldridge, Dying Adam with his Multiethnic Family: Understanding the Greek Life of Adam and Eve (SVTP 16; Leiden: Brill, 2001), 24. 70 De Jonge and Tromp, Life of Adam and Eve, 87. 248 Minov of Sedrach, and the Coptic Encomium on Saint Michael the Archangel, ascribed to Theodosius of Alexandria.71 Of particular interest for us is the version of Satan’s fall that appears in the Questions of Bartholomew, a New Testament apocryphon that was produced in the second or third century. Here, in a scene where Bartholomew engages in a dialogue with Satan, the latter discloses to the apostle, among other things, the history of his own fall, relating it to the episode of the veneration of Adam by the angels:

And he (God) showed him (Adam) reverence for his own sake, because he was his image. And Michael also worshipped him. And when I (i.e., Satan) came from the ends of the world, Michael said to me: “Worship the image of God (προσκύνησον τὴν εἰκόνα τοῦ θεοῦ) which he has made in his own likeness.” But I said: “I am fire of fire (πῦρ ἐκ πυρός), I was the first angel to be formed (πρῶτος ἄγγελος πλασμένος ἤμην), and shall I worship clay and matter (πηλὸν καὶ ὕλην)?” And Michael said to me: “Worship, lest God be angry with you.” I answered: “God will not be angry with me, but I will set up my throne over against his throne, and shall be as he is.” Then God was angry with me and cast me down, after he had commanded the windows of heaven be opened.72

There are several elements that point in the direction of this narrative having been derived from the Life’s version of Satan’s fall, such as the role of the archangel Michael in the conflict and the claim by Satan that he had been created first. At the same time, however, this story also features an element that brings it close to the version of the Cave; namely, Satan’s appeal to the hierarchical difference between himself and Adam in their physical natures. This combination of the two different versions of Satan’s claim of superiority over Adam in the Questions strengthens the possibility that the author of the Cave was

71 The latter source is found (with translation) in E. A. W. Budge, Miscellaneous Coptic Texts in the Dialect of Upper Egypt (Coptic Texts 5; London: British Museum, 1915), 334–38, 904– 6. In the Apocalypse of Sedrach 5:1–2, the story of Satan’s fall appears, but the reason for his rebellion is not made explicit. 72 Questions of Bartholomew 4:53–55; ed. G. N. Bonwetsch, “Die apokryphen Fragen des Bartholomäus,” Nachrichten von der Königlichen Gesellschaft der Wissenschaften zu Göttingen. Philologisch-historische Klasse (1897): 1–42 (25–26); the translation used here is from New Testament Apocrypha (ed. W. Schneemelcher; trans. R. McL. Wilson; 2 vols.; rev. ed.; Cambridge: Clarke; Louisville: Westminster John Knox, 1991–1992), 1:549. Satan’s Refusal to Worship Adam 249 dependent for the myth of Satan’s fall on an intermediary Christian work that adopted and further reworked the story from the Life. In a recent discussion of the possible sources behind the account of the fall of Satan in the Cave, Alexander Toepel draws attention to the description of Satan’s conspiracy against Adam found in the works of Narsai, a fifth-century East Syrian poet;73 Toepel suggests that both Syriac writers relied upon the same source in their depiction of this event.74 Among the various instances in which Narsai evokes Satan’s enmity against humans, the most developed discussions of this theme are found in two mēmrē, i.e., metrical homilies: On the Making of Creatures (1.221–240); and On the Making of Adam and Eve, and on the Transgression of the Commandment (4.101–125).75 According to the Syriac poet, “envy” (ܐܡܣܚ) was the main driving force that caused Satan, who is presented as the “chief of the aerial realm” (ܪܐܐܕ ܐܫܪ), to plot against the first man. This envy was caused by the fact that God had endowed Adam with His own “image” (ܐܡܠܨ), which elevated the human being over the rest of the created world, making him “the king upon the earth” (ܐܥܪܐܒ ܐܟܠܡ) and “the lord over all that exists” (ܐܘܗܕ ܠܟ ܠܥ ܐܪܡ). The poet states explicitly on more than one occasion that because of this elevated status, and as a result of his “kinship” (ܐܬܘܢܝܚܐ) with both the material and immaterial realms, Adam’s dominion extended over the angelic forces as well. Like the rest of God’s creatures, the angels are submitted to “the yoke of toiling for him (i.e., Adam)” (ܐܪܝܢ ̈ ܗܠܡܥܕ). Entrusted with “the service for his life” (ܝܗܘܝܚ ܢܚܠܘܦ), they “were ̈ stirred up at his service gladly” (ܘܘܗ ܢܝܕܚ ܕܟ ܗܬܫܡܫܬܒ ܐܟܐܠܡ ܘܘܗ ܢܝܠܝܕܡ), while “praising (or carrying) him as a king” (ܐܟܠܡܠܕ ܟܝܐ ܗܠ ܢܝܚܝܙܡ). It is the divine act of elevating this lowly creature, made of “this contemptible fine dust of the earth” (ܐܥܪܐܕ ܐܚܝܚܕ ܐܪܦܥ ܐܛܝܫ ܐܢܗ), over Satan, who is “of spirit” (ܐܢܚܘܪ) and has other angels under his own command, that fuels Satan’s resentment and makes him refuse “to submit himself to the authority of love for Adam” (ܡܕܐܕ ܗܒܘܚܕ ܐܬܘܪܡܒ ܗܫܦܢ ܢܕܟ), and to begin to intrigue against the first man. One cannot deny that the portrayal of Satan’s fall in the Cave has several elements in common with the homilies of Narsai. Thus, as has been pointed out

73 See, on Narsai, A. Vööbus, History of the School of Nisibis (CSCO 266, Subsidia 26; Louvain: Secrétariat du CorpusSCO, 1965), 57–121; P. Gignoux, “Narsaï,” in Dictionnaire de spiritu- alité ascétique et mystique: doctrine et histoire (Paris: Beauchesne, 1982), 11:39–42. 74 A. Toepel, Die Adam- und Seth-Legenden im syrischen Buch der Schatzhöhle: Eine quellen- kritische Untersuchung (CSCO 618; Subsidia 119; Louvain: Peeters, 2006), 89. 75 Ed. P. Gignoux, Homélies de Narsaï sur la création (PO 34.3–4; Turnhout: Brepols, 1968), 540–41 and 616–17 respectively. In his analysis, Toepel relies upon the first account only. 250 Minov by Toepel, both authors establish a connection between the notion of Satan’s “envy” towards Adam and the motif of the angels’ subjection to the first man.76 To this, one might add that Satan’s denigration of Adam as a creature of “dust” and his self-aggrandizement as a “spiritual” being in Narsai’s homilies come very close to his explanation for the refusal to bow down to Adam in the Cave. Nevertheless, it must be stressed that there are also significant differences between these two accounts of the fall of Satan. Thus, contrary to the author of the Cave, Narsai draws no connection between Satan’s rebellion and the episode of the naming of animals. Even more important is that in distinction from the author of the Cave, Narsai makes no explicit mention of Satan’s refusal to commit a ritual act of obeisance before Adam. In Narsai’s account Satan resents the general subjection of angels to the human race, presented as an obligation to perform tasks of an unspecified kind on behalf of Adam. These considerations, I believe, make less plausible the possibility that the Cave and Narsai rely upon the same common source in their treatment of this episode. I would like instead to suggest an alternative explanation for the similarities in the treatment of Satan’s fall by these two authors. In my opinion, the portrayal of this event found in the works of Narsai reflects the standard interpretation of this issue that was current in the scholastic milieu of Edessa, to which this author belonged.77 At the cornerstone of this approach to the fall of Satan lies the idea that it took place on the sixth day of creation, as a result of his envy towards the newly created Adam. We find this notion already in the works of Ephrem the Syrian, the famous fourth-century poet and theologian, whose legacy had a significant impact on the tradition of biblical exegesis in the “school” of Edessa.78 In his Commentary on Genesis (2.32.1), Ephrem notes

76 Toepel, Die Adam- und Seth-Legenden, 90. 77 On the so-called “school of Edessa,” see H. J. W. Drijvers, “The School of Edessa: Greek Learning and Local Culture,” in Centres of Learning: Learning and Location in Pre-Modern Europe and the Near East (ed. H. J. W. Drijvers and A. A. MacDonald; Brill’s Studies in Intellectual History 61; Leiden: Brill, 1995), 49–59; A. H. Becker, Fear of God and the Beginning of Wisdom: The School of Nisibis and the Development of Scholastic Culture in Late Antique Mesopotamia (Divinations: Rereading Late Ancient Religion; Philadelphia: University of Pennsylvania Press, 2006), 64–68. 78 It has been suggested by Gary Anderson that already in the fourth century, Ephrem the Syrian might have been acquainted with a version of the story of Satan’s fall similar to that in the Life. See G. A. Anderson, “The Fall of Satan in the Thought of St. Ephrem and John Milton,” Hugoye: Journal of Syriac Studies 3/1 (2000): 3–27 (14–26). Yet, regardless of whether or not one agrees with Anderson’s reading of the sources, it should be emphasized that the corpus of Ephrem’s genuine writings provides no explicit references to this motif. Satan’s Refusal to Worship Adam 251 that until the sixth day Satan “was fair” (ܐܘܗ ܪܝܦܫ), and that it was on this same day that he “secretly became Satan” (ܐܝܣܟܒ ܐܘܗ ܐܢܛܣ).79 Ephrem mentions Satan’s envy towards Adam on several occasions in his madrāšē hymns.80 Concerning another idea shared by Narsai and the Cave—i.e., that the angels were subjected to the first man through the necessity of serving him—it appears that this motif was also a part the traditional Edessene exegetical repertoire. An early expression of the subordinated status of angels vis-à-vis humans is found in the works of Ephrem, in the second part of the Nisibene Hymns. This collection contains a series of dramatic dialogues, in which Death and Satan are the main dramatis personae. In one of these dialogues (Nis. Hym. 68.3–4), Death is reproached in the following words: “Adam was chosen and endowed with authority, and under his yoke you, Death, and the Evil One, your companion, were made slaves.” Death responds to that by boasting, “We pride ourselves in that the slaves have become masters; Death and his companion Satan have trampled upon Adam.”81 Slightly further along, in Nis. Hym. 68.7, Adam’s authority over all created beings is evoked again: “Tremble, O Death, before man, for even if he himself is a servant, the yoke of his lordship reigns over the creatures.”82 We find this notion made even more explicit in the interpretation of Satan’s fall by another representative of the Edessene “school,” Jacob of Sarug, a prominent West Syrian hymnographer and a younger contemporary of Narsai.83 In a mēmrā on the sixth day of creation, Jacob touches briefly upon the subject

79 Ed. R. M. Tonneau, Sancti Ephraem Syri in Genesim et in Exodum commentarii (2 vols.; CSCO 152–153, Scriptores Syri 71–72; Louvain: Durbecq, 1955), 1:44; trans. E. G. Mathews and J. P. Amar, St. Ephrem the Syrian. Selected Prose Works: Commentary on Genesis, Commentary on Exodus, Homily on Our Lord, Letter to Publius (FC 91; Washington, D.C.: The Catholic University of America Press, 1994), 121. 80 Cf. Eccl. 48.11.1; Ieiun. App. 1.1.1–4; Fid. 50.6.2; Nat. 21.15.4. For a discussion of this subject, see T. Kronholm, Motifs from Genesis 1–11 in the Genuine Hymns of Ephrem the Syrian, with Particular Reference to the Influence of Jewish Exegetical Traditions (Uppsala: Almqvist & Wiksell, 1978), 89–92. ̈ 81 ܢܪܗܒܘܫܕ .ܢܘܬܝܘܗ ܐܕܒܥ ܟܬܢܟ ܐܫܝܒܘ ܐܬܘܡ ܬܢܐ ܗܪܝܢ ܬܝܚܬܘ ܛܠܫܡܘ ܡܕܐ ܐܘܗ ܐܒܓ ̈ ̈ ܡܕܐܠ ܝܗܘܫܕ ܗܬܢܟ ܐܢܛܣܘ ܐܛܘܡ ܐܝܖܡ ܘܘܗ ܐܕܒܥ ܐܗܕ ܝܗ; ed. E. Beck, Des heiligen Ephraem des Syrers Carmina Nisibena, II (2 vols.; CSCO 240–241, Syr. 102–103; Louvain: Secrétariat du CorpusSCO, 1963), 1:108. ̈ 82 ܐܬܝܖܒ ܠܥ ܟܠܡܡ ܗܬܘܪܡܕ ܐܪܝܢ ܘܗ ܐܕܒܥ ܕܟ ܦܐܕ ܐܫܢܪܒ ܢܡ ܐܬܘܡ ܥܘܙ; ed. Beck, Des heiligen Ephraem des Syrers Carmina Nisibena, II, 1:109. 83 See F. Graffin, “Jacques de Saroug,” in Dictionnaire de spiritualité ascétique et mystique: doctrine et histoire (Paris: Beauchesne, 1982), 8:56–60; C. Lange, “Jakob von Sarug, † 521,” in Syrische Kirchenväter (ed. W. Klein; Urban-Taschenbücher 587; Stuttgart: Kohlhammer, 2004), 217–27. 252 Minov of Satan’s fall.84 He follows the general lines of the “Edessene interpretation” of this event by connecting it with the story of the creation of Adam; he states that Satan’s downfall was caused by his envy of the “greatness” (ܐܬܘܒܪ) and “honor” (ܐܪܩܝܐ) that were given to Adam. As he develops the subject further, Jacob adds that a particular reason for Satan’s resentment against Adam was that Satan himself had originally been entrusted with the authority over this “tenebrous world” (ܟܘܫܚܕ ܐܡܠܥ); as a result of Adam’s “superiority” (ܐܬܘܢܫܪ), Satan was downgraded to a subordinate status, which entailed performance of the “service that he (i.e., Adam) might need” (ܗܠ ܐܝܥܒܬܡܕ ܐܬܫܡܫܬ). Later on, we come across a similar interpretation of Satan’s fall in the chronicle Ktābā d-rēš mellē by Yoḥannān bar Penkāyē, a seventh-century East Syrian historiographer.85 The first book of his chronicle deals with the biblical narratives of the creation and fall of Adam and Eve. There, Bar Penkāyē relates that when Satan, who originally had dominion over the aerial realm, saw Adam become the “heir of all creation,” he was overtaken by an “irruption of jealousy,” and said to himself: “If this is the heir, whom even I have to serve, then I shall kill him and snatch from him his inheritance.”86 A similar explanation of the origins of Satan’s animosity towards Adam is found in the so-called Diyarbakir Commentary. This biblical commentary, which covers the books of Genesis and Exodus, was composed by an unknown East Syrian exegete sometime in the eighth century.87 According to this composition, the spiritual beings receive their respective duty assignments at the same time that Adam receives

84 Ed. P. Bedjan, Homiliae selectae Mar-Jacobi Sarugensis (5 vols.; Leipzig: Harrassowitz, 1905–1910), 3:126–27. For a French translation, see B. M. B. Sony, “Hymne sur la création de l’homme de l’Hexaméron de Jacques de Saroug,” Parole de l’Orient 11 (1983): 167–200 (196–97). 85 See on this writer and his chronicle, P. Bruns, “Von Adam und Eva bis Mohammed— Beobachtungen zur syrischen Chronik des Johannes bar Penkaye,” OrChr 87 (2003): 47–64. ̇ 86 ܡܕܐܠ ܝܗܝܙܚ ܕܟ ܗܠܝܕ ܤܝܣܟܛ ܗܠܟ ܡܥ ܪܐܐܕ ܐܬܘܢܫܝܪ ܗܠ ܬܒܗܝܬܐܕ ܐܕܘܪܡ ܪܝܓ ܐܨܪܩܠܟܐ ܐܠܘ ܗܠܕ ܐܬܪܝ ܘܢܗ ܢܝܐܕ ܘܗ̣ ܦܐ ܗܒܠܒ ܪܡܐܘ .ܗܠܒܘܩܠ ܐܢܢܛܕ ܐܦܐܚ ܗܠܩܫ ܐܬܝܪܒ ܗܠܟܕ̇ ܐܬܪܝ ܗܬܘܬܪܝ ܗܢܡ ܐܕܥܐܘ .ܗܘܝܠܛܩܐ ܐܢܐ ܫܡܫܐܕ ܝܠ ܦܐ. This part of John’s chronicle has not yet been published. I quote the text according to the manuscript Mingana Syr. 179, f. 4v. For a description of this manuscript, see A. Mingana, Catalogue of the Mingana Collection of Manuscripts now in the Possession of the Trustees of the Woodbrooke Settlement, Selly Oak, Birmingham (3 vols.; Woodbrooke Catalogues 1–3; Cambridge: Heffer, 1933–1939), 1:395–96. The translation is my own. 87 For a critical edition of the Syriac text and a comprehensive introduction to this work, see L. van Rompay, Le commentaire sur Genèse–Exode 9,32 du manuscrit (olim) Diyarbakir 22 (2 vols.; CSCO 483–484; Scriptores Syri 205–206; Louvain: Peeters, 1986). Satan’s Refusal to Worship Adam 253 the commandment to keep the Garden of Eden (Gen 2:15). Satan, who is one among them, is entrusted with governing the air. Yet, Satan falls because he becomes envious of Adam, endowed with the “image” of God, and is ashamed that he has to “serve” him.88 This evidence suggests that at least from the fifth century on there existed in Syriac-speaking biblical exegesis a well-established tradition of interpreting the fall of Satan as connected to the creation and elevation of the first man. Developed most likely in the milieu connected with the city of Edessa, this tradition also employed the motif of Satan’s refusal to acknowledge the subordinate status of the angels vis-à-vis Adam. This subordination, however, was primarily expressed in terms of the angels’ general obligation to “serve” humanity. The author of the Cave was an heir to this tradition, and that explains the similarity between his narrative and the poems of Narsai. However, he also developed this tradition one step further, by introducing the apocryphal motif of Satan’s refusal to acknowledge the superiority of the first man by participat- ing in an act of obeisance.

As a result of the considerable influence exercised by the Cave of Treasures on subsequent Syriac tradition, its account of Satan’s fall became an important part of the exegetical repertoire of Syriac-speaking Christians generally. Thus, we find it used by various authors from each of the two main factions into which Syriac Christianity divided from the sixth century forward: West Syrian, or “Jacobite,” Christianity; and East Syrian, or “Nestorian,” Christianity. For instance, this version of the story appears in the so-called Chronicle up to the Year 1234. Composed by an unknown West Syrian author, this chrono- graphic work is universal in its perspective and covers the whole course of the history of the world, from the creation until the thirteenth century.89 In the first chapter, the author of the Chronicle presents, among other things, a detailed account of the story of Adam and Eve. This narrative of the life of the first couple includes the story of the adoration of Adam by the angels, as well as Satan’s refusal to participate and subsequent fall.90 The story of Satan in the

̇ ̈ ̈ ̇ 88 ܐܬܐ ܐܛܝܫ ܐܫܚܖܠܘ ܐܬܚܪܦܠܘ ܐܪܝܥܒܠܘ ܐܫܢܝܢܒܠܕ ܝܗܒܘ ܗܠܝܕ ܐܬܘܒܪܒ ܝܩܒܬܐ ܪܝܓ ܕܟ ܠܦܢ ܐܢܕܥܒ ܗܒ .ܝܗܘܝܚܠܦܢܕ ܬܡܐܟ ܪܥܡܛܨܐܘ .ܝܪܩܬܐܕ ܐܫܢܪܒܒ ܡܣܚ ܒܘܬܘ .ܚܠܦ.; ed. van Rompay, Le commentaire, 1:37. 89 On this work, see D. Weltecke, “Les trois grandes chroniques syro-orthodoxes des XIIe et XIIIe siècles,” in L’historiographie syriaque (ed. M. Debié; Études syriaques 6; Paris: Geuthner, 2009), 107–35 (118–23). 90 For the Syriac text, see J. B. Chabot, Anonymi auctoris Chronicon ad annum Christi 1234 pertinens (3 vols.; CSCO 81, 82, 109; CSCO Scriptores Syri 36, 37, 56; Paris: Typographeo Reipublicae, 1916, 1920, 1937), 1:29. 254 Minov

Chronicle closely resembles that in the Cave. Numerous coincidences of wording and style between the two accounts prove beyond doubt that the author of the Chronicle used the Cave as one of his main literary sources for this part of his history. The influence of the Cave on representations of Satan’s fall in Syriac sources may be discovered not only in historiographic compositions, but also in works that belong to other genres, such as liturgical poetry. Thus, we find the story evoked by Yōḥannān bar Zōʿbī, an East Syrian theologian and grammarian active during the late twelfth and early thirteenth centuries.91 Bar Zōʿbī makes explicit use of this tradition in the prologue to his versified commentary on the liturgy, the Interpretation of the Mysteries, where he offers a concise overview of the history of God’s providential care for humanity, from the first days of creation until the coming of Jesus. After a vivid description of the creation of Adam in the image and likeness of God, the poet relates that God positioned this creature of his in order to test the reasoning power of the angelic forces.92 Whereas the majority of the angels pass this test successfully by giving due honor to the divine image, Satan does not. As in the Chronicle of Bar Penkāyē and the Diyarbakir Commentary, Satan is characterized as the ruler of the air; he refuses to worship Adam and, as a result, loses his place in heaven:

Like a furnace the Existing One set up him, his creature, In order to test the power of discernment among the heavenly ones. The love of the heavenly ones was discovered to be (like) pure gold, As they honored with love the image of the Hidden Being.

The gold of the chief of the air was discovered to be (like) hateful rust, As he refused and did not subject himself to the worship. Like lightning the hater of humanity fell from his rank, And like a serpent he crawled into the depth of the earth.93

91 See on this writer H. G. B. Teule, “Yoḥannān bar Zoʿbī,” in Christian–Muslim Relations: A Bibliographical History, Volume 4 (1200‒1350) (ed. D. R. Thomas and A. Mallett; HCMR 17; Leiden: Brill, 2012), 365–67; N. N. Seleznyov, Yōḥannān Вar Zō‘bī and his “Explanation of the Mysteries”: Critical Text, Russian Translation from Syriac, and Investigation (Moscow: Russian State University for the Humanities, 2014) [in Russian], 5–16. 92 Compare the passage from 2 Enoch discussed above (pp. 235–36). ̈ ̇ 93 ܐܝܟܕ ܐܒܗܕ .ܐܢܝܡܫܕ ܢܘܗܬܘܫܘܪܦܕ ܐܪܨܝ ܪܘܚܒܢܕ ܆ܗܬܘܠܝܒܓܠ ܐܝܬܝܐ ܗܡܣ ܐܪܘܟ ܬܘܡܕܒ ܚܟܬܫܐ ܐܬܝܢܣ ܐܬܚܘܫ .ܐܙܝܢܓ ܐܝܬܝܐܕ ܗܡܠܨ ܘܪܩܝ ܐܒܘܚܒܕ ܝܗܒ̇ ܆ܐܢܝܡܫܕ̈ ܐܒܘܚ ܚܟܬܫܐ ܗܓܪܕ ܢܡ ܠܦܢ ܐܩܪܒ ܬܘܡܕܒ .ܘܫܡܫܡܠ ܕܒܥܬܫܐ ܐܠܘ ܐܘܗ ܝܨܥܕ ܝܗܒ̇ ܆ܪܐܐܕ ܐܫܪܕ ܐܒܗܕ ܐܝܘܚ ܬܘܡܕܒ ܐܥܪܐܕ ܐܩܡܘܥܒ ܕܠܚ ܐܘܗܘ ܆ܐܫܢܐܕ ܐܐܢܣ.; ed. Seleznyov, Yōḥannān Вar Zō‘bī, 34–36. I am grateful to Nikolai Seleznyov for bringing this reference to my attention. Satan’s Refusal to Worship Adam 255

The story of the fall of Satan is employed in a context similar to that of Bar Zōʿbī by another East Syrian poet, Joseph of Telkepe, who lived during the seventeenth century and wrote in a Neo-Aramaic dialect of Syriac. In the prologue to one of his poems, On Revealed Truth (17–19), where he deals briefly with the creation of the world, the author touches also upon the matter of the angelic hierarchy and the story of Adam’s life. The representation of the conflict between Adam and the demonic forces and the explanation of the reason for their enmity towards the humans closely resemble the myth of Satan’s fall in the Cave:

He created the world with this, His Word and He called Adam His image. He ordered that which is on earth and in the height to greet Adam.

The angels greeted him. He brought the animals before him and he gave names to all of them. But those devils did not assent to Him.

They did not assent and they did not accept His order. And they fell from their positions because of that. They hated him and they devised evil against him. They sought a pretext against him.94

Thus, as we have seen, the story of the fall of Satan is found across multiple genres of Syriac literature. In the representations we have thus far surveyed, the details of this tradition remain remarkably stable and faithful to the Cave’s recrafting of the earlier tradition. However, as we will see in the next section, one strand of this transmitted tradition underwent a more radical transformation, under the influence of Islam.

94 .ܐܡܘܪܒ ܬܝܐܕ ܐܡܘ ܐܥܪܐܒ ܬܝܐܕ ܐܡ .ܐܡܠܨ ܗܝܝܕ ܐܠܝܪܩ ܡܕܐܘ .ܐܡܠܥ ܐܠܪܒ ܗܬܝܢܬ ܕܒ ̈ ̈ ̈ ܐܢܡܫ ܢܝܗܝܠܘܟܘ .ܐܠܝܬܘܡ ܗܬܘܠ ܐܬܘܝܚܘ .ܐܠܝܠܪܕ ܐܡܠܫ ܐܟܐܠܡ ܀ܐܡܠܫ ܡܕܐܠ ܝܪܕܕ ܐܠܕܩܦ ܢܝܗܝܓܖܕܡ̈ ܒܒܣ ܕܡܘ .ܝܠܒܩ ܐܠ ܗܡܟܘܚܘ ܝܠܐܒܬ ܐܠܘ ܀ܐܠܝܠܐܒܬ ܐܠ ܐܢܛܣ̈ ܢܐܘ .ܐܠܝܪܩ ܀ܝܠܓ ܗܠܐ ܐܬܫܘܚ ܐܕܟܠܘ .ܝܠܫܟ ܗܠܐ ܐܬܫܝܒܘ ܝܠܢܣܘ .ܝܠܦܢ; A. Mengozzi, Israel of Alqosh and Joseph of Telkepe: A Story in a Truthful Language. Religious Poems in Vernacular Syriac (North Iraq, 17th Century) (2 vols.; CSCO 589–590; Scriptores Syri 230–231; Louvain: Peeters, 2002), 1:56 (Syr.), 2:170–71 (translation). 256 Minov

3 Fall out of Favor: Transformations of Syriac Christian Traditions in a Muslim Milieu

It has been demonstrated, on the one hand, that the legend of Satan’s refusal to worship Adam appeared in Syriac Christian tradition for the first time during late antiquity and continued to enjoy a certain vogue during the medieval period, when the prevailing majority of Syriac-speaking Christians lived under Islamic rule. On the other hand, this period also saw new developments in the fortunes of this legend: it was marginalized or even completely rejected in certain contexts as a result of the renegotiation of Syriac Christian identity in the new socio-cultural context of Muslim domination. The most important factor that conditioned such reversals in the attitude to this interpretation was the prominent position that it gained in Muslim tradition. The story of Satan’s fall as a result of his refusal to worship Adam was incorporated in the foundational text of the new Abrahamic religion, the Qurʾān itself. The sūra Ṣād presents the following account of the fall of Iblīs (the name generally employed by Muslim authors to identify the Devil of Judeo-Christian tradition):95

Your Lord said to the angels, “I will create a man from clay. When I have shaped him and breathed from My Spirit into him, bow down before him.” The angels all bowed down together, but not Iblis, who was too proud. He became a rebel. God said, “Iblis, what prevents you from bowing down to the man I have made with My own hands? Are you too high and mighty?” Iblis said, “I am better than him: You made me from fire, and him from clay.” “Get out of here! You are rejected: My rejection will follow you till the Day of Judgment!”96

A detailed discussion of this and other Qurʾānic accounts of the conflict between Iblīs and Adam, including the problem of ultimate origin of these accounts, lies beyond the scope of the present investigation.97 It is impossible, however, not to notice the basic similarity between the narrative of the Qurʾān

95 For a recent discussion of its etymology and origins, see J. P. Monferrer-Sala, “One More Time on the Arabized Nominal Form Iblīs,” Studia Orientalia 112 (2012): 55–70. 96 Qurʾān 38:71–78; trans. M. A. Abdel Haleem, The Qurʾān (Oxford World Classics; Oxford: Oxford University Press, 2004), 293. Cf. also sūras 2:30–36; 7:10–19; 15:26–35; 17:61–65; 18:50; 20:116–17. 97 See E. Beck, “Iblis und Mensch, Satan und Adam: Der Werdegang einer koranischen Erzählung,” Le Muséon 89 (1976): 195–244. Satan’s Refusal to Worship Adam 257 and that of the Cave. In both cases, Satan falls because he refuses to bow down to the newly created Adam, giving as the reason his ontological superiority over the human. In addition, the Qurʾān, like the Cave, connects the incident of Satan’s fall with the scriptural account of Adam giving names to creatures (cf. Qurʾān 2:30–34). In light of these observations, I tend to agree with those scholars who suggest that the Qurʾānic story of Iblīs’s fall represents an instance of the considerable influence exercised by Syriac Christianity upon the formative text of Islam.98 What is of foremost importance for our subject, however, is that the Qurʾānic interpretation of the fall of Iblīs became standard for the subsequent Muslim tradition of demonology. It is well attested in a wide range of Arabic historiographic compositions, tafsīr commentaries on the Qurʾān, and collections of ḥadīth, as well as works of other genres.99 The popularity of the Qurʾānic legend of Satan’s fall had a direct bearing on the fortunes of this story among Christians and other religious minorities in the Islamic world. The almost canonical status among Muslims of the story of Satan’s refusal to worship Adam did not go unnoticed by the Christians who lived under their rule. One comes upon various examples of the treatment of this tradition by Syriac Christian intellectuals in the context of Christian polemic against

98 See most recently G. S. Reynolds, “Redeeming the Adam of the Qurʾān,” in Ara- bische Christen: Christen in Arabien (ed. D. Kreikenbom, F.-C. Muth, and J. Thielmann; Nordostafrikanisch/Westasiatische Studien 6; Frankfurt am Main: Lang, 2007), 71–83. For a general discussion of the Syriac background of the Qurʾān, see I. Shahîd, “Islam and Oriens Christianus: Makka 610–622 AD,” in The Encounter of Eastern Christianity with Early Islam (ed. E. Grypeou, M. Swanson, and D. R. Thomas; HCMR 5; Leiden: Brill, 2006), 9–31; S. K. Samir, “The Theological Christian Influence on the Qurʾān: A Reflection,” in The Qurʾān in its Historical Context (ed. G. S. Reynolds; Routledge Studies in the Qurʾān; Lon- don: Routledge, 2007), 141–62; J. Witztum, “The Syriac Milieu of the Quran: The Recast- ing of Biblical Narratives” (Ph.D. diss., Princeton University, 2011). An attempt by Wilson Bishai (“A Possible Coptic Source for a Qurʾānic Text,” JAOS 91/1 [1971]: 125–28) to derive the origins of the Qurʾānic myth of Satan’s fall from Coptic Christian sources, such as the Encomium on Saint Michael the Archangel, attributed to Theodosius of Alexandria, can hardly be accepted as satisfactory. Bishai’s attempt fails to do justice to the complicated problem of the source-criticism of the Qurʾān because it unjustifiably privileges one narrative element, i.e., the dialogue between God and Satan, over others. 99 For examples, see G. Calasso, “Intervento di Iblīs nella creazione dell’uomo: L’ambivalente figura del ‘nemico’ nelle tradizioni islamiche,” Rivista degli Studi Orientali 45/1–2 (1970): 71–90; P. J. Awn, Satan’s Tragedy and Redemption: Iblīs in Sufi Psychology (SHR 44; Leiden: Brill, 1983), 33–40; M. Kister, “Ādam: A Study of Some Legends of Tafsīr and Ḥadīṯ Literature,” IOS 13 (1993): 113–74 (135–42); G. Schoeler, “Iblis in the Poems of Abū Nuwās,” ZDMG 151/1 (2001): 43–62 (44–46). 258 Minov

Islam—where it was either perceived as problematic and marginalized; or, on the contrary, integrated into Christian arguments aimed at subverting claims made by Muslims. It should be noted at this point that the earliest example of negative reaction to the story of Satan’s refusal to worship Adam, in the context of Christian polemic against Islam, comes from the Greek-speaking milieu and belongs to Anastasius of Sinai, a seventh-century theologian. In his book of Questions and Answers (#80), Anastasius tackles the question of why some people say that “Satan fell away because of his not paying homage to Adam,” by asserting that “such silly myths belong to the Pagans and Arabs” (Ἑλλήνων καὶ Ἀράβων εἰσὶν οἱ τοιοῦτοι μάταιοι μῦθοι); he offers instead an orthodox version of Satan’s fall, based on the motif of his pride vis-à-vis God and supported by the scriptural authority of Ezekiel 28.100 A relatively early instance of Syrian Christian use of the story of Satan’s fall for the purposes of polemic against Islam comes from the works of Timothy I (727–823), one of the most prominent figures in the history of the East Syrian church.101 He served as the patriarch of the Church of the East during the last decades of the eighth and first decades of the ninth century, first in Seleucia- Ctesiphon and afterwards in Baghdad. Of particular interest for us is one of his letters, number 34, addressed to “the priests and the believers in the cities of Baṣrah and Huballat,” which offers an extensive exposition of the doctrines of the Christian faith. As has been demonstrated by Thomas Hurst, this letter together with letters 35, 36, and 40 form a distinct group, dealing primarily with anti-Muslim polemic, within the patriarch’s epistolary legacy.102 The main

100 Ed. M. Richard and J. A. Munitiz, Anastasii Sinaitae Quaestiones et Responsiones (CCSG 59; Turnhout: Brepols, 2006), 131; trans. J. A. Munitiz, Anastasios of Sinai: Questions and Answers (Corpus Christianorum in Translation 7; Turnhout: Brepols, 2011), 195. For general information on Anastasius and his polemic against Islam, see A. Binggeli, “Anastasius of Sinai,” in Christian–Muslim Relations: A Bibliographical History, Volume 1 (600‒900) (ed. D. R. Thomas and B. H. Roggema; HCMR 11; Leiden: Brill, 2009), 193–202. 101 On this figure, see B. H. Roggema and M. Heimgartner, “Timothy I,” in Thomas and Roggema, Christian–Muslim Relations 1, 515–31; V. Berti, Vita e studi di Timoteo I, Patriarca cristiano di Baghdad: Ricerche sull’epistolario e sulle fonti contigue (Cahiers de Studia Iranica 41; Chrétiens en terre d’Iran 3; Paris: Association pour l’avancement des études iraniennes, 2009). 102 See T. R. Hurst, “The Syriac Letters of Timothy I (727–823): A Study in Christian–Muslim Controversy” (Ph.D. diss., The Catholic University of America, 1986), 32–68; idem, “The Epistle-Treatise: An Apologetic Vehicle. Letter 34 of Timothy I,” in IV Symposium Syriacum, 1984: Literary Genres in Syriac Literature (ed. H. J. W. Drijvers et al.; OrChrAn 229; Rome: Pontificium Institutum Studiorum Orientalium, 1987), 367–82. Satan’s Refusal to Worship Adam 259 thrust of letter 34 is to demonstrate the superiority of the Christian doctrine of Jesus as the Son of God against the Muslim understanding of Jesus as a mere human. The Muslim doctrine is based on the expression “the servant of God” (ʿabd Allāh), found in the Qurʾān as a self-description of Jesus. To refute this doctrine, Timothy brings forward a number of arguments, one of which is based on the story of Satan’s refusal to worship Adam. In this passage, addressing his rhe- torically constructed Muslim interlocutor, Timothy draws a parallel between the refusal of some angels to worship Adam and the denial by Muslims of Jesus’s divine nature. The connection between the two parts of this analogy is established through the New Testament notion of Adam as the “type” of Christ (cf. Rom 5:14), reformulated by Timothy as “the image of the Messiah”:

For if there were among the angels those who did not worship Adam, according to what you say, the one who was the image of the Messiah, as the divine scripture says, they became resisters, as you say, and they were condemned. How much more, then, will you be condemned, since you do not worship God in the clothing of your own body?103

What is remarkable about this piece of anti-Muslim polemic is that Timothy is trying to subvert the objectionable Muslim doctrine from within. He does that by exploring the possibility of tension between the Muslim view of Jesus as merely human and the Qurʾānic tradition of the fall of Iblīs. It is not clear, however, from this relatively short passage, whether the patriarch considered the Muslim tradition of Iblīs’s fall to be ultimately incompatible with the Christian doctrine. In fact, from the way he handles this tradition by reinterpreting it in the context of Adam–Christ typology, one may conclude that there was no inherent obstacle that would prevent Christians from holding to it. A more or less contemporary example of the polemical handling of the Qurʾānic myth of Iblīs’s fall by Christians in Syria is found in the writings of the famous Arab Christian theologian Theodore Abū Qurrah (ca 750–ca. 825), who served as the Melkite bishop of the city of Ḥarrān.104 In chapter 9 of his

̇ ̈ 103 ܐܚܝܫܡܕ ܗܡܠܨ ܝܗܘܬܝܐܕ ܘܗ :ܟܬܠܡ ܟܝܐ ܡܕܐܠ ܘܕܓܣ ܐܠܕ ܢܝܠܝܐ :ܐܟܐܠܡ ܢܡ̣ ܪܝܓ ܢܐ ܬܢܐ ܐܪܐ ܐܡܟ .ܘܒܝܚܬܐܘ ܬܢܐ ܪܡܐܕ̇ ܟܝܐ ܐܝܨܥ̈ ܢܡ̇ ܘܘܗ :ܐܝܗܠܐ ܐܒܬܟ ܪܡܐܕ̇ ܟܝܐ ̇ ܟܪܓܦ ܫܘܒܠܒ ܐܗܠܐܠ ܬܢܐ ܕܓܣ ܐܠܕ :ܬܢܐ ܒܝܚܬܡ ܬܝܐܪܝܬܝ.; ed. O. Braun, Timothei patriarchae I epistulae (2 vols.; CSCO 74, 75; Scriptores Syri 30, 31; Paris Typographeo Reipublicae, 1914–1915), 1:201. 104 See on Abū Qurrah, J. C. Lamoreaux, “Theodore Abū Qurra,” in Thomas and Roggema, Christian–Muslim Relations 1, 439–91. 260 Minov treatise, On the Veneration of the Holy Icons, written in Arabic at the request of a certain Yannah (who was an official of the “Church of the Image of Christ” in Edessa), Abū Qurrah defends the controversial Christian devotional practice of prostration (or bowing down) in front of sacred images; his arguments are aimed at Jews and Muslims, in whose eyes this practice was tantamount to idolatry.105 Whereas during most of this chapter, the Christian apologete develops his argumentation vis-à-vis an imaginary Jewish interlocutor, at one point he also rebuffs supposedly Muslim objections to this practice. Abū Qurrah wants to argue that the “act of prostration” (suǧūd) does not necessarily always implies “worship” (ʿibāda), which is appropriate to God alone, but might be legitimately used to express “honor” (karāma) toward sacred objects and clergy. He thus points out at what he perceives to be an internal contradiction in the Muslim position: on the one hand, the Muslims “mock the Christians for their practice of making prostration to the icons and to people” and “maintain that making the act of prostration is worship”; while on the other hand, their own holy scripture presents God himself as authorizing this same action in the story about the refusal of Iblīs to prostrate himself before Adam.106 In this way,

105 On the complicated question of possible connections between Abū Qurrah’s defense of icons and the contemporary iconoclastic controversy in the Byzantine empire, see S. H. Griffith, “Theodore Abū Qurrah’s Arabic Tract on the Christian Practice of Venerat- ing Images,” JAOS 105/1 (1985): 53–73 (71–73). 106 Ed. J. P. Arendzen, Theodori Abu Ḳurra de Cultu imaginum libellus e codice arabico nunc primum editus latine versus illustratus (Bonn: Drobnig, 1897), 17; trans. S. H. Griffith, A Treatise on the Veneration of the Holy Icons Written in Arabic by Theodore Abû Qurrah, Bishop of Harrân (c. 755–c. 830 AD) (Eastern Christian Texts in Translation 1; Louvain: Peeters, 1997), 52. Abū Qurrah’s distinction between ʿibāda and karāma corresponds to the distinction between λατρεία, “worship,” and προσκύνησις, “veneration,” used by such Greek-speaking defenders of the cult of icons as John of Damascus and Theodore of Studios. See K. Parry, Depicting the Word: Byzantine Iconophile Thought of the Eighth and Ninth Centuries (The Medieval Mediterranean 12; Leiden: Brill, 1996); A. Giakalis, Images of the Divine: The Theology of Icons at the Seventh Ecumenical Council (Studies in the History of Christian Traditions 122; Leiden: Brill, 2005). On the pertinence of the polemic against iconoclasm for Christians living under Muslim rule, see S. H. Griffith, “Images, Islam, and Christian Icons: A Moment in the Christian/Muslim Encounter in Early Islamic Times,” in La Syrie de Byzance a l’Islam, VIIe–VIIIe siècles: Actes du Colloque international, Lyon– Maison de l’Orient Méditerranéen, Paris—Institut du Monde Arabe, 11–15 Septembre 1990 (ed. P. Canivet and J.-P. Rey-Coquais; Publications de l’Institut français de Damas 137; Damascus: Institut Français de Damas, 1992), 121–38; idem, “Crosses, Icons and the Image of Christ in Edessa: The Place of Iconophobia in the Christian–Muslim Controversies of Early Islamic Times,” in Transformations of Late Antiquity: Essays for Peter Brown (ed. P. Rousseau and E. Papoutsakis; Farnham: Ashgate, 2009), 63–84. Satan’s Refusal to Worship Adam 261 the Qurʾānic account of Satan’s fall provides Abū Qurrah with the convenient leverage that allows him to subvert the Muslim criticism of Christian devotion: “If prostration is an act of worship, then without a doubt, according to what you say, God in that case commanded the angels to worship Adam! Far be it from God to do that!”107 The power of the anti-Muslim argument employed by the Christian apologete is thus derived from the inner tension between the prohibition of all prostration except for the worship of God in the Muslim tradition of devotion, and the residual traces of the practice of secular prostration preserved in the Qurʾānic myth of Iblīs and Adam.108 Over time, we see a more pronounced negative attitude towards the Muslim myth of Satan’s fall among Syriac-speaking Christian writers. One such example comes from the polemical works of Dionysius bar Ṣalībī, a prominent Western Syrian theologian and polemicist of the twelfth century.109 In the beginning of chapter 26 of his apologetic composition, Response to the Arabs, Dionysius quotes at length a fragment from al-Baqarah, the second sūra of the Qurʾān, which narrates the story of the worship of Adam by the angels and of Satan’s disobedience and fall. In his commentary on this passage, Bar Ṣalībī seeks to expose the error of this Muslim tradition:

As soon as God said that the angels should bow down to him [i.e., Adam], Satan did not bow down, and he did well. How shall a fiery being bow down to a mortal? And how shall secondary lights bow down to a tertiary light? Therefore, those who say that the perfect ones and the hermits are more excellent than the angels are of the mind of Muḥammad. It is quite appropriate to wonder at how the words of this book are arranged so haphazardly, so that from this fact, its error may be revealed. While scripture says that Eve transgressed the commandment, Muḥammad said that Adam transgressed. If Adam was driven out of

107 Ed. Arendzen, Theodori Abu Ḳurra de Cultu, 17; trans. Griffith, Treatise on the Veneration, 52. 108 On the practice and conceptualization of prostration in Islam, see R. Tottoli, “Muslim Attitudes towards Prostration (sujūd): I. Arabs and Prostration at the Beginning of Islam and in the Qurʾān,” Studia Islamica 88 (1998): 5–34; idem, “Muslim Attitudes towards Prostration (sujūd): II. Prominence and Meaning of Prostration in Muslim Literature,” Le Muséon 111/3–4 (1998): 405–26; idem, “The Thanksgiving Prostration (sujūd al-shukr) in Muslim Traditions,” BSOAS 61/2 (1998): 309–13. On polemical aspects of this practice, see R. Tottoli, “Muslim Traditions against Secular Prostration and Inter-Religious Polemic,” Medieval Encounters 5/1 (1999): 99–111. 109 See H. G. B. Teule, “Dionysius bar Salibi,” in Christian–Muslim Relations: A Bibliographical History, Volume 3 (1050‒1200) (ed. D. R. Thomas and A. Mallett; HCMR 15; Leiden: Brill, 2011), 665–70. 262 Minov

Paradise because he transgressed, what penalty did Satan incur, who rebelled and did not obey his Lord and bow down to Adam?110

One may observe here how Dionysius subverts the Qurʾānic representation of the fall of Iblīs by pointing out that in fact, the ontological argument put forward by Satan was sound, and that he whose nature was of fire indeed should not have had to bow down to Adam, who was only a mortal. Another argument against the Muslim interpretation of Satan’s fall is based on what is perceived by Dionysius to be an inner inconsistency between the Qurʾānic account and the biblical narrative. Thus, Dionysius wonders why one should blame Satan for his refusal to worship Adam, if Adam himself disobeyed God by transgress- ing the commandment. In an interesting aside, Dionysius castigates as Muslim-minded those among his coreligionists who consider Christians ascetics and monks to be “more excellent” than the angels. This accusation sounds rather unusual in light of the centuries-old anthropological tradition, influential among Syriac-speaking Christians, which understood the perfect nature of Adam before the fall to have been equal to that of the angels. It developed under the influence of the saying of Jesus, transmitted by the Synoptics (Matt 22:29–30; Mark 12:24–25; Luke 20:34–36), that the institution of marriage is to be abolished in the new world following the eschatological consummation, at which time those men and women who are found worthy of resurrection will become “equal to angels” (ἰσάγγελοι in Lk 20:36).111 This notion served as the backbone of Syriac ascetic ideology from the fourth century onward. Reaching the perfect, angel-like state of prelapsarian and/or eschatological humankind was regarded by many holy

̇ ̈ 110 ܕܓܣ ܐܢܪܘܢ ܢܟܝܐ .ܕܒܥ ܪܝܦܫܘ ܕܓܣ ܐܠ ܐܢܛܣܘ .ܐܟܐܠܡ ܗܠ ܢܘܕܓܣܢܕ ܐܗܠܐ ܪܡܐ ܐܕܚܡܘ ܐܖܝܡܓܕ̈ ܢܝܪܡܐܕ ܢܝܠܝܐ ܢܝܕܡ .ܐܝܬܝܠܬ ܐܪܗܘܢܠ ܢܝܕܓܣ ܐܢܝܖܬ̈ ܐܖܗܘܢ̈ ܢܟܝܐܘ .ܐܢܪܦܥܠ ܐܘܗ ܢܡܝܣ̈ ܢܟܝܐ ܩܕܙ̇ ܘܪܡܕܬܡܠ .ܢܘܗܝܬܝܐ ܕܡܚܡܕ ܐܬܝܥܪܬ ܢܡ .ܐܟܐܠܡ̈ ܢܡ ܢܝܪܬܝܡ ܐܝܕܝܚܝܘ̈ ܐܘܚܕ ܪܡܐ ܐܒܬܟ ܕܟܘ .ܗܬܘܝܥܛ ܐܠܓܬܐ ܐܕܗ ܢܡ ܢܟܝܐ .ܐܢܗ ܐܒܬܟܕ ܐܠܡ̈ ܬܝܐܪܕܒܡ ܐܣܝܕܪܦ ܢܡ ܕܪܛܬܐ ܐܢܕܩܘܦ ܪܒܥܕ ܠܥ ܡܕܐ ܢܐܘ .ܪܒܥ ܡܕܐܕ ܪܡܐ ܕܡܚܡ :ܐܢܕܩܘܦ ܠܥ ܬܪܒܥ ܀ܠܒܩ ܐܫܝܪܒܡܣܡ ܢܘܡ .ܡܕܐܠ ܕܓܣܘ ܗܪܡܠ ܥܡܫ ܐܠܘ ܐܢܝܨܥܒ ܡܩܕ ܐܢܛܣ; for the text, see J. P. Amar, ed., Dionysius bar Ṣalībī. A Response to the Arabs (2 vols.; CSCO 614–615; Scriptores Syri 238–239; Louvain: Peeters, 2005), 1:116 (Syr.), 2:107–9 (translation). 111 See on this U. Bianchi, “The Religio-Historical Relevance of Lk 20:34–36,” in Studies in Gnosticism and Hellenistic Religions Presented to Gilles Quispel on the Occasion of His 65th Birthday (ed. R. van den Broek and M. J. Vermaseren; EPRO 91; Leiden: Brill, 1981), 31–37; D. E. Aune, “Luke 20:34–36: A “Gnosticized” Logion of Jesus?” in Geschichte—Tradition— Reflexion: Festschrift für Martin Hengel zum 70. Geburtstag. Band 3: Frühes Christentum (ed. H. Lichtenberger; Tübingen: Mohr Siebeck, 1996), 187–202. Satan’s Refusal to Worship Adam 263 men of Syria–Mesopotamia as the ultimate goal of their ascetic efforts.112 For example, already in the discussion of monastic life by Aphrahat (fourth century) one finds a description of an ascetic as one, who “assumes the likeness of ̈ the angels” (ܐܟܐܠܡܕ ܐܬܘܡܕ ܠܩܫ).113 Later on, in the metrical Life of Rabban Bar ‘Idtā (an East Syrian monastic leader of the seventh century), composed by a certain Abraham Zabāyā, it is related that this holy man had a disciple named Yāwnān, who persevered in the ascetic way of life for twenty-nine years “until he became an angel” (ܐܟܐܠܡ ܐܘܗܕ ܐܡܕܥ).114 The list of similar examples can be easily continued.115 These examples, however, do not help us much with understanding why, in his rejection of the notion of ascetics as superior to the angels, Dionysius evokes Muslim tradition. In order to provide an answer to this question we shall turn to the tradition of Muslim mysticism. A relevant discussion of ange- lology in connection with asceticism is found in the tractate Kašf al-Maḥǧūb by ‘Alī b. ‘Uthmān al-Huǧwīrī, an eleventh-century Persian author. In the section that deals with the Sufi order of Ḥakīmīs, i.e., the followers of the ninth- century mystic al-Ḥakīm al-Tirmidhī, this writer devotes a long passage to defending the opinion that “the prophets and such of the saints as are guarded from sin are superior to the angels,” presented as a notion shared by “the whole community of orthodox Moslems and all the Sufi Shaykhs”.116 According to

112 For more on this ascetic notion in Syria and beyond, see A. Vööbus, History of Asceticism in the Syrian Orient: A Contribution to the History of Culture in the Near East (3 vols.; CSCO 184, 197, 500; Subsidia 14, 17, 81; Louvain: Secrétariat du CorpusSCO, 1958, 1960, 1988), 2:298–307; S. Frank, Ἀγγελικὸς βίος: Begriffsanalytische und Begriffsgeschichtliche Untersuchung zum “Engelgleichen Leben” im frühen Mönchtum (Beiträge zur Geschichte des alten Mönchtums und des Benediktinerordens 26; Münster: Aschendorff, 1964), esp. 141–62; M. Aveta, “Ad instar angelorum: Per un’analisi storico-religiosa dell’antropologia del Liber graduum,” Cristianesimo nella storia 8 (1987): 481–500; R. M. M. Tuschling, Angels and Orthodoxy: A Study in their Development in Syria and Palestine from the Qumran Texts to Ephrem the Syrian (STAC 40; Tübingen: Mohr Siebeck, 2007), 163–66; E. Muehlberger, “Ambivalence about the Angelic Life: The Promise and Perils of an Early Christian Discourse of Asceticism,” JECS 16/4 (2008): 447–78. 113 Dem. 6.1; see J. Parisot, ed., Aphraatis Sapientis Persae Demonstrationes (Patrologia Syriaca 1.1; Paris: Firmin-Didot, 1894), 248. Cf. also Dem. 6.6, 19; 18.12. 114 Ed. E. A. W. Budge, The Histories of Rabban Hôrmîzd the Persian and Rabban Bar-‘Idtâ (3 vols.; Luzac’s Semitic Text and Translation Series 9–11; London: Luzac, 1902), 1:148 (Syr.), 2.1:221 (translation). 115 Cf. Ephrem, Hymns on the Nativity 21.15.4; Commentary on the Diatessaron 16.22; Theodoret, Hist. relig., Prologue 2–3; John of Dalyatha, Letter 4.3–5; Barhebraeus, Ethicon 1.6.1, 5. 116 Ed. V. Žukovskij and M. ʿAbbāsī, Kašf al-maḥǧūb li-Abi-ʾl-Ḥasan ʿAlī Ibn-Uṯmān Ibn-Abi- ʿAlī al-Huǧwīrī al-Ġaznawī. Az rūy-i matn-i taṣhīḥ šuda-i Wālintīn Čūkūfskī (2d ed.; Tehran: 264 Minov al-Huǧwīrī, the “superiority” (tafḍīl) of the prophets over the angels is derived from the Qurʾānic stories about the latter worshiping Adam. The superiority of the “saints” (awliyā’) is based on the fact that contrary to the angels, who by their nature are “instinctively obedient to God,” humans, who have a natural propensity to commit sins, can reach the state of being “protected (i.e., from sin)” (maḥfūẓ) only as a result of prolonged ascetic struggle, aimed at mortify- ing their “lower soul” (nafs), which incites them to “all manner of wickedness.”117 The genuine writings of al-Tirmidhī provide us with a proof that this notion is not an invention of al-Huǧwīrī, but reflects the views of the Ḥakīmī Sufis. Thus, in his Kitāb ‘Ilm al-awliyā’, al-Tirmidhī discusses the divine light that emanates through the efforts of all beings who praise God, angels as well as humans, and states that from God’s point of view the lights that come forth “from a form of earth that is found amongst lusts and passion,” i.e., from the human “Friends of God” (awliyā’), are superior to those that come “from interiors with a light- nature that contain no passion, no lust and no enticement from the Enemy,” i.e., the angels.118 We have seen, thus, that the notion of holy men as superior to the angels enjoyed popularity at least in one fraction of the Sufi mystical movement. This observation throws additional light on the accusation of “Muslim-mindedness” aimed by Dionysius at those among his coreligionists who regarded Christian ascetics in a similar fashion. It should be noted here that the idea of holy men as superior to the angels in the Sufi tradition of mysticism might itself have originated under Christian influence. Most of the Christian expressions of the ideology of imitatio angelorum mentioned above envision the ascetic as equal to the angels, and thus would seemingly be exempt from Dionysius’s cutting accusation. However, it is possible that at least some Syriac-speaking Christians might have taken this idea one step further and claimed superiority to the angels as the final goal of spiritual progress. One may recognize a hint of such a vision of spiritual perfection in the treatise On the Kinds of Prayer, transmitted under the name of

Amīr Kabīr, 1957), 307–11; trans. R. A. Nicholson, The Kashf al-Maḥjúb: The Oldest Persian Treatise on Ṣufiism by ‘Alí B. ‘Uthmán al-Jullábí al-Hujwírí (E. J. W. Gibb Memorial Series 17; London: Luzac, 1911), 239–41. 117 Ed. Žukovskij and ʿAbbāsī, Kašf al-maḥǧūb, 308–9; trans. Nicholson, Kashf al-Maḥjúb, 240. 118 Trans. B. Radtke and J. O’Kane, The Concept of Sainthood in Early Islamic Mysticism: Two Works by al-Ḥakīm al-Tirmidhī (Curzon Sufi Series; Richmond, Surrey: Curzon, 1996), 228– 29. On this aspect of al-Tirmidhī’s anthropology, see also B. Radtke, Al-Ḥakīm at-Tirmiḏī: Ein islamischer Theosoph des 3./9. Jahrhunderts (Islamkundliche Untersuchungen 58; Freiburg im Breisgau: Schwarz, 1980), 63, 149 n. 179. Satan’s Refusal to Worship Adam 265

Abraham of Nathpar, an East Syrian spiritual writer from the seventh century.119 It is asserted in this work that when an ascetic attains the level of spiritual prayer, “his worship is more perfect than that of the company of Gabriel and Michael.”120 While this subject certainly deserves a more thorough investigation than can be carried out here, I would like to suggest that it is the notion of the imago Dei as the unique quality of human beings that might serve as a possible catalyst behind the development of this ascetic idea. In support of this suggestion, I may adduce the opinion of an otherwise unknown Syriac Christian author of two mēmrē on the Hexaemeron, which circulated under the name of Jacob of Sarug, mentioned above. The content of this work is known to us only partially, from the polemic waged against it in one of the letters by the West Syrian theologian Jacob of Edessa (ca. 640–708). According to the latter, the author of these homilies claimed that “Adam is greater and more excellent than Michael and Gabriel” (ܠܝܐܟܝܡܘ ܠܝܐܝܪܒܓ ܢܡ ܡܕܐ ܪܬܝܡܘ ܒܪ); this declaration seems connected in the writer’s thought with the fact that the angels lack the image of God: “these rational minds and secondary luminaries are not in the image ̈ ̈ ̈ ̈ of God, their creator” (ܐܡܠܨܒ ܐܢܝܖܬ ܐܖܗܘܢܘ ܐܠܝܠܡ ܢܘܢܗ ܐܢܘܗ ܢܘܗܝܬܝܐ ܐܠ ܢܘܗܕܘܒܥ ܐܗܠܐܕ).121 Unfortunately, the brief and hostile summary of Jacob of Edessa does not allow us to establish whether the author of the homilies relied on the narrative of the creation of Adam in the Cave in his treatment of these matters.

Bar Ṣalībī’s expressly negative attitude toward the Muslim interpretation of the fall of Satan, seems to have had no particular influence on his great successor, Barhebraeus, a famous thirteenth-century West-Syrian theologian and scholar.122 In his monumental theological compendium, the Lamp of the Sanctuary (7.1.1.1), Barhebraeus discusses, among other things, the topic of Satan’s downfall. At some point he refers to the Muslim way of dealing with

119 See on this work G. M. Kessel and K. Pinggéra, A Bibliography of Syriac Ascetic and Mystical Literature (Eastern Christian Studies 11; Leuven: Peeters, 2011), 32. 120 Trans. S. P. Brock, The Syriac Fathers on Prayer and the Spiritual Life (Cistercian Studies Series 101; Kalamazoo: Cistercian Publications, 1987), 192. 121 For the Syriac text and German translation, see R. Schröter, “Erster Brief Jacob’s von Edessa an Johannes den Styliten,” ZDMG 24 (1870): 261–300 (270, 275); for a French translation, see F. Nau, “Cinq lettres de Jacques d’Édesse à Jean le Stylite (traduction et anal- yse),” Revue de l’Orient chrétien 14 (1909): 427–40 (435). 122 See H. Takahashi, “Barhebraeus,” in The Orthodox Christian World (ed. A. M. Casiday; Routledge Worlds; London: Routledge, 2012), 279–86; H. G. B. Teule, “Barhebraeus,” in Thomas and Mallett, Christian–Muslim Relations 4, 588–609. 266 Minov this subject, noticing that this treatment stands close to what had been taught much earlier by Jacob of Sarug:

On the one hand, the Muslims say that because God told Satan to worship Adam, who had just been created in the image of God, and he refused and did not yield, he fell together with the whole band of his; on the other hand, Mar Jacob of Sarug says something similar to this in his Hexaemeron. However, the venerable Mar Jacob (of Edessa) rejects that homily, as we have already said above.123

Barhebraeus glosses the fact of harmony between the Christian and Muslim traditions on the fall of Satan by pointing out that the Syriac mēmrē transmitted under the name of Jacob of Sarug, where this opinion supposedly appears, are not authentic and, in fact, contain heterodox teaching. This statement, however, betrays a certain confusion on Barhebraeus’s part, since neither the genuine Jacob’s mēmrā on the sixth day of creation nor the two mēmrē on the Hexaemeron falsely ascribed to him, which were mentioned above, contain the notion of Satan’s refusal to worship Adam.124 Even so, what strikes us in connection with Barhebraeus’s handling of the Muslim story of Satan’s fall is his basically nonjudgmental attitude towards it. This might be not so surprising, however, when one recalls the great indebtedness of this Syriac polymath to Muslim traditions of science and philosophy.125

̇ ̇ ̇ ̈ 123 ܬܝܐܬܕܚ ܕܟ ܆ܡܕܐܠ ܕܘܓܣܢ ܠܦܢܕ ܆ܐܢܛܣܠ ܐܗܠܐ ܪܡܐܕ ܝܗܠܛܡܕ ܢܝܪܡܐ ܢܡ ܐܢܡܠܫܡܘ ܓܘܪܣܕ ܒܘܩܥܝ ܝܪܡܘ .ܗܠܗܝ ܗܠܟ ܡܥ ܠܦܢ ܆ܦܫܐ ܐܠܘ ܠܐܬܫܐ ܘܗܘ̣ ܆ܐܗܠܐ ܡܠܨܒ ܝܪܒܬܐ ܐܢܗ ܐܪܡܐܡܠ ܗܠ ܐܠܣܡ̇ ܒܘܩܥܝ ܢܝܕ ܐܝܣܚ .ܗܡܘܝ ܬܬܫܐܒ ܪܡܐ ܢܟܗܠ ܐܡܕܕ̇ ܡܕܡ ܒܘܬ ܢܢܪܡܐ ܠܥܠ ܢܡܕ ܟܝܐ; ed. M. Albert, Le Candélabre du Sanctuaire de Grégoire Abouʾlfaradj dit Barhebræus. Septième base: Des démons (PO 30:2; Paris: Firmin-Didot, 1961), 284–86. 124 In fact, the author of these mēmrē shares with Jacob the general framework of the “Edessene interpretation” of Satan’s fall; i.e., that it occurred on the sixth day as a result of Satan’s envy of Adam’s greatness. For the Syriac text and German translation, see Schröter, “Erster Brief Jacob’s von Edessa,” 270, 275; for a French translation, see Nau, “Cinq lettres de Jacques,” 436. 125 See on this H. Takahashi, “Barhebraeus und seine islamischen Quellen: Têgrat têgrâtâ (Tractatus tactatuum) und Gazâlîs Maqâsid al–falâsifa,” in Syriaca: Zur Geschichte, Theologie, Liturgie und Gegenwartslage der syrischen Kirchen, 2. Deutsches Syrologen- Symposium (Juli 2000, Wittenberg) (ed. M. Tamcke; Studien zur Orientalischen Kirchengeschichte 17; Münster: Lit, 2002), 147–75; H. G. B. Teule, “The Transmission of Islamic Culture to the World of Syriac Christianity: Barhebraeus’ Translation of Avicenna’s kitâb al-išârât wa l-tanbîhât. First Soundings,” in Redefining Christian Identity: Cultural Interaction in the Middle East since the Rise of Islam (ed. J. J. van Ginkel, H. L. Murre-van den Berg, and T. M. van Lint; OLA 134; Leuven: Peeters, 2005), 167–84; Satan’s Refusal to Worship Adam 267

As these examples show, the story of Satan’s refusal to worship Adam and consequent fall could be seen simultaneously in medieval Syriac Christianity as an authentic part of the tradition and as an element of heterodox teaching formed under Muslim influence. That same influence made itself felt in a different way in medieval rabbinic tradition.

4 Satan’s Fall Revisited in Medieval Jewish Tradition

To conclude this overview of the fascinating odyssey of the story of Satan’s fall through various confessional traditions, I bring forward several examples that demonstrate how the interpretation of Satan’s fall as a result of the refusal to worship Adam found its way back in to Jewish literary tradition as a result of contact with the Islamicate milieu. As has been mentioned above, no explicit indication of this interpretation seems to be found in the corpus of rabbinic writings from late antiquity. It is only in a few medieval Jewish sources that one comes across portrayals of Satan’s fall similar to that of the Life of Adam and Eve. One of the earliest attestations of this myth in Jewish literature comes from Sefer Eldad ha-Dani,126 the author of which appeared during the late-ninth century in the Jewish community of Kairouan in North Africa. Eldad claimed descent from the tribe of Dan, now supposedly resident in an (East African) Jewish kingdom comprised of some of the lost tribes of Israel, in the biblical land of Havilah. The Sefer presents an account of Satan’s fall: After God had created Adam, he commanded all the ministering angels to bow down before his new creation. While most of the angels complied with this order, Satan refused to worship Adam, under the pretext that he himself had been created from the “radiance of Shekhinah,” whereas the human being had been formed from the “dust of the earth.” To settle the conflict, God challenged Satan to participate in a contest whereby he and Adam would demonstrate

idem, “The Interaction of Syriac Christianity and the Muslim World in the Period of the Syriac Renaissance,” in Syriac Churches Encountering Islam: Past Experiences and Future Perspectives (ed. D. W. Winkler; Pro Oriente Studies in the Syriac Tradition 1; Piscataway: Gorgias, 2010), 110–28. 126 On Eldad and his works, see A. Epstein, Eldad ha-Dani: Seine Berichte über die X Stämme und deren Ritus in verschiedenen Versionen nach Handschriften und alten Drucken mit Einleitung und Anmerkungen (Pressburg: Alkalay, 1891) (in Hebrew); M. Schloessinger, The Ritual of Eldad ha-Dani Reconstructed and Edited from Manuscripts and a Genizah Fragment (Leipzig: Haupt, 1908). 268 Minov their wisdom by naming animals; the loser was to worship the winner. As one might expect, Satan lost the competition after failing to name the animals that were brought before him, whereas Adam succeeded in this task.127 It is not possible to deal here in detail with the complex literary genealogy of this variation on Satan’s refusal to worship Adam. Yet, the fact that Eldad ha-Dani circulated in communities under Islamic rule, as well as the peculiar linguistic profile of his Hebrew, which betrays influence of Arabic and Syriac,128 make it a strong possibility that this Jewish author became acquainted with the myth of Satan’s refusal worship Adam in the context of a Muslim-dominated cultural milieu.129 Another example of Jewish acquaintance with the motif of Satan’s refusal to worship Adam comes from the literary tradition of a diaspora community located at the other extremity of the Muslim world, in Iran. The fourteenth- century Judaeo-Persian poet Shāhīn-i Shirāzi makes use of this motif in his poetic reworking of the book of Genesis, Bereshit-nāmah. After a description of the superiority of Satan (also referred to as ʿAza‌ʾzel and Iblīs) to the rest of the angelic forces, the poet offers a vivid description of his negative reaction to God’s decree: when God commands the angels to venerate Adam, ʿAzaʾzel‌ and refers to his (סגדה פישי תוֹדהי כאךּ) ”refuses to “bow before a lump of clay The poet continues with a series of 130.(נוּרי פאכּם) ”own nature of “pure light dialogues between Satan and God, in which the former tries unsuccessfully to argue with his Creator from the position of a strict monotheism that presumes no other object of worship besides the Divinity. As a result, God punishes Satan for his disobedience and curses him “till the Day of Resurrection,” while grant- ing him the right to tempt Adam and his posterity. As has been demonstrated by Vera Moreen, Shāhīn’s representation of the figure of Satan and his conflict

127 Ed. Epstein, Eldad ha-Dani, 66–67. 128 See on this S. Morag, “Eldad Haddani’s Hebrew and the Problem of His Provenance,” Tarbiz 66/2 (1997): 223–46 (in Hebrew). Cf. also L. I. Rabinowitz, “Eldad Ha-Dani and China,” JQR n.s. 36/3 (1946): 231–38. 129 Through the mediation of Sefer Eldad ha-Dani the myth of Satan’s fall reached Jewish intellectual circles of southern Europe. Thus, Bereshit Rabbati, a late midrashic collection associated with R. Moshe ha-Darshan of Narbonne (eleventh century) presents an account of Satan’s rebellion against Adam which betrays its dependence on the work of Eldad ha-Dani. For the Hebrew text, see C. Albeck, Midraš Berešit Rabbati ex libro R. Mosis Haddaršan collectus e codice Pragensi cum annotationibus et introductione (Jerusalem: Mekize Nirdamim, 1940), 24–25. 130 Ed. S. Ḥakham, Sefer Sharḥ Shahin Torah (4 vols.; Jerusalem: Luntz, 1902–1905) (in Hebrew), 1:5a–6b; trans. V. B. Moreen, In Queen Esther’s Garden: An Anthology of Judeo- Persian Literature (Yale Judaica Series 30; New Haven: Yale University Press, 2000), 32–33. Satan’s Refusal to Worship Adam 269 with Adam reveals the poet’s familiarity with the contemporary Sufi interpretation of the figure of Iblīs and of the theme of his fall.131 And finally, a very interesting example of Jewish appropriation of this myth comes from Safavid Persia. It is found in the so-called “Morgan Bible,” known also as the “Crusader Bible” (New York, Pierpont Morgan Library, Ms. M. 638), a medieval picture Bible that depicts Old Testament scenes.132 Produced in thirteenth-century France under the patronage of King Louis IX (1226–1270), this lavishly executed manuscript was brought by Carmelite missionaries as a gift to Shah ʿAbbās I (1587–1629) in the year 1607. Upon receiving the book, the Safavid monarch commanded that explanatory notes in Persian be added in the margins. Later on, after the Afghan invasion of Isfahan in 1722, when the manuscript fell into the hands of a Jewish buyer, a second set of explanatory glosses was added to it, this time in Judeo-Persian.133 It is the first out of the four pictures on f. 1r that grabs our attention. This image depicts the separation of light from darkness on the first day of creation, along with the fall of the angels. In this picture, God (iconographically pic- tured as Christ triumphant) is surrounded by angelic choirs, while trampling on Lucifer and the fallen angels below. This image is accompanied by a Persian gloss on the folio’s upper left margin that says: “This is the theme of the angels who submitted to God except for the Devil, who did not prostrate himself and was cursed.” Below the Persian gloss the following comment in Judeo-Persian נקל שטן) ”is added: “The story of Satan, who did not make obeisance to Adam What is remarkable about these explanations is that .(כה באדם סגדה נכרד whereas the original Christian image has no iconographic references to the myth of Satan’s refusal to worship Adam, both the Muslim Persian and Judeo-

131 See V. B. Moreen, “A Dialogue between God and Satan in Shāhīn’s Bereshit Nāmah,” in Irano-Judaica III (ed. S. Shaked and A. Netzer; Jerusalem: Yad Izhak Ben-Zvi, 1994), 127–41. 132 For a facsimile reproduction of the manuscript, see D. H. Weiss, ed., Die Kreuzritterbibel = The Morgan Crusader Bible = La Bible des croisades (Lucerne: Faksimile Verlag / New York: Pierpont Morgan Library, 1998). On the manuscript’s arrival in Persia, see M. Shreve Simpson, “The Morgan Bible and the Giving of Religious Gifts between Iran and Europe/ Europe and Iran during the Reign of Shah ʿAbbas I,” in Between the Picture and the Word: Manuscript Studies from the Index of Christian Art (ed. C. Hourihane; University Park: Penn State University Press, 2005), 141–50. I thank Dr. Ruth Clements for drawing my attention to this evidence. 133 For English translation and discussion of Persian and Judeo-Persian glosses, see D. H. Weiss et al., The Morgan Crusader Bible: A Commentary (Lucerne: Faksimile Verlag, 1999). Both the image and the translations of the glosses are available online at http:// www.themorgan.org/collection/crusader-bible/1. 270 Minov

Persian glossators choose to bring the image into the line with the explanation of Satan’s fall that was familiar to them, notwithstanding the violation of the inner-biblical chronology involved in such reinterpretation. These glosses bear witness to how the thirteenth-century Christian imagery could be assimilated into Iranian Jewish and non-Jewish culture, when the foreign visual language is appropriated through the act of misreading, a powerful and common strategy of intercultural communication.

5 Conclusion

In the course of this investigation it has been argued that the peculiar interpretation of the fall of Satan as a result of his refusal to acknowledge the superiority of the first man by bowing down (or prostrating himself) before him, found in the Life of Adam and Eve, developed on the basis of an ancient Jewish tradition about the veneration of Adam by the angels. Although this interpretation of the fall of Satan was marginalized in the tradition of rabbinic Judaism, early Christianity turned out to be particularly receptive to it, and constituted the primary channel of its transmission during late antiquity. It was through the mediation of Christians, most likely Syriac-speaking ones, that this myth became an integral part of the demonological tradition of Islam, where its popularity reached new peaks. We have seen that this account of Satan’s undoing enjoyed a certain popularity among the Christians of Syria and adjunct regions, from the sixth century onward. It should be emphasized, however, that even among Syriac-speaking Christians this account did not become the par excellence explanation of Satan’s fall. One may propose several reasons for this. First of all, the fact of the appearance of this tradition in the Cave of Treasures, a work of doubtful pedigree in the eyes of at least some Syriac Christian theologians, might have negatively affected the perceived “orthodoxy” of this tradition. Second, as I have demonstrated above, the central place of this myth in the Muslim tradition could also have rendered it unacceptable to some Syriac-speaking Christians, especially those concerned with drawing clear boundaries between the two communities of faith. I have also shown, however, that notwithstanding the efforts of such polemically driven intellectuals as Dionysius bar Ṣalībī to sup- press this tradition, the myth of Satan’s refusal to worship Adam retained its appeal among Syriac-speaking Christians. Finally, the wide appeal of the account of Satan’s refusal to worship Adam is also attested by its reemergence in medieval Jewish sources, as well as by its appearance in the literary traditions of other religious minorities of the Satan’s Refusal to Worship Adam 271

Dar al-Islam, such as the Mandaeans of Southern Iraq,134 or the Yezidis of Kurdistan.135 These instances, along with the continuing popularity of this story among the Syriac-speaking Christians of the Middle Ages, demonstrate that this story became an integral part of the biblically inspired cultural imagi- naire of the Islamicate world, shared by the followers of all three Abrahamic religions.

134 Cf. Right Ginza 1.88; 2.21; in J. H. Petermann, ed., Thesaurus s. Liber magnus vulgo “Liber Adami” appellatus: opus Mandaeorum summi ponderis (2 vols.; Leipzig: Weigel, 1867), 1:13, 34; German translation by M. Lidzbarski, Ginza, der Schatz: oder das Grosse buch der Mandäer (Quellen der Religionsgeschichte 4.13; Göttingen: Vandenhoeck & Ruprecht, 1925), 16, 34. 135 See I. Joseph, “Yezidi Texts,” AJSL 25/2–3 (1909): 111–56, 218–54 (235–36). Stars of the Messiah

Hillel I. Newman

Light, both in its purest form and as embodied by heavenly luminaries, exerts its fascination in the biblical narrative from its opening verses. In Jewish eschatological thought of the Second Temple period and beyond, divine light and celestial bodies also play a role in the imagery and drama of the final redemption. These themes are too broad to exhaust in a single paper. Here I will focus on one aspect of the topic: the anticipated appearance of a messianic star or other attendant divine light at the End of Days. I am interested in examining not only the literary expressions of this motif in the context of apocalyptic speculation, but also the manner in which this speculation has generated historical outbreaks of messianic fervor and informed the presentation of astronomical events as realized eschatology. Without ignoring the evolution of tradition over time, I will trace several paths of continuity in the development of this motif, extending from Second Temple times to the early Middle Ages. Let me clarify further the parameters of this paper by mentioning an important issue that I will address only tangentially. Some texts use astral language or imagery as metaphor or symbol with reference to a messiah (Davidic as well as priestly), but without meaning to suggest that extraordinary astronomical events will actually take place at the End of Days. The distinction between figurative language and speculation about concrete astronomical phenomena is often subtle, as both may share the same biblical and exegetical foundations and employ similar terminology. The boundaries between the two categories are permeable, and the sense of a given image may be ambiguous. Though more figurative texts will not be ignored, they do not stand at the center of this discussion.

Two biblical loci feature prominently (though not always explicitly) in the sources on messianic stars and light. The first is Balaam’s prophecy in Num 24:17:

What I see for them is not yet, What I behold will not be soon: A star rises from Jacob,

* My thanks to Prof. Rina Talgam, of the Hebrew University Department of the History of Art, for her assistance with the illustrations for this article.

;comes forth from Israel 1(ׁשבט) A scepter It smashes the brow of Moab, The foundation of all children of Seth.

The second is a cluster of verses in Isaiah 60, addressed to the city of Jerusalem:

Arise, shine, for your light has dawned; The Presence of the Lord has shone upon you! Behold! Darkness shall cover the earth, And thick clouds the peoples; But upon you the Lord will shine, And his presence be seen over you. And nations shall walk by your light; Kings, by your shining radiance.2

In what follows we will return repeatedly to these passages.3

1 Messianic Stars in Second Temple Sources

The traces in Second Temple literature of the messianic interpretation of Balaam’s prophecy in general and Num 24:17 in particular are well known. I use the term “messianic” loosely in this part of the discussion, to include all divinely ordained eschatological heroes. It will suffice here to survey the major sources.4

1 Or: “a meteor” (thus the NJPS version; all biblical translations here follow that version, unless otherwise stated). See B. A. Levine, Numbers 21–36: A New Translation with Introduction and Commentary (AB 4A; New York: Doubleday, 2000), 199–202. 2 Isa 60:1–3. Cf. Isa 60:19–20; 62:1. 3 For a selection of other biblical verses taken by exegetes in antiquity to allude to the light of the Messiah, see W. Horbury, Jewish Messianism and the Cult of Christ (London: SCM Press, 1998), 92–94, 99. 4 From the abundant bibliography see, for example, J. L. Kugel, Traditions of the Bible: A Guide to the Bible As It Was at the Start of the Common Era (Cambridge, Mass.: Harvard University Press, 1998), 803–8, 820–23; K. J. Cathcart, “Numbers 24:17 in Ancient Translations and Interpretations,” in The Interpretation of the Bible: The International Symposium in Slovenia (ed. J. Krašovec; JSOTSup 289; Sheffield: Sheffield Academic Press, 1998), 511–19; The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity, and Islam (ed. G. H. van Kooten and J. van Ruiten; TBN 11; Leiden: Brill, 2008); J. J. Collins, The Scepter and the Star: Messianism in the Light of the Dead Sea Scrolls (2d ed.; Grand Rapids: Eerdmans, 2010), 71–75, 85–90. 274 Newman

Let us begin with the Septuagint’s rendering of the verse: “A star shall rise out of Jacob, and a man (ἄνθρωπος) shall arise out of Israel.” While the “scepter” of the prophecy has become explicitly personified and may be taken to refer to an eschatological figure, the translation with regard to the star is essentially literal, providing little indication of the translators’ precise understanding of the image in context.5 The evidence from Qumran is more explicit. According to one manuscript of the Damascus Document in the Cairo Genizah (CD 7:18–20 [Ms. A]; corroborated by two Qumran fragments: 4Q266 3 iii 19–22 and 4Q269 5),6 the star of and (דורש התורה) ”Num 24:17 is identified with the “interpreter of the Torah These .(נשיא כל העדה) ”the scepter with the “prince of the whole congregation are both commonly interpreted as eschatological titles; in the opinion of some scholars the first refers to a priestly messiah and the second to a royal one.7 The verse is also cited in eschatological contexts in 4Q175 12–13 and 1QM 11:5–6, though not in a manner that suggests a clear distinction between messianic figures.8 In any case, nothing indicates that the verse is taken to refer to an astronomical event. The religious identity of the authors of the Testaments of the Twelve Patriarchs is still contested, though the fact that the Testaments as we have them contain both Jewish and Christian material is not. Without resolving these problems, I note two passages pertaining to our topic. In T. Jud. 24:1 the patriarch declares: “And after these things a star will arise to you from Jacob in peace, and a man will arise from my seed like the sun of righteousness, walking with the sons

5 For various points of view on the LXX of this verse see W. Horbury, “Monarchy and Messianism in the Greek Pentateuch,” in The Septuagint and Messianism (ed. M. A. Knibb; BETL 195; Leuven: Leuven University Press and Peeters, 2006), 121–24; and in the same volume: J. J. Collins, “Messianism and Exegetical Tradition: The Evidence of the LXX Pentateuch,” 144–47; M. Rösel, “Jakob, Bileam und der Messias: Messianische Erwartungen in Gen 49 und Num 22–24,” 168–74. Each of these takes issue to some degree with J. Lust, Messianism and the Septuagint: Collected Essays (ed. K. Hauspie; BETL 178; Leuven: Leuven University Press and Peeters, 2004), 69–86, 147–50. Lust is inordinately skeptical of any “messianic” reading by the LXX here. 6 See J. M. Baumgarten, Qumran Cave 4.XIII: The Damascus Document (4Q266–273) (DJD 18; Oxford: Clarendon, 1996), 44 (for 4Q266 3 iii 19–22), 128 (for 4Q269 5). 7 See for example Collins, The Scepter and the Star, 79–91. Note, however, that the “Interpreter of the Torah” of Qumran features initially as one of the community’s historical leaders. On this ambiguity see, for example, G. J. Brooke, “The Messiah of Aaron in the Damascus Document,” RevQ 15 (1991): 224–25. 8 See the literature in n. 4 above; and especially F. García Martínez, “Balaam in the Dead Sea Scrolls,” in van Kooten and van Ruiten, The Prestige of the Pagan Prophet Balaam, 71–82. Stars Of The Messiah 275 of men in meekness and righteousness, and no sin whatever will be found in him.”9 The passage, alluding to a Davidic messiah, appears, together with subsequent verses, to betray the hand of a Christian author or redactor. It is not clear if the reference to the star is to be taken literally, but whatever the author’s original intent, the verse would ultimately have been taken by some Christian readers to signify the star of the magi in Matthew 2. The Testament of Levi, on the other hand, tells of an eschatological priest:

And his star will arise in heaven, as a king, lighting up the light of knowledge as by the sun of the day; and he will be magnified in the world until his assumption. He will shine as the sun on the earth and will remove all darkness from under heaven, and there will be peace on all the earth.10

In this passage, the star in heaven corresponding to the earthly, eschatological priest gives the impression of being a genuine celestial body.11 Scholars have recognized the affinity of these verses to 4Q541 9, which alludes to the universal sunlight that will accompany a figure who likewise seems to be the eschatological priest.12

Before turning to those topics which constitute the major part of this paper, I would like briefly to take a closer look at the star of the magi in Matt 2:1–12, perhaps the most famous of messianic stars. I make no pretense of adding something new to the vast body of scholarship on the subject, but it is important for my purposes to highlight several important conclusions.13

9 H. W. Hollander and M. de Jonge, eds., The Testaments of the Twelve Patriarchs: A Commentary (SVTP 8; Leiden: Brill, 1985), 226–27. 10 T. Levi 18:3–4. Hollander and de Jonge, Testaments of the Twelve Patriarchs, 177–80. 11 It is understood metaphorically, however, by A. Hilhorst, “Biblical Metaphors Taken Literally,” in Text and Testimony: Essays on New Testament and Apocryphal Literature in Honour of A. F. J. Klijn (ed. T. Baarda; Kampen: Kok, 1988), 124–25. 12 See Collins, The Scepter and the Star, 99–100; M. Kister, “Wisdom Literature and its Relation to Other Genres: From Ben Sira to Mysteries,” in Sapiential Perspectives: Wisdom Literature in Light of the Dead Sea Scrolls (ed. J. J. Collins et al.; STDJ 51; Leiden: Brill, 2004), 42–43. 13 I am particularly indebted to W. D. Davies and D. C. Allison, A Critical and Exegetical Commentary on the Gospel According to Saint Matthew (3 vols.; ICC 26; Edinburgh: T&T Clark, 1988–1997), 1:224–56; T. Hegedus, Early Christianity and Ancient Astrology (Patristic Studies 6; New York: Peter Lang, 2007), 201–22; T. Nicklas, “Balaam and the Star of the Magi,” in van Kooten and van Ruiten, The Prestige of the Pagan Prophet Balaam, 233–46. 276 Newman

The first is astronomical. Given that the appearance of stars and comets is anticipated in various eschatological scenarios, it is natural that extraordinary astronomical phenomena such as novas or comets should generate apocalyptic excitement. Historical examples are not lacking, and we will encounter several of them below. For hundreds of years historians and astronomers have hypoth- esized about “scientific” explanations for the star of the nativity. That star, however, makes a poor candidate for such speculation. For one thing, the problems and contradictions that beset the chronology of nativity narratives are rife, so that any suggestion of a dateable comet, nova, or planetary conjunction inevitably entails making arbitrary choices and doing more than a little violence to our written sources. For another, much in Matthew’s account suggests that the narrative draws on extant typology that has fostered the introduction of a messianic star. It has been recognized, for example, that the opening chapters of Matthew are modeled on the early life of Moses, both as related in the book of Exodus and as embroidered in postbiblical tradition. In these later expansions, Pharaoh’s wise men foresee the birth of the savior of the Jews; in rabbinic tradition the wise men are identified explicitly as astrologers.14 We must not forget, of course, the star of Balaam. While there is no explicit reference to Balaam’s prophecy in Matthew itself, interpreters from the second century till the present make this connection and point to its influence.15 We should note as well the stamp of Isaiah 60 on the account of the gifts of the magi in Matt 2:11. As we saw above, the prophet portrays the divine radiance that will accompany the redemption of Jerusalem. He continues by describing the riches of the nations which will be borne there by camels: “They shall bear gold and frankincense, and shall herald the glories of the Lord” (Isa 60:6). In the context of Matthew, the visit of the magi and their offerings are by implication the fulfillment of Isaiah’s prophecy, which is understood to refer to the messianic star. We will return below to star of the magi, but in a very different setting and with some surprising embellishments. This does not exhaust the topic of messianic light in the New Testament. Particularly important is the

14 See Davies and Allison, Commentary, 190–95 (esp. 195 n. 22); R. D. Aus, Matthew 1–2 and the Virginal Conception: In Light of Palestinian and Hellenistic Judaic Traditions on the Birth of Israel’s First Redeemer, Moses (Lanham, Md.: University Press of America, 2004), 15–18. According to a medieval Samaritan tradition, the conception of Moses was marked by the appearance of a star. See S. J. Miller, The Samaritan Molad Mosheh (New York: Philosophical Library, 1949), 256–57. 15 For patristic exegesis see G. Dorival, “Un astre se lèvera de Jacob: L’interprétation ancienne des Nombres 24,17,” Annali di storia dell’esegesi 13/1 (1996): 295–352; J. Leemans, “‘To Bless with a Mouth Bent on Cursing’: Patristic Interpretations of Balaam (Num 24:17),” in van Kooten and van Ruiten, The Prestige of the Pagan Prophet Balaam, 287–99. Stars Of The Messiah 277 lamp of the Lamb in Jerusalem according to Rev 21:23, in a chapter rich in allusions to Isaiah 60.16 The following discussion will be organized episodically. First I will examine Josephus’s report of the stellar omens which preceded the Great Revolt; then I will turn to the stellar imagery associated with the Bar Kokhba Revolt. Each of these topics will serve as an occasion to analyze a range of apocalyptic themes. Finally, I will take a close look at the motif of the messianic star in several Jewish apocalyptic sources of late antiquity and the Middle Ages.

2 Josephus on Celestial Omens

In his description of the circumstances leading to the outbreak of the Great Revolt against Rome, Josephus includes a list of portents which the Jews— imbued, he says, with vain hopes cultivated by false prophets—understood to be signs of their imminent redemption.17 He writes:

Thus it was that the wretched people were deluded at that time by char- latans and pretended messengers of the deity. . . . So it was when a star, resembling a sword, stood over the city (= Jerusalem), and a comet which continued for a year. So again when, before the revolt and the commotion that led to war, at the time when the people were assembling for the feast of unleavened bread, on the eighth of the month Xanthicus (= 8 Nisan), at the ninth hour of the night, so brilliant a light shone round the altar and the sanctuary that it seemed to be broad daylight; and this continued for half an hour. By the inexperienced this was regarded as a good omen, but by the sacred scribes it was at once interpreted in accordance with after events.18

16 See D. Flusser, Judaism and the Origins of Christianity (Jerusalem: The Hebrew University Magnes Press, 1988), 454–65; D. E. Aune, Revelation 17–22 (WBC 52C; Nashville: Thomas Nelson, 1998), 1169–73, 1181. Cf. Luke 1:78–79; 2 Pet 1:19; Rev 2:26–28, 22:16. 17 See the historiographical survey of F. Schmidt, “Signes et prodiges chez Flavius Josèphe et Tacite (Guerre des Juifs VI, 288–315; Histoires V, 13),” in La Raison des signes: Présages, rites, destin dans les sociétés de la Méditerranée ancienne (ed. S. Georgoudi et al.; Religions in the Graeco-Roman World 174; Leiden: Brill, 2012), 253–89. Schmidt notes the scholarly debate between those who believe Josephus’s description of the omens was appropriated from Roman models and those who maintain that it should be read as emerging from within the context of Second Temple Judaism. 18 Jewish War 6:288–291 (trans. H. St. J. Thackeray, Josephus III: The Jewish War, Books IV–VII [LCL; Cambridge, Mass.: Harvard University Press, 1928], 459–61). In the fourth-century 278 Newman

The chronology of these events is obscure. Josephus is generally understood to mean that the light round the altar and the sanctuary appeared in the days before the final Passover festival preceding the outbreak of the revolt, i.e., in 66 CE.19 Even if this is the case, it is not clear how much time passed— assuming Josephus’s sequence is chronological—between the appearance of the “star,” the comet which followed it, and the subsequent Passover festival. Astronomical calculations might be of some assistance, but Josephus does not make it easy for us.20 First, though a comet might vaguely resemble a sword, a star cannot; if Josephus attests to a genuine astronomical event, we must assume that he uses “star” loosely and actually refers to a comet—an assumption bolstered by the fact that sword imagery is indeed attested for comets in other sources and that comets are often understood to be a variety of star. Second, no comet is ever visible for a full year. Since Josephus lumps the omens together, it is at least plausible—but not imperative—to suppose that he describes a sequence of two comets not far removed from Passover of 66 CE. In fact, two comets are known to have to have appeared within that range: the first from July to September of 65 CE; the second—Halley’s Comet— from January to April of 66 CE, leading up to Passover. These are attractive possibilities, though we still lack a satisfactory explanation for the alleged year- long duration of the second comet. I am, I confess, inclined towards this astronomical reconstruction of events, but I do not want to give the impression that what follows is contingent upon it. In fact, I am much more interested here in Josephus’s literary presentation of these phenomena—real, “improved,” or imagined. Several details point to the construction and reception of the memory of these omens along traditional and ideological lines. First, when we observe a comet (among other celestial phenomena), we share the experience with other observers around the planet.

Latin translation of Pseudo-Hegesippus 5.44 (Hegesippi qui dicitur historiae libri V [ed. V. Ussani; CSEL 66; Vienna: Hölder, Pichler, Tempsky, 1932], 391), Josephus’s account has been conflated: a single sword-like comet appears over the Temple (not the city) for a year prior to the outbreak of the revolt. This is taken over by the Yosippon (ed. D. Flusser; Jerusalem; Bialik Institute, 1981), 1:413 (in Hebrew). 19 See D. R. Schwartz, “Portents of Destruction: From Flavian Propaganda to Rabbinic Theodicy,” in From Despair to Solace: A Memorial Volume for Ziporah Brody z”l on the Tenth Anniversary of Her Passing (ed. S. T. Brody; Jerusalem: Midreshet Lindenbaum, 2009), 11. 20 For the following astronomical discussion see especially J. T. Ramsey, A Descriptive Catalogue of Greco–Roman Comets from 500 BC to AD 400 (Syllecta Classica 17; Iowa City: University of Iowa Press, 2006), 148–58. Cf. W. Horowitz, “Halley’s Comet and Judaean Revolts Revisited,” CBQ 58 (1996): 456–59. Stars Of The Messiah 279

For many it will appear “overhead.” Hence the description of a star standing over the city of Jerusalem tells little about the star or comet (if indeed there was one), but, apart from indicating the location of certain witnesses, it tells us a great deal about the belief system shared by those who interpreted it as a propitious omen.21 Second, comets do not really look like swords; by describing them as such one betrays a predisposition towards a particular visual interpretation laden with symbolism.22 Third, we should ask what overtones would be carried by an omen appearing on the verge of Passover.23 The appearance of the star over Jerusalem and the blaze of light around the altar and the Temple as an omen of salvation naturally bring to mind eschatological imagery harking back to Isaiah 60.24 We will see other examples below of the spatial connection between the messianic star and Temple. Leaving aside the sword imagery for a moment, let us first consider the implications of the Passover setting. Within the Bible itself, the salvation of Israel in the future is often modeled typologically on the exodus from Egypt.25 In rabbinic and targumic literature we find the notion that the final redemption, prefigured by the redemption from Egypt, will take place on Passover. It is possible that already in Second Temple times some Jews approached this festival with heightened eschatological anticipation, though explicit evidence from that period for the

21 Josephus does not state unambiguously that the year-long comet appeared over Jerusalem, but ὑπὲρ τὴν πόλιν may pertain to it as well. 22 For a different brand of symbolic interpretation compare Cyril of Jerusalem’s description of the luminous cross which appeared over Jerusalem in 351 CE. See O. Irshai, “Cyril of Jerusalem: The Apparition of the Cross and the Jews,” in Contra Judaeos: Ancient and Medieval Polemics between Christian and Jews (ed. O. Limor and G. Stroumsa; TSAJ 10; Tübingen: Mohr Siebeck, 1996), 85–104. A similar staurophany is reputed to have taken place in 363 CE, marking the failure of Julian’s attempt to rebuild the Temple. See J. W. Drijvers, “The Power of the Cross: Celestial Cross Appearances in the Fourth Century,” in The Power of Religion in Late Antiquity (ed. A. Cain and N. Lenski; Surrey: Ashgate, 2009), 237–48. 23 Note that the Passover aura is not the final omen recorded by Josephus, who proceeds to describe others which continued until the Shavuʿot festival. 24 For Isaiah 60 see O. Michel and O. Bauernfeind, Flavius Josephus: De bello Judaico. Der jüdische Krieg (Darmstadt: Wissenschaftliche Buchgesellschaft, 1969), 2.2:180–82. Cf. M. Kister, “Legends of the Destruction of the Second Temple in Avot de-Rabbi Nathan,” Tarbiz 67 (1998): 521–22 (in Hebrew). Schalit interprets the omens in terms of Num 24:17. See A. Schalit, “Die Erhebung Vespasians nach Flavius Josephus, Talmud und Midrasch: Zur Geschichte einer messianischen Prophetie,” ANRW 2:243–46. 25 See M. Fishbane, Biblical Text and Texture: A Literary Reading of Selected Texts (Oxford: Oneworld, 2003), 121–40. 280 Newman expectation of the Messiah’s appearance on Passover is admittedly slim.26 In the New Testament, the Passover of Jesus’s crucifixion is, of course, rich in sote- riological significance. It will be helpful at this point to cite a passage from the Divine Institutes of Lactantius, in which we find a different combination of some of the apocalyptic elements we have just seen in Josephus. Describing the second coming of Christ, Lactantius writes:

Then, at darkest midnight, the centre of heaven will open, so that the light of God descending is visible throughout the world like lightening. The Sibyl has announced this in verse as follows: “When he comes, there will be lurid fire at black midnight.”27 This is the night that we shall celebrate watching for the advent of our king and God. It has a double meaning: on that night he regained life after his passion, and on that night he will regain his kingship of the earth. . . . Before he descends he will give the following sign. A sword will suddenly fall from the sky, so that the just may know that the leader of the holy army is about to descend, and he will come with angels accompanying him to the centre of the earth, and in front of him will go an inextinguishable flame, and the virtue of the angels will put into the hand of the just all that host which besieged their mountain.28

26 See especially R. Le Déaut, La nuit pascale: Essai sur la signification de la Pâque juive à partir du Targum d’Exode XII 42 (AnBib 22; Rome: Institut Biblique Pontifical, 1963). Cf. the critique of C. Leonhard, The Jewish Pesach and the Origins of the Christian Easter: Open Questions in Current Research (SJ 35; Berlin: Walter de Gruyter, 2006), 317–424. In favor of an early provenance for the expectation that the final redemption will take place on Passover see M. Kister, “Legends of the Destruction,” 513 n. 174; note his reference to Jer 38:8 (LXX) (= MT 31:7). On the connection between such expectations and the date of the omen in Josephus, see C. Mézange, “Josèphe et la fin des temps,” in Le Temps et les Temps dans les littératures juives et chrétiennes au tournant de notre ère (ed. C. Grappe and J.-C. Ingelaere; JSJSup 112; Leiden: Brill, 2006), 215. 27 The passage is not found in the extant Sibylline Oracles. Paul Alexander compares it to the Oracle of Baalbek on the coming of Christ after the death of Enoch and Elijah: “And then he who was crucified on the wood of the cross will come from the heavens, like a great and flashing star, and he will resurrect these two men.” See P. J. Alexander, The Oracle of Baalbek: The Tiburtine Sibyl in Greek Dress (Washington, D.C.; Dumbarton Oaks, 1967), 22, 29, 116–17; P. F. Beatrice, Anonymi Monophysitae Theosophia: An Attempt at Reconstruction (VCSup 66; Leiden: Brill, 2001), xviii–xix, 71. 28 Lactantius, Divine Institutes 7.19.2–5. The translation is taken from Lactantius, Divine Institutes (trans. A. Bowen and P. Garnsey; Liverpool: Liverpool University Press, 2003), 428. Stars Of The Messiah 281

According to Lactantius, Christ will return to earth on the night of Easter. At midnight, the center of heaven will open and “the light of God descending” will illuminate the world like lightening. Before his descent, Christ will send a sword falling from the sky as a sign. Then he will descend to Jerusalem (“the centre of the earth”),29 accompanied by an army of angels and preceded by fire. We recognize some of this, mutatis mutandis, from Josephus’s omens, which were spread over an uncertain span of time. In both texts we encounter a heavenly sword as a portent of approaching salvation. Lactantius writes that the brilliant light of God will appear on Easter—the equivalent of Passover— though it illuminates the entire world, not merely Jerusalem or the Temple, as in Josephus.30 Next he says that Christ and his angelic army will descend to Jerusalem on Easter preceded by fire. Josephus reported the appearance of celestial soldiers and chariots in the skies of Judea shortly after Passover of 66 CE.31 Though Lactantius is reminiscent of Josephus, nothing suggests that he is dependent on him. Many have correctly noted the affinity of his description to Revelation 19, but even that is not sufficient to explain all that we find in his apocalypse, which draws on a broader range of tradition. Upon closer examination, we find that the sword of Lactantius indeed has its roots in an early Jewish Passover tradition—one that is attested in Second Temple literature. In an analysis of the passage in the Divine Institutes, David Flusser demonstrated the popularity of the motif of the heavenly sword at the End of Days, bringing numerous examples from texts spanning a period from Second Temple times to the end of late antiquity, though he failed to mention Josephus. Yet none of the sources he presented associates the sword specifically with Passover; some even date its appearance to the month of Kislev.32 There are, however, other texts which indicate a connection between

See the edition, with commentary, of S. Freund, Laktanz, Divinae institutiones. Buch 7: De vita beata (Texte und Kommentare 31; Berlin: de Gruyter, 2009), 166–67, 489–98. 29 On Jerusalem as the center of the earth see Divine Institutes 7.24.6, where Lactantius also quotes the Sibyl on the brilliant light of the eschatological city. 30 For the anticipation of the parousia on Easter see Freund, Laktanz, 493–95. 31 Jewish War 6:298. Cf. Tacitus, Histories 5.13.1: “Contending hosts were seen meeting in the skies, arms flashed and suddenly the Temple was illuminated with fire from the clouds.” See M. Stern, Greek and Latin Authors on Jews and Judaism (3 vols.; Jerusalem; Israel Academy of Sciences and Humanities, 1974–1984), 2:23, 31, 60; Schmidt, “Signes et prodiges chez Flavius Josèphe et Tacite,” 273–74. 32 Flusser, Judaism and the Origins of Christianity, 430–33. For the appearance of the sword in Kislev see Flusser’s references to the Book of Elijah—a Hebrew apocalypse of late antiquity—and to the liturgical poem, “In Those Days and at That Time,” attributed by some to Eleazar berabbi Kalir. Freund, Laktanz, 495, hastily dismisses Flusser’s analysis 282 Newman the divine sword and the first Passover, one of the typological models for the final redemption. Ironically, it was Flusser himself who elsewhere called attention to these passages, citing remarks of his teacher Hans Jacob Polotsky, but without reference to Lactantius.33 Flusser analyzed the obscure midrash in the Passover Haggadah on Deut 26:8. The verse reads: “The Lord freed us from Egypt by a mighty hand, by an outstretched arm and awesome power, and by signs and portents.” The Haggadah explains “by an outstretched arm” to mean “the sword,” bringing as a proof text 1 Chr 21:16: “David looked up and saw the angel of the Lord standing between heaven and earth, with a drawn sword in his hand directed against Jerusalem.” Polotsky and Flusser observed that the notion of the Logos34 bearing the divine sword to smite Egypt on the night of the Exodus is found already in Wis 18:14–16, which even alludes to the same verse in Chronicles:

While all things were enveloped in peaceful silence and night was mid- way through her swift course, your all-powerful Logos, out of the heavens, from the royal throne, leaped like a relentless warrior into the midst of the land marked for destruction, bearing your unambiguous decree as a sharp sword. Standing it filled all things with death; it touched the heavens, yet stood poised upon the earth.35

From this comparison with earlier Jewish materials we learn that Lactantius’s paschal sword is in effect nothing less than the sword of the first Exodus,

of the sword motif as being derived from what he cavalierly describes as a “spätjüdischen Tradition.” The parallels he brings from the Sibylline Oracles (3:672–673; 798–799; 4:173– 174) are not as apt as those garnered by Flusser and are not grounds for their dismissal. Freund correctly notes in this context the omen reported by Josephus. 33 D. Flusser, “Not by Means of an Angel . . .,” Turei Yeshurun 29 (1972): 18–21 (in Hebrew). 34 On the Logos here see M. Kister, “A Contribution to the Interpretation of Ben Sira,” Tarbiz 59 (1990): 364 n. 225 (in Hebrew). 35 D. Winston, The Wisdom of Solomon (AB 43; New York: Doubleday, 1979), 313. Scholars have noticed the similarity of this passage to Rev 19:13–15, 21, which describes the sword in the mouth of the Logos (the “king of kings”), who appears with his celestial cavalry at the End of Days to do battle. The midrash of the Haggadah is found in a different form in Sifre Num. 115 (Sifre on Numbers and Sifre Zuta [ed. H. S. Horovitz; Leipzig: Fock, 1917), 128) and its parallels. Some regard the tradition in the Haggadah as secondary; see E. D. Goldschmidt, The Passover Haggadah: Its Sources and History (Jerusalem: Bialik Institute, 1969), 45 (in Hebrew). Regardless of which came first, it is clear, as Polotsky and Flusser have shown, that the midrash in the Haggadah is in its own way coherent, and that its origins must be sought in pre-Destruction tradition. Stars Of The Messiah 283

reenlisted as an instrument of the final redemption.36 Both the celestial sword of Lactantius and that of Josephus emerge from the context of eschatological motifs of Second Temple Judaism. I have deliberately deferred till now any discussion of Lactantius’s sources. He himself appeals repeatedly in Book 7 of the Divine Institutes to the Sibylline Oracles, to Hermes Trismegistus, and—less frequently—to the so-called Oracles of Hystaspes.37 The extent of his debt to Hystaspes and his faithful- ness to that putative source is contested by scholars. This debate feeds into a vicious circle of disagreement over the very nature and unity of the Oracles of Hystaspes, which by analogy with the Sibylline Oracles are perhaps better construed as a literary cluster. There is no agreement over their origins. They are identified variously as a Persian Zoroastrian work, a syncretistic Hellenistic–Oriental composition, a Jewish apocryphon, or some combination of these; there is further evidence of a Christian revision of the earlier text or texts.38 Flusser argues that large eschatological portions of Book 7 of the Divine Institutes—including the passage quoted above—were taken virtually verbatim from Hystaspes, which, he contends, was a Jewish apocalypse of Second Temple times to which was added a thin Christian veneer.39 Because of the highly speculative nature of Flusser’s source criticism, I am reluctant to rely upon his conclusions, though they would be congenial to my argument. To my mind, the more important contribution of Flusser’s reading of the Divine Institutes is his demonstration of the indispensability of Jewish sources in uncovering the tradition history of Lactantius’s eschatology. That is the path I have pursued here, whatever the degree of dependence—if any—on the lost book of Hystaspes. Reflecting back on Josephus, we should recognize the importance of reading his account in aggregate as a collection of omens that

36 Earlier in the same book Lactantius alludes to the paschal typology of the second coming: “Just as signs were made at the time to warn the Egyptians of the disaster threatening them, so at the end of time there will be extraordinary portents in every element of the world so that all the nations may know of the imminence of the end” (Divine Institutes 7.15.6 [trans. Bowen and Garnsey, 422]). 37 On Lactantius’s sources see Freund, Laktanz, 33–71. 38 See the survey of W. Sundermann, “Hystaspes, Oracles of,” in Encyclopaedia Iranica (London: Routledge & Kegan Paul, 1982–), 12:606–9; P. F. Beatrice, “Le livre d’Hystaspe aux mains des Chrétiens,” in Les syncrétismes religieux dans le monde méditerranéen antique (ed. C. Bonnet and A. Motte; Études de Philologie, d’Archéologie et d’Histoire Anciennes 36; Brussels: Brepols, 1999), 357–82. 39 D. Flusser, “Hystaspes and John of Patmos,” in Judaism and the Origins of Christianity, 390– 453. For a discussion of Lactantius and Hystaspes, including a critique of Flusser’s thesis, see Freund, Laktanz, 53–69; Beatrice, “Le livre d’Hystaspe aux mains des Chrétiens.” 284 Newman has its own coherence within the context of known patterns of apocalyptic expectation in Second Temple times. These patterns also inform Lactantius’s source or sources.

3 The Star of Bar Kokhba

With the possible exception of Jesus, no messianic claimant of antiquity was identified more consistently with a messianic star than Shimʿon b. Kosba (or Kosiba), popularly known by the epithet attributed to him in patristic literature: Bar Kokhba, the “Son of the Star.” The earliest Christian testimonies are almost contemporary with the events themselves. Justin Martyr refers once by name to Βαρχωχεβας, leader of the then recent Jewish rebellion.40 Eusebius refers twice more to Bar Kokhba: once in the Ecclesiastical History, where he apparently draws on a lost account by Ariston of Pella, and once in the Chronicle, as preserved in Jerome’s translation.41 According to a famous rabbinic tradition, Rabbi Akiba, the preeminent sage of his generation, identified Bar Kosba (or rather Bar Kozba, “the liar,” as he was dubbed retroactively in talmudic literature in light of the failure of the revolt) as the Messiah and saw in him the fulfillment of Balaam’s prophecy of the eschatological star in Num 24:17: “R. Shimʿon b. Yoḥai (said): My master Akiba interpreted: ‘A star shall come forth from Jacob’—Kozba shall come forth from Jacob. When R. Akiba saw Bar Kozba, he would say: ‘This is the messianic king.’ ”42

40 Justin, 1 Apol. 31.6. Justin is quoted by Eusebius, Hist. eccl. 4.8.4. Wishing to dispel any suspicion that Bar Kokhba’s cause was one of revolutionary messianism, Leo Mildenberg clutches at straws to cast doubt on the reading of the name of the leader of the revolt in the Apology; see L. Mildenberg, The Coinage of the Bar Kokhba War (Aarau: Sauerländer, 1984), 79–80. I see no merit to Stefan Beyerle’s assertion that the patristic sources are fatally unreliable a priori; see S. Beyerle, “‘A Star Shall Come out of Jacob’: A Critical Evaluation of the Balaam Oracle in the Context of Jewish Revolts in Roman Times,” in van Kooten and van Ruiten, The Prestige of the Pagan Prophet Balaam, 170 n. 30. 41 Eusebius, Hist. eccl. 4.6.2; for the Chronicle see Die Chronik des Hieronymus (ed. R. Helm; GCS 47; Berlin: Akademie-Verlag, 1956), 201. On Eusebius and Ariston of Pella see A. Carriker, The Library of Eusebius of Caesarea (VCSup 67; Leiden: Brill, 2003), 190–93. 42 Y. Taʿanit 4:8 (68d). Cf. Lam. Rab. 2:4 = Lam. Rab. 2:2 (ed. S. Buber; Vilna: Romm, 1899), 101. Peter Schäfer argues that the attribution of the messianic declaration to Akiba is a secondary textual development, though he, too, dates the notion itself to the time of the revolt. See P. Schäfer, Der Bar Kokhba-Aufstand (TSAJ 1; Tübingen: Mohr Siebeck, 1981), 168–69; idem, “Bar Kokhba and the Rabbis,” in The Bar Kokhba War Reconsidered: New Perspectives on the Second Jewish Revolt against Rome (ed. P. Schäfer; TSAJ 100; Tübingen: Stars Of The Messiah 285

Let us examine the passage in Eusebius’s Ecclesiastical History:

At that time, the leader of the Jews was a man by the name of Barchochebas, which means “star.” For the rest, he was a murderous man and a brigand, but according to his epithet, as if before captive slaves (οἷα ἐπ᾿ ἀνδραπόδων),43 he spoke marvels as a luminary having descended to them from heaven to shine upon the afflicted (κακουμένοις τε ἐπιλάμψαι).44

According to this account, Bar Kokhba presents himself as a luminous celestial redeemer who has descended to liberate the Jews, described as captive slaves. In context, he is a heavenly warrior, come to free the Jews from Roman subju- gation. We easily recognize here a variation of the motif we encountered in the Divine Institutes, where Lactantius writes of “the light of God descending” upon Jerusalem and of Christ the divine warrior leading a band of angels, who deliver into the hands of the just the enemy who has besieged them. William Horbury has compared this passage to a later midrash in Pesiqta Rabbati 36, where towards the end of a lengthy homily on the eschatological light of Isaiah 60 we read:

Our sages taught:45 When the Messiah is revealed, he will come and stand on the roof of the Temple and announce to Israel and say to them: the time of your redemption has come.” And should ,(ענוים) Humble ones“ you not believe it, look at his46 light shining upon you, as it says: “Arise, shine, for your light has come; the glory of the Lord has shone upon you” (Isa 60:1); and upon you alone it shines and not on the nations of the world, as it says: “Behold, darkness will cover the earth and clouds the

Mohr Siebeck, 2003), 3–4; cf. Mildenberg, Coinage of the Bar Kokhba War, 45, 73–76. I remain unconvinced by Schäfer’s redactional analysis of the passage and his reconstruction of its textual evolution, but the question of attribution to Akiba is of no consequence for the present discussion. 43 See K. L. Gaca, “The Andrapodizing of War Captives in Greek Historical Memory,” TAPA 140 (2010): 117–61. 44 Eusebius, Hist. eccl. 4.6.2. -it is doubtful that this is a genu ,תנו רבנן ,Notwithstanding the opening formula 45 ine Tannaitic tradition. See: J. N. Epstein, Introduction to the Mishnaic Text (2 vols.; 3d ed.; Jerusalem: The Hebrew University Magnes Press, 2000), 2:883 n. 5 (in Hebrew); W. Horbury, Jewish Messianism and the Cult of Christ (London: SCM Press, 1998), 189 n. 100. .of Yalqut Shimʿoni באורי of the manuscripts against באורו Reading 46 286 Newman

nations, but upon you the Lord will shine and his glory will be seen upon you” (Isa 60:2).47

Horbury sees in the astral imagery of both Eusebius and Pesiqta Rabbati inti- mations of the angelic nature of the Messiah.48 Scholars have observed that the effulgent Messiah’s announcement of redemption to the “humble ones” in Pesiqta Rabbati alludes to Isa 61:1, the verse immediately following the (ענוים) extended prophecy on the light of the End of Days:

The spirit of the Lord God is upon me, because the Lord has anointed me; to bind ,(לבשר ענוים שלחני) he has sent me as a herald of joy to the humble לקרא לשבוים) up the wounded of heart, to proclaim release to the captives .liberation to the imprisoned ,(דרור

It seems to me that the nexus of chapters 60 and 61 in Isaiah should likewise be seen as the biblical inspiration for the Eusebian portrayal of Bar Kokhba as a stellar messiah who proclaims, as if to captive slaves, that he has come to shine upon the afflicted.49 This description, as noted above, was probably taken from Ariston of Pella, but we have no idea who Ariston’s source was. I would like to draw attention to another element in the passage from Pesiqta Rabbati. The midrash graphically describes the radiant Messiah standing on the roof of the Temple, something we have not encountered till now. In light of this I wish to consider another body of evidence: the iconography of Bar Kokhba coinage. Silver tetradrachms minted during the revolt bear the image of what is almost universally identified as the facade of the Jerusalem Temple, above which may be found one of a number of symbols. Of greatest interest to

47 Pesiqta Rabbati 36 (ed. M. Friedmann; Vienna: n.p., 1880), 162a–b = (ed. R. Ulmer; 3 vols.; Atlanta: Scholars Press, 1997–2002), 2:836. See A. Goldberg, Erlösung durch Leiden: Drei rabbinische Homilien über die Trauernden Zions und den leidenden Messias Efraim (PesR 34, 36, 37) (Frankfurt am Main: Gesellschaft zur Förderung Judaistischer Studien, 1978), 232–35. 48 Horbury, Jewish Messianism, 92. On stars and angels see J. Ben-Dov, Head of All Years: Astronomy and Calendars at Qumran in Their Ancient Context (STDJ 78; Leiden: Brill, 2008), 27–28; B. H. Reynolds, Between Symbolism and Realism: The Use of Symbolic and Non-Symbolic Language in Ancient Jewish Apocalypses 333–63 BCE. (JAJSup 8; Göttingen: Vandenhoeck & Ruprecht, 2012), passim. 49 On Isa 61:1–3 see also J. J. Collins, “A Herald of Good Tidings: Isaiah 61:1–3 and Its Actualization in the Dead Sea Scrolls,” in The Quest for Context and Meaning: Studies in Biblical Intertextuality in Honor of James A. Sanders (ed. C. A. Evans and S. Talmon; Biblical Interpretation Series 28; Leiden, Brill, 1997), 225–40. Stars Of The Messiah 287

Figure 1 Silver tetradrachm of the Bar Kokhba Revolt, 134–135 CE. The Israel Museum, Jerusalem (IMJ 98.96.15174). Used by permission.

us are those commonly identified as representing stars (see Fig. 1); this identification is repudiated, however, by Leo Mildenberg, dean of Bar Kokhba numis- matists. I turn now to this fundamental problem. Mildenberg raises several objections to the identification of the coin symbols as stars.50 One argument is that the putative stars are in fact “rosettes,” a contention based on the false premise that there is a clear-cut iconographic distinction in antiquity between two discrete symbols: floral rosettes and celestial stars. This is not the case. There are some “rosettes” whose context attests unequivocally to their being stars. One of the closest parallels to the

50 Mildenberg, Coinage of the Bar Kokhba War, 43–45; idem, “Bar Kokhba Coins and Documents,” HSCP 84 (1980): 313–15. He is followed by Y. Meshorer, A Treasury of Jewish Coins (trans. R. Amoils; Jerusalem: Yad Izhak Ben-Zvi, 2001), 152–53, 158. 288 Newman more elaborate Bar Kokhba rosettes (with or without scare quotes), which are characterized by alternating spokes and petals, is found in the zodiac mosaic of the Sepphoris synagogue, where the image is patently stellar, but there are also earlier examples.51 To demonstrate the alleged distinction between rosettes and stars, Mildenberg invokes the case of the rosettes adorning temple pediments in the wall paintings of the synagogue at Dura Europos. This, however, begs the question. What do those rosettes in fact represent?52 By analogy, one could cite the numerous Roman coins portraying temples with spiked stars (or rosette-like stars) on their pediments. The most famous of these bear the

51 See Z. Weiss, The Sepphoris Synagogue: Deciphering an Ancient Message through Its Archaeological and Socio-Historical Contexts ( Jerusalem: Israel Exploration Society and the Institute of Archaeology, Hebrew University, 2005), 111–21. For an earlier example, note the rosette-like stars in the zodiac of the first-century temple of Bel in Palmyra; see H. G. Gundel, Zodiakos: Tierkreisbilder im Altertum (Mainz am Rhein: van Zabern, 1992), 104, 219 (no. 45 in the catalogue). See also A. Houghton, “The Seleucid Mint of Mallus and the Cult Figure of Athena Magarsia,” in Studies in Honor of Leo Mildenberg: Numismatics, Art History, Archaeology (ed. A. Houghton et al.; Wetteren, Belgium: Editions NR, 1984), 107–9. Rosettes may also function as solar symbols. On the difficulty of interpreting rosettes, see E. R. Goodenough, Jewish Symbols in the Greco–Roman Period, (13 vols.; New York: Pantheon Books, 1953–1968), 7:179–98. Mildenberg also objects that the fully developed rosette appears only on later dies of the Bar Kokhba tetradrachms, before which time the coins bore what he calls a “cross-rosette.” Assuming, plausibly, that all these symbols represent the same thing, we still have no cause to dismiss their equivalence with a star. For the use of a cross to represent a star see for example H. J. W. Drijvers, Cults and Beliefs at Edessa (Leiden: Brill, 1980), 130, 137 n. 34. Stars with four rays (including the sidus Iulium) are well documented on ancient coins. Note also the motif of a cross set in a diadem, found on certain Herodian coins; I suggest that this derives from the pattern of a star set in a diadem found on earlier Hasmonean coinage. For an assessment of the meaning of the cross (or saltire) and diadem on Herodian coins see D. T. Ariel and J.-P. Fontanille, The Coins of Herod: A Modern Analysis and Die Classification (AJEC 79; Leiden: Brill, 2012), 126. Finally, it should be noted that the star/rosette of the Bar Kokhba tetradrachms is strikingly similar in appearance to the so-called asteriskos, or “little star,” used as a siglum in contemporary Greek literary papyri. See K. McNamee, Sigla and Select Marginalia in Greek Literary Papyri (Papyrologica Bruxellensia 26; Brussels: Fondation Égyptologique Reine Élisabeth, 1992), 11; G. Nocchi Macedo, “Formes et fonctions de l’astérisque dans les papyrus littéraires grecs et latins,” Segno e Testo 9 (2011): 3–33. As for the mysterious wavy line appearing over the Temple facade on other coins, it is a design unto itself—whatever its meaning—and poses no obstacle to a “stellar” reading of the symbols in question. 52 For a solar interpretation see C. H. Kraeling, The Synagogue (Excavations at Dura Europos, Final Reports 8.1; New Haven: Yale University Press, 1956), 236 n. 941. Stars Of The Messiah 289 image of the temple of Divus Iulius decorated with the sidus Iulium, but there are later examples as well.53 What really troubles Mildenberg, however, is the very notion that Shimʿon b. Kosba would have perceived himself or would have been perceived by his followers as a messianic figure. He casts doubt on the reliability of the name given to the leader of the revolt in the earliest Christian sources and summar- ily dismisses all relevant rabbinic evidence as a late invention.54 He argues that the documentary evidence of Shimʿon b. Kosba’s letters, attesting to his preoc- cupation with logistics and military affairs, is incompatible with any form of eschatological leadership. He is captive to a rigid notion of the incompatibility of earthly authority and messianic charisma.55 Plainly uncomfortable with the very idea of Jewish apocalypticism, he concludes that Shimʿon b. Kosba had no messianic aspirations, but was instead “a good Jew.”56 Even allowing for a certain ambiguity of the interpretation of the star/ rosette when viewed in isolation, I contend that there is no mistaking its meaning within the iconographic context of the Bar Kokhba coins, suspended over the roof of the Temple. This meaning emerges not only from the literary sources describing the leader of the revolt (though they are certainly corroborative), but also from everything we have learned concerning the eschatological anticipation—or even alleged realization—of a divine light or luminary over the Temple. The two elements on the silver tetradrachms—the Temple facade and over it the star/rosette—constitute a single coherent iconographic tableau. A striking iconographic parallel to this scheme in a wall painting in the cata- combs of the Villa Torlonia in Rome deserves special attention (see Fig. 2). The date of the painting, found on the back wall of an arcosolium in the upper catacomb, is uncertain; Rutgers is inclined to date it to the years 350–370 CE,

53 For the comparison of these coins with Bar Kokhba coins see P. Romanoff, “Jewish Symbols on Ancient Jewish Coins,” JQR 33 (1942): 3 n. 11; M. Küchler, “Jesus von Nazaret und Schimeʿon ben Kosiba: Zwei ‘Könige der Juden’ und ihre Sterne in Texten und auf Münzen,” in Jesus—Gestalt und Gestaltungen: Rezeptionen des Galiläers in Wissenschaft, Kirche und Gesellschaft (ed. P. von Gemünden et al.; Göttingen: Vandenhoeck & Ruprecht, 2013), 317–45 (esp. 333, 343). Romanoff and Küchler also cite the stars on the coins of Alexander Jannaeus and Herod the Great as antecedents of Bar Kokhba coinage. 54 See above, nn. 40 and 42. 55 This is a common but inadequate distinction. See the well-taken criticism of Schäfer, “Bar Kokhba and the Rabbis,” 17–19. 56 Mildenberg, Coinage of the Bar Kokhba War, 45. See the detailed critique in M. Hengel’s review of Mildenberg in Gnomon 58 (1986): 326–31. 290 Newman

Figure 2 Wall painting in upper catacomb of Villa Torlonia, fourth century. After N. Zimmermann, Werkstattgruppen römischer Katakombenmalerei (JACSup 35; Münster, Westfalen: Aschendorff, 2002), Tafel XV, Abb. 73.

though others put it earlier.57 At the center of the painting stands a Torah shrine. The image is partially damaged, but it is still possible to make out the schematic portrayal of rolled scrolls seen head on as circles resting on shelves, common in Diaspora drawings of Torah shrines. It is flanked by several typical Jewish symbols, including a shofar, palm branch, and citron, but most prominent are two large menorahs, one on either side. Centered above the apex of the Torah shrine is a star with eight rays; to its left and right appear the sun and the moon, both partially obscured by dark clouds painted in black and red.58 If there is ambiguity here, it does not pertain to the star, but rather to the interpretation of the shrine below it. The iconographic significance of what

57 L. Rutgers, The Hidden Heritage of Diaspora Judaism (2d ed.; CBET 20; Leuven: Peeters, 1998), 64–66. 58 See H. W. Beyer and H. Lietzmann, Die jüdische Katakombe der Villa Torlonia in Rom (Studien zur spätantiken Kunstgeschichte 4; Jüdische Denkmäler 1; Berlin: de Gruyter, 1930), 15–27. An excellent color photograph is found in S. Laderman and Y. Furstenberg, “Jewish and Christian Imaging of the ‘House of God’: A Fourth-Century Reflection of Religious and Historical Polemics,” in Interaction between Judaism and Christianity in History, Religion, Art, and Literature (ed. M. J. H. M. Poorthuis et al.; JCP 17; Leiden: Brill, 2009), Figure 4. Laderman and Furstenberg do not discuss the star. Stars Of The Messiah 291 seems to be a Torah shrine or ark flanked by menorahs—so common in synagogal and Jewish funerary art—is a well-known scholarly crux. The image is usually interpreted either as a stereotypic representation of the furnishings of the synagogue or as a highly stylized picture of the Temple—or both.59 I do not wish to generalize, but there are reasons to believe that at least the painting in the catacomb of the Villa Torlonia is meant to evoke the Temple of Jerusalem. This identification rests on more than just the similarity to the Bar Kokhba coins. Beyer and Lietzmann indeed note this connection and conclude from it not only that the shrine of the Villa Torlonia painting represents the Temple,60 but also that by analogy the star above is a messianic symbol.61 No less significant than the similarity to the coins, however, is the manner in which the star suspended over the ark/Temple dominates the overcast sun and moon on either side.62 I suggest that this scene is inspired by the description of the eschatological light of Jerusalem in Isa 60:19–20:

No longer shall you need the sun for light by day, nor the shining of the moon for radiance; for the Lord shall be your light everlasting, your God shall be your glory. Your sun shall set no more, your moon no more withdraw; for the Lord shall be a light to you forever, and your days of mourning shall be ended.63

If this interpretation is correct, then the image also implicitly contains a message of comfort to the mourners of the deceased laid to rest just below the wall painting.64

59 L. Levine, The Ancient Synagogue: The First Thousand Years (2d ed.; New Haven: Yale University Press, 2005), 232–36. 60 In fact, they take it to represent the heavenly Temple, an unlikely hypothesis given the other evidence we have seen attesting to the anticipation of divine or messianic light over earthly Jerusalem at the End of Days. 61 Beyer and Lietzmann, Die jüdische Katakombe, 21–24. 62 Might there also be symbolic significance to the way in which the two menorahs visually dominate the sun and moon from below? 63 See also Isa 24:23: “Then the moon shall be ashamed, and the sun shall be abashed. For the Lord of Hosts will reign on Mount Zion and in Jerusalem, and the Presence will be revealed to His elders.” On Isa 60:19 in Qumran texts see M. Kister, “A Qumran Fragment (4Q392 1) and the Conception of Light in ‘Qumran Dualism,’ ” Meghillot 3 (2005): 138–39 (in Hebrew). 64 I cannot agree with the interpretation of the painting proposed by P. Maser, “Darstellungen des Olam hab-ba in der spätantik-jüdischen Kunst Roms?” in Jenseitsvorstellungen in Antike und Christentum: Gedenkschrift für Alfred Stuiber (JACSup 9; Münster, Westfalen: Aschendorff, 1982), 233–34. Maser believes that the image is inspired by Isa 30:26, but 292 Newman

These verses in Isaiah describe the eschatological light of God, but with the emergence of messianic ideologies, that light came to be associated particularly with the Messiah himself. Returning to Bar Kokhba, it is important to stress the validity of the conclusion that Mildenberg sought so vigorously to avoid: the deliberate iconographic message of the silver tetradrachms confirms that the leader or leaders of the revolt identified in one form or another with the eschatological theme of stellar messianism which we have found in our literary sources.65 In the cases of both Jesus and the omens of 66 CE, I addressed the question of independent evidence for contemporary, irregular astronomical events.

he fails to take notice of the deliberate obscuring of the sun and moon. For the same reason, I am unconvinced by the suggestion of G. Noga-Banai, “Between the Menorot: New Light on a Fourth-Century Jewish Representative Composition,” Viator 39/2 (2008): 21–48 (esp. 34), who proposes that the star, sun, and moon are introduced, together with the pair of menorahs, to lend the scene chronometric significance. See also S. Fine, Art and Judaism in the Greco–Roman World: Toward a New Jewish Archaeology (Cambridge: Cambridge University Press, 2005), 127–28. The star of the Villa Torlonia catacomb should also be compared to numerous representations in early Christian iconography of the star of the magi and, more generally, to stars in other biblical scenes. On these see R. J. Pillinger, “Dead Sea Scrolls and Early Christian Art,” in The Dead Sea Scrolls in Context: Integrating the Dead Sea Scrolls in the Study of Ancient Texts, Languages, and Cultures (ed. A. Lange et al.; 2 vols.; VTSup 140; Leiden: Brill, 2011), 2:758–59. Dr. Norbert Zimmermann has kindly drawn my attention in particular to the similarity between the star of the Villa Torlonia and stars in two paintings in Cubiculum O of the Christian catacomb of the Via Latina: one of the splitting of the Red Sea, the other of an unidentified figure (perhaps Balaam) pointing to a star overhead. See N. Zimmermann, Werkstattgruppen römischer Katakombenmalerei (JACSup 35; Münster, Westfalen: Aschendorff, 2002 ), 105, with figures 64 and 65 on Plate XII. 65 This conclusion is consistent with the messianic overtones of Bar Kokhba’s title of choice: nasi. The best discussion of this term remains D. Goodblatt, “The Title Nasiʾ and the Ideological Background of the Second Revolt,” in The Bar Kokhba Revolt: New Studies (ed. A. Oppenheimer and U. Rappaport; Jerusalem: Yad Izhak Ben-Zvi, 1984), 113–32 (in Hebrew). In contrast to the coins, a case can be made that the rosettes which decorated the lead weights produced by Bar Kokhba’s administration did not carry any eschatological meaning but were merely copied from a Hadrianic model. See A. Kloner, “Lead Weights of Bar Kokhba’s Administration,” IEJ 40 (1990): 58–67; R. Deutsch, “A Lead Weight of Hadrian: The Prototype for the Bar Kokhba Weights,” Israel Numismatic Journal 14 (2000–2002): 125–28. Cf. also the grafitto of a star or pentagram discovered by Joseph Patrich in a refuge cave in Wadi Suweinit. Patrich dated the find to the Great Revolt, whereas Hanan Eshel argued that it should be dated to the Bar Kokhba Revolt. See H. Eshel, “Gleaning of Scrolls from the Judean Desert,” in The Dead Sea Scrolls: Texts and Context (ed. C. Hempel; STDJ 90; Leiden: Brill, 2010): 77–78. Stars Of The Messiah 293

What, if anything, do we know of such phenomena at the time of the Bar Kokhba Revolt?66 The most precise records of astronomical observations from that period come from the Far East. We have, it turns out, reliable evidence in Chinese records for the appearance of a suitable comet in January of 132, that is to say, coinciding roughly with the outbreak of the revolt.67 There was no short- age of contributing causes to the Bar Kokhba Revolt, and I am not suggesting that this was one of them, or even that it was a precipitating event that tipped the scales towards war when all other pieces were in place. I assume the uprising would have taken place with or without a comet. I do suspect, however, that its appearance at that time contributed to the cultivation of the particular messianic typology which characterized the public persona of the leader of the ensuing rebellion. There was at the time another figure associated with a miraculous star. Antinous of Bithynia, Hadrian’s young lover, drowned mysteriously in the Nile in October of 130.68 Following his death—so says Cassius Dio—Hadrian claimed to have seen a new star in the heavens, which he understood to be none other than the spirit of Antinous.69 Hadrian deified the dead youth and established a religious cult in his honor; coins were struck bearing the image of Antinous, and on some appeared a star signifying his apotheosis.70 We do not know what Hadrian really observed in the night sky or precisely when he saw it. The comet of 132 has been suggested as a possibility, though that seems unlikely since Cassius Dio implies an earlier date.71

66 What follows is a revised version of an argument I first made in a short paper entitled “The Star of Bar Kokhba,” in New Studies on the Bar Kokhba Revolt (ed. H. Eshel and B. Zissu; Ramat Gan: Bar-Ilan University, The Martin [Szusz] Department of Land of Israel Studies, Jeselsohn Epigraphic Center of Jewish History, 2001), 95–99 (in Hebrew). 67 For the Chinese evidence see Ramsey, A Descriptive Catalogue of Greco–Roman Comets, 165–66. I wish to express my thanks to Prof. Christopher Cullen of the Needham Research Institute for responding to my queries about the Chinese astronomical records. 68 See R. Lambert, Beloved and God: The Story of Hadrian and Antinous (London: Weidenfeld and Nicolson, 1984); H. Meyer, Antinoos: Die archäologischen Denkmäler unter Einbeziehung des numismatischen und epigraphischen Materials sowie der literarischen Nachrichten (Munich: Fink, 1991); A. R. Birley, Hadrian: The Restless Emperor (London: Routledge, 1997), 235–58. 69 Cassius Dio, Hist. 69.11.4. 70 Lambert, Beloved and God, 150; Meyer, Antinoos, 137–50. On a bust of Antinous found in Caesarea Philippi (Banias), see J. F. Wilson, Caesarea Philippi: Banias, the Lost City of Pan (London: Tauris, 2004), 41. 71 See J. R. Rea, ed., The Oxyrhynchus Papyri (London: Egyptian Exploration Society, 1898–), 63 (1996): 13–14, on POxy 4352; Ramsey, A Descriptive Catalogue of Greco–Roman Comets, 165–66. 294 Newman

We find ourselves faced by two remarkably contrapuntal sets of circumstan- tial evidence begging to be compared. Each leader of this conflict happened to be invested to some degree in the same powerful symbol. After Hadrian sent Antinous to heaven to become a star and a god, Shimʿon b. Kosba descended, as it were, from his own star as Bar Kokhba to lead a revolt against Rome. We have no explicit evidence for the consequences of this combination of antago- nistic events and symbols, but we may pose the question—and wonder. Is this mere happenstance—coincidental cases of A Star is Born—or do our sources also betray a touch of Star Wars?72

4 The Year of the Messianic Star and the Star of the Magi, Once Again

There is another Jewish source which apparently dates to the period between the two Judean revolts and which refers to the anticipation of an eschatological, redemptive star that is to descend and punish Rome. In the fifth Sibylline Oracle we read:

But when, after the fourth year (ἀλλ᾿ ὅταν ἐκ τετράτου ἔτεος), a great star shines which alone will destroy the whole earth, because of the honor which they first gave to Poseidon of the sea, a great star will come from heaven to the wondrous sea and will burn the deep sea and Babylon itself and the land of Italy, because of which many holy faithful Hebrews and a true people perished.73

Babylon serves here, as elsewhere, as a transparent epithet for Rome, which will be destroyed for what it has done to the Jews. The oracle mysteriously dates this retribution to a time “after the fourth year.” In the opinion of John Collins, the prophecy alludes to the eschatological heptad of years, the fundamental

72 Newman, “The Star of Bar Kokhba,” 98–99. Cf. the semi-fictional account of E. Speller, Following Hadrian: A Second-Century Journey through the Roman Empire (Oxford: Oxford University Press, 2003), 276. On Hadrian, Antinous, Bar Kokhba and the Warren Cup—a silver vessel found at Beitar, the final stronghold of the revolt—see H. Eshel, “A Silver Chalice with Homoerotic Images Discovered in Beitar in 1906,” Jerusalem and the Land of Israel 8–9 (2013): 233–40 (in Hebrew). 73 Sibylline Oracles 5:155–161. The translation is that of J. J. Collins, OTP 1:397. See also J. J. Collins, The Sibylline Oracles of Egyptian Judaism (SBLDS 13; Missoula, Mont.: Society of Biblical Literature, 1974), 89–92. For various opinions on questions of dating, sources, and redaction see S. Felder, “What is the Fifth Sibylline Oracle?” JSJ 33 (2002): 363–85; Beyerle, “A Star Shall Come out of Jacob,” 179–88. Stars Of The Messiah 295 unit of time in Daniel’s calculations of the eschaton. Specifically, Collins takes the expression to refer to the three-and-one-half years of the dominion of the Fourth Kingdom (here identified with Rome), after which, according to Dan 7:25–26, its power will pass to Israel.74 This is a plausible interpretation; in fact, I hope to show that there are later Jewish sources which essentially corroborate it. It is important to note, however, that four is not really equivalent to three and a half. The oracle plainly states that the star will appear only after the fourth year, that is to say in the fifth year, or perhaps later. The eschatological heptad appears repeatedly in rabbinic literature and in extrarabbinic, Jewish apocalyptic sources of late antiquity and the Middle Ages, and it is to some of these that we will turn for comparison and thematic analysis.75 The most important of the Jewish apocalypses of late antiquity, Sefer Zerubbabel, is generally dated to the seventh century. I have argued elsewhere that the first recension of the book actually dates to the sixth century and that the best witnesses to that recension, though occasionally betraying a later hand, are the editio princeps (Constantinople, 1519) and an unpublished manuscript in the Jewish Theological Seminary of New York.76 In Sefer Zerubbabel there are fragmented hints of a seven-year sequence, but there is not a discrete, coherent account of events for each year. While the book has its own eschatological star, it does not conform to the patterns we have examined till now. We read of a special star that precedes Hefzibah, mother of the Davidic Messiah, and lights her way as she marches forth to kill two enemy kings who threaten Israel. Hefzibah makes her appearance, together with the star, in the sixth year of the heptad. The narrative invention of the star of Hefzibah is perhaps inspired by Isa 62:1–4, where we read:

74 Collins, The Sibylline Oracles of Egyptian Judaism, 89; cf. ibid., OTP 1:397 n. m2. 75 There are many different versions of the seven-year scheme of salvation; in each, the stages of apocalyptic redemption are listed year by year. For references, see D. Sperber, Masekhet Derekh Eretz Zuta and Perek Hashalom (3d ed.; Jerusalem: Tzur-Ot, 1994), 158– 59 (in Hebrew). There is evidence that at least some Jews expected this final heptad to coincide with the sabbatical year cycle; see B. Z. Wacholder, “Chronomessianism: The Timing of Messianic Movements and the Calendar of Sabbatical Cycles,” HUCA 46 (1975): 201–18. Wacholder’s article must be read with caution due to the idiosyncrasy of his chronology for the sabbatical year cycle. My own attempt at chronomessianic speculation about that cycle in Newman, “The Star of Bar Kokhba,” 97, is best ignored. 76 H. I. Newman, “Dating Sefer Zerubavel: Dehistoricizing and Rehistoricizing a Jewish Apocalypse of Late Antiquity,” Adamantius 19 (2013): 324–36. The text of the first edition is reprinted, with occasional errors, in Y. Even-Shemuel, Midreshei Geʾulah (2d ed.; Jerusalem: Bialik Institute, 1954), 379–82. 296 Newman

For the sake of Zion I will not be silent; for the sake of Jerusalem I will not be still, till her victory emerge resplendent and her triumph like a flaming Nevermore shall you be . . . .(עד יצא כנגה צדקה וישועתה כלפיד יבער) torch called “Forsaken,” nor shall your land be called “Desolate”; but you shall be called Hefzibah (= “I delight in her”), and your land “Espoused.”

In the Secrets of Rabbi Shimʿon b. Yoḥai, a Hebrew apocalypse of the eighth century, we find a systematic description of the eschatological heptad:

And this will be your sign, that you will see one week [i.e., of years]: At its beginning—rain. And in the second (year), arrows of hunger are fired. And in the third, great famine. And in the fourth, neither famine nor plenty. And in the fifth, great plenty, and a certain star will rise from the ,and that is the star of Israel ,(ובראשו שבט) East with a staff at its head as it says: “A star shall come forth from Jacob.” If it lingers, it is to Israel’s advantage, and then the Messiah Son of David will arise.77

of years, but the (שבוע) The opening of the passage refers to a Danielic week details of the list itself extend only as far as the fifth year. Here, as in the fifth Sibylline Oracle, the star appears in the fifth year. The Prayers of R. Shimʿon b. Yoḥai is a composite Jewish apocalypse of contested provenance whose final compilation extends into the period of the Crusades but which contains earlier material, including the following description of the heptad:

77 The most accessible edition of the text may be found in BHM 3:82, which was copied from the edition of Salonika, 1743 (this was not the first edition, as is repeatedly claimed in the scholarly literature; it was in fact copied in turn from the editio princeps, printed in Ferrara in 1555). For an introduction and translation see J. C. Reeves, Trajectories in Near Eastern Apocalyptic: A Postrabbinic Jewish Apocalypse Reader (SBLRBS 45; Atlanta: Society of Biblical Literature, 2005), 76–89. I have not followed Reeves’s translation here; note that he mistakenly understands the text to refer to a sequence of weeks, not a “week” of years. Remarkably, an eschatological sequence of famine and infertility followed by a period of fecundity and abundance and finally by the appearance of a star is also found in an Armenian apocryphon, “The Vision of Enoch the Just.” See J. Issaverdens (trans.), The Uncanonical Writings of the Old Testament Found in the Armenian Mss. of the Library of St. Lazarus (Venice: Armenian Monastery of St. Lazarus, 1901), 319: “Then shall appear a certain star, having a tail toward the east, which means that there shall be more peace in those parts. And the people of the Jews shall gather together in Mesopotamia and toward the country of Palestine.” The narrative context and chronology in that text differ, however, from those of the Jewish sources cited here. Stars Of The Messiah 297

At the beginning of the one week (of years) there will be no rain. And in the second (year), arrows of famine. And in the third, there will be great famine, and there will be no rain. And in the fourth, medium. And in the fifth there will be great plenty. And in the sixth a certain star will rise from ובראשו שבט של) the East, and at its head will be a staff of fire like a spear The nations of the world will say: “This star is ours.” But that .(אש כמו רומח is not so, rather it is Israel’s, as it says: “A star shall come forth from Jacob.” The time of its shining will be the first watch of the night for two hours, and it will set (and linger) for fifteen days in the East and turn to the west and linger for fifteen days, and if for longer—it is good for Israel.78

The passage is similar but not identical to what we saw in the Secrets of Rabbi Shimʿon b. Yoḥai. What was found there in the fifth year has been split, and here the messianic star appears in the sixth year. There is also an additional section describing the reaction of the Gentiles and details of the star’s course. A short apocalyptic midrash known as Aggadat Hamashiaḥ is incorporated into the eleventh-century Midrash Lekaḥ Tov of Tuviah b. Eliezer. Its origin is unknown. There we a find complete heptad with the star in the fifth year:

The week (of years) in which the Son of David arrives—the first year there is insufficient food. The second, arrows of famine are fired. The third, great famine. In the fourth, neither famine nor plenty. In the fifth, great plenty, and a star will rise from the East that is the star of the Messiah; and it will linger in the East for fifteen days, and if it stays longer, it is to Israel’s advantage. The sixth, thunder and noises. The seventh, wars. Following the seventh, expect the Messiah.79

All of this is reminiscent of what we saw in the previous sources. To these passages we may add one more found in an apocalyptic text preserved in a fragment of an unpublished Judeo-Arabic manuscript from the Cairo Genizah (T–S NS 261.111).80 Though much of the page is missing, from

78 BHM 4:121. See Reeves, Trajectories in Near Eastern Apocalyptic, 89–106 (Reeves again renders this as a series of weeks). On the anthological character of the text as it has reached us see E. E. Urbach, “A Midrash of Redemption from the Last Days of the Crusades,” Eretz Israel 10 (1971): 58–63 (in Hebrew). 79 Pesikta Zutarta (= Midrash Lekaḥ Tov), Balak (Venice, 1546), 58a = BHM 3:141. See Reeves, Trajectories in Near Eastern Apocalyptic, 144–48. 80 The manuscript has been mistakenly described as containing a “[c]alendrical discussion concerning a request for rain”; see A. Shivtiel and F. Niessen, eds., Arabic and Judaeo- Arabic Manuscripts in the Cambridge Genizah Collections: Taylor–Schechter New Series, (Cambridge: Cambridge University Press, 2006), 319. Prof. Gideon Bohak, however, has 298 Newman what remains there is no doubt that it, too, contains a description of the eschatological heptad, towards the end of which appears a star that lingers for fifteen days in the East and seventeen days in the West. Let us examine another element in these texts. In the Prayers of R. Shimʿon b. Yoḥai we encountered the phenomenon of the nations of the world claiming the messianic star for themselves and saying: “This is our star.” A much more elaborate version of such a claim by the Gentiles is found in the anthological Midrash Haggadol on Num 24:19, where, however, it is projected onto the days of Samuel and David:

Num 24:19)—At first a star rose) ”(וירד מיעקב) He shall rule from Jacob“ and Israel saw it ,(ובראשו חרב) from the East and at its head was a sword and they said to one another: “What is this?” The nations of the world asked the astrologers among them and said to them: “What sort of star is this?” They told them: “This is the star of Israel, this is the king who shall arise for them.” When Israel heard that, they approached Samuel the prophet and said to him: “Give us a king to judge us like all the nations” (1 Sam 8:5)—just as the nations said. And concerning him [i.e., David] it says: “A star shall come forth from Jacob” (Num 24:17). Similarly at the end a star will rise from the East which is the star of the Messiah, as it says: “He shall rule from Jacob.” R. Yose said: “In the language of the Arameans they call the East yerd.”81 And it shall linger in the East for fifteen days, and if it stays longer, it is to Israel’s advantage. From here on, expect the footsteps of the Messiah.82

correctly identified this as an apocalyptic text (personal communication). Its serial number in the Friedberg Genizah Project is C 389688/9. אמר רבי יוסה בלשון ארמיא קרן למדנחא :The sentence is odd. Midrash Haggadol reads 81 Cf. Y. T. Langermann, Yemenite Midrash: Philosophical Commentaries on the Torah .ירד (New York: HarperCollins, 1996), 175–76. 82 Midrash Haggadol on Num 24:19; see Midrash Haggadol on the Pentateuch (ed. Z. M. Rabinowitz; Jerusalem: Mosad ha-Rav Kook, 1983), 4:431–32. The last portion of the passage reappears in a Yemenite midrash analyzed by E. Schlossberg, “An Anonymous Yemenite Midrash on the Torah from the Early Sixteenth Century,” Sidra 18 (2003): 144 (in Hebrew). The midrash is found in London: British Library Or. 1481 (F5988 in the Institute of Microfilmed Hebrew Manuscripts of the National Library of Israel). There are errors in Schlossberg’s transcription of the passage, in his punctuation of the text, and in his identification of one of the key biblical verses. The next passage in the manuscript attests to the sighting of a comet, which is taken to be that predicted by our text. On other “messianic” stars reported by the Jews of Yemen in the seventeenth century, see J. Tobi, The Jews of Yemen: Studies in their History and Culture (Leiden: Brill, 1999), 51–56. Stars Of The Messiah 299

There are many points of interest in this passage. Its source is unknown, but it resembles the late antique Hebrew apocalyptic texts of the sort we saw above. In it, Balaam’s prophecy in Numbers 24 is understood to refer to both David and the Messiah, and the midrash establishes an analogy between the two; this Davidic interpretation is known from medieval Hebrew sources.83 In some of the previous texts we saw the star of the Messiah described as a staff or spear, but here we find ourselves confronted once again by a stellar sword, even if it is the sword announcing the kingship of David, not of the Messiah. Remarking on the traces of ancient motifs in this late source, Menahem Kister has already drawn attention inter alia to the similarity between the sign of the sword and what Josephus wrote of the omens before the Great Revolt. Kister also points to the affinity of the interest of the Gentile astrologers in the stellar sword to that of the magi in the Star of the Nativity.84 How are we to interpret this similarity to the story of the magi? In theory it could reflect independent use of an ancient topos; but there is, as I hope to demonstrate, other evidence of a Jewish response in late antiquity to the Christian story of the magi, and I am inclined to read the passage in Midrash Haggadol as a similar response. Commenting on the claim of the Gentiles in the Prayers of R. Shimʿon b. Yoḥai that “this star is ours,” John Reeves has suggested that this “seems to reflect the Christian prophecies associated with the ‘star of the magi’ found in Syriac sources like the Cave of Treasures.”85 This, I believe, is fundamentally correct, but the most compelling evidence has been missed. Let us return to the Secrets of Rabbi Shimʿon b. Yoḥai, where we find in the text of the printed edition: “And in the fifth (year), great plenty, and a certain star will rise from

83 See references in M. M. Kasher, Torah Shelemah (Jerusalem: Beit Torah Shelemah, 1991), 42:140–45 (in Hebrew). Numbers 24:17 was already taken to refer to David and his descendents by the Emperor Julian in his Contra Galilaeos; see Iuliani imperatoris librorum contra Christianos quae supersunt (ed. K. J. Neumann; Leipzig: Teubner, 1880), 212–13 (261E). Julian sought to refute the interpretation of Christians who associated the verse with Christ and the star of the magi. See Dorival, “Un astre se lèvera de Jacob,” 340–41; J. G. Cook, The Interpretation of the Old Testament in Greco–Roman Paganism (STAC 23; Tübingen: Mohr Siebeck, 2004), 301–2. I have encountered no evidence among Jews of similar exegesis identifying the star of Balaam with David prior to the medieval Hebrew sources. I cannot say if the similarity to Julian’s argument is purely coincidental or if it indicates dependence upon a common tradition. 84 Kister, “Legends of the Destruction,” 522–23. 85 Reeves, Trajectories in Near Eastern Apocalyptic, 97 n. 120. 300 Newman

and that is the star of Israel, as it ;(ובראשו שבט) the East with a staff at its head says: ‘A star shall come forth from Jacob.’ ” If we look in the manuscripts, however, we find that an important part of the passage has been omitted. In the Munich manuscript of the Secrets we read:

And in the fifth (year), great plenty, and a certain star will rise from the East with a staff at its head; and the nations of the world will ascend to the top of the mountains and say: “This is our star.” But it is rather the star of Israel, as it says: “A star shall come forth from Israel [sic].”86

A similar reading is also found in other manuscripts.87 This resembles what we have already seen in the Prayers of R. Shimon b. Yoḥai, with one important addition: the ascent of the Gentiles to the top (or tops) of the mountains to claim the star for their savior. This, it turns out, is a key to identifying the pedigree of the tradition. Mountain climbing at the appearance of the star is not an obvious response. We do not catch the Jews doing it, and we must wonder where it comes from. The most elaborate apocryphal legend of the magi in late antiquity is found, with variations, in two sources. The first is the Latin Opus imperfectum in Matthaeum, the work of an Arian author of the mid-fifth century. The second is the Syriac Chronicle of Zuqnin (once commonly known as the Chronicle of Pseudo-Dionysius), an eighth-century composition.88 In the Opus imperfectum we read:

ובחמישי שובע גדול, ויצמח כוכב :München–Bayerische Staatsbibliothek, Cod. Hebr. 222 86 אחד ממזרח ובראשו שבט, ואומות העולם עולין לראש ההרים ואומ' זה כוכבינו הוא, .ואינו אלא כוכבו של ישראל, שנ' דרך כוכב מישראל ואומות העולם רואין אותו ומנירים )?( ועולים :Thus Oxford–Bodleian Library Heb.d.46/11 87 Parma–Biblioteca Palatina Cod. Parm. 3122 .לראשי ההרים ואומרים זה הוא כוכבינו (= De Rossi 1240; F12284 in the Institute of Microfilmed Hebrew Manuscripts), fol. 228b, the nations ;מנירים measure”—in place of“—מודדים :has the more plausible reading of the world measure the position of the star and thereby determine where to ascend. Cf. Parma–Biblioteca Palatina Cod. Parm. 2342 (= De Rossi 541; F13218 in the Institute of Microfilmed Hebrew Manuscripts), fol. 201b. For a transcription of the Parma manuscripts by A. Jellinek see I. Desheh, ed., Sefer Yalkut Midrashim: Otzar Midreshei Hazal (4 vols.; Tzefat: Or Olam, 2007), 4:77. 88 The most recent full length studies of these texts are by B. C. Landau, “The Sages and the Star Child: An Introduction to the Revelation of the Magi, an Ancient Christian Apocryphon” (Ph.D. diss., Harvard Divinity School, 2008) (I am grateful to Sergey Minov for sharing with me a copy of this work); idem, Revelation of the Magi: The Lost Tale of the Wise Men’s Journey to Bethlehem (New York: HarperCollins, 2010). See the review of Stars Of The Messiah 301

[T]here was a certain race situated at the beginning of the East next to the ocean, in whose possession was a certain writing ascribed to Seth concerning the appearance of this star and the kind of gifts to be offered it which was considered to have been handed down from fathers to their sons through generations of studious men. . . . They were called magi in their own tongue, because they glorified God in silence and unspoken voice. Year after year, after the threshing harvest, they would ascend on top of a certain mountain situated there, which was called in their language “Mount of Victory.” . . . So they were doing for generations, always waiting, if perchance in their generation that star of blessedness would arise, until at last the star appeared to them, descending on top of that Mount of Victory, containing the form of a small boy and having the image of the cross above itself.89

The tale in the Chronicle of Zuqnin is much longer, but it will suffice to quote a short excerpt here:

And those books of hidden mysteries were placed on the Mountain of Victories in the east of Shir, our country, in a cave, the Cave of Treasures of the Mysteries of the Life of Silence. And our fathers commanded us as they also received from their fathers, and they said to us: “Wait for the light that shines forth to you from the exalted East of the majesty of the Father, the light that shines forth from on high in the form of a star over the Mountain of Victories and comes to rest upon a pillar of light within the Cave of Treasures [of] Hidden Mysteries.”90

Indeed, when the star finally appeared, the magi ascended the Mountain of Victories to the Cave of Treasures.

Landau’s book by A. Y. Reed in Sino–Platonic Papers 208 (February 2011): 36–54. Reed is particularly critical of Landau’s assessment of the legend’s origins. She challenges his early dating and his denial of Eastern influences. 89 Opus imperfectum in Matthaeum, Homily 2 (PG 56:637–38). The translation is taken, with modifications, from J. A. Kellerman (trans.) and T. C. Oden (ed.), Incomplete Commentary on Matthew (Opus Imperfectum) (2 vols.; Downers Grove, Ill.: InterVarsity Press, 2010), 1:32. 90 For the Syriac see Landau, “The Sages and the Star Child,” 30 = Chronicon anonymum Pseudo-Dionysianum vulgo dictum (ed. J.-B. Chabot; 4 vols.; CSCO 91, Scriptores Syri 43; Louvain: Imprimerie Orientaliste, 1953), 1:59. The translation is that of Landau, Revelation of the Magi, 38. 302 Newman

According to the Syriac Christian legend, the magi await the appearance of the star of Christ, which is to be revealed on the Mountain of Victories in the East;91 at its epiphany they ascend to the Cave of Treasures atop the mountain.92 The legend was incorporated into Jewish eschatological tradition with a polemical twist. The Secrets of Rabbi Shimʿon b. Yoḥai predicts the advent of the star as a prelude to the coming of the Messiah. When the star appears in the East, the Gentiles will ascend to the top of the mountains and claim the star as their own, but they will of course be wrong, for it is the star of Israel. This thematic confrontation lends credibility to speculation about literary responses to the star of the magi in some of our other sources, such as Midrash Haggadol. Ironically, we have come round full circle to the star of the nativity. Though it emerged out of Second Temple Judaism, it comes back like so much else as a Doppelgänger in a battle of competing eschatologies.

5 Conclusions

As I explained at the outset, my purpose in this study has been primarily to give an account of the role of stars in Jewish eschatology, not as metaphors or abstractions, but as agents or omens of salvation. I have sought to describe both the continuity and the transformation of their role over time in apocalyptic speculation and realized eschatology, beginning with Second Temple Judaism and continuing through late antiquity and beyond. We have been able to recognize traditional concepts which found their way into new textual and historical contexts, thereby acquiring new significance. The comparison of Jewish and Christian sources has revealed shared conceptions, but what they hold in common leads almost inevitably to confrontation and to competition for divine validation.

91 In Matt 2:2 the magi report seeing a star ἐν τῇ ἀνατολῇ. Commentators have noted that this is better understood as “at its rising,” than “in the East.” See for example Davies and Allison, Commentary, 236. Nevertheless, in the Latin and Syriac versions and dependent sources we find it interpreted in the latter sense; that, of course, is how it is taken in the Chronicle of Zuqnin. It is thus legitimate to draw an analogy between these Christian sources and the Jewish texts cited above which speak of the appearance of a messianic star in the East. What that signifies in astronomical terms is a separate question, which cannot be addressed here. 92 On hypotheses of Zoroastrian inspiration for the notion of the Mountain of Victories see M. Boyce and F. Grenet, A History of Zoroastrianism (3 vols.; Leiden: Brill, 1991), 3:451–52. Stars Of The Messiah 303

According to Franz Rosenzweig, “[i]n the innermost narrows of the Jewish heart there shines the Star of Redemption.”93 Rosenzweig’s star is undoubtedly more accessible to modern sensibilities than those I have described here, but that should not prevent us from appreciating the power of the star of the Messiah as more than a figure of speech. We are all children of the scientific revolution, which has demystified the nighttime sky. When we do bother to look up, city lights blind us to its grandeur. Yet with enough effort, we can imagine the awe of earlier generations that stared into the heavens at night and, finding them inhabited by angels, searched longingly for the point of light her- alding redemption.

93 F. Rosenzweig, The Star of Redemption (trans. B. E. Galli; Madison: University of Wisconsin Press, 2005), 434. Retelling Biblical Retellings: Epiphanius, the Pseudo-Clementines, and the Reception-History of Jubilees

Annette Yoshiko Reed

1 Introduction

In the heresiological writings of Epiphanius, one finds preserved a poignant moment in the late antique Christian reception of so-called “Old Testament Pseudepigrapha.” In the course of his magisterial mapping of the genealogy of error from the primeval age to his present, Epiphanius comes to the Sethians—the thirty-ninth of the eighty poisonous sects for which he claims to provide antidotes in his medicine chest, the Panarion. Epiphanius’s derision, in this particular case, centers on what he mocks as the “melodrama, mythic nonsense, and fictitious clap-trap” of their beliefs about primeval history.1 As exemplar and cause of the fanciful folly of their ideas about Abel, Cain, and Seth, he accuses the Sethians of partaking in the practice and products of pseudepigraphy:

Composing books in the names of great men, they say that there are seven books in the name of Seth. . . . And another in the name of Abraham— which they also claim to be revelation—quite full of all sorts of wickedness. And another in the name of Moses. And still others [in names of] others. (Pan. 39.5.1)2

1 Translations of Epiphanius in this paper follow that of F. Williams, The Panarion of Epiphanius of Salamis (2 vols.; NHS 35–36; Leiden: Brill, 1987–1994), occasionally revised in consultation with the Greek in the edition of K. Holl and J. Dummer, Epiphanius (GCS 10, 13; Berlin: Akademie-Verlag, 1980). 2 The importance of this passage was noted already by Johann Albert Fabricius, who quotes it as the first of his witnesses to what he calls the Parva Genesis in Codex pseudepigraphus veteris testamenti (Hamburg: Liebezeit, 1713), 849–64. On the passing reference to “Allogenes” elsewhere in Pan. 39.5.1, compare Pan. 20.2.2, and see W.-P. Funk, P.-H. Poirier, and M. Scopello, eds., L’allogène: NH XI, 3 (Bibliothèque Copte de Nag Hammadi: Textes 30; Louvain: Peeters, 2004), 1–4.

At first sight, Epiphanius’s statement—written in the 370s CE—might seem simply to echo the sentiment more famously expressed roughly a decade earlier by the Alexandrian bishop Athanasius. In the Thirty-Ninth Festal Letter of 367 CE, Athanasius had presented pseudepigraphy as virtually coterminous with “heresy.”3 Not only is this letter often celebrated as the first to define a closed biblical canon of the same scope as would later become common in Western Christendom, but Athanasius denounces, in the process, what he calls “apocrypha” and what we might call “Old Testament Pseudepigrapha.” Whereas earlier Christians such as Clement, Origen, and Tertullian had expressed a range of views about the value of such works for preserving nonpagan wisdom and pre-Christian witnesses to Christ,4 Athanasius reframes the issue. For him, the circulation of books in the names of figures like Enoch, Isaiah, and Moses is not a question about the Jewish literary heritage of the church, but rather a problem of Christian “heresy” (especially Ep. fest. 39.21). Athanasius’s Thirty-Ninth Festal Letter was as innovative as it was influential, and its heresiological thrust has become even more evident with the discovery of an additional Coptic fragment, as presented in a new edition in 2010 by David Brakke.5 Not only is Athanasius’s argument for the ultimate sufficiency of the “canonized” scriptures rooted in his citation of biblical verses to counter Marcionites, Manicheans, and so on, but his argument for the undue dangers of “apocrypha” is rooted in the claim that “heretics” of this very sort are the ones really responsible for creating and circulating such writings. Parabiblical literature is thus presented as the pernicious and duplicitous opposite of what Athanasius here defends, in a carefully-constructed contrast, as the canon of scriptures that ensures “orthodoxy.” Athanasius further contends that any deviance from what he here seeks to promote as a self-evidently closed set of exclusively sacred scriptures is tantamount to the hard-hearted transgression of Deut 4:2: “You shall not add to the word that I commanded you.”6 Just as Athanasius thus invokes words

3 D. Brakke, “Canon Formation and Social Conflict in Fourth-Century Egypt: Athanasius of Alexandria’s Thirty-Ninth Festal Letter,” HTR 87 (1994): 395–419. 4 J. Ruwet, “Les apocryphes dans le oeuvres d’Origène,” Biblica 25 (1944): 143–66; W. Adler, “The Pseudepigrapha in the Early Church,” in The Canon Debate (ed. L. M. McDonald and J. A. Sanders; Peabody: Hendrikson, 2002), 214–24; A. Y. Reed, Fallen Angels and the History of Judaism and Christianity (Cambridge: Cambridge University Press, 2005), 160–232. 5 D. Brakke, “A New Fragment of Athanasius’s Thirty-Ninth Festal Letter: Heresy, Apocrypha, and the Canon,” HTR 103 (2010): 47–66. 6 For a recent iteration of the same problem, posed with the same proof text, see J. L. Kugel, How to Read the Bible (New York: Free Press, 2007), 684–85. 306 Reed attributed to Moses to counter the production of further works in Moses’s name, so his argument against pseudepigraphy in the names of Enoch and others also pivots on Mosaic authorship. Certainly—Athanasius argues—no Scripture existed before Moses, whether by Enoch or anyone else: “How could Moses have an apocryphal book? He is the one who published Deuteronomy with heaven and earth as witnesses!” (cf. Deut 4:26).7 One might assume that Epiphanius would argue along the same lines, not least because he takes aim at Sethian, Abrahamic, and Mosaic pseudepigraphy within an even more explicitly heresiological context.8 Earlier in the Panarion, he, too, provides the reader with a list of those Jewish Scriptures that are authentically “sacred writings” (Pan. 8.6.1–4), alluding in passing to a contrasting group of “certain apocrypha.” In his subsequent catalogue of sects, moreover, the Sethians are hardly the only “heretics” to be associated with pseudepigraphical forgery and “apocrypha.” So too with the Archontics (40.2.1), Bardaisan (56.2.2), the Hieracites (67.3.4), and others.9 A closer look at the passage quoted above, however, reveals a somewhat different take. In Pan. 39.5.1, Epiphanius does indeed connect the “heresy” of the Sethians with their use of books circulating under the names of Seth, Abraham,

7 On the resonance of Athanasius’s assertions with modern notions of pseudepigraphy, see my essay, “Pseudepigraphy, Authorship, and the Reception of ‘the Bible’ in Late Antiquity,” in The Reception and Interpretation of the Bible in Late Antiquity (ed. L. DiTommaso and L. Turcescu; The Bible in Ancient Christianity 6; Leiden: Brill, 2008), 467–90. 8 Traditionally, the Panarion has been treated primarily as a mine of information, but fresh attention to its form and rhetoric have made clear that Epiphanius is here engaged in the project of constructing “orthodoxy”; see A. Pourkier, L’Hérésiologie chez Épiphane de Salamine (Paris: Beauchesne, 1992); J. Schott, “Heresiology as Universal History in Epiphanius’s Panarion,” ZAC 10 (2007): 546–63; Y. R. Kim, “Reading the Panarion as Collective Biography: The Heresiarch as Unholy Man,” VC 64/4 (2010): 382–413. 9 In some cases, Epiphanius’s denunciation of “heretics” for reading or producing “apocrypha” is paired with explicit references to known works that modern scholars call “Old Testament pseudepigrapha.” Epiphanius reports of Archontics, for instance, that they “have forged some apocrypha of their own . . . and they take cues from the Ascension of Isaiah, and from still other apocrypha” (40.2.1); the Ascension of Isaiah is associated with the Hieracites, as well (67.3.4). In other reports, reference is made to works that modern scholars call “New Testament Apocrypha.” Hence, of the Origenists, he says that “they use various scriptures of the Old and New Testaments and certain apocrypha, especially the so-called Acts of Andrew” (63.2.1); while Encratites are said to use “the so-called Acts of Andrew, and [Acts] of John, and [Acts] of Thomas, and certain apocrypha” (67.1.5). At times, allusion to such books is perhaps deliberately vague: Severans use “certain apocrypha . . . but also the canonical books in part” (45.4.1); Bardaisan is said to draw from “the Law and the Prophets and the Old and New Testaments, besides certain apocrypha” (56.2.2). Retelling Biblical Retellings 307 and Moses, and he implies that some or all of these books have been spuriously produced by the Sethians themselves. Yet what is so striking, in this particular passage, is how Epiphanius goes on to answer them. Where Athanasius quotes Deuteronomy, Epiphanius cites the Book of Jubilees. He appeals, in other words, to a Second Temple Jewish text, attributed to Moses, outside of what he himself knows and lists as the Jewish Scriptures.10 And he makes no effort to hide this move. Not only does he cite Jubilees by name—informing the reader that he reports what he discovered “in the Jubilees (ἐν τοῖς Ἰωβηλαίοις), which is also called the Little Genesis (τῇ καὶ λεπτῇ Γενέσει καλουμένῃ)” (Pan. 39.6.1)—but he does so without any explanation, despite the fact that he is among the earliest known Christian authors to cite the work by name. With Jerome and Didymus of Alexandria, in fact, Epiphanius marks the beginning of our extant evidence for the explicit engagement with this Second Temple Jewish text in Christian sources.11 In modern scholarship, Athanasius’s Thirty-Ninth Festal Letter has been much discussed, often cast as a turning-point in the transmission of Christianity’s literary heritage from Second Temple Judaism.12 In the patristic literature of the second and third centuries CE, one finds references, citations, and allusions to the parabiblical literature of Second Temple times, consistent with the Jewish background of Jesus and earliest Christianity. In the fourth century, however, such fluidity is seemingly foreclosed. Or, rather, so it seems when we focus on Athanasius and consider his canonizing efforts in light of our evidence for the

10 See his list of “books of prophets” (προφητῶν βίβλους) that the Jews had at the time of their return from the Babylonian Exile at Pan. 8.6.1–2; Epiphanius there distinguishes these “twenty-seven books given to the Jews by God,” and counted by the Jews as twenty-two (8.6.3), from two other disputed books (Wisdom of Ben Sira; Wisdom of Solomon), as well as “certain other hidden books” (ἄλλων τινῶν βιβλίων ἐναποκρύφων; 8.6.4). 11 W. Adler, “Reception History of the Book of Jubilees: A Prime Example,” unpublished paper presented at the Philadelphia Seminar on Christian Origins, 13 March 2003. The relevant quotations and allusions from Epiphanius, Jerome, and various chronographers were gathered already by Fabricius (Codex pseudepigraphus VT, 849–64), and expanded by H. Rönsch, Das Buch der Jubiläen oder die kleine Genesis (Leipzig, 1874), 251–382, with materials from catenae, etc. Newer assessments include those of A.-M. Denis, Fragmenta pseudepigraphorum quae supersunt graeca (PVTG 3; Leiden: Brill, 1970), 70–102; and J. C. VanderKam, “The Manuscript Tradition of Jubilees,” in Enoch and the Mosaic Torah: The Evidence of the Book of Jubilees (ed. G. Boccaccini and G. Ibba; Grand Rapids: Eerdmans, 2009), 12–15. 12 This tendency is noted, e.g., in Brakke, “Canon Formation and Social Conflict”; L. M. McDonald, The Biblical Canon: Its Origin, Transmission, and Authority (Peabody: Hendriksen, 2007), esp. 51–52, 355. 308 Reed afterlives of certain Second Temple texts, such as the Books of Enoch—for which there are a number of references, citations, and allusions in the first three centuries CE, but dwindling and negative notices precisely in the fourth century and following.13 What I would like to suggest here is that Epiphanius’s citation of Jubilees may also have something important to tell us about the Christian transmission of texts and traditions from Second Temple Judaism, precisely because it does not fit quite so neatly into conventional scholarly narratives about the creation of the canon and the reception of “pseudepigrapha.” Why is it that explicit reference to Jubilees begins precisely when canonical boundaries start to tighten, and when similar Second Temple Jewish texts, like the Books of Enoch, are coming under sharpened suspicion? And why is Epiphanius—hardly a cham- pion of “apocrypha” and far from a paragon of conscientious source-citation— among the first Christians to cite Jubilees by name? It is this double puzzle that I would like to consider in this essay. To do so, I shall reflect upon the fourth-century reception of Jubilees in light of its own self-presentation in relation to the Torah; but I would also like to ask how Jubilees’ practice of “retelling” Genesis traditions about primeval times (especially Genesis 1–11) relates to Epiphanius’s own acts of “retelling” that same history, in much the same terms, from Genesis, Jubilees, and other sources. With Jubilees, as we shall see, Epiphanius shares the concern to create a universal primeval history that is also a preface to a genealogy of error, with the person- ages and lineages from Genesis 1–11 serving as base and backbone. To highlight some of what is at stake in this shared concern, I shall adduce another fourth-century example, namely, the Pseudo-Clementine Homilies. The material about primeval history in the eighth and ninth Homilies offers some interesting points of parallel, contrast, and counterpoint to the Panarion—as a fourth-century collection of traditions possibly dependant on Jubilees, possibly known to Epiphanius, and certainly sharing halakhic concerns with the former and heresiological concerns with the latter. Through the triangulation of these three sources—each of which inextricably interweaves “biblical retellings,” universal histories, and genealogies of error—I hope to shed light on the Christian transformation of Second Temple Jewish texts and traditions, but also on the continuities that connect Jubilees with its late antique readers.14

13 Reed, Fallen Angels, 122–232. 14 I here use the term “biblical retelling” in the broadest sense, so as to highlight conceptual and discursive continuities. For a reading of Epiphanius’s Panarion within a tradition of Christian chronographical discourse indebted to Jubilees, see now Y. R. Kim, “The Imagined Worlds of Epiphanius of Cyprus” (Ph.D. diss., University of Michigan, 2006), Retelling Biblical Retellings 309

2 Jubilees and Its Early Reception

The Book of Jubilees, composed in Hebrew in the second century BCE, presents itself as a record of divine revelations delivered through an angel to Moses on Mt. Sinai. The work begins rather remarkably, as Martha Himmelfarb notes, “with a story of its own revelation that provides an account of its relationship to the Torah. . . . The Torah is apparently identified with the tablets of the law, while Jubilees itself is the transcript of the revelation that took place during the forty days and nights.”15 The result, as James VanderKam has observed, is that

The writer leaves no doubt that he has placed his story at Sinai and, within that episode, in the action described in Exodus 24 where Moses ascends the mountain the day after the revelation on the same mountain (Exod 24:4). . . . And, far from mentioning this circumstance only at the beginning of the book and ignoring it afterwards, he reminds the reader of it in a whole series of passages. In fact, the book ends where it began, with Moses at Mt. Sinai ( Jub. 50:2) . . .16

What this schema serves to authorize is an expansive and detailed account of events also described in Genesis and at the beginning of Exodus. In the process, the author seems to grapple with a number of the same textual, chronological, halakhic, and other issues discussed in more explicitly exegetical terms by later Jews and Christians. Indeed, it is not for naught that James Kugel can laud the author of Jubilees as “one of the heroes” of The Bible As It Was.17 A great number of the motifs found in Jubilees appear in later sources as well, from the Second Temple period well into the Middle Ages—whether because the author was perhaps especially inventive or influential, or because Jubilees just happens to

196–237. For an application of the category “rewritten Bible” to Epiphanius’s work, see Schott, “Heresiology,” 547–48; but compare the terminological debates in relation to Second Temple materials in H. Najman, Seconding Sinai: The Development of Mosaic Discourse in Second Temple Judaism (JSJSup 77; Leiden: Brill, 2003), esp. 1–12; M. Segal, “Between Bible and Rewritten Bible,” in Biblical Interpretation at Qumran (ed. M. Henze; Grand Rapids: Eerdmans, 2005), 10–28; M. J. Bernstein, “Rewritten Bible: A Generic Category Which Has Outlived Its Usefulness?” Textus 22 (2005): 169–96. 15 M. Himmelfarb, “Torah, Testimony, and Heavenly Tablets: The Claim to Authority of the Book of Jubilees,” in A Multiform Heritage: Studies on Early Judaism and Christianity in Honor of Robert A. Kraft (ed. B. G. Wright; Atlanta: Scholars Press, 1999), 19–29 (19). For a range of opinions, see n. 18 below. 16 J. C. VanderKam, “The Scriptural Setting of the Book of Jubilees,” DSD 13/1 (2006): 61–72 (61). 17 J. L. Kugel, The Bible As It Was (Cambridge, Mass.: Harvard University Press, 1999), 38. 310 Reed preserve and transmit, in writing, an unusually dense deposit of the oral interpretative traditions surrounding the Torah in Second Temple times. For some modern readers, the apparently interpretative character of much of Jubilees has only served to sharpen the seeming paradox of its own self- presentation as angelic speech, Mosaic writing, and extrapentateuchal revelation. Jubilees is indeed a “parade example” of what can seem so very strange from a modern perspective about the parabiblical literature of Second Temple Judaism. How could a Jewish author in the second century BCE know so much about the Torah, seemingly care so much about the problems of its proper interpretation, and yet deign to write in the name of Moses, claiming a status for his own work equal to that of the Torah? How could he dare to “add to the words,” unless with some prideful or deceptive aim to replace, displace, or supersede? This is certainly how Athanasius framed the issue, and it is also how some modern scholars have reacted to the authorial and revelatory claims of this and other parabiblical writings therefore labeled as “pseudepigrapha.”18 More recently, however, scholars such as Martha Himmelfarb and Hindy Najman have drawn attention to the supplementarity tacit in the authorizing claims and self-presentation of Jubilees.19 Jubilees may relativize what it calls “the first Torah,” but it lays out a “division of labor” for the two to coexist side by side, as earthly selections from the divine archive of the heavenly tablets. Himmelfarb proposes, moreover, that this is precisely how Jubilees was

18 Esp. B. Z. Wacholder, “Jubilees as the Super Canon: Torah-Admonition versus Torah- Commandment,” in Legal Texts and Legal Issues: Proceedings of the Second Meeting of the International Organization for Qumran Studies (ed. M. J. Bernstein, F. García Martínez, and J. Kampen; STDJ 23; Leiden: Brill, 1997), 195–211. For a variety of positions on this issue, see Himmelfarb, “Torah, Testimony”; F. García Martínez, “The Heavenly Tablets in the Book of Jubilees,” in Studies in the Book of Jubilees (ed. M. Albani, J. Frey, and A. Lange; TSAJ 65; Tübingen: Mohr Siebeck, 1997), 243–60; H. Najman, “Interpretation as Primordial Writing: Jubilees and its Authority-Conferring Strategies,” JSJ 30/4 (1999): 379–410; C. Werman, “The Torah and the Teʿudah Engraved on the Tablets,” DSD 9/1 (2002): 75–103; J. C. VanderKam, “Moses Trumping Moses: Making the Book of Jubilees,” in The Dead Sea Scrolls: Transmission of Traditions and Production of Texts (ed. S. Metso, H. Najman, and E. Schuller; STDJ 92; Leiden: Brill, 2010), 25–44. 19 Esp. Himmelfarb, “Torah, Testimony”; Najman, “Interpretation as Primordial Writing”; eadem, “Reconsidering Jubilees: Prophecy and Exemplarity,” in Boccaccini and Ibba, Enoch and the Mosaic Torah, 229–43. For a reassessment, see now J. J. Collins, “Changing Scripture,” in Changes in Scripture: Rewriting and Interpreting Authoritative Traditions in the Second Temple Period (ed. H. von Weissenberg, J. Pakkala, and M. Marttila; BZAW 419; Berlin: de Gruyter, 2011), 23–45, esp. 34–38 on Jubilees. Retelling Biblical Retellings 311 received at Qumran: “For the Damascus Covenant, the Torah of Moses contains commandments, while Jubilees contains the history of Israel’s failure to fulfill those commandments.”20 Such truth claims might seem bold to us now, but they were plausible and persuasive—she suggests—when “there was not yet a fixed form of the biblical text, the final contours of the canon had not yet been delineated, and . . . the very notion of a canon, a body of literature with exclusive claims to authority, had not yet emerged.”21 Yet by this logic, as even Himmelfarb admits, “the existence of a canon makes the peaceful coexistence of Jubilees and the Torah more difficult to maintain.”22 How, then, can we explain the case of Epiphanius, who seems to use the work with much the same ease as the Damascus Document, albeit writing in the fourth century CE, in the wake of Athanasius, and with a closed canon seemingly already in mind? And why is he among the first known Christian authors to cite this work explicitly by name?

3 Jubilees, Genesis, and Primeval History in Epiphanius’s Panarion

The challenge of answering these questions is compounded by the lack of explicit references to Jubilees in the centuries between the Damascus Document, on the one hand, and Epiphanius, Jerome, and Didymus, on the other.23 Hence, when we find parallel exegetical and other traditions, it remains unclear whether we can posit any influence per se—whether direct or indirect, oral or written, narrow or diffuse.24 Perhaps some early Christians consulted the text,

20 Himmelfarb, “Torah, Testimony,” 23, pointing also to the Hebrew fragments of Jubilees and related materials discovered at Qumran; cf. D. Dimant, “Two ‘Scientific’ Fictions: The So-Called Book of Noah and the Alleged Quotation of Jubilees in CD 16:3–4,” in Studies in the Hebrew Bible, Qumran, and the Septuagint Presented to Eugene Ulrich (ed. P. Flint, E. Tov, and J. C. VanderKam; VTSup 101; Leiden: Brill, 2006), 242–48. 21 Himmelfarb, “Torah, Testimony,” 28–29. 22 Himmelfarb, “Torah, Testimony,” 29. 23 As Adler (“Reception History”) notes, even authors like Clement and Origen, who typically “weigh in” on such works, are silent in the case of Jubilees. Compare the argument of A. Kreps, “From Secret Knowledge to Public Paideia: Citations of Jubilees in Epiphanius’s Panarion,” unpublished paper presented at the Second Enoch Graduate Seminar, Princeton Theological Seminary, 16 June 2008. Kreps interprets this silence as signaling early Christian treatment of Jubilees as an “authoritative hidden book”; I thank her for sharing a prepublica- tion draft of her article with me. 24 I here follow Adler, “Reception History”; contrast the maximalist reading of the evidence in J. M. Scott, Geography in Early Judaism and Christianity: The Book of Jubilees (SNTSMS 312 Reed in whole or in part, and just did not mention it. Or perhaps the parallels primarily reflect a common store of traditions. Or maybe, however innovative the author of Jubilees may have been in the second century BCE, however bold his revelatory claims, and however idiosyncratic in other ways, he was so successful in interweaving text and interpretation that some of his inventions came to be received, at least in some times and places, as if simply the self-evident or traditional meaning of Genesis itself. The last option, in fact, is what we might have suspected if we only had the beginning of Epiphanius’s Panarion. In the first three sections, Epiphanius retells the early history of humankind as the preface to his catalogue of “heretical” sects. Even as he draws on Genesis 1–11, his account is clearly indebted to traditions from Jubilees, such as the treatment of Noah’s progeny, the account of the Tower of Babel, and the attribution of the origins of idolatry to the age of Serug, all found in Jubilees 10–11. No source, however, is here mentioned by name. Instead, the information paralleled in Jubilees is introduced [1] as information encompassed in “the tradition (παράδοσις) that came down to us,” in the case of the “mischief” that appeared in the world with “sorcery, witchcraft, licentiousness, adultery, and iniquity” in the lifetime of Jared (Pan. 1.3; cf. Jub. 4:15, 22); and [2] as “the knowledge (γνῶσις) that came down to us,” in the case of the origins of idolatry with Serug (3.4; cf. Jub. 11:4–6).25 For our purposes, the introductory sections of the Panarion also prove significant because they remind us that the work as a whole is framed as a sort of “biblical retelling” in its own right. Epiphanius sets the stage for his catalogue of sects by “retelling” Genesis 1–11 to make a point about the prehistory of “heresy.”26 His point, more specifically, is that humankind sprang from a single lineage with a single language, living in unity of belief as well, prior to the diversification in the age of Serug that birthed idolatry and thus the “protoheresy” of Hellenism.

113; Cambridge: Cambridge University Press, 2002). Interestingly, there is an even larger gap in the evidence for the Jewish Nachleben of the work, with traditions clearly connected to Jubilees not reappearing again until the early Middle Ages in works like Pirqe de Rabbi Eliezer, Bereshit Rabbati, etc.; see, e.g., R. Adelman, The Return of the Repressed: Pirqe de Rabbi Eliezer and the Pseudepigrapha (JSJSup 140; Leiden: Brill, 2010), 109–208. 25 I.e., ὡς ἡ ἐλθοῦσα εἰς ἡμᾶς γνῶσις περιέχει in the former case, and ὡς δὲ ἡ παράδοσις ἡ εἰς ἡμᾶς ἐλθοῦσα περιέχει in the latter. Interestingly, similar phraseology (ὡς ἡ εἰς ἡμᾶς ἐλθοῦσα παράδοσις ἔχει) accompanies the description of Moses’s burial by angels in Pan. 9.4.12. 26 Schott, “Heresiology,” 547–50. Schott posits that “for Epiphanius, heresy exists in radical opposition to an a-historical orthodoxy that is entirely dissociated from historical processes of cultural development” (p. 547). Retelling Biblical Retellings 313

Nevertheless, if the reference to Serug betrays something of the debt to Jubilees, analysis of its form and context exposes the debt to be likely indirect— as has been established, most recently and decisively, by William Adler.27 It is to the lost third-century Chronicle of Julius Africanus, he suggests, that Epiphanius is ultimately indebted for his material from Jubilees, albeit likely through the mediation of another source that has further adapted it.28 Indeed, if Epiphanius’s knowledge of Jubilees has been mediated through one or more Christian sources, it might help to explain how he can so readily appeal to the work, particularly if the excerpts known to him were already preselected for their value in enhancing the utilization of Scripture as scaffolding for universal history. In that case, the Panarion may provide us with an early example of a pattern that Adler highlights with respect to the later reception of Jubilees, into Byzantium and beyond—that is, whereby excerpts from Jubilees were integrated into catenae, interpolated into Josephus’s Antiquities, and cited across a broad range of Christian chronicles. Perhaps already by the late fourth century CE, the historiographical use of Jubilees by Africanus and others was helping to “normalize” the work, to draw attention to its value as a source for filling the gaps in Genesis, and to disseminate its circulation in Greek translation, whether also or exclusively in the form of excerpts preselected for their value for the concerns of Christian historiography and exegesis.29 Even so, the problem of Epiphanius’s explicit citation of the work by name, in the context of the Sethians, still remains. It is surely tempting to speculate that someone who knew Jubilees only from excerpts could have had no sense of its revelatory claims. Yet Jerome is quite clear about placing the book within the category of “apocrypha.”30 And, consistent with the diffusion of Jubilees’

27 See esp. W. Adler, “The Origins of the Proto-Heresies: Fragments from a Chronicle in the First Book of Epiphanius’s Panarion,” JTS 41 (1990): 472–501. Adler here shows how “Epiphanius’s story of Serug and the rise of Hellenism is a reworking of historical and chronological traditions from Eusebius’s Canons and Julius Africanus, combined with apocryphal expansions based loosely on Jewish sources, most notably the Book of Jubilees” (481–82). 28 Adler, “Origins of the Proto-Heresies,” 492–93, 498. 29 See further Adler, “Reception History”; idem, “Abraham and the Burning of the Temple of Idols: Jubilees’ Traditions in Christian Chronography,” JQR 77/2–3 (1986–1987): 95–117; idem, “The Chronographiae of Julius Africanus and its Jewish Antecedents,” ZAC 14/3 (2011): 496–524, esp. 510–14. 30 See Jerome, Ep. 78.20, which has been central to the discussion of the Christian Nachleben of Jubilees since Fabricius and Rönsch. For recent assessments of Didymus’s knowledge of traditions from Jubilees, particularly in light of the Tura papyri, see D. Lührmann, “Alttestamentliche Pseudepigraphen bei Didymos von Alexandrien,” ZAW 104 (1992): 314 Reed self-authenticating claims throughout the book, reference to such statements survives even in the excerpts we now know secondhand through George Syncellus and others.31 It is thus all the more striking, in my view, that Epiphanius’s explicit reference to Jubilees is marked by a seeming lack of canonical consciousness. He does not hesitate to denounce others for using “apocrypha,” even as he himself draws on Jubilees without remark. That he so often integrates his sources, without any signal or citation, makes the pointedness of his choice, in this particular case, even more surprising. Attention to the context and function of the citation, however, may give us some of sense of his perception and presentation of Jubilees. The immediate context here is Epiphanius’s argument against Sethian claims that Cain and Abel are sons of two different fathers, and that Seth is the product of the planting of a divine seed, equivalent to Christ and fathering the line of the chosen (Pan. 39.2.1–3.5). Against the Sethians, Epiphanius thus seeks to establish that “one man was formed, Adam, and Cain, Abel, and Seth came from Adam” (39.4.2); and that “Seth was a real man . . . the real brother of Cain and Abel, from one father and mother” (39.5.4). To do so, he first cites what “Scripture says,” quoting from Genesis (4:1–2, 25; Pan. 39.5.5–8). He then asserts that “it is clear that Cain and Seth took wives” (39.5.9). It is to establish this point that Epiphanius turns to cite Jubilees. Its genealogical material serves to fill the gaps in Genesis—those same gaps that, for Sethians and others, seem to have become like chasms into which to interpret archons and powers.32 What is presented as valuable about Jubilees, then, is its extensive and concrete detail. “The book even contains the names of both Cain’s and Seth’s wives!” (Pan. 39.6.1; cf. Jub. 4:7–11), Epiphanius exclaims, when introducing it as

231–49 (239–45); R. A. Layton, Didymus the Blind and His Circle in Late Antique Alexandria (Urbana: University of Illinois Press, 2004), 60–61. 31 When Syncellus (Chron. 33.1–18; F14b), for instance, mentions the twenty-two works of creation, twenty-two Hebrew letters, twenty-two patriarchs, and so on, in relation to Africanus and with reference to Jubilees (cf. Jub. 2.23), he notes concerning “the Little Genesis” that “some say [it] is also a revelation of Moses.” 32 My concern here is not with the accuracy of Epiphanius’s report, but rather with the Sethians as he represents them in the course of his appeal to Jubilees. For considerations of his statements about the Sethians in the broader context of the surviving evidence for texts and sects of this sort, see A. F. J. Klijn, Seth in Jewish, Christian, and Gnostic Literature (NovTSup 46; Leiden: Brill, 1977), 81–117; G. G. Stroumsa, Another Seed: Studies in Gnostic Mythology (NHS 24; Leiden: Brill, 1984), 49–53; J. D. Turner, “The Gnostic Seth,” in Biblical Figures Outside the Bible (ed. M. E. Stone and T. A. Bergen; Harrisburg: Trinity, 1998), 33–58. Retelling Biblical Retellings 315 an accurate witness. Its accuracy is further implied by the contrast with “these people who have recited their myths to the world” (39.6.1). The two are presented in terms as different as myth and history. Jubilees’ specificity with respect to names seems to be useful for Epiphanius, as does another key concern in the work, which has been noted by Betsy Halpern-Amaru—namely, the orderly pattern of its presentation of the evolution of marriage practices from brother-sister marriages onwards to the endogamy prescribed in the author’s present.33 Countering what appears to have been an argument for allegorization grounded in the logic that the Torah would not condone incest, Epiphanius explains that Cain and Seth were married to their sisters, as was lawful in that age (Pan. 39.6.2–4; 39.7.1–3). After adding that Adam also had nine other sons, he turns back to the Torah, asserting that “you also have the suggestion of them in the Genesis of the World, the first Book of Moses, which says ‘And Adam lived 930 years, and begat sons and daughters, and died’ ” (39.6.6; cf. Gen 5:3–5). The material from Jubilees, in other words, is here presented as flowing naturally out from Genesis and then back again. Even if Epiphanius never consulted Jubilees itself, it remains significant that he implies to the reader that he does know the work, just as he implies that he knows about the Sethians from perhaps having met some in Egypt (Pan. 39.1.2). The citation of the text and title, in the context of his argument here, functions as a claim that he himself knows where “true scriptures” end and where their supplements begin. Even though the work from which he ultimately draws is one in which the two are so famously blurred, Epiphanius is thus able, on the level of argument, to set up his own textually grounded practice in contrast to that of Sethians and others, who “mix their own invention” with the truth (39.9.1). It is this very signaling, moreover, that helps to authenticate what immediately follows, wherein citations from Genesis blur into his own retelling of primeval history. Epiphanius cites traditions from Genesis again, outlining Noah’s construction of the ark by divine decree, as “the true scriptures tell us” (Pan. 39.7.5), and the entry into the ark of Noah, his wife, his three sons, and their three wives as what “the same book of truth says” (39.7.5; cf. Gen 7:7). This leads into a selective summary of the account of primeval history at the very beginning of the Panarion, focusing on the differentiation of human languages and lineages after the Flood (39.8.1–5). Having aligned his own “biblical retelling” with Scripture and history, in contrast to the myths and invention of the

33 B. Halpern-Amaru, The Empowerment of Women in the Book of Jubilees (JSJSup 60; Leiden: Brill, 1999). 316 Reed

Sethians, he reveals what is at stake in a manner that places his practice of “retelling” on the side of right doctrine as well: “Once the origin [of humankind] is shown to be one,” he asserts, “[the Sethians] will return to the confession that the Master of all, the Creator and Maker of the whole, is one” (39.10.6).

4 Jubilees and the Torah in the Pseudo-Clementine Homilies

Interestingly, this concern to argue for the singularity of God, against those who read multiple deities from and into Torah, is what ultimately motivates the “retelling” of primeval history in the Pseudo-Clementine Homilies as well.34 As in Epiphanius’s account, moreover, heresiology is the very point and context.35 The Homilies’ account of primeval history occurs in a series of ser- mons on the origins of polytheism, sacrifice, and idolatry, which are placed in the mouth of the apostle Peter, set in Tripolis, and framed as part of the content of his debates with Simon Magus.36 Like the Panarion, the Pseudo-Clementine Homilies describe the very first era of human history as prior to any error or differentiation; Adam, who is called a prophet, is also associated with the true religion later revealed by Jesus

34 This version of the Pseudo-Clementine novel is extant in the original Greek (ca. 300– 320 CE?), while the other main version, the Recognitions (ca. 360–380 CE?) survives in full only in Rufinus’s Latin translation of 407 CE; both are of probable Syrian provenance, and survive in part in Syriac translation. For the purposes of the present inquiry, I do not focus on the relationship between them, although I do treat as especially characteristic of the aims of the authors/redactors of the Homilies those passages, themes, etc., which are not directly paralleled in the Recognitions and which are thus less likely to reflect a shared source. 35 I make this point in more detail, with Epiphanius’s Panarion as a main intertext, in “Heresiology and the (Jewish-)Christian Novel: Narrativized Polemics in the Pseudo- Clementines,” in Heresy and Self-Definition in Late Antiquity (ed. E. Iricinschi and H. Zellentin; TSAJ 119; Tübingen: Mohr Siebeck, 2008), 273–98. See also F. S. Jones, “Jewish Christians as Heresiologists and as Heresy,” Rivista di storia del cristianesimo 6/2 (2009): 333–47. 36 On parallels of content and differences of emphasis in the Tripolis material in Hom. 8–11 and Rec. 4–6, see G. Strecker, Das Judenchristentum in den Pseudoklementinen (2d rev. and exp. ed.; TUGAL 70; Berlin: Akademie-Verlag, 1981), 70–75; also A. Y. Reed, “‘Jewish Christianity’ after the ‘Parting of the Ways’: Approaches to Historiography and Self- Definition in the Pseudo-Clementine Literature,” in The Ways that Never Parted: Jews and Christians in Late Antiquity and the Early Middle Ages (ed. A. H. Becker and A. Y. Reed; TSAJ 95; Tübingen: Mohr Siebeck, 2003), 188–231 (213–17). Retelling Biblical Retellings 317 and his apostles.37 Whereas Epiphanius tersely notes some “mischief” that occurred in the days of Jared (see above), the Homilies explain how the lack of any difficulties led the first humans to respond with ingratitude (8.11). The angels of the lowest heavens sought to teach them to act otherwise, and thus asked God for permission to descend to earth (8.12). Although they intended to serve as models for proper action, they were overtaken by lust upon embodiment, whereupon they took wives, revealed forbidden knowledge, and sired giants (8.13–15). The story in the Homilies (cf. Rec. 1.29; 4.26–27) is readily recognizable as a retelling of the account of the fallen angels first famous from the Enochic Book of the Watchers (especially 1 Enoch 6–16; cf. Gen 6:1–4), expanded with themes from its Christian Nachleben as well as from Greco-Roman discourses about the origins of civilization. The positive element of the angels’ teaching (Hom. 8.12) echoes Jubilees (4:15; 5:6) in particular,38 even as its articulation in terms of human–angelic rivalry brings to mind later rabbinic versions of the tale.39 More significant, for our present purposes, is the Homilies’ treatment of the giants, which includes some of the most pointed parallels of content and concern between the Homilies and Jubilees.40 Here, the sins of the giants center not

37 On the prophethood of Adam, which is closely tied to the idea of the True Prophet, see Hom. 2.16–18; 3.17–21; 8.10; Strecker, Judenchristentum, 145–53; H. J. W. Drijvers, “Adam and the True Prophet in the Pseudo-Clementines,” in Loyalitätskonflikte in der Religionsgeschichte: Festschrift für Carsten Colpe (ed. C. Elsas and H. Kippenberg; Würzburg: Königshausen & Neumann, 1990), 314–23. For the Homilies’ depiction of monotheistic piety as the original state of humankind, to which polytheistic corruptions later accrued, see also Hom. 1.18; 3.23–25; 8.11–20; 9.2–18; 10.7–23. This theme finds ample parallels in the Recognitions, including its early strata; see, e.g., Rec. 1.24–38 and discussion in F. S. Jones, An Ancient Jewish Christian Source on the History of Christianity: Pseudo- Clementine Recognitions 1.27–71 (SBLTT 37; Atlanta: Scholars Press, 1995). 38 So already Rönsch, Das Buch Jubiläen, 322–25. 39 See, e.g., 3 Enoch 4 (P. Schäfer, ed., Synopse zur Hekhalot-Literatur [Tübingen: Mohr Siebeck, 1981], §§5–6); Pesiqta Rabbati 34:2 (M. Friedmann, ed., Pesikta Rabbati: Midrasch für den Fest-Cyclus und die ausgezeichneten Sabbathe [repr.ed. Tel Aviv, 1963], 159 a–b); Aggadat Bereshit praef. on Gen 6:2 (Aggadat Bereshit [ed. and trans. L. M. Teugels; Leiden: Brill, 2001], 254–55); Bereshit Rabbati on Gen 6:2 (C. Albeck, ed., Midraš Berešit Rabbati ex libro R. Mosis Haddaršan collectus e codice Pragensi cum annotationibus et introductione [ Jerusalem: Mekitse Nirdamim, 1940], 29–31); Midrash Petirat Moshe Rabbeinu in BHM 1:129; Reed, Fallen Angels, 261. But compare also Lactantius, Inst. 2.15. 40 See now E. J. C. Tigchelaar, “Manna-Eaters and Man-Eaters: Food of Giants and Men in the Pseudo-Clementine Homilies 8,” in The Pseudo-Clementines (ed. J. Bremmer; Studies on Early Christian Apocrypha 10; Leuven: Peeters, 2010), 92–114. 318 Reed so much on their violence, but on their consumption of blood (cf. 1 En. 7:5).41 God rains manna upon them (cf. Ps 78:24–25), but the giants desire to taste blood and thus eat flesh, tempting humankind to do the same (Hom. 8.15–16).42 As in the summary of the Flood attributed to Noah in Jub. 7:21–25, the consumption of blood looms large, and cannibalism is added to their sins. The resultant impurity, moreover, is depicted as a primary reason for the Flood: the shedding of blood, according to the Homilies, defiled even the air, causing the spread of disease (8.17; cf. Hippocrates, On Airs, Waters, and Places 5.6). Following the Flood, blood sacrifice and bodily disease are, in turn, central to the operation of demons—those evil spirits who sprang from the dead bodies of the giants (8.18–20; cf. 7.8; 9.12–14; cf. Jub. 10:1–5; 11:4; 22:17). Apart from any explicit citation or direct quotation, of course, we can only speculate about the channels through which the authors/redactors of the Homilies came to know these traditions. By the fourth century, stories about fallen angels were certainly widespread. With regard to some traditions about giants, parallels to Jubilees also have precedents in the Book of the Watchers, and independent exegetical development remains a possibility.43 The Homilies then goes on, however, to describe “the law upon the demons” in a manner perhaps most readily explained as a recasting of traditions in Jub. 10:3–10. Not only do the souls of giants survive as demons, but they are bound by a special law, whereupon their violence is brought under angelic control for the discipline and punishment of human sinners (Hom. 8.18–19). The connection with Jubilees is, at the very least, intriguing, not least because the Homilies here develop traditions not discussed elsewhere, to my knowledge, in any such detail. That this material is unparalleled within the Recognitions suggests that

41 On blood in Jubilees, see W. K. Gilders, “Blood and Covenant: Interpretive Elaboration on Genesis 9.4–6 in the Book of Jubilees,” JSP 15/2 (2006): 83–118. On blood in the Homilies, see R. S. Boustan and A. Y. Reed, “Blood and Atonement in the Pseudo-Clementines and the Story of the Ten Martyrs,” Henoch 30/2 (2008): 111–42 (114–30). 42 For this passage in the context of “pagan” etiologies of meat-eating, defenses of vegetari- anism, and so on, see Tigchelaar, “Manna-Eaters,” 105–7; Boustan and Reed, “Blood and Atonement,” 126–27. 43 Jones posits that thematic connections reflect direct dependence on 1 Enoch and Jubilees at various stages in the development of the Pseudo-Clementine tradition (e.g., Ancient Jewish Christian Source, 138–39). Tigchelaar weighs the possibility of the circulation of Enochic traditions within the Pseudo-Clementines’ Syrian milieu (“Manna-Eaters,” 100– 102), while stressing with respect to Jubilees that “all these motifs are also attested in other compositions” (99); although his point is well taken, I am not sure that it applies to the motif of the “law upon the giants/demons.” Retelling Biblical Retellings 319 it belongs to the fourth century CE, rather than to earlier sources or strata in the Pseudo-Clementine tradition. If so, we may find here a further clue to the seemingly new popularity of Jubilees among Christians in the fourth century, as well as a glimpse into yet another trajectory in its late antique afterlife, apart from those chronographical and related traditions so richly discussed by Adler. “[I]n the Panarion,” as Adler notes, “legends from Jubilees appear in a highly denatured and rational- ized form.”44 But if Jubilees was known in some form to the fourth-century authors of the Homilies, it was in a form that retained precisely the demonological concerns that struck Africanus and others as so problematic.45 Africanus is the first known Christian author to suggest a euhemeristic reading of the “sons of God” of Gen 6:1–4 (Syncellus, Chron. 19.24–20.4), and Epiphanius seems to follow his lead.46 By contrast, the Pseudo-Clementine Homilies construct primeval history with precisely the demonological element as pivot. The result is a genealogy of error that falls closer to Jubilees itself. No less intriguing is the manner in which the very practice of “biblical retelling”—and the retelling of “retellings”—fits with the theory of the Torah distinctive to the Homilies. According to the Homilies, what Moses received from God on Mt. Sinai was not the Torah alone, but the “Law with the explanations” (2.38). Both, moreover, are in oral form (3.47). These explanations are said to have been faithfully transmitted by the Jews in perfect succession from Moses, among the Pharisees as well as among Jewish apostles like James and Peter.47 It is upon them that one must depend when interpreting the Written Law, which

44 Adler, “Origins of the Proto-Heresies,” 477. 45 Interestingly, the circulation of the work in more than mere excerpts may be attested by POxy 4365; Simon Franklin suggests that this manuscript “contains the earliest surviving Greek manuscript reference to the Book of Jubilees . . .; extremely valuable evidence— the only evidence—for the period between Sextus Julius Africanus and Epiphanius of Salamis . . . that the Book of Jubilees circulated among Greek-speaking Christians in early- or mid-fourth-century Egypt”; see his “A Note on a Pseudepigraphical Allusion in Oxyrhynchus Papyrus No. 4365,” VT 48/1 (1998): 95–96. For more recent reflections on this possibility, see A.-M. Luijendijk, Greetings in the Lord: Early Christians in the Oxyrhynchus Papyri (HTS 60; Cambridge, Mass.: Harvard University Press, 2008), 72. 46 Adler, “Origins of the Proto-Heresies,” 478–79. 47 See further Hom. 3.18–19; 11.29; A. Y. Reed, “‘Jewish Christianity’ as Counter-history? The Apostolic Past in Eusebius’ Ecclesiastical History and the Pseudo-Clementine Homilies,” in Antiquity in Antiquity: Jewish and Christian Pasts in the Greco-Roman World (ed. G. Gardner and K. L. Osterloh; TSAJ 123; Tübingen: Mohr Siebeck, 2008), 173–216 (190–95). This sentiment is echoed in the Epistle of Peter to James 1.2 but finds no parallel in the Recognitions. 320 Reed contains points that have been added and exploited by “heretics” to denigrate the divine Creator.48 Such a view of the Torah offers an unusual but effective solution to much the same problem that Epiphanius seeks to solve with reference to Jubilees— namely, the breadth of interpretative possibilities within the text of Genesis, as exploited by those whom both would deem “heretics.” In the Homilies, the oral transmission of true interpretations and the proper succession of the Law’s interpreters are privileged, and the result is intriguingly resonant (perhaps not coincidentally) with the Oral Torah of contemporaneous rabbinic Judaism.49 Yet one wonders, as well, if such beliefs about the true transmission of Mosaic teaching might have facilitated the use of Second Temple Jewish sources, like Jubilees, which claim precisely to preserve Mosaic teachings not found in the written text of Torah.50

5 Conclusions

For our understanding of the fate of Jubilees in the fourth century CE, however, what may be most illuminating is what the Pseudo-Clementine Homilies share with both Epiphanius and Athanasius. All three are preoccupied with “heresy.” For Athanasius, the closing of the biblical canon seems to provide one solution, and it is one that would seem to find the loss of a work like Jubilees an adequate price to pay for asserting the epistemological monopoly and heresiological sufficiency of Scripture. Even though Athanasius’s own reference to Mosaic pseudepigraphy more likely takes aim at a book like the Assumption of Moses, it takes its power from the categorical dismissal of parabiblical literary production. The authors/redactors of the Homilies take up the opposite

48 See further Hom. 2.38–52; 3.4–6, 9–11, 17–21, 37–51; 16.9–14; 18.12–13, 18–22; Strecker, Judenchristentum, 166–86; K. E. Shuve, “The Doctrine of the False Pericopes and Other Late Antique Approaches to the Problem of Scripture’s Unity,” in Nouvelles intrigues pseudo-clémentines (ed. F. Amsler et al.; Publications de l’Institut romand des sciences bibliques 6; Lausanne: Zèbre, 2008), 437–45. 49 A. I. Baumgarten, “Literary Evidence for Jewish Christianity in the Galilee,” in The Galilee in Late Antiquity (ed. L. I. Levine; New York: Jewish Theological Seminary of America; Cambridge, Mass.: Harvard University Press, 1992), 43. 50 In light of the “Jewish-Christian” profile of the Homilies and the echoes of Jubilees in earlier strata of Pseudo-Clementine tradition, one might further wonder whether Christians of this sort could have played some part in the translation of the work from Hebrew to Greek, whether before Africanus, or in the wake of his dissemination of extracts from the text. On these fronts, unfortunately, the evidence permits nothing beyond speculation. Retelling Biblical Retellings 321 position, relativizing the Torah (at least in its written forms) out of a sense of the dangers of interpretation; for them, the threat of “heretical” exegesis has become so pointed that they are willing to set aside the primacy of the written text of the Torah, so as to retain its perfection with respect to the oneness of God and the piety of his prophets. Consistent with the Homilies’ emphasis on orality, no written sources are mentioned by name, even as the content suggests possible dependence on Jubilees or similar traditions, as well as an openness to the self-presentation of such works as supplementary records of Sinaitic revelation. Epiphanius, then, falls somewhere in between. He seems aware that there are gaps in Genesis that allow for “heretical” interpretations, and that the text of Genesis might not always suffice to answer them. Hence, for him, the information in a book like Jubilees could prove especially useful, particularly when received as presifted, such as in the rationalistic framework of the Christian chronographical tradition. Today, Athanasius’s comments are so widely cited perhaps in part because they sound so familiar, and their repetition can function to add an aura of inevitability to the modern notion of the natural and inviolable bounds of “the Bible,” in comparison with which practices like “biblical retelling” or pseudepigraphy might seem like hubris or “heresy.” Yet, the example of Epiphanius cautions us against assuming that the story of the closing of the Christian canon comes to a tidy conclusion in 367 CE with Athanasius. Likewise, as we have seen, attention to the reception history of Jubilees stands as a reminder that modern labels like “Old Testament Pseudepigrapha” can conflate diverse texts that often had very different afterlives in the many centuries between their origins in Second Temple times and their integration into modern collections such as those of Johann Albert Fabricius, R. H. Charles, and James Charlesworth.51 Rather, here as elsewhere, the Nachleben of Second Temple Jewish texts and traditions remains stubbornly multivalent, reflecting the complexity and continued vitality of Christianity’s literary heritage from Second Temple Judaism into late antiquity and beyond.

51 See A. Y. Reed, “The Modern Invention of ‘Old Testament Pseudepigrapha,’ ” JTS 60/2 (2009): 403–36. Why is “A” Placed Next to “B”? Juxtaposition in the Bible and Beyond

Avigdor Shinan and Yair Zakovitch

The phenomenon of juxtaposition can be recognized as a literary or compositional technique from the early biblical strata, through Second Temple literature, to the rabbis. Moreover, juxtaposition of biblical passages is seen as an interpretative key both by the biblical writers themselves and by their later exponents. The most explicit recognition of juxtaposition as a literary feature of biblical texts, and exploitation of this feature as a hermeneutical key, occurs in rabbinic literature. We will begin with some examples from this corpus and work back through more implicit usages of juxtaposition in Second Temple literature, finishing with an exploration of the use of juxtaposition as both compositional technique and interpretive key in the Bible itself.

1 Juxtaposition in Rabbinic Literature

Rabbinic literature pays much attention to the contiguous placement of different biblical literary units—laws, stories, psalms, wisdom sayings, and so forth— and the purpose behind these arrangements. Rabbinic writings thus brim with questions that explicitly voice the wonder elicited by juxtaposed texts, and confirm our sense of the exegetical significance of these juxtapositions.1

* The Pentateuch is quoted according to the translation of R. Alter, The Five Books of Moses (New York: Norton, 2004); the Psalms according to R. Alter, The Book of Psalms: A Translation with Commentary (New York: Norton, 2007). Other biblical translations are from Tanakh: A New Translation of the Holy Scriptures according to the Traditional Hebrew Text (Philadelphia: Jewish Publication Society, 1985). The Babylonian Talmud is quoted according to The Babylonian Talmud (ed. I. Epstein; 35 vols.; London: Soncino, 1935–1952); and Midrash Rabbah according to: Midrash Rabbah (ed. H. Freedman and M. Simon; 10 vols.; London: Soncino, 1939). Josephus’s Jewish Antiquities is quoted according to the Loeb Classical Library translation ( Josephus in Nine [Ten] Volumes [trans. H. St. J. Thackeray et al.; LCL; London: Heinemann, 1934–1976], vols. 4–10). Occasional changes may have been made in these translations when they do not agree with our understanding of the sources. All other translations are ours unless otherwise indicated. 1 See I. Heinemann, Darkhe ha-Aggadah (Jerusalem: The Hebrew University Magnes Press, 1949), 140–43 (in Hebrew); Y. Fraenkel, Darkhe ha-Aggadah veha-Midrash (2 vols.; Givʿatayim:

A clear awareness of juxtaposition as a feature of biblical literature, reflecting a conscious recognition of the legitimacy of using this feature as basis for exegesis and indicative of the fundamental approach of the rabbis to the Bible, is found in b. Ber. 10a:

A certain min2 said to R. Abbahu: “It is written, ‘A psalm of David when he fled from Absalom his son’ (Ps 3:1). And it is also written, ‘A mikhtam of David when he fled from Saul in the cave’ (Ps 57:1). Which event happened first? Did not the event of Saul happen first? Then let him write it first!” He replied to him: “For you who do not derive interpretations from juxtaposition, there is a difficulty, but for us who do derive interpretations from juxtaposition there is no difficulty. For R. Johanan said: ‘How do we know from the Torah that juxtaposition (semikhut) counts? Because it says, “They are joined (semukhim) for ever and ever, they are done in truth and uprightness.” ’ (Ps 111:8).”3

In R. Abbahu’s opinion, Psalm 3 (the superscription of which speaks about a son’s rebellion against his father) was placed immediately following Psalm 2 (which this sage interprets as dealing with the war of Gog and Magog, i.e., humankind’s future war against God),4 to convey a specific teaching: “So that if one should say to you, ‘is it possible that a slave should rebel against his master?’ you can reply to him: ‘Is it possible that a son should rebel against his father?’ Yet this happened; and so this too [will happen].” In the halakhic midrashim, we find equally explicit statements on juxtaposition. For example, R. Aqiba concludes that the significance of the juxtaposition of literary units must be sought in each and every case: “R. Aqiba says: Every passage that is juxtaposed to another learns from it” (Sifre Num. 131). This principle is applicable, in the opinion of the rabbis, with regard to both halakhah and aggadah. Thus, for example, the midrash asks, “Why is the passage about

Yad la-Talmud, 1991), 2:182–84 (in Hebrew); and especially I. B. Gottlieb, Order in the Bible: The Arrangement of the Torah in Rabbinic and Medieval Jewish Commentary (Jerusalem: The Hebrew University Magnes Press, 2009), 38–73 (in Hebrew). 2 The identity of this min (heretic) is unclear. The fact that he is in conversation with R. Abbahu, who lived in Caesarea in the third century and who was in contact with both pagans and Christians, opens a few possibilities. See further, Gottlieb, Order in the Bible, 40–41, who supports the opinion that he was a Jewish–Christian. 3 See Ps 111:7, which specifies the subject of these verses as God’s works and commands. 4 This idea is based on the mention in Psalm 2 of “nations,” “peoples,” “kings of the land and their princes,” who set out to do battle against God and His Messiah. For this tradition of interpreting Psalm 2 see also b. Ber. 7b; b. ʿAbod. Zar. 60b; Tanḥ. Lekh Lekha 9. 324 Shinan and Zakovitch the wood-gatherer (Num 15:32–36) followed immediately by the passage about the fringes (vv. 37–41)?” The answer: “In order to teach you that the dead are obliged to wear fringes” (Sifre Zuta to Num 15:36). In this example, the connection between the anecdote of the wood-gatherer, who was condemned to death by stoning, and the commandment regarding the making of a fringed garment, serves as the basis for a halakhic ruling concerning a burial practice.5 Another halakhic example may be brought from the laws pertaining to levirate marriage:

R. Shesheth stated in the name of R. Elʿazar, who stated it in the name of R. Elʿazar b. ʿAzariah: “Whence is it proved that a sister-in-law who falls to the lot of a levir who is afflicted with boils is not muzzled? From the biblical text, ‘You shall not muzzle an ox when it threshes’ (Deut 25:4); and in close proximity to it is written, ‘Should brothers dwell together,’ etc. (5–10).” (b. Yebam. 4a)

The command “You shall not muzzle” applies to the muzzling of an ox’s mouth in order to prevent it from eating the crops it has just threshed. From the juxtaposition of this command with the text regarding levirate marriage, the sages conclude that, just as the ox’s mouth must not be muzzled, so, too, the woman obligated to a levirate marriage must not be silenced if she does not wish to undertake such a marriage, on the grounds that the man is ill. The legal ruling on such a matter demands that the man be required to release the woman, through ḥaliṣah, from her obligation to marry him.6 Juxtaposition is used also in order to draw conclusions in the field of ethics. In talmudic literature, for example, we find this:

It has been taught: Rabbi says, “Why does the section of the Nazirite (Num 6:1–21) follow immediately that of the unfaithful wife (Num 5:5– 31)? To teach you that anyone who sees an unfaithful wife in her evil ways should completely abstain from wine.” (b. Ber. 63a)

At the base of this saying stands the assumption that the drinking of wine is what caused the woman (and the man with whom she sinned) to leave the

5 For a discussion of this halakhah, see Gottlieb, Order in the Bible, 49 n. 56. He views it as an “enigmatic halakhah” and points to exegetical problems in regard to it, as emerges from the Tosafot and elsewhere. 6 In the continuation of this discussion, the Talmud states the opinions of various sages regarding the phenomenon of juxtaposition, and supplies additional examples. Why is “A” Placed Next to “B”? 325 righteous path. Thus, anyone who observes such a woman must take this into consideration and abstain from wine. In Moses’ blessing upon the tribes of Israel, the verse about Judah is juxtaposed with that about Reuben, in a departure from the birth order of the sons. The midrash derives meaning from this juxtaposition:

What has one thing to do with the other? It is because Judah did what he did and stood and confessed, “She is more righteous than I” (Gen 38:26). When Reuben saw that Judah had confessed, he stood and confessed, too, about his own deeds; one must therefore conclude that Judah made Reuben carry out repentance. (Sifre Deut. 348)7

According to this midrash, Reuben confesses to his sin with Bilhah (Gen 35:21– 22) following and because of Judah’s similar confession regarding Tamar. Along the same lines, Moses’ blessing upon the tribe of Reuben (“Let Reuben live and not die”; Deut 33:6) is then interpreted as the nullification of a curse that had been pronounced against him by his father.8

Until now we have seen examples in which the significance of juxtaposition has been questioned explicitly. More examples can be found, however, where the exegetical principle of learning from juxtaposition is in use even though it is not recognized outright, as in the following two cases:

Example 1:

“And he set up an altar there and called it El-Elohe-Israel” (Gen 33:20). He [Jacob] declared to Him: “You are God in the celestial spheres and I am a god in the terrestrial sphere.” R. Huna commented in the name of R. Simeon b. Lakish: “[God reproved him:] ‘Even the synagogue superin- tendent cannot assume authority of himself, yet you did take authority

7 J. Neusner, Sifre to Deuteronomy: An Analytical Translation (2 vols.; BJS 98, 101; Atlanta: Scholars Press, 1987), 2:422 (slightly altered). 8 The curse arises from the words of Jacob to Reuben in Gen 49:4 (“Unsteady as water, you’ll no more prevail!”). That said, it may be that a curse upon Reuben was originally included in the narrative at Gen 38:22 but was replaced with a mid-verse break (following the words “And Israel heard”), in order to prevent its repetition. The writer of Deuteronomy 33 may have been aware of such a curse against Reuben and intended that his blessing would cancel it. See A. Shinan and Y. Zakovitch, The Story about Reuben and Bilhah: Gen 35:21–26 in the Bible, the Old Versions and the Ancient Jewish Literature ( Jerusalem: Hebrew University, 1983), 12–13 (in Hebrew). 326 Shinan and Zakovitch

to yourself!? Tomorrow your daughter will go out and be dishonored!’ ” Hence it is written, “And Dinah, Leah’s daughter, whom she had borne to Jacob, went out . . .” etc. (Gen. Rab. 79:8)

For what reason did the calamity of Dinah’s rape befall Jacob? A variety of answers to that question are offered in rabbinic literature, along with that of R. Huna’s, who viewed it as punishment for Jacob’s prior prideful behavior.9

Example 2:

The sons of Aaron died only because they entered the Tent of Meeting drunk with wine. . . . Thus [it is said], “And fire came out from before the Lord and consumed them, and they died before the Lord” (Lev 10:2); but we would not know why they died, except for His commanding Aaron, “Wine and strong drink you shall not drink” (Lev 10:8). We know from this that they died precisely on account of the wine. (Lev. Rab. 12:1)

Nadab and Abihu’s enigmatic transgression when they offered “alien fire” is thus interpreted as having been an act of depravity brought on by their drinking wine, which is mentioned in the law that follows the account of the death of the two brothers.10 Many other examples of the use of juxtaposition, evoked either explicitly or implicitly, may be found in the rabbinic corpus.11 As we have seen from the foregoing discussion, juxtaposition was understood by the rabbis as a key to expansion of the narrative or halakhic meaning of the biblical text. In fact, juxtaposition allowed the crossing of generic boundaries in order to derive additional meanings from the text, as in the example cited above of the narrative of the wood-gatherer and the commandment to wear fringes (Num 15:32–40). Although juxtaposition appears in rabbinic literature as a fundamental and sophisticated principle of rabbinic exegesis, it did not originate with the rabbis, as the following discussion will show.

9 For more connections made in the ancient sources between the statement, “And Jacob came to Shalem” (Gen 33:18–20) and the story of the rape of Dinah (ch. 34), see A. Shinan and Y. Zakovitch, And Jacob Came “Shalem”: Gen 33:18–20 in the Bible, the Old Versions and the Ancient Jewish Literature (Jerusalem: Hebrew University, 1984), 78–79 (in Hebrew). 10 For other explanations for the death of Aaron’s sons, see A. Shinan, “Nadab and Abihu’s Transgression in Rabbinic Literature,” Tarbiz 48 (1979): 201–14 (in Hebrew). 11 They are brought in abundance in Gottlieb, Order in the Bible, who discusses other formu- las for the sages’ explicit identification of juxtapositions and brings examples of each. Why is “A” Placed Next to “B”? 327

2 Juxtaposition as an Interpretive Key in First-Century Sources

The use of juxtaposition as an interpretive key may also be found outside of rabbinic literature. In fact, we find examples of this phenomenon in Greek sources that are contemporary with the earlier sages, such as the writings of Flavius Josephus and the New Testament. In one such example, Josephus explains the juxtaposition of the story of the arrival at Elim (Exod 15:27) with the story that follows it, that of the Israelites’ complaints about Moses and Aaron (16:1–3). Josephus seems to be saying that the Israelites’ dissatisfaction with what they found in Elim drove them to resent Moses and Aaron. He describes Elim as a meager and poor oasis, in which there were insufficient water and fruit to nourish the great multitudes that had just left Egypt:

Departing thence, they reached Elis, a spot which from a distance made a good show, being planted with palm-trees, but on approach proved bad; for the palms, numbering no more than seventy, were dwarfed and stunted through lack of water, the whole place being sandy. For from the springs which existed, to the number of twelve, there oozed no liquid sufficient to water them: impotent to gush forth or rise to the surface these yielded but a few drops. . . . So they fell to accusing and denouncing their general, declaring that this misery and experience of woe which they were undergoing were all due to him. For it was now their thirtieth day on the march, the provisions which they had brought with them were all exhausted, and, lighting upon nothing whatever, they were in utter despair. (Ant. 3.1.3)

Was this Josephus’s own idea, or does it perhaps echo a tradition he heard in one of the study houses in Jerusalem, where he studied? An allusion to this tradition may also be found in Mekhilta d’Rabbi Shimon bar Yoḥai, which states that Elim was “a place more spoiled (mequlqal; another version has: “cursed” [mequlal]) than any place,” based on the fact that “there were twelve springs there but they only sustained seventy palm trees.”12 That is, the springs could support only a small number of trees.13 It is worth noting that despite this

12 J. N. Epstein and E. Z. Melamed, Mekhilta d’Rabbi Šimʿon b. Jochai ( Jerusalem: Mekitse Nirdamim, 1955), 105. See the translation of W. D. Nelson (Philadelphia: Jewish Publica- tion Society, 2006), 165. 13 In the Mekhilta d’Rabbi Ishmael it is said that Elim was a place “extolled (mehulal) for its water”; but this does not appear to be the original version, and mequlal should be 328 Shinan and Zakovitch

characterization of the place and in contrast to Josephus, in the continuation of the tradition in Mekhilta d’Rabbi Shimon bar Yoḥai it is said that, because of a miracle, the Israelites enjoyed an abundance of water there. In another place,14 Josephus seeks to explain the reason for the juxtaposition of the law concerning the red heifer (Numbers 19) with the story of Miriam’s death (Num 20:1).

And now it was that death overtook his sister Mariamme. . . . They buried her at the public expense in state on a mountain which they call Sin, and when the people had mourned for her thirty days, they were purified by Moses on this wise. A heifer, yet ignorant of the plough and of husbandry, without blemish and entirely red, was conducted by the high priest a little way outside of the camp to a place of spotless purity, where he sacrificed it and sprinkled with his finger drops of its blood seven times in the direction of the tabernacle of God. Next the heifer was burnt whole. . . . Its ashes were then all collected. . . . When therefore any had been polluted by contact with a corpse, they put a little of these ashes in running water, dipped hyssop in to the stream and sprinkled such persons therewith on the third and on the seventh day. . . . This ceremony Moses charged them to continue when they had entered upon their allotted territories. After a purification held in such wise in consequence of the mourning for the sister of their chief, he led his forces away. . . . (Ant. 4.4.6–7)

Contact with a corpse requires that one be purified, something that may be achieved with the use of ashes from the red cow. The midrash also saw a lesson implicit in the juxtaposition of these two passages, although a different one:

Abba b. Abina enquired: For what reason was the section recording the death of Miriam placed in close proximity to that dealing with the ashes of the red heifer? Simply this: to teach that, as the ashes of the heifer effect atonement, so the death of the righteous effects atonement. (Lev. Rab. 20)

Josephus is not alone among first-century nonrabbinic interpreters in assign- ing significance to juxtaposition. The New Testament, too, shows awareness of the significance of the proximity of different literary units. An example occurs

read. H. S. Horovitz (Mechilta d’Rabbi Ismael [ed. H. S. Horovitz and I. A. Rabin; Frankfurt: Kauffmann, 1928–1931], 158 [in Hebrew]), correctly suggests that this explanation of the name Elim may have been derived from ʾalah (“curse”). 14 As Heinemann, Darkhe ha-Aggadah, 143, has already noted. Why is “A” Placed Next to “B”? 329 in the Letter to the Hebrews, where the writer makes a connection between Exod 2:10, in which Moses is given his name, and 2:11, in which he sets out to see the suffering of his people:

By faith, when he grew up, Moses refused to be called the son of Pharaoh’s daughter, choosing rather to be ill-treated with the people of God than to enjoy sin’s fleeting pleasure. (11:24–25)15

In all three of the cases detailed here, juxtaposition is recognized as supplying a link of cause and effect between two otherwise disparate narrative episodes. It should be noted in addition that, again in all three cases (and in contrast to significant examples noted above from the rabbinic corpus), this recognition functions at an entirely implicit level. But again, the interpretive technique in itself is not an innovation of the first-century writers, as will be seen in the following section.

3 Juxtaposition in the Book of Jubilees

Going back in time to earlier Second Temple literature, we find the same phenomenon. The apocryphal Book of Jubilees establishes a connection between Genesis 15, the story of the covenant between the pieces, and Genesis 16, the narrative of Sarai and Hagar:

And on that day we made a covenant with Abram. . . . And Abram rejoiced, and he told all of these things to Sarai his wife. And he believed that he would have seed, but she did not give birth. And Sarai advised Abram, her husband, and she said to him, “Go into Hagar, my Egyptian maid. It may be that I will build seed for you from her.” (14:20–21)16

Sarai’s words derive from her confidence that Abraham will have progeny, as promised by God, and since she herself is unable to bear children, she offers him Hagar—a conclusion that emerges from the juxtaposition of the two stories. The writer of Jubilees also creates a relationship between Esau’s ravenous hunger when he returns from the field (Gen 25:29) and the famine that is described in the next chapter, chapter 26. For that reason, he first tells us of the

15 The Jewish Annotated New Testament: New Revised Standard Version (Oxford: Oxford University Press, 2011), 424. 16 Translation by O. S. Wintermute, OTP 2:85. 330 Shinan and Zakovitch famine, before then relating the story of the sale of the birthright—in contrast to the order in Genesis.

And in the first year of the fourth week a famine began in the land other than the first famine which occurred in the days of Abraham. And Jacob was cooking lentil soup, and Esau came in from the field hungry, and he said to Jacob, his brother, “Give me some of this reddish-colored soup.” (24:2–3)17

As we saw with the examples in the previous section, Jubilees taps the phenomenon of juxtaposition in order to read links of cause and effect between separate narrative units. The last example, however, represents a new twist on this technique: by reversing the narrative (and therefore cause-and-effect) sequence of the biblical account, Jubilees might be seen as creating its own juxtaposition, thus moving into the realm of using juxtaposition as a literary technique.

4 Juxtaposition as a Literary Technique in the Bible

Thus far, we have moved from later texts to earlier ones, from the literature of the sages and their earlier contemporaries (Flavius Josephus, the New Testament) to texts that were clearly written earlier in the Second Temple period, such as Jubilees. But it is clear that already in the biblical period, writers were aware that an exegetical-ideological dimension is added by the juxtaposition of two texts. Of course, in the Bible we cannot expect questions to be asked explicitly, as we saw in rabbinic literature; but the purposeful use of juxtaposition as a literary technique to create meaning, or the recognition of its significance for interpreting existing traditions, is certainly apparent. We will demonstrate these phenomena using texts from different literary genres: 1) poetry; 2) law; 3) narrative; and 4) prophecy.

4.1 Poetry We begin with an example that comes from the very edge of the biblical corpus, Psalm 151. This Psalm, not included in the MT, is found in the Septuagint and in a slightly different version in the Dead Sea Scrolls (11QPsa 28:3–14).18

17 Ibid., 102. 18 J. A. Sanders, The Psalms Scroll of Qumrân Cave 11 (11QPsa) (DJD 4; Oxford: Clarendon, 1965), 54–64; idem, The Dead Sea Psalms Scroll (Ithaca: Cornell University Press, 1967), 96. Why is “A” Placed Next to “B”? 331

The reader of the story of David’s anointing (1 Sam 16:1–13) may wonder why God chose David—what it was that God discerned in him—for, as God himself said to Samuel, the grounds for such a choice were “not as man sees, for man sees what is visible, while God sees into the heart” (1 Sam 16:7). Psalm 151 answers this question on the basis of the juxtaposition in chapter 16 of the story of David’s anointing and the story that follows it (vss. 14–23), which relates how David came to play music for King Saul. According to the psalm, God knew that David, of his own accord, wished to praise Him, and that it was for this reason that David created musical instruments. The psalm also offers an explanation of how a young herdsman could have been a talented musi- cian, worthy of appearing before the king. We cite here the Qumran version:19

A Hallelujah of David, son of Jesse. I was smaller than my brothers and the youngest of my father’s sons; He made me shepherd of his flock and ruler over his kid goats. My hands made a flute, my fingers a lyre, And I gave glory to yhwh. I said to myself: the mountains do not witness to me, nor do the hills proclaim on my behalf, [Nor] the trees my words, [nor] the flock my deeds. Who, then, is going to announce and who will speak and who will recount my deeds? The Lord of all saw; the God of all, he heard, and he listened. He sent his prophet to anoint me; Samuel to make me great; My brothers went out to meet him, handsome of figure and handsome of appearance. Though they were tall of stature, handsome by their hair, yhwh God did not choose them, but sent to fetch me from behind the flock and anointed me with holy oil, and made me leader of his people and ruler over the sons of his covenant.

See more recently M. Segal, “The Literary Development of Psalm 151: A New Look at the Septuagint Version,” Textus 21 (2002): 139–58 (and the vast bibliography therein). 19 The translation is drawn (slightly altered) from F. García Martínez and E. J. C. Tigchelaar, Dead Sea Scrolls Study Edition (2 vols.; Leiden: Brill, 1997–1998), 2:1179. Their reading of the Hebrew, which differs somewhat from that of Sanders, follows that of P. Auffret, “Structure litteraire et interpretation du Psaume 151,” RevQ 9 (1977): 163–88. See also D. Amara, “Psalm 151 from Qumran and its Relation to Psalm 151 LXX,” Textus 19 (1998): 1*–35* (in Hebrew); and Segal, “Literary Development,” esp. 140. 332 Shinan and Zakovitch

It should be noted that, in the Psalms Scroll from Qumran, this psalm was immediately followed by another, of which only the superscription and beginning of the first verse survive: “Beginning of David’s po[w]er, after God’s prophet had anointed him. Then I saw a Philistine threatening from the ran[ks of the Philistines.] I [. . .]”20 Thus, the sequence of the Psalms Scroll suggests an implicit link between the events of 1 Samuel 16 and the story of Goliath, told in the following chapters. In fact, the Septuagint’s version of Psalm 151 makes that link explicit. Verses 6–7 of LXX 151 refer to David’s encounter with the “foreigner,” who cursed David “by his gods” (1 Sam 17:43), and whom he beheaded with the warrior’s own sword (1 Sam 17:51). And indeed, although the MT Book of Psalms contains no psalm with a superscription connecting it with the story of Goliath—unlike the thirteen other events that did achieve such status there—an allusion to David’s victory over the giant has been preserved in Psalm 144, a psalm that bears the title “For David” and that opens with the words, “Blessed is the Lord, my rock, Who trains my hands for battle, my fingers for the fray.” Within the lines of this psalm, whose opening refers to both “battle” and “fray,” we find a clear repetition: verses 7–8 (“Send forth Your hand from on high, redeem me and save me from the many waters, from the foreigners’ hand, whose mouth speaks falsely, and whose right hand is a right hand of lies”) are duplicated almost verbatim in verse 11 (“Redeem me and save me from the foreigners’ hand, whose mouth speaks falsely, and whose right hand is a right hand of lies”). Between these two repeated phrases we find two verses: “God, a new song I would sing to You, on a ten-stringed lute I would hymn to You, Who grants rescue to kings, redeems David His servant from the evil sword” (vv. 9–10). These two verses would appear to be an addition, which engendered the repetition in an attempt to repair the break caused by the added lines. The addition stemmed from a reader who understood the previous verses as though they spoke about David’s victory over Goliath, since they contain a prayer for deliverance from a foreigner who speaks falsely. As we know, the story of David and Goliath emphasizes more than once that the Philistine enemy taunted the Israelite army and God (1 Sam 17:10, 26, 45). The added verses (vv. 9–10) appear to have been motivated by the juxtaposition of the story of David’s music playing (1 Sam 16) and his victory over Goliath (1 Sam 17). The passage begins with the words, “a new song I would sing to You, on a ten-stringed lute I would hymn to You.” It is worth noting that “lyre” (kinor) appears in 1 Sam 16:16, 23; “lute” (nevel) appears in Psalm 144;

20 11QPsa 28:13-14; see Sanders, Psalms Scroll, 98. Why is “A” Placed Next to “B”? 333 and they appear together, in parallel, in Psalm 151.21 A few of the expressions in Ps 144:9–10 (“grants rescue” [ poṣeh], “redeems” [hanoten teshuʿah]) likewise allude to other biblical stories that speak of David and his victories. In 1 Sam 19:5—when Jonathan describes David’s victories to Saul—he says, “He took his life in his hands and killed the Philistine, and the Lord wrought a great victory [teshuʿah] for all Israel.” David himself says twice, “As the Lord lives, who has rescued [padah] me from every trouble” (2 Sam 4:9; 1 Kgs 1:29); paṣah is an alternative pronunciation of padah.22 The connection with the Goliath story is further strengthened by the explicit reference to David in v. 10; and, even more by the mention of the sword from which David was saved (which, as we saw above, is particularly connected with the encounter with Goliath; cf. 1 Sam 17:50–51). Thus, the connections between 1 Sam 16 and 17, expounded explicitly by the author of Psalm 151 LXX, already constitute part of the literary fabric of MT Psalm 144. It is no wonder, then, that the Septuagint’s superscription to Psalm 143 (MT 144) states explicitly that the psalm is “For David about Goliath.” The writer of Midrash Tehillim (to this psalm) makes a similar connection, noting that David

went down to battle and won, and said, “Not from my heroism was I triumphant but for that He came to my aid, and He caused me to triumph and I triumphed, and created me to be a warrior.” . . . And so he said, “Blessed is the Lord, my rock, Who trains my hands for battle, my fingers for the fray. And when did He train my fingers? When I killed Goliath.”23

4.2 Law The relationship between biblical narrative and biblical law is apparent in the two examples we now offer, which illustrate opposite phenomena. In the first,

21 The kinor is a stringed instrument in the lyre family. The nevel, also a stringed instrument, is mentioned in the Bible almost always together with the kinor. According to the Mishnah, its strings were thicker than those of the kinor (m. Qinnim 3:6), and so its sound was louder and its range deeper than that of the kinor. See B. Bayer, “Playing and Singing,” Encyclopaedia Biblica, 5:763–71 (in Hebrew). 22 On the interchanging of p-d-h and p-ṣ-h, see Y. Zakovitch, “On Interchanges of Dalet and Ṣade in Biblical Hebrew,” in Language Studies 11–12 (= Avi Hurvitz Festschrift) (ed. S. E. Fassberg and A. Maman; Jerusalem: The Department of Hebrew and Jewish Languages, The Hebrew University of Jerusalem, 2008), 113–20 (in Hebrew). 23 The translation is that of W. G. Braude, The Midrash on Psalms (2 vols.; New Haven: Yale University Press, 1959), 2:358. 334 Shinan and Zakovitch the juxtaposition of two laws forms the basis for a story, while in the second, we find the reverse.

Example 1: The law in Exod 22:4 that discusses letting one’s flock loose to graze in another’s field is phrased in an unusual way: ki yavʿer ʾish sadeh ʾo kerem, veshilaḥ ʾet beʿiroh, u-viʿer bisdeh ʾaḥer: “Should a man [let his beast loose] to graze in a field or vineyard, and let his beast loose to graze, and it grazes in another’s field. . . .” Verbs from the root b-ʿ-r are common in the context of fire, but are rarely used with the meaning of “letting livestock loose to graze.”24 The law was likely phrased this way in order to assimilate it to the law that follows (v. 5); here, the person responsible for the destruction of a field by fire is called hamavʿir et-habʿerah. The shared terminology creates a short collection dealing with damages caused to a piece of land either by allowing an animal to stray or by causing a fire. The similarity between the two laws caused some readers to interpret v. 4 as though it, too, dealt with fire damage. Such an interpretation is found in the Fragment Targum: “Should a man clear a field or a vineyard and send forth his fire, which [then] burns another’s field . . .,” and in Targum Ms. Neofiti 1: “Should a man set a fire in a field or a vineyard and leave the fire to spread and (thus) cause fire in the field of another . . .”25 It seems to us that the interpretation of Exod 22:4 as dealing with fire is much older than its overt expressions in the Aramaic targumim. We find our proof in the story of Samson, when he sets fire to the fields of the Philistines (Judg 15:4–5): “setting fire to stacked grain, standing grain, vineyards, and olive trees.” The narrator, in order to convey the severity of Samson’s deed, created

24 Instances in piʿel: Isa 3:14; 5:5; 6:13; in hiphʿil, the term carries this meaning only in our verse—all other occurrences refer to fire (see Exod 22:5, and also Judg 15:5; Ezek 5:2; Nah 2:14). On the problematics of this verse and its possible explanations, see B. S. Jackson, “A Note on Exodus 22:4 (MT),” JJS 27 (1967): 138–41. 25 See M. Klein, The Fragment-Targums of the Pentateuch: According to Their Extant Sources (2 vols.; Rome: Biblical Institute Press, 1980), 1:ad loc; A. Díez Macho, Neophyti 1: Targum Palestinense ms. de la Biblioteca Vaticana (6 vols.; Madrid: CSIC Press, 1968–1979), 2:ad loc. G. Schelbert has found evidence for this interpretation in rabbinic halakhah as well, in the mishnaic expression ha-sholeaḥ ʾet habeʿarah (“the one who sets fire”; m. B. Qam. 6:4–5); see G. Schelbert, “Exodus 22:4 in Palästinischen Targum,” VT 8 (1958): 262. J. Heinemann found two halakhic midrashim dealing with fire damages that use Exod 22:4—but not 22:5—as a prooftext! Indeed, all these pieces of evidence lead to the inevitable conclusion that we have here a clear example of a premishnaic halakhah. See J. Heinemann, “Early Halakhah in the Palestinian Targumim,” JJS 25 (1974): 114–22. Why is “A” Placed Next to “B”? 335 associations with the language of Exod 22:5: “Should a fire go forth and catch in thorns, and stacked or standing grain or the field be consumed . . .” The similarities emphasize the difference between the two passages: while the law deals with two separate cases—either stacked or standing grain is consumed—Samson’s foxes destroy both. Furthermore, the law speaks of damage caused as a result of negligence, while Samson sets his fire intentionally. The law’s assumption is that only one field is damaged, while Samson destroys all of the Philistines’ cultivated land. Reading the story of Samson carefully, one finds that the narrator did not limit himself to Exod 22:5, but made use also of the verse that precedes it. Compare Exod 22:4 and Judg 15:5:

Exod 22:4 Judg 15:5

וַ ַ ּיבְ עֶ ר־ ֵׁש ֹאּבַּלַּפִידִ ים ּכִ י יַבְעֶ ר־ ִ ֹאיׁש ׂשָ דֶ ה ֹאו־כֶרֶ ם ַ ו ְ י שַׁ ּ ַ לח ּבְקָ ֹמות ּפְ לִ ׁשְ ּתִ י ם וְשִׁ ּלַח אֶ ת־בעירה וַ ַ ּיבְ עֵ ר ּג ֹמִ ָ י דִ ׁש וְעַד־קָ ָ ֹמה וְעַד־ּכֶרֶ ם זָ� ֽ יִ ת ּו בִ עֵ ר ּבִ ׂשְ דֵ ה ַ ֹאחֵ ר . . .

the torches and [וַּיַבְעֶ ר־ ֵ ֹאׁש] Should a man [let his beast loose] to and he lit into the [ ְוַי ׁשַ ּלַ ח] in a field or vineyard, let [the foxes] loose [יַבְעֶ ר־ ִ ֹאיׁש] graze [ וַ ּיַ בְ עֵ ר] standing grain . . . setting fire ׁשִ וְ ּלַ ח] and let his beast loose to graze ,in to the stacked grain, standing grain [ ּו בִ עֵ ר] and it grazes ,[ ֶ ֹאת־בעירה another’s field . . . vineyards, and olive trees.

Example 2: An example of the reverse phenomenon, in which the juxtaposition of two stories is used in the creation of a law, is found in relation to the stories about Massah and Meribah (Exod 17:1–7) and the war against Amalek (17:8–16). The story of Massah and Meribah concludes with an etymology: “And he called the name of the place Massah and Meribah [Testing and Dispute], for the disputation of the Israelites, and for their testing of the Lord” (17:7a). This etymology was already alluded to in the account of the people’s complaint at the story’s beginning: “And Moses said to them, ‘Why do you dispute with me and why do you test the Lord?’ (17:2).” To this etymology have been appended a few explanatory words, which open with the formulaic “saying [lʾemor]”: “saying, ‘Is the Lord in our midst or not?’ ” (v. 7b). These words are not an integral part of the etymology: When lʾemor appears in etymologies, it usually introduces an explanation that 336 Shinan and Zakovitch

contains an easily distinguishable sound-play on the name.26 In our verse, on the other hand, a different formula was used for the explanation: “for [the disputation] . . . and for [their testing] . . .” The added question construes the test at Massah and Meribah as though the Israelites’ request did not stem from their need for water but from a desire to prove whether or not God was really with them. In the ensuing story of the war with Amalek (vv. 8–16), when Moses (who holds God’s staff in his hand; see v. 9) “would raise his hand, Israel prevailed, and when he would put down his hand, Amalek prevailed” (v. 11). In the original version of the story, Moses’ lowering of his hand was a sign of physical weakness, so that he required the support of Aaron and Hur (v. 12). According to the exegetical viewpoint of the addition in 7b, however, this action provides a proof for the Israelites that God is in their midst; when Moses raises the staff, he signifies God’s presence. It thus seems that the question, “Is the Lord in our midst or not?” was added to the story of Massah and Meribah in order to bind it to the story that follows, and to combine the two stories with a unifying interpretation. The added words in verse 7 lend to each story a new layer of meaning that is not present in either episode when read separately. The war with Amalek becomes a sort of punishment for the complaints at Massah and Meribah, and shows the Israelites that God is indeed among them. The sages, therefore, correctly inferred a connection between the two stories:

It is written: “Is the Lord in our midst or not?”—“And Amalek came.” What has one to do with the other? This resembles a toddler who rode on his father’s shoulders and saw his father’s friend. He said to him, “Did you see my father?” His father said to him, “You ride on my shoulders, and you ask about me? I throw you off and let the enemy come and dominate you.” So said the Holy One, blessed be He, to Israel: “I carried you on clouds of glory and you say, ‘Is the Lord in our midst or not?’ Therefore, let the enemy come and dominate you.” And Amalek came. . . . (Exod. Rab. 22:3)

Indeed, within the Pentateuch itself, in Deuteronomy, a law makes explicit what is expressed only implicitly in Exodus: “Remember what Amalek did to you on the way when you came out of Egypt, how he fell upon you on the way and cut down all the stragglers, and you were famished, and exhausted, and not fearing God” (25:17–18). The last clause, “and you were famished and

26 See, e.g., Gen 5:29 (Noah); 30:24 (Joseph); Judg 6:32 (Yerubaʿal); 1 Sam 4:21 (Ichabod); 1 Chr 4:9 (Yabeṣ). Why is “A” Placed Next to “B”? 337 exhausted and not fearing God,” is interpreted by most readers as though the first two verbal phrases speak about Israel (“famished and exhausted”) and the last speaks about Amalek (“he did not fear God”). This reading is evident already in the LXX, the midrash, and in medieval exegesis. It is similarly interpreted by many modern translators and commentators.27 This division of the clause into two parts, with the presumed switch in the subject of the verbs, has nothing to support it. The entire verse speaks of the Israelites: “not fearing God” refers to their confrontational question in Exod 17:7, “Is the Lord in our midst or not?”—the question that resulted in their punishment by way of the war with Amalek. Furthermore, it is well known that the word ʿayef can mean “thirsty”;28 it is used here in order to allude to the thirst of the Israelites in the story of Massah and Meribah.

4.3 Narrative In the story of Balak, the King of Moab is told that he—along with the prophet Balaam—will not be able to harm Israel as long as God wishes for their well being. God’s blessing of the Israelites is not arbitrary but results from their righteousness: “I have perceived no guilt in Jacob, have seen no perversity in Israel. The Lord his God is with him, and a royal acclamation to greet him” (Num 23:20–21). Although Balaam’s own words suggest that God will not retract his blessing, the immediate continuation of the story suggests otherwise. As soon as Balaam and Balak part company (Num 24:25), the Israelites sin with the daughters of Moab–Midian, through whoring accompanied by idolatry (Num 25:1–3); a plague breaks out, killing 24,000. The juxtaposition implies that the blessing is given conditionally, dependent on the nation’s behavior. The moment that the Israelites sin, God’s blessing may be removed. The biblical writer’s awareness of the implications of this juxtaposition is made clear in Numbers 31, which relates the revenge of Israel against Midian. To that chapter were added two passages, which interrupt the narrative unity: 1) “. . . and Balaam son of Beor they killed by the sword” (v. 8b; cf. Josh 13:22); and 2) “These are the ones who led the Israelites by Balaam’s word to betray the Lord’s trust in the affair of Peor, and there was a scourge against the Lord’s

27 See for example, Midrash Tanḥuma, Ki Teṣe 16 (ed. S. Buber; Vilna: Romm, 1885); Rashi ad loc.: “And not fearing—Amalek”; ibn Ezra ad loc.; and many others. Among translations, note KJV, RSV, NIV, NJPS, to name a few. 28 See M. Z. Kaddari, A Dictionary of Biblical Hebrew (Ramat Gan: Bar-Ilan University, 2007), s.v: ʿayef, 796 (in Hebrew). 338 Shinan and Zakovitch

community” (v. 16).29 These two verses portray Balaam as responsible for Israel’s sin, as though he himself had suggested to Balak that as long as the Israelites’ behavior was righteous, God’s blessing would be with them; but that if Balak were to find a way to pull them from the righteous path, the blessing would be removed. This tradition is made explicit in Pseudo-Philo’s Biblical Antiquities:

And then Balaam said to him [Balak]: “Come and let us plan what you should do to them. Pick out the beautiful women who are among us and in Midian,30 and station them naked and adorned with gold and pre- cious stones before them. And when they see them and lie with them, they will sin against their Lord and fall into your hands; for otherwise you cannot fight against them.” And on saying this, Balaam turned away and returned to his place. And afterward the people were seduced after the daughters of Moab. For Balak did everything that Balaam had showed him (L.A.B. 18:13–14).31

A similar description appears in Josephus’s Antiquities:

Balak, furious because the Israelites had not been cursed, dismissed Balaam, dignifying him with no reward. But he, when already departing and on the point of crossing the Euphrates, sent for Balak and the princes of Madian and said: “Balak and you men of Madian here present—since it behooves me despite God’s will to gratify you—doubtless this race of Hebrews will never be overwhelmed by utter destruction, neither through war, nor through pestilence and dearth of the fruits of the earth, neither shall any other unlooked-for cause exterminate it. For God is watching over them to preserve them from all ill and to suffer no such calamity to come upon them as would destroy them all. Yet misfortunes may well befall them of little moment and for a little while, whereby they will appear to be abased, though only thereafter to flourish once more to the terror of those who inflicted these injuries upon them. You then, if you yearn to gain some short-lived victory over them, may attain that end by acting on this wise. Take of your daughters those who are come- liest and most capable of constraining and conquering the chastity of

29 For the secondary nature of these verses, see M. Noth, Numbers (OTL; trans. J.D. Martin; London: SCM, 1968), 230–31. 30 This is an attempt to reconcile Num 25:1 (which mentions the Moabite women) with the rest of chapter 25, which speaks about Midianite women. 31 Translation by D. J. Harrington, OTP 2:326. Why is “A” Placed Next to “B”? 339

their beholders by reason of their beauty, deck out their charms to add to their comeliness, send them to the neighborhood of the Hebrews’ camp, and charge them to company with their young men when they sue their favors. Then, when they shall see these youths overmastered by their passions, let them quit them and, on their entreating them to stay, let them not consent . . . [until] they have induced their lovers to renounce the laws of their fathers and the God to whom they owe them, and to worship the gods of the Madianites and Moabites. For thus will God be moved to indignation against them.” And, having propounded to them this scheme, he went his way. Thereupon the Madianites having sent their daughters in accordance with his advice, the Hebrew youths were captivated. . . . (Ant. 4.6.6)

The midrash continues this interpretative direction:

“These are the ones who led the Israelites by Balaam’s word” (Num 31:16). What was Balaam’s word? He said to them: “Even if you bring all the multitudes in the world, you cannot overcome them. . . . But come and let me give you advice, what you should do. Their God hates whoredom, so let them have your wives and daughters for whoring and they will be possessed with whoredom and their God will overwhelm them. This is the rule: As long as they do His will, He fights for them. As it is said, “The Lord shall do battle for you.” And when they don’t do His will, He fights them. (Sifre Num. Matot 157)

4.4 Prophecy

Example 1: Much of Genesis 10–11 has to do with the topic of names. In chapter 10:21–31 we find the list of names of Shem’s descendants (shem is the Hebrew word for “name”). Following this comes the story of the Tower of Babel (11:1–9), in which the building of the Tower is motivated by the desire of the descendents of Noah to “make a name for ourselves”; the story concludes with the statement, “for this reason the name of it was called Babel, for there the Lord made the language of all the earth babble” (v. 9). The rest of chapter 11 (vv. 10–32) recapit- ulates the names of Shem’s descendants, in order to describe the generations that lead to Abram. Chapter 12 opens with God’s command to Abram to “Go forth from your land,” and with God’s promise to Abram to “make your name great” (v. 3); Abram, for his part, builds an altar to the Lord and “invoke[s] the name of the Lord” (v. 8). Abram’s actions contrast with those of the generation 340 Shinan and Zakovitch of Babel. That generation builds a tower in order to glorify its own name and prevent its dispersion throughout the earth (pen naphuṣ al-pnei kol ha-areṣ; 11:4), but their actions result in the dispersion they fear (11:8–9). Abram, who answers a divine command to wander from his home, is awarded God’s promise of a land, descendents, and a name; he builds an altar in the name of God and invokes God’s name there. The prophet Zephaniah, it seems, availed himself of the juxtaposition of these two stories in the prophecy in Zeph 3:9–13, where echoes of both may be heard:32

For then I will make the peoples pure of speech [safah berurah], so that they all invoke the Lord by name [liqro’ kulam beshem Yhwh] and serve Him with one accord. From beyond the rivers of Cush, My suppliants who were scattered [ʿatarai bat puṣai] shall bring offerings to Me. In that day, you will no longer be shamed for all the deeds by which you have defied Me. For then I will remove the proud and exultant within you, and you will be haughty no more on My sacred mount. But I will leave within you a poor, humble folk, and they shall find refuge in the name of the Lord. The remnant of Israel shall do no wrong and speak no falsehood; a deceitful tongue [leshon tarmit] shall not be in their mouths.

The first verse, “For then I will make the peoples pure of speech,” should be con- trasted with the story of the scrambling of the languages in Genesis 11: “Come, let us go down and baffle their language” (11:7); “for there the Lord made the language of all the earth babble” (11:9). The God who punished humankind with babbled (belula) speech will in the future grant them “pure” (berurah) speech. Furthermore, the pronouncement, “For then I will remove the proud and exultant within you, and you will be haughty no more on My sacred mount” (Zeph 3:10), alludes to the growing tower and the hubris of those who built it (for the establishment of their own name, rather than the name of the Lord). The words ʿatarai bat puṣai (3:10) allude to the dispersion of the builders at the conclusion of the Tower story, “And the Lord scattered them [va-yapheṣ] from there over all the earth. . . . And from there the Lord scattered them [hephiṣam]

32 Some of the similarities between this prophecy and the story of the Tower of Babel have been discussed by E. Eshel and M. E. Stone, “The Holy Language At the End of Days in the Light of a New Fragment Found At Qumran,” Tarbiz 62 (1993): 175–96 (in Hebrew); and more recently by S. Aḥituv, “Zephaniah,” in Nahum, Habakkuk, Zephaniah (M. Cogan and S. Aḥituv; Mikra le-Israel; Tel Aviv: Am Oved; Jerusalem: The Hebrew University Magnes Press, 2006), 49 (in Hebrew). Why is “A” Placed Next to “B”? 341 over all the earth” (vss. 8–9). The prophet promises the return of those who were scattered. But the prophecy also contains an allusion to the story about Abram in Genesis 12. The prophet says that in the future, the nations will, “all invoke the Lord by name and serve Him with one accord,” a formulation that echoes the description of Abram, who invokes the name of the Lord (Gen 12:8). Note also Zeph 3:12, where the prophet promises the people that they will find “refuge in the name of the Lord,” an allusion to humility and the recognition of God’s benevolence. Thus, in the prophet’s reading, the future redemption reverses both the sin and the punishment of the Babel generation, when God recreates the speech of the nations so that they, like Abram, will call on the name of the Lord.

Example 2: We conclude with an example in which a prophecy is based on the juxtaposition of two other prophetic texts. At the end of the book of Amos, we find two distinct literary units. The first deals with the reestablishment of Judah and the House of David, along with the fall of Edom (9:11–12); the second speaks of the great abundance with which those returning to Israel from captivity will be blessed (vv. 13–15).

In that day, I will set up again the fallen booth of David. . . . So that they shall possess the rest of Edom and all the nations once attached to My name—declares the Lord who will bring this to pass. A time is coming—declares the Lord—when the plowman shall meet the reaper, and the treader of grapes him who holds the [bag of] seed; When the mountains shall drip wine and all the hills shall wave [with grain]. I will restore My people Israel. They shall rebuild ruined cities and inhabit them; they shall plant vineyards and drink their wine; they shall till gardens and eat their fruits. And I will plant them upon their soil, nevermore to be uprooted from the soil I have given them. (Amos 9:11–15)

The prophet Joel, who knew Amos’s prophecy,33 concludes his book with a prophecy which not only references the earlier prophecy but also unites the two independent passages in Amos into one literary unit. The new prophecy contrasts the fecundity of Judah with the desolated wildernesses of Edom and Egypt. In this way, Joel intensifies his portrayal of the revenge on Edom and widens the gap between Judah’s fate and that of Edom.

33 So, too, argues J. L. Crenshaw, Joel (AB 24c; New York: Doubleday, 1995), 198. 342 Shinan and Zakovitch

And in that day, the mountains shall drip with wine, the hills all flow with milk, and all the streambeds of Judah shall flow with water. A spring shall issue from the House of the Lord and shall water the Wadi of the Acacias. Egypt shall be a desolation, and Edom a desolate waste, because of the outrage to the people of Judah, in whose land they shed the blood of the innocent. But Judah shall be inhabited forever, and Jerusalem throughout the ages. Their unavenged blood shall be avenged, and the Lord shall dwell in Zion. (Joel 3:18–21)

In the foregoing examples, we have seen that at various compositional levels, the biblical writers perceived and used juxtaposition as a meaningful element in the crafting of biblical texts. Juxtaposition of narrative episodes is seen to imply a relation of cause and effect between them, which can then be acti- vated in other contexts and genres, as we saw in the case of Psalms 144 and 151 and the David cycle. Juxtaposition of legal texts can form a basis for both legal expansion and development of a narrative, as we saw in the case of Exod 22:4–5 and the story of Samson.34 And prophetic literature freely draws on both pen- tateuchal material and extant prophetic collections, using juxtaposition to create new “narratives” of redemption.

5 Conclusion

In the course of this paper we have explored both the literary phenomenon of juxtaposition and the awareness of this phenomenon as a hermeneutical key, in biblical and biblically related literature. Beginning with rabbinic literature, which features the most explicit recognition of the phenomenon and the conscious use of it as a basis for interpretation, we have seen it used as an interpretive key to the biblical texts in earlier literature of the late first century and the earlier Second Temple period. As the examples in the last section have shown, such interpretations, though extrabiblical, take their cues from literary features inherent in the composition of the biblical text. It is thus evident that an awareness of juxtaposition and the desire to use it or even interpret it were not postbiblical innovations; we find early evidence of the technique already within the pages of the Hebrew Bible itself.

34 And note the similarity of this technique to the later rabbinic connection of the narrative of Nadav and Abihu (Leviticus 10) with the injunction against priests drinking wine, discussed above. The Reception and Reworking of Abraham Traditions in Armenian

Michael E. Stone

In this paper, I shall discuss a corpus of mostly narrative texts about Abraham that are preserved in late medieval manuscripts in Armenian. The dates of origin of these traditions are not explicit, but they can in general be set prior to the tenth century, when their constituent elements appear in a number of sources.1 These texts cannot be older in Armenian than the fifth century, at which time the Armenians began to write their own language. S. P. Brock, in a fine study of the Syriac story of Abraham and the ravens,2 isolates points at which the Syriac tradition differs from the version of the same events in Jubilees. At virtually all these points, the Armenian Abraham saga resembles the Syriac, though I would not claim it is derived from Syriac. The story of Melchizedek, of which I shall speak below, is most closely cognate to the Greek Melchizedek tale, while certain other distinctive traditions have not been found in any language but Armenian. Of course, the date of constitutive traditions is not necessarily the date of any particular literary formulation of that tradition.

1 The Character of the Armenian Abraham Traditions

The Armenians showed a deep interest in Abraham, expressed not only in apocryphal narratives, but in poetry, art, and exegesis as well. To trace all this

* Much of the material presented in this paper was subsequently incorporated into the “General Introduction” to my book, Armenian Apocrypha Relating to Abraham (SBLEJL 37; Atlanta: Society of Biblical Literature, 2012). 1 Naturally, direct manuscript evidence before the tenth century is not available, since the oldest literary manuscript, located in the Matenadaran, Institute of Ancient Manuscripts, Erevan is dated to 981 (M2679); see A. Mathévossian, A Book of Knowledge and Belief by Priest David: The Oldest Armenian Manuscript on Paper, 981 (2 vols.; Erevan: Matenadaran-Nairi, 1995, 1997). In addition, we find traditions distinctive to this corpus in authors from the tenth century, such as Samuel Kamrǰajorecʿi (MH tenth century, 742–43 [in Armenian]); and Grigor Narekac‘i (Book of Lamentations [ed. P. M. Xač‘atryan and A. A. Łazinyan; Erevan: Academy of Sciences, 1985], 622 [93.5] [in Armenian]). 2 S. P. Brock, “Abraham and the Ravens: A Syriac Counterpart to Jubilees 11–12 and Its Implications,” JSJ 9 (1978): 135–52.

© koninklijke brill nv, leiden, ��5 | doi ��.��63/9789004299139_015 344 Stone abundance is beyond our scope here, but please bear in mind that the corpus of fifteen documents, mainly narratives, which I shall discuss is far from exhaustive.3 After all, in its simple, biblical4 form, Abraham’s is a very dramatic story, moving from one exciting incident to another—Abraham’s migration to the Land of Israel, his battle against the four kings, the double narratives of Sarah in the palaces of pagan monarchs, the story of Lot, the burning of Sodom and Gomorra, and the binding of Isaac (the Aqedah). Above all, in Christian thought, the visit of the three “men” and their annunciation of Isaac’s birth to Abraham, as well as Abraham’s offering of Isaac, played a pivotal role. Here our attention will be directed to the tradition, transmission, and transformation of these stories in their Christian, specifically Armenian, retelling. Which elements of the biblical narrative were emphasized and which omitted, and why? Which nonbiblical episodes were introduced? Which ideas were reinterpreted? It is to such questions that we shall direct our gaze. The Armenians narrated the biblical story, inviting grist for the mill of any storyteller, with clearly Christian reformulations. Christians emphasized Abraham’s role as the father of all believers (cf. Rom 4:16) and the idea of the bosom of Abraham as the resting place of the righteous souls (Luke 16:23). In Jewish and Christian stories, Abraham’s discovery of God was a focus of fascination. Numerous retellings of this event occur in varied sources as far back as Jubilees, and in the Armenian stories retailed here ancient traditions are mixed with newer ones. The strange story, taken as paradigmatic yet always puzzling, of Abraham offering Isaac5 is not connected solely with a trial of the patriarch’s faith. For Christians, it foreshadowed God’s offering of his Son, and so the central mystery of Christian understanding of the world.

3 See Stone, Armenian Apocrypha Relating to Abraham, for editions and translations of these fifteen narratives; the translations used in this article are drawn from that volume. See ibid., xv–xvi, for lists of the works and their manuscripts. In this article, the texts are cited by short titles derived from those lists, and by section number within the manuscript. 4 I use “Bible” and “biblical” to denote the books of the Hebrew Bible. Old and New Testaments together are designated “Scripture” and “scriptural.” 5 For two interesting perspectives, see D. Shulman, The Hungry God: Hindu Tales of Filicide and Devotion (Chicago: University of Chicago Press, 1993); and J. Licht, Trial in the Bible and in Second Temple Judaism ( Jerusalem: The Hebrew University Magnes Press, 1973) (in Hebrew). Obviously, the scholarly and exegetical literature on this topic is enormous. The stark narrative in Genesis is finely presented by E. Auerbach, Mimesis: the Representation of Reality in Western Literature (New York: Doubleday, 1957), 3–23. Reception of Abraham Traditions in Armenian 345

Below, I will discuss several key episodes in the Abraham story. Some of them are readings of incidents in the biblical text, while others represent apocryphal episodes added to the narrative, either in Armenian circles or in prior Syriac or Greek recountings. In each instance, incidents from the biblical stories were read and integrated into the Christian Heilsgeschichte. Themes were developed in the retelling of these episodes and woven into an expanded biblical narrative; this narrative in turn mediated between past events and the transtemporal history of salvation, as viewed through a Christian lens.6 These interpretations thus transformed the Abraham narrative. Instead of having a single, punctiliar significance, it became a multilayered, perpetual foreshadowing and reflection of the redemptive dynamic of the cosmos. Such interpretations arose from a Christian worldview that regarded all of Scripture as one cohesive and timeless revelation.

2 The Abraham Saga

Characteristically, this approach regarded the biblical story as a unified history of redemption from creation to crucifixion, resurrection, and parousia. Narrative sequence governs the surface relation between the episodes of the story; but in fact, the central redemptive event imposes an atemporal unity that supersedes narrative sequence. This approach gave rise to certain specific Christian interpretations or exegeses of biblical events or texts and to their reformulation as prefiguring, indeed enfolding, the salvific life and death of Christ, in which their true meaning was to be found. Such tendencies appear in the Armenian Abraham narratives. Once we enter into this worldview, the modern contrast of “Old Testament” and “New Testament” apocrypha has no meaning;7 yet, of course, the origin and content of the various narratives can, and indeed should, be considered.

6 The techniques used are usually typological or paradigmatic, but I have also seen the use of identifications and parallels, methods often used in rabbinic exegesis. 7 See also M. E. Stone, “Two Armenian Manuscripts and the Historia Sacra,” in Apocryphes arméniens: Transmission, Traduction, Création, Iconographie: Actes du Colloque International sur la Littérature apocryphe en Langue arménienne, Genève, 18–20 Septembre, 1997 (ed. C. Calzolari Bouvier, J.-D. Kaestli, and B. Outtier; Prahins: Zèbre, 1999), 30–31; idem, “Biblical and Apocryphal Themes in Armenian Culture,” in Proceedings of Strasbourg Conference of January 2010 (ed. R. Gounelle; Prahins: Zèbre, forthcoming). 346 Stone

2.1 Specific Exegetical and Narrative Traditions in the Abraham Saga Of course, the traditions and interpretations that constituted the building blocks of these developed narratives were not exclusively Christian in origin. Scholars have remarked that for the first millennium CE at least, it is misleading to treat the various religious and literary traditions that derive from the Bible—Judaism, Christianity, Islam, and others—as if they grew and lived her- metically sealed off from one another.8 Instead, the relations between them are complex and dynamic, and involve not only diachronic transmission of shared “parabiblical” material, but also mutual borrowing and influence over centuries. In the course of these complex interactions, an ongoing process of interpretation and reinterpretation took place.9 Consequently, it is often difficult exhaustively to trace the exact genealogical derivation of specific elements of tradition, though some striking parallels between Jewish, Christian, and Muslim sources exist. Indeed, often the attempt to achieve genetic certainty is misleading, since the data at our disposal is, by the nature of things, partial, and the fit of the material is only probable and not probative. Not only do difficulties attend the attempt to clarify the genetic origins of specific units of the narrative tradition, but the issues of literary interrelation- ship of the Armenian Abraham texts as such are equally problematic.10 Of course, certain texts stand in obvious literary relationship with one another.

8 An attempt to describe a similarly complex process of transmission in the Arabian pen- insula is made by R. Firestone, Journeys in Holy Lands: The Evolution of the Abraham– Ishmael Legends in Islamic Exegesis (Albany: State University of New York Press, 1990), 3–21; on the general issue, see also papers by J. C. Reeves cited in the next note. 9 See, e.g., J. C. Reeves, “Exploring the Afterlife of Jewish Pseudepigrapha in Medieval Near Eastern Religious Traditions: Some Initial Soundings,” JSJ 30 (1999): 148–77; idem, Trajectories in Near Eastern Apocalyptic (SBLRBS 45; Atlanta: Society of Biblical Literature, 2005). An interesting example is the material on Jewish and Christian “encounters” collected in E. Grypeou and H. Spurling, The Exegetical Encounter between Jews and Christians in Late Antiquity (JCP 18; Leiden: Brill, 2009). At another level, the movement of traditions between different Christian channels, often widely separated in time and place, is significant. Such is illustrated in M. E. Stone, Adam’s Contract with Satan: The Legend of the Cheirograph of Adam (Bloomington, Ind.: Indiana University Press, 2002). 10 It would be difficult to trace in detail the sources, growth, and development of these Armenian biblical retellings without an investigation of most of the Armenian literary tradition, for Abraham material in one form or another is very widespread. It is also quite unclear whether such research would actually uncover literary relationships. I did some analogous research into the very rich Adam tradition, which I hope will be paradigmatic for those interested in Armenian biblical retellings. See M. E. Stone, “Adam and Eve Traditions in Fifth-Century Armenian Literature,” Le Muséon 119/1–2 (2006): 89–120; and idem, “Satan and the Serpent in the Armenian Tradition,” in Beyond Eden: The Biblical Reception of Abraham Traditions in Armenian 347

Yet, most attempts to establish precise relations of literary dependence or derivation between the narrative texts seem fated to fail. Some sort of relationship is evident: some elements are shared, some expressions or turns of speech are common, but in as many other points, the texts differ in this or that way, and one text form is not obviously more pristine than another. Such a pattern of commonality and difference between text forms typifies what I have called “textual clusters.” This term designates a group of thematically related texts comprising “multiple versions of the same textual material”; I have invoked it to explain the complex relationships between the Adam books and equally between the Esdras apocalypses.11 I suggest that this concept is also appropriate for describing the relationship between these Armenian Abraham texts. Having said this, I readily admit that the description or naming of the phenomenon is not an explanation of it. The chief advantage of the description is that it helps us to distinguish this phenomenon of textual affiliation from other types of relationships between texts. Textual clusters clearly represent a type of textual transmission; but the phenomenon does not yield to conventional stemmatic analysis,12 and thus we must consider alternative paradigms of textual development.13 The aetiology of textual clusters may lie in the way the documents were created and used. To resolve this issue is a challenge that lies ahead, as the study of medieval texts that stand in similar relations to one another advances.14

2.2 Selected Elements within the Abraham Saga The remainder of this paper will deal with specific episodes from the expanded Abraham saga, into which various traditions and literary sources are interwoven. In the texts I have studied, the range and selection of incidents, their combination and recombination, their inclusion and exclusion, changed and

Story of Paradise (Genesis 2–3) and Its Reception History (ed. K. Schmid and C. Riedweg; FAT 2.34; Tübingen: Mohr Siebeck, 2008), 141–86. 11 See M. E. Stone, Ancient Judaism: New Visions and Views (Grand Rapids: Eerdmans, 2011), chapter 6, 151–71; quotation p. 151. 12 See Stone, Ancient Judaism, 157 n. 16, in relation to Johannes Tromp’s strained textual genealogy of the primary Adam books. 13 It may well be that if the context of usage of the Abraham texts can be clarified, it may contribute to resolving this issue. There may also be something to learn from the types of manuscripts preserving these texts, and also from their (non-Armenian) Vorlagen. For that, a complete (or as complete as possible) inventory of all Armenian Abraham texts is required. 14 More needs to be learned about how such groups of allied literature were created, used, and transmitted. 348 Stone changed again. The phrase “expanded Abraham saga” designates a reservoir of traditions, which, while maintaining a measure of stability, nonetheless combined differently in each document.15 We must bear in mind that this saga as a whole is a conceptual construct, not existing in any complete textual crystalli- zation.16 Even the fullest Abraham texts do not encompass all the episodes, nor do the texts agree with one another. This, of course, leads me to regard these narratives as a textual cluster rather than as descendants of a single archetype. Table 1 enumerates the chief narrative units of the Armenian Abraham saga. The list follows the biblical order of events and introduces the nonbiblical incidents at the junctures at which they occur in the apocryphal texts. A blank in the second column indicates a nonbiblical incident or episode that the Abraham saga introduced de novo into the biblical narrative line.

Table 1 Chief narrative units of the Armenian Abraham saga

1. Abraham’s background 2. Idols and the recognition of God 3. Story of the ravens: recognition of God 4. Both stories combined in some versions: recognition of God 5. Abraham contemplates the luminaries: recognition of God 6. Abraham burns the idolatrous temple 7. His brother Aṙan dies (for the misdeed of breeding Gen 11:28 the mule)17 8. Terah dies in Haran Gen 11:32 9. Abraham goes to Canaan Gen 12:1–5 10. Abraham and Sarah go to Egypt; the incident with Gen 12:10–20 Pharaoh 11. Excursus: List of the ten trials of Abraham 12. Abraham increases in wealth; separation from Lot Gen 13:1–12

15 I have dealt with the complexities of tradition transmission in Ancient Judaism, chapter 6. 16 At least in terms of those published in Armenian Apocrypha Relating to Abraham or otherwise consulted for this project. 17 This misdeed of Aṙan (biblical Haran; the name appears in the Armenian texts as Aṙan) is not mentioned in the Bible. Reception of Abraham Traditions in Armenian 349

13. Incident of the Four Kings and Melchizedek Gen 14 14. Stories about Melchizedek of Salem (not as an ascetic) 15. Hagar and the birth of Ishmael Gen 16 16. Circumcision of Abraham Gen 17 17. The story of Mamrē, 18. Abraham’s hospitality Gen 18:1–5 19. Appearance of the Three Visitors Gen 18:2 20. The meal Gen 18:6–8 21. Annunciation to Abraham Gen 18:9–15 22. Destruction of Sodom and Gomorra; story of Lot Gen 18:20–33, 19:1–29 23. Typology of Abraham and the Visitors’ meal 24. Abimelech of Gerar Gen 20, 21:25–34 25. Isaac Gen 21 26. Aqedah—binding/sacrifice of Isaac Gen 22 27. Story of Melchizedek (M. as an ascetic) 28. Renaming of Abram18 Gen 17:5 29. Sarah’s death and burial Gen 23 30. Story of Rebecca19 31. Isaac marries Rebecca 32. Prophecy 33. Descendants of Abraham 34. Armenization of the genealogy 35. Death of Abraham Gen 25:8–10

It is instructive to compare this list of incidents with the biblical narrative. The Abraham saga adds major incidents to the biblical narrative; these incidents are presented in Table 2. Particularly notable among these are the stories about the ravens and Abraham’s recognition of God; the story of Mamrē; and the Story of Melchizedek the Ascetic.

18 Note, however, that the Armenian texts use the form Abraham throughout the cycle. 19 The incidents denoted here are not related in the Bible; of course Rebecca herself and her marriage to Isaac occur in the Bible. So also in the next table. 350 Stone

Table 2 Incidents added to the biblical story

1. Abraham’s background 2. Idols and the recognition of God 3. Story of the ravens: recognition of God 4. Both stories combined in some versions: recognition of God 5. Abraham contemplates the luminaries: recognition of God 6. Abraham burns the idolatrous temple 11. Excursus: List of the ten trials of Abraham 14. Stories about Melchizedek of Salem (not as an ascetic) 17. The story of Mamrē, Abraham’s servant 23. Typology of Abraham and his slaughter of the calf for the Three Visitors 27. Story of Melchizedek (M. as an ascetic) 30. Story of Rebecca 31. Isaac marries Rebecca 32. Prophecy 33. Descendants of Abraham 34. Armenization of the genealogy

Notes on Table 2: (1) These items give background to Abraham, in some texts extending back to the Flood; they explain the idolatry against which he reacted as due to the degeneration of the postdiluvian generations. Humans forgot God or forgot the books and the law. (2–5) There exist several differing versions of the incident of Abraham’s recognition of God, a major theme in all readings of this material. (17) The story of Mamrē, Abraham’s black shepherd, precedes the story of the Annunciation (of Isaac’s birth) to Abraham.20 This story provides a basis for Abraham’s famed hospitality as well as an aetiology of the Oak of Mamrē. The source of the story is unclear. (27) The story of Melchizedek as an ascetic was taken from an external source. In those Abraham narratives that contain this story, it follows the Aqedah, which itself follows the incidents of Sodom and Lot. Texts that do not feature the story of Melchizedek as ascetic contain the different (and more biblically connected) incident of the Four Kings, along with Melchizedek’s wel- come and blessing of Abraham; in these texts, this episode precedes the Story of Mamrē.

2.2.1 Melchizedek and the Story of Melchizedek The apocryphal “Story of Melchizedek,” most probably drawn from a Greek source,21 occurs in two documents, the Poem on Abraham, Isaac, Melchizedek,

20 See the more detailed discussion below. 21 It was also known later in Slavonic. See J. Dochhorn, “Die Historia De Melchisedech (Hist Melch): Einführung, editorischer Vorbericht, und Editiones praeliminares,” Le Muséon 117 (2004): 7–48; P. Piovanelli, “Much to Say and Hard to Explain: Melchizedek in Early Christian Literature, Theology, and Controversy,” in New Perspectives on 2 Enoch: No Longer Slavonic Only (ed. A. A. Orlov, G. Boccaccini, and J. Zurawski; Studia Judaeoslavica 4; Leiden: Brill, 2013), 411–29. Reception of Abraham Traditions in Armenian 351

and Lot, and the Story of Terah and Father Abraham.22 Where this apocryphal story occurs, the Melchizedek encounter of Genesis 14 is not mentioned; the biblical and apocryphal Melchizedek stories never occur in the same text (see Table 3).23

Table 3 Melchizedek incidents in the texts

Text Melchizedek stories Story of Melchizedek Other Melchizedek based on Genesis 14 (no. 27 in Table 1) (no. 14 in Table 1)

Story of Father Melchizedek meets Abraham Abraham—brief

Biblical Melchizedek as priest Paraphrases consulted by Rebecca

Memorial of Predicts preeminence the Forefathers of Jacob

Poem Melchizedek in forest

The Tree of Sabek Bread & wine, type Some elements of Story and Melchizedek of Christ—brief of Melchizedek (see n. 22)

Story of Terah and Melchizedek in forest Father Abraham

Sermon on the Meets Abraham— People of Sodom close to the biblical account

22 The text called The Tree of Sabek and Melchizedek has a unique form of these incidents: it combines some features of the Genesis 14 account with the main features of the Story of Melchizedek, along with some unparalleled features. 23 We also encounter a third type of Melchizedek material, in which he is consulted for counsel or as an oracular source of knowledge; see Biblical Paraphrases on Abraham and the Memorial of the Forefathers (Table 3). In these instances, however, Melchizedek’s introduction cannot be situated within the narrative sequence. 352 Stone

The traditions about Melchizedek referred to at no. 14 in Table 1 comprise expansions related to Gen 14:18–20 which are not found in the Story of Melchizedek related in the Poem and the Story of Terah and Father Abraham (Table 1 no. 27). This complex, found in the Story of Father Abraham, the Tree of Sabek, and the Sermon on the People of Sodom (see Table 3) is directly connected with Melchizedek’s importance for Christians as a non-Levitical priest, a view that is anchored in the biblical text (“a priest of the Most High God” [Gen 14:18]; and note Ps 110:4), and that already occurs in the Letter to the Hebrews. Hebrews makes Jesus the scion of a high-priestly line founded by Melchizedek (5:6, 10; 6:20; 7:1–17) and argues for the primacy of Melchizedek’s sacerdotal line over the Levitical line descended from Abraham. In Jewish sources of the Second Temple period, Melchizedek takes on a heavenly character, and he is identified in 11QMelchizedek as a savior figure.24 In 2 Enoch he is a divine man, born without a father and taken to heaven before the Flood, to be brought down again in Abraham’s time.25 His connection with the days of Noah is highlighted in the widespread view that he was Shem, son of Noah.26 Thus Hebrews is building on a Jewish tradition connected with Genesis 14 and bolstered by Ps 110:4.27 For the Christian authors of the Armenian Abraham texts, Hebrews’ view is supported by the bread and wine that Melchizedek the priest offers to Abraham (see Gen 14:18); this offering was readily viewed as a sacrificial, eucharistic act, one of several found in the Armenian Abraham saga.28 Poem 3–4 says of the meal Abraham prepared for the Three Visitors:

Unleavened bread, wine, and calf he slaughtered for the meal, A type of unleavened (wafer) (and) chalice of the Mass.

24 See A. Steudel, “Melchizedek,” in EDSS 1:535–37. 25 See 2 En. 71:37–72. 26 See Stone, “Introduction,” in Armenian Apocrypha Relating to Abraham, 9 n. 23. 27 D. M. Hay, Glory at the Right Hand: Psalm 110 in Early Christianity (SBLMS 18; Nashville: Abingdon, 1973). In addition, it might be the case that the Story of Melchizedek itself subsumes some older Melchizedek traditions from the Second Temple period, presented now in a Christian form. It seems problematic, however, to derive the whole Story of Melchizedek complex from exegetical expansions of Ps 110 and Hebrews, not to speak of Genesis. Its elements are too distinctive and differ in too many ways from those biblical traditions. 28 Eucharistic connections are stressed, sometimes indeed created, as in, for example, the story of the meal of the Abraham and the Three Visitors (see discussion below); and in the story of Melchizedek’s offering, which highlights his foreshadowing sacerdotal function (Sermon Concerning Hospitality 55). Reception of Abraham Traditions in Armenian 353

In The Tree of Sabek, it says of Melchizedek’s offering to Abraham that:

13. Melchizedek took from the grapes of the tree and made wine. And having brought it he offered it to Abraham. And he broke unleavened bread beneath (it) when he came from cutting down the kings. 14. Abraham took it, (and) he himself communicated, and his 318 soldiers with him.

Similar reasons lay behind the incorporation of the apocryphal Story of Melchizedek (no. 27) into the Poem and the Story of Terah and Father Abraham.29 The eucharistic reading of Abraham’s story thus occurs thrice: in the meal offered by Abraham to the Three Visitors; in the bread and wine offered by Melchizedek to Abraham; and in the Aqedah, where it is less explicit. The Story of Melchizedek serves to highlight the Christian perception of events but, except for the name of the priest Melchizedek, shows no connections with the biblical text beyond the association of Melchizedek with Abraham. This story evokes a number of well-known incidents. Melchizedek discovers God in an encounter very much like those related about Abraham in Story of Terah and Father Abraham 6, and Sermon Concerning Hospitality 6. He lives as a “hairy ascetic” in a forest on Golgotha, in Jerusalem;30 he is found after eight years by Abraham, with God’s help, and becomes Abraham’s household priest.31 In the composite Abraham saga, the Story of Melchizedek follows the story of Abraham’s ordeal in the Binding of Isaac. In that ordeal, Abraham’s will- ingness to sacrifice his son is seen typologically to foreshadow God’s sacrifice of his only-begotten Son.32 This is, of course, in fruitful tension with the wish of Melchizedek’s father, King Melkʿi to sacrifice his son to idols. Moreover,

29 See Dochhorn, “Die Historia De Melchisedech”; and the incisive remarks of W. Adler concerning this complex, in his introduction to “Palaea Historica (‘The Old Testament History’),” in Old Testament Pseudepigrapha: More Noncanonical Scriptures (ed. R. Bauckham, J. R. Davila, and A Panayotov; Grand Rapids: Eerdmans, 2013), 1:592–93. The Palaea Historica itself is no earlier than the ninth century. The story has been taken into chapter 72 of the long version of 2 Enoch, though there the connection with Abraham is not made. This leads me to speculate that it was an independent piece associated with Melchizedek, utilizing traditions lying outside Genesis 14. 30 On Golgotha’s location on a mountain in or near Jerusalem, see Tree of Sabek 18 and Synaxarium, Constantinople 1730 13. Compare the discussion by D. Satran, “Early Jewish and Christian Interpretation of the Fourth Chapter of the Book of Daniel” (Ph.D. diss., Hebrew University of Jerusalem, 1985), 345–69. 31 See Story of Terah and Father Abraham 54, and Sermon Concerning Hospitality 48. 32 See the discussion below in connection with this theme. See also A. A. Orlov, “Melchizedek Legend of 2 (Slavonic) Enoch,” JSJ 31 (2000): 23–38; reprinted in idem, From Apocalypticism 354 Stone in the Story of Melchizedek, Melchizedek himself intervenes to prevent his father’s sacrifice of his brother; at his prayer, the earth opens up to swallow the idolatrous temple and all the worshipers.33 The incident is reminiscent of the episode of Korah in Numbers 16: just as the swallowing up of Korah and his associates purified the Israelite camp in the desert, so here the swallowing up of the temple guarantees the purity of Jerusalem and Golgotha. The apocryphal Story of Melchizedek, then, forms a typology of redemption in intertextual conversation with the Abraham narrative: Melchizedek, like Abraham, recognizes God; Melchizedek’s father sacrifices his other son to idols and Melchizedek is saved (a reverse Aqedah); instead of redemption, as with the sacrifice of the Son of God, the sacrifice of Melkʿi’s son leads to a swallowing up of idolaters, while Melchizedek, saved from slaughter, offers the eucharistic sacrifice on behalf of the children of Adam. Christian themes of an eremitic character, such as that of the hairy ascetic living alone in the forest, are introduced; and Melchizedek, instead of being King of Salem (he is never that in the Story of Melchizedek!), becomes Abraham’s family priest. The forest to which Melchizedek withdraws is on the Mount of Olives (Poem 13) or in Jerusalem (Tree of Sabek 2).34 Redemption is on that mountain, identified with Golgotha and with Adam’s burial. One chief problem raised by the Story of Melchizedek is its integration, not into the particular Abraham narratives that include it, but into the sequence of events inferred from Genesis. In those two Abraham narratives that include the Story, the Poem and the Story of Terah and Father Abraham, the material relating to the four kings (Genesis 14) is not found. This produces anomalies in the biblical timeline, in which the Story of Melchizedek seems to be independent of Genesis.35 As far as the Armenian Abraham saga is concerned, the Story of Melchizedek is an import, and the primary questions of its origin and purpose must be answered within its original context. Here we can only be concerned with its function within the Armenian Abraham texts.

to Merkabah Mysticism: Studies on the Slavonic Pseudepigrapha (JSJSup 114; Leiden: Brill, 2007), 423–39; who discusses this legend. 33 Story of Terah and Father Abraham 52–53; Sermon Concerning Hospitality 53–54. 34 According to Synaxarium 13, the Aqedah was on Golgotha, in the mountains of the Jebusites, which once more ties the place of sacrifice to Golgotha; thus one has the burial of Adam, sacrifice of Isaac, life of Melchizedek, and the crucifixion all in the same, central place. Since Adam was created at the navel of the world and buried there, Golgotha = Zion = the site of Adam’s creation and burial, and of the New Adam’s crucifixion and resurrection. The geography of these texts is located on the plane of sacred place. 35 This causes one to wonder about its origins. Reception of Abraham Traditions in Armenian 355

2.2.2 The Story of Mamrē Item 17 on the list of episodes is the apocryphal story of Mamrē, Abraham’s black slave. He sets out to pasture the sheep, carrying three loaves of bread. In three incidents en route he generously gives his loaves away to starving men and is blessed, particularly by the last one. Then he reaches his goal and falls asleep, after sticking his oak staff into the ground. When he awakes, the staff has become a great oak tree, the Oak of Mamre:36 he, a black slave, has turned white, together with his sheep. He returns to Abraham, who recognizes the miraculous nature of the event, and praises him.37 Here the theme of Abraham’s famed hospitality is introduced, leading into the story of the Annunciation to Abraham. Abraham observed the miracles that followed upon Mamrē’s hospitality and swore never to eat again without a guest at his table. The sequence of the Story of Mamrē, followed by Abraham’s oath, Satan’s subsequent blocking of the way to Abraham’s tent, and eventually the arrival of the Three Visitors, is found, mutatis mutandis, in the major narrative texts.38 The story is, among other things, an aetiology of the “oaks of Mamrē” in Genesis.

2.2.3 Abraham’s Offering and Abel’s Offering When Abraham kneels down, after offering the ram as Isaac’s substitute, the Sermon Concerning Hospitality (47) relates that, “there was a voice from the heavens which said, ‘Thus, I too did not pity my beloved Son for your sake who, having come, will free all the children of Adam from Hell because of your goodness.’ ” Similarly, according to the Synaxarium, Constantinople 1730 13, Isaac was offered on Golgotha.39 So Abraham’s willing sacrifice of his son foreshadows and atemporally represents God’s sacrifice of his Son for the sake of Adam’s

36 The “oaks of Mamrē” are mentioned in Gen 13:18; 14:13; and 18:1. Sextus Iulius Africanus, Fragment 30, knows the tradition that a tree sprang from the staff of Abraham’s servant (see M. Wallraff, W. Adler et al., Sextus Iulius Chronographiae: The Extant Fragments [GCS n.s. 15; Berlin: de Gruyter, 2007], 66–67). See also Story of Father Abraham 8. 37 In another version Abraham comes to him, and equally recognizes the miraculous nature of the event. The oak may have taken on some characteristics of the World Tree; see Stone, Armenian Apocrypha Relating to Abraham, 99. 38 Cf. Story of Father Abraham, Poem, Story of the Holy Father Abraham, and of Isaac, his Son, and of Mamrē, his Servant, Story of Terah and Father Abraham, Sermon on the People of Sodom, and Sermon Concerning Hospitality. 39 Note likewise the statement of Anania Katʿołikos (tenth century) that Isaac showed forth the Lord’s passion; see MH tenth century, 255 and 297–98. See also Biblical Paraphrases on Abraham 18; Isaac was “a likeness and type of Christ” (see Memorial of the Forefathers 24, and Story of Terah and Father Abraham 46). Isaac as a type of Christ is commonplace in early Christian literature. 356 Stone offspring, and therefore, in a Christian perspective, the central meaning of the world. The Poem on Abraham, Isaac, Melchizedek, and Lot 9, telling the story of the Aqedah, says: “He showed him the tree of Sabek, / which is in that place, / And a ram was hung by its two horns,40 / He went forth upon the flat rock,41 / He took Abel’s ram,42 slaughtered it in the place.” Thus, the ram in the thicket (Gen 22:13), which replaced Isaac, is here identified with the “firstling of the flock” offered by Abel in Gen 4:4. Abel’s offering and Abraham’s ram are the same beast, an equation that highlights the role of this beast as a special sacrificial animal. The ram is Isaac’s substitute, and Isaac being offered is a type of Christ on the cross. Here two different exegetical techniques are used. The first, also used by the midrash, is the making of exegetical connections and identifications within the biblical stories. The other is typology; that is, seeing events in the Bible as foreshadowing or hinting at those in the New Testament. Another text explains the chronological anomaly inherent in this identification by saying, “And this ram [i.e., Abraham’s—MES] was Abel’s ram that the Lord accepted as Abel’s offering; and Cain’s (offering) He did not accept.43 And Heaven accepted the ram alive and it was preserved until Sahak’s birth, whom Abraham has promised to God” (Story of the Holy Father Abraham 32).44 This is reminiscent of the Melchizedek theme in 2 Enoch, where Melchizedek is taken to heaven in order to return from time to time;45 it is likewise similar to the biblical and midrashic treatment of Elijah (cf. 2 Kgs 2:1–11). In the present context this supernatural ram, taken to heaven in Abel’s day and returned in Isaac’s, is the type of Christ as a sacrifice. The story does not end here, however, but continues in the next episode for our consideration, the story of Abraham and the Three Visitors.

2.2.4 The Annunciation to Abraham: The Three Visitors The annunciation of Isaac’s birth to Abraham by the Three Visitors (items 18–22 in Table 1) is also characterized by Christian typology and is thus integrated

40 Gen 22:13. 41 This feature is not found in Genesis 22. 42 Abel’s ram is elsewhere identified not only, as here, with that offered by Abraham, but also with the calf that Abraham slaughtered for the Three Visitors. 43 Gen 4:3–5. 44 This promise is not mentioned in the other Armenian Abraham texts. 45 See 2 Enoch 71–72. Reception of Abraham Traditions in Armenian 357 into the Christian history of redemption.46 Like the Annunciation to the Virgin Mary, this Annunciation to Abraham also foreshadows redemption. Much attention is paid to the calf that Abraham slaughtered for the Three Visitors, according to Gen 18:7. Like Isaac’s ram, the calf, too, is identified with Abel’s sacrifice.47 This calf is noted to be particularly glorious, patouakan, a term which also hints at its supernatural nature. In some versions of the story, Christ, who is one of the visitors, revives the dead calf by making the sign of the cross, because he is moved by its mother’s lowing.48 Since one of the Three Visitors is the Lord himself, the calf slaughtered by Abraham also takes on a sacrificial character.49 At the same time it foreshadows Christ’s death and resurrection, which association is heightened by the use of the sign of the cross for the resurrection of the calf. The lowing mother cow surely prefigures the women weeping at the crucifixion. Another dimension of this sacrificial animal is evident in Story of the Holy Father Abraham 32: “The marrow of this ram is the sweet oil with which they anointed you.” “You” is Abraham, so the point of the anointing is unclear, for Abraham’s anointing is not mentioned in the Bible.50 Yet, the special nature of the animal is quite evident. The line may evoke the transformatory oil of 2 En. 22:8–9 and 56:2, and perhaps also the “fat things full of marrow” mentioned in Isa 25:6.51 Of course, at another level this imagery may refer to the myron, the oil of chrism used in baptism and unction.52

46 Sometimes this is extremely explicit, as for example in Story of the Holy Father Abraham 29. 47 See on this identification, Story of the Holy Father Abraham 32; Story of Terah and Father Abraham 27; Concerning Abraham’s Hospitality 28; Sermon Concerning Hospitality 37. The point appears to be the perfect nature of the sacrificial victim in each instance—Abel’s ram, Abraham’s son, and then the ram as substitute, and God’s Son. 48 See Story of the Holy Father Abraham 29. 49 Moreover, the bones of the calf Abraham slaughtered are not to be broken, a motif which evokes the Paschal lamb and introduces another way of talking about the crucifixion as sacrifice. 50 But see Yalqut Shimʿoni ad loc. §62. 51 Cf. 3 Bar. 15:1–2; and see A. Kulik, 3 Baruch: The Greek–Slavonic Apocalypse of Baruch (CEJL; Berlin: de Gruyter, 2010), 366–68. See also M. E. Stone, “The Angelic Prediction in the Primary Adam Books,” in Literature on Adam and Eve: Collected Essays (ed. G. A. Anderson, M. E. Stone and J. Tromp; SVTP 15; Leiden: Brill, 2000), 111–32 (118, 124–27). 52 Compare the imagery of the oil of joy or gladness in Isa 61:3; Ps 45:8. The marrow is associated with fatness and plenty, see Isa 25:6; Ps 66:15; and Job 21:24. The use in anointing of oil deriving from marrow is not discussed by M. Dudley and G. Rowell, The Oil of Gladness: Anointing in the Christian Traditions (London: SPCK, 1993). 358 Stone

3 The Fusing of Biblical and Non-Biblical Episodes in the Abraham Saga

The question may now be raised as to which elements of the biblical narrative have been omitted from the apocryphal Armenian Abraham saga and which elements of the saga have no point d’appui in the biblical narrative. First, let us consider those episodes and incidents that are present in the biblical narrative but not in the Armenian apocryphal Abraham texts. When omitted episodes or incidents share features that are completely absent from the apocryphal material, it seems reasonable to assume that these shared features were the motive. A good example is furnished by the texts relating to the promise of the gift of the land to Abraham and his descendants. These are all omitted, as is the promise of the perdurance of Abraham’s seed: Gen 12:7; 15; and 18:17–19. These omissions, I suggest, issue from a Christian reading of the Abraham stories, for they are all the specific biblical promises to or about the bodily descendants of Abraham, the “old Israel.” In the writers’ perspective, these divine undertakings were superseded by the revelation through Christ and the understanding of the Christians as the “new Israel.”53 Abraham’s bodily role as ancestor of Israel is replaced by his role as Father of All Believers. In none of the apocryphal Abraham texts is any attempt made to handle the promises to Israel; they are just left out of the retelling, which is, of course, a way of handling them. If our primary observations here are to the point, they confirm the conclusion that the Abraham stories were of interest for their role in the history of salvation and not simply because they are found in Scripture.54 What is more surprising is that the battle of the Four Kings, and Melchizedek’s meeting with Abraham, are given only in brief compass. The chronological problems involved in relating the two Melchizedek sequences, the one connected with Genesis 14 and the other being the apocryphal story, have been dealt with above. The typological potential of the Four Kings story is not fully

53 Interestingly, other early Armenian authors stress these very texts, reinterpreting “Israel” to mean the Church. 54 There are Armenian Abraham texts that are basically scholarly and learned. One such is in Ms. Galata 154, the Genealogy of Abraham, translated in Stone, Armenian Apocrypha Relating to Abraham. Probably the list of the Ten Trials of Abraham also belongs to this category, though it came to be included within narrative texts: note the reuse of the Ten Trials of Abraham in the Genealogy. Numerous other copies of this list exist. On Armenian learned literature related to the Bible, see M. E. Stone, “The Armenian Apocryphal Literature: Translation and Creation,” in Il Caucaso: Cerniera Fra Culture Dal Mediterraneo Alla Persia (Secoli IV–XI) (Spoleto: Presso la Sede del Centro, 1996), 627–28. Reception of Abraham Traditions in Armenian 359 realized, perhaps precisely because of the enhancement of Melchizedek’s role by the inclusion of the ascetic Melchizedek story within the Abraham cycle, even though the two Melchizedek sequences do not occur together in any single text.55 Thus, the general outline of the Armenian Abraham saga shows a pro- foundly Christian selection and editing of material drawn for the main part from the biblical Abraham texts and certain apocryphal sources. A Christian perspective is expressed not only by the (to us) anachronistic use of “Christ” for “God” (e.g., Story of the Holy Father Abraham 6), but in the introduction of the Melchizedek material and in many typological exegeses.56 Certain biblical passages, relating in particular to Abraham’s physical descendants, are omitted completely. All these techniques result in a story line that is quite exciting and which is read naturally as part of a divine revelation that is deemed unitary and seamless.

55 Above, I observed that in Tree of Sabek, which is unlike all the other texts, we do find a certain mixing of elements. 56 These are pointed out in the notes to the individual texts in Stone, Armenian Apocrypha Relating to Abraham.

Index of Ancient Texts

Hebrew Bible: Masoretic Text 10–11 339–340 10:21–31 339 Genesis 8–9, 10, 32, 71, 202, 11–25 343–359 203, 252, 268, 309, 11:1–9 339–341 320–321 11:7 340 1 145–148, 157–159, 165, 11:8–9 339–341 167, 171 11:10–32 339 1–2 230–271 11:28 10, 348 1–11 6, 308, 311–316 11:31 10 1:1–7 73 11:32 348 1:2 145–146, 148, 166, 12:1 10 169–171 12:1–5 348 1:3 73–74, 164, 165–172 12:1–8 339–341 1:4 74, 168 12:7 358 1:14–19 165 12:8 341 1:26 83–84, 156 12:10–20 348 1:26–27 237–238 13:1–12 348 1:27 240 13:18 355 1:28 158 14 349, 351–354, 1:31 184–185, 186, 188, 358–359 193–194, 203, 205 14:1–17 17 2 145 14:13 355 2:3 173–178 14:18 17, 31 2:6–7 155 14:18–20 3, 16–19, 18, 352 2:7 154–155, 240, 247 15 329, 358 2:15 252–253 15:6 129 2:19–20 247 15:8–21 140 2:21 238 16 329, 349 3:19 153, 155 17 349 3:21 186, 188, 193, 17:5 349 203–204, 205 18 355–357 3:23 155 18:1 355 3:24 75, 136–138 18:1–5 349 4:1–2 314 18:1–16 4 4:3–5 356 18:2 137, 349 4:4 356 18:6–8 349 4:15 26 18:7 357 4:23 27 18:9–15 349 4:25 314 18:17–19 358 5:3–5 315 18:20–33 349 5:29 336 18:21 187, 189, 197, 205 6:1–4 317, 319 19:1–29 349 7:7 315 20 349 8:20–22 22 20–21 99 9:2 158, 160 21 349 362 Index of Ancient Texts

21:25–34 349 15:15 20 22 344, 349, 355–356 15:27 2, 327–328 22:13 356 16:1–3 327–328 23 349 17:1–16 335–337 24:2 3 17:7 335–337 24:7 187, 188, 195–196, 205 19:3 187, 189, 206 25:8–10 349 20:7 177 25:29 329–330 20:8 173 25:31 196–197 20:11 174–176 25:32 196–197 22:4–5 334–335, 342 25:33 187, 188, 196–197, 205 23:7 65 25:34 196–197 24:4 309 26 99, 329–330 25:22 74 27:2 187, 188, 192, 205 26:27 187, 189, 206 27:7 193, 194 31:13 182, 186, 187, 188, 190, 27:10 8–9 198, 200, 206 27:27 187, 188, 198, 205 34:29 176, 177 28:1 45 30:24 336 Leviticus 33:18–20 326 4:34 187, 188, 190–192, 206 33:20 325 10 342 34 326 10:2 326 35:21–22 325 10:8 326 36:5 187, 188, 190, 198, 200, 11:28 199 205 14:10 187, 188, 206 36:10 187, 188, 192, 206 15:8 182, 186, 187, 189, 196, 36:14 187, 188, 190, 198, 200, 206 205 15:13 196 38:22 325 19:7 142–143 38:26 325 43:15 187, 189, 190, 197, 198, Numbers 133, 202 200, 205 4:3 187, 189, 197, 200, 206 45:8 181, 182, 187, 189, 197, 5:5–31 324 200, 203–204, 206 6:1–21 324 46:8 187, 189, 197, 200, 206 6:35 177 46:23 203 7:89 74 48:7 187, 188, 190–192, 198, 14:30 30 206 15:21 187, 188, 207 49:4 325 15:32–36 324 49:5 196–197 15:32–40 326 15:37–41 324 Exodus 32, 252, 309 16 30, 126–128, 130–132, 2:10–11 328–329 354 2:23–25 202, 276 18:1 128 7–12 24 19 328 12:37 187, 188, 195–196, 206 19:9 48 12:42 169 20:1 328 13:16 194 21 135 15:1–9 20 22:23–35 24 Index Of Ancient Texts 363

23:20–21 337 16 26 24 299 19 25 24:17 272–275, 279, 284, 298–299 1 Samuel 31 24:19 298–299 2:1 6, 20 24:25 337 2:1–10 20 25 127–131, 338 4:21 336 25:1–3 337, 338 6:8 226 25:10–13 128 8:5 298 25:14 128 9:1–10:8 13, 30–31 31 130 16–17 332–333 31:2 187, 207 16:1–13 331–333 31:8 337–338 16:14–23 83, 331–333 31:12 187, 189, 190, 198, 207 16:16 332 31:16 337–338, 339 16:23 332 36:1 187, 188, 190, 198, 207 17:10 332 17:26 332 Deuteronomy 306, 307 17:43 332 1:26 187, 188, 190–192, 207 17:45 332 1:27 187, 189, 207 17:50–51 333 3:20 187, 188, 190–192, 207 17:51 332 4:2 305–306 19:5 333 4:26 306 5:21 193 2 Samuel 24 8:15 135 4:9 333 22:6 187, 194, 207 6:2–8 23 25:4 324 6:7 29 25:5–10 324 11–12 26–27 25:17–18 336–337 12:10 26–27 26:8 282 24 3 28:36 222, 226 24:16–17 13 29:22 188, 207 32:1–43 20 1 Kings 32:13 6, 20 1:29 333 32:26 188, 207 8:8 221, 228–229 32:48–50 215–216 16:8–20 130 33 325 33:6 325 2 Kings 34:5–6 216 2:1–11 356 34:6 37 9:30 130 20:17 221, 229 Joshua 22:11–20 222 13:22 337 25:13–17 213

Judges Isaiah 4–5 3 2:22 238 6:32 336 3:14 334 9 99 5:5 334 15:4–5 334–335, 342 6 91, 134, 135, 138 364 Index of Ancient Texts

6:2 135 1:5–12 140 6:3 237 1:13 135 6:6 135 5:2 334 6:13 334 10:4 74 14:13–14 243 28 258 14:29 135 16:6 193 Joel 21:11 203 3:18–21 341–342 24:23 291 24:24 79 Amos 25:6 357 9:11–15 341 30:6 135 30:26 165, 168, 175, 291 Jonah 30:33 139 1:8 192 34:7 203 39:6 221, 228–229 Nahum 59:5 194 2:14 334 60 276–277, 279, 285–286 Zephaniah 60–61 286 3:9–13 340–341 60:1 285 3:10 340–341 60:1–3 273 3:12 341 60:2 286 60:6 276 Zechariah 60:19 168, 291 5:9 137 60:19–20 165, 273, 291–292 14:5 73 61:1 192, 286 61:1–3 286 Psalms 6, 19–21, 34 61:3 357 2 323 62:1 273 3 323 62:1–4 295–296 3:1 323 62:4 192 8 161, 163 63:2 193 8:5–6 161 45:8 357 Jeremiah 51 26–27 3:16 217–218 51:5 35 22:29 218–219 57:1 323 25:11 103 66:15 357 27:19–22 213 69:3 141 29:10 103 78:24–25 318 31:7 280 83:11 23 31:15 6 89:20 191 38:7–13 99 89:31 191 39:15–18 99 93:2 91 52:17–22 213 103:20 136 104:1–5 143 Ezekiel 104:2 165–167 1 91, 134 104:2–4 144 1:4 135 104:4 134–149 1:5 135 passim, 247 Index Of Ancient Texts 365

105:29–36 24 2 Chronicles 106:28–31 129 5:9 221 110:4 352 26:19 31–32 111:7 323 34:19–28 222 111:8 323 35:3 220–221, 224, 126:1 102–105 226–227 144 332–333, 342 36:7 213 144:7–11 332–333 36:10 213, 228–229 36:18 213 Proverbs 6:23 170 8 169–170 Hebrew Bible: Other Versions and 8:22–31 89 Translations 16:15 166 Samaritan Pentateuch 179, 186, 195, 198, 200 Job Genesis 21:24 357 1:31 186 38:1 191 3:21 186 38:4–7 73 18:21 187 38:7 85 24:7 187, 195 40:6 191 25:33 187, 195 27:2 187, 192 Daniel 294–295, 296 27:7 193 1:2 213 27:27 187, 198 1:17 164 36:5 187 5:2 213 36:10 187, 192 5:11 164 36:14 187 7:9–10 139 43:15 187, 197, 200 7:25–26 295 45:8 187, 197 46:8 187, 197 Ezra–Nehemiah 59, 64 48:7 187, 190, 198

Ezra Exodus 1:7–10 213 12:37 187 9:1–12 59 19:3 187 9:2 46 26:27 187 31:13 187 Nehemiah 2:13 191 Leviticus 9:38 58 14:10 187 10:29–31 57–59 15:8 187, 196

1 Chronicles Numbers 1:35 187 4:3 187, 197, 200 4:9 336 15:21 187 13:10 29 31:2 187 16:36 88 31:12 187 21:16 282 36:1 187 21:26–27 27 366 Index of Ancient Texts

Deuteronomy Jeremiah 1:27 187 38:7–13 99 22:6 187 38:8 280 29:22 188 39:15–18 99 32:26 188 39:16 99

Septuagint 186, 193, 195, 197–198, Psalms 200 7:1 24 Genesis 8:6–7 160–161 1:31 186 26:1 171 2:6 155 50 26–27 18:21 187, 197 50:5 35 25:33 187, 196 82:11 23 27:27 198 103:4 135 36:5 187 104:29–36 24 36:14 187 143 333 45:8 187, 197, 200 151 330–333, 342 48:7 187, 198 Peshitta Exodus Psalms 12:37 187 93:2 91 19:3 187 104:4 247 26:27 187 Targumim Leviticus Targum Neofiti 14:10 187 Exodus 15:8 187, 196 12:42 169, 171 22:4 334 Numbers 4:3 187, 197, 200 Palestinian Targum Genizah mss. 24:17 274 Exodus 25:14 128 12:42 169, 171 31:2 187 20:8 173 31:12 187 36:1 187 Fragment Targum Exodus Deuteronomy 12:42 169, 171 1:27 187 22:4 334 29:22 188 32:26 188 Targum Pseudo-Jonathan Genesis 1 Sam (1 Kgdms) 1:5 168 1:1 23 1:26 71 2:7 153 1 Kings (3 Kgdms) 4:5 193 Vulgate Psalms Isaiah 103:4 135 60:13 15 Index Of Ancient Texts 367

New Testament Titus 1:12 106 Matthew 2:1–12 275–276 Hebrews 352 2:2 302 1:3–6 237 2:11 276 1:6 234 2:18 6 1:7 135 9:9 65 5:6 352 22:29–30 262 5:10 352 23 130 6:20 352 7:1–17 352 Mark 7:3 16 2:15 65 7:9 17 12:24–25 262 11:24–25 328–329

Luke 2 Peter 1:78–79 277 1:19 277 5:27 65 10:1 2 Jude 10:18 243 1:9 4, 36 16:23 344 20:34–36 262 Revelation 2:26–28 277 John 19 281 1:1–5 169, 171–172 19:13–15 282 1:9 172 19:21 282 21:23 276–277 Acts 22:16 277 17:28 106

Romans Apocrypha and Pseudepigrapha 4:16 344 5:12 235 Ahiqar 164 5:14 259 Apocalypse of Abraham 1 Corinthians 17:1 140 6:19 155 17:18–19 165–167 15:45–49 240 18:2–3 140 19:4–6 140–141 2 Corinthians 4:6 168 Apocalypse of Moses 8

Colossians Apocryphon of Lot 16, 37 1:15–18 240 Ascension of Isaiah 306 2 Timothy 3:8 4 Ascension of Moses 4

Assumption of Moses 4, 36–37, 320 368 Index of Ancient Texts

Baruch 101 10:13 139 1:8 213 12–16 28 14:11–22 138–139 2 Baruch 97, 101 17:1 139 6:7–9 212, 213, 214, 218–219 17:5 139 20:82 213 18:11–13 139 21:6 90–91, 144–145 18:15 139 48:6–8 144–145 21:7–10 139 48:39 139 37–71 (Similitudes) 176–177 48:43 139 38:4 176 80:2 213 39:7 139 54:1–2 139 3 Baruch 101 54:5 139 1:1 100 67:4–7 139 15:1–2 357 69:13–25 80–81 69:15 84 4 Baruch (Paraleipomena Jeremiou) 69:15–16 81 97–102, 110, 213, 214, 71:1–2 139 215 71:5–7 139 3:9–20 213–214, 215 90:24–27 139 3:10 104 98:3 139 3:11 97 100:7 139 3:15 97 100:9 139 5 97–100, 101, 104–106 103:8 139 5:9 104 108:3–6 139 5:21 100–101 9 100 2 Enoch 35, 82, 161–162, 231 7:3–4 236 Bel and the Dragon 18:3 236 1:33–36 14 21–22 235–237, 238, 241, 254 Ben Sira 82, 307 22:6–7 235–237 24:2 89 22:8–9 357 24:3–5 89 25–26 81, 82 43:1 166 25:3–4 74–75 43:4–5 166 29:3 [A] 142 44:16 28 29:3 [J] 82, 140, 141, 142, 143, 50:24 129 147 29:4–5 234 Coptic Jeremiah Apocryphon 30:9–15 34 12, 22, 38–39, 101, 110, 30:13 152 214 56:2 357 71–72 356 1 Enoch 81, 82, 139, 142, 177, 71:37–72 352 308, 318 72 353 1–36 (Book of the Watchers) 138, 317, 318 3 Enoch (Sefer Hekhalot) 6–16 317 4 241–242, 317 7:5 318 26 135–136 10:6 139 Index Of Ancient Texts 369

Epistle of Jeremiah 101 Liber Antiquitatum Biblicarum (Pseudo-Philo) 4 Ezra 18:13–14 338 6:42 81–82, 83, 84 26:13 214 6:54 159 60:1–3 82–83 7:36 139 8:22 137–138 Life of Adam and Eve 8, 152, 156, 231–271 10:22 222 passim 11–17 232–243, 247–249, Jannes and Jambres 4, 37 267, 270 14:3 161–162 Joseph and Aseneth 150 15:3 243 36:1–44:3 14 Jubilees 4, 7–12, 35, 43, 82, 92, 304–321 passim, Lives of the Prophets 329–330, 343, 344 Jeremiah 2:11 213 2:1–3 145–147 2:1–23 9 1 Maccabees 125 2:2 9, 166 2:1 124 2:2–3 71–72, 82 2:26 128 2:3 85 2:51–54 128 2:23 34, 314 2:70 125 4:7–11 314 9:19 125 4:15 312, 317 13:25 125 4:22 28, 312 16:4 125 4:31 8, 35 5:6 317 2 Maccabees 7:21–25 318 1:19–22 210 8:10–18 9 2:1 217 10–11 312 2:1–8 212–214, 215 10:1–5 318 2:4 217 10:3–10 318 11:4 318 Oracles of Hystaspes 283 11:4–6 312 12 10–11 Paraphrase of Shem 12:1–5 10 8 167 12:1–8 10 12:16–17 10 Prayer of Joseph 12:28–31 11 1–4 89–90 14:20–21 329 22:16–22 43–45, 46, 57–59, 60 Psalm 151 330–333, 342 22:17 318 24:2–3 329–330 Sibylline Oracles 34, 35, 280, 281, 283 26:34 8 3:24–27 34–35, 152 37–38:3 8 3:672–673 282 50:2 309 3:798–799 282 4:173–174 282 Judith 5:155–161 294–296 11–13 3 370 Index of Ancient Texts

Testaments of the Twelve Patriarchs 1QM (War Scroll) 101, 274–275 10:12 193 11:5–6 274 Testament of Judah 11:10 139 24:1 274–275 1QHa (Hodayot) 7:29 166 Testament of Levi 12:6 168 18:3–4 275 12:6–7 176 12:23 168 Testament of Abraham 37, 150 16:11–12 136–137 6 4 21:14 168

Testament of Job 150 2Q2 (2QExoda) 187 2Q6 (2QNuma) 187, 197 Testament of Moses 36–37 4Q1 (4QGen–Exoda) 187 Tobit 6 4Q3 (4QGenb) 199 4Q6 (4QGenf) 187, 190, 198 Wisdom of Solomon 4Q22 (4QpaleoExodm) 187 2:23–24 235 4Q23 (4QLev–Numa) 187, 197 18:14–16 307 4Q32 (4QDeute) 199 4Q35 (4QDeutg) 199 4Q88 (4QPsf) Judean Desert Documents 8:10 (Apostrophe to Zion) 193–194 Qumran 4Q98g (4QPsx) 1QIsaa 183–191 passim, 197, 1 191 200 2 191 13:23 193 8 191 47:16 194 4Q134 (4QPhyl G) 50:28 193 1:26 193 1QpHab (Pesher Habakkuk) 4Q169 (Pesher Nahum) 126 10:5 139 3–4 ii 9–10 130 10:13 139 4Q175 (Testimonia) 1QS (Rule of the Community) 12–13 274 40–47 passim 4Q180 (Ages of Creation A) 1:3–10 43 2–4 ii 6 187, 197 2:3 177 4Q185 (Sapiential Work) 2:8 139 1–2 8–9 141–142 3:2–5 41 4Q213–214 (Aramaic Levi) 3:4–5 51 13:4 (4Q213 1 + 2 ii 8–9) 164 4:13 139 4Q216 (Jubilees) 5:7–20 43, 45–46 5:6 146 5:9–10 44 5:9 146 5:13 44 4Q255 (4QpapSa) 5:13–14 41, 45 2 5 191 5:15 65 4Q256 (4QSb) 6:24–7:25 41 4 43 Index Of Ancient Texts 371

4Q258 (4QSd) 11Q5 (11QPsa; Psalms Scroll) 1 43 22:1 (Apostrophe to Zion) 194 1 i 5–7 57–59 26:11–12 (Hymn to the Creator) 1 i 5–10 43–46 73, 86 1 i 7 44 27:2–4 164 1 i 10 65 28:3–14 (Ps 151) 330–333 4Q266 (Da) 28:13–14 (Ps 151B) 332 3 iii 19–22 274 11Q13 (Melchizedek) 352 4Q269 (Dd) 5 274 CD (Damascus Document) 311 4Q274 (4QTohorot A) 2:5 139 2 i 2 191 7:18–20 274 2 i 8 191 3 ii 9 194 Wadi Murabbaʿat 12 4Q365a (4QTa?) Mur1 (Gen, Exod, Num) 192, 199 2 ii 6 191 Mur3 (Isa) 192 4Q372 (Narrative and Poetic Compositionb) Mur88 (Minor Prophets Scroll) 3 10–12 129 199, 201 4Q384–385 101 10:14 192 4Q392 168 1 5–7 167–168 Wadi Sdeir 4Q394–399 (MMTa–f ) Sdeir1 (SdeirGen) 199 C 7–9 132 C 31 129 Naḥal Ḥever 4Q403 (ShirShabbd) 5/6ḤevPs 192 1 i 45 168 1 ii 7–9 141 Masada 4Q404 (ShirShabbe) Mas1 (MasGen) 5 4–5 168 Gen 46:8 187, 197, 200 4Q405 (ShirShabbf) Mas1b (MasLevb) 199 20 ii 9–10 141 Mas1e (MasPsa) 192 4Q427 (4QHodayota) 7 ii 10 191 XLevc (Schøyen 4611) 199 7 ii 17 191 4Q448 (Apocryphal Psalm and Prayer) 2:2 191 Hellenistic Jewish Writers 2:3 191 2:8 191 Aristobulus 150, 152 2:9 191 Eusebius, Praep. ev. 3:1 191 13.12.10–11 4Q511 (Songs of the Sageb) (Fragment 5) 163–165, 172–173, 18 8 168–169 176, 178 4Q530 (EnGiantsb ar) 2:16–18 139 Eupolemos 4Q541 (Apocryphon of Levib [?] ar) Eusebius, Praep. ev. 9 275 9.39.5 213–214, 215

6Q13 (Priestly Prophecy) 4–5 129 372 Index of Ancient Texts

Josephus 6 1.36 125 Jewish Antiquities 8–9, 11, 12, 112–115, 1.67–68 116–120 313 1.106 116, 118–120, 127 1.5 114 1.203–211 114 1.68–71 28 1.659–660 114 2.233–236 11 1.666 114 2.243–248 11 2.150 46–47, 52, 53 3.9–12 (3.1.3) 327–328 2.184–203 115 4.15–19 130 2.409–417 113 4.78–81 (4.4.6–7) 328 5.219 210 4.126–130 (4.6.6) 338–339 6.199–218 113 4.146 130 6.278–280 113 11.302–347 114 6.288–291 277–283, 299 12.237–240 115 6.298 281 12.265–271 125 6.423 114 12.282–283 114 7.150 180 12.285 125 7.162 180 12.387–388 115 7.409–417 115 12.402–412 114 7.423–432 115 12.432 125 13 115, 125 Philo 150–178 passim 13.62–74 115 Allegorical Interpretation 13.210–212 125 1.16–18 173–174 13.230–300 126 1.31 153 13.282–283 123 On the Confusion of Tongues 13.288–299 114, 116–126 146–147 240 13.372 125 On the Creation of the World 13.398–404 114, 116, 119–120, 31 170–171 127–131 73–75 84 14–15 115 77 157 14:22–24 104 77–78 242 14.22–28 114 77–79 158–160 14.145–149 130 82–88 158–160 14.158–184 114 83 162 14.488–491 114 84 161–162 15.2–7 114 131–133 155 15.262–266 114 134–150 156 15.366–370 114 136 154–155 15.380–381 114 137 154–156 15.388–391 114 On Dreams 15.425 114 1.75 171 17.41–42 130 On the Posterity of Cain 17.165–166 114 64 173 17.173–181 114 On the Special Laws 18.256–309 115 2.59 173 Jewish War 113, 114 That God is Unchangeable 1.18 114 11 173 1.31–33 115 Index Of Ancient Texts 373

Whether Animals Have Reason m. Yoma 156–157 5:1 210–211 16 156 5:2 210–211

Tosefta Other Jewish Literature t. ʿAbodah Zarah 3:9 56–60 Mishnah t. ʾAhilot m. Baba Qamma 9:2 53 6:4–5 334 t. Baba Batra 10:1 65 7:3 50 m. Demai t. Demai 2:3 42, 55–64 2:2 62–64 m. ʿEduyot 2:2–14 55, 67–68 1:14 40 2:10–13 61, 62 m. Ḥagigah 2:12 42, 60–62 2:5–7 48–54 2:17 63 2:7 56, 67 3:1 63 3:5 123 3:4 65 3:6 64–68 3:7 63 m. Ketubbot t. Ḥagigah 2:9 124 3:1–2 49 4:8 124 3:3 48 m. Makhshirin t. Ḥullin 2:3 53 2:20–21 60 m. Niddah t. Ketubbot 4:3 53 5:1 50 m. Parah t. Kippurim 10 48 1:4 115 m. Pesaḥim 3(4):20 53 9:2 123 t. Mikvaot m. Qiddushin 6:7 53 4:5 50 t. Parah m. Qinnim 3:8 53 3:6 333 t. Qiddushin m. Sheqalim 5:15 158 6:1 210, 219, 223 t. Sanhedrin m. Soṭah 8:9 159 3:4 127 t. Shabbat m. Ṭeharot 1:13 42 4:5 53 t. Sheqalim 5:7 53 2:18 219, 222–223, 5:8 53 227–229 7:1–4 66 t. Soṭah 7:5–6 53, 66 13:1 212, 213, 219, 221–224, 7:6 66–68 227–229 8:1 40, 53 13:5 115, 123 374 Index of Ancient Texts t. Ṭeharot b. Berakhot 5:2 53 7b 323 5:4 53 10a 323 6:11 53 29a 124 t. Yoma 63a 324–325 2:15 219, 227–229 b. Giṭṭin t. Zabim 14a 50 2:1 52 55b–56a 113, 115 67a 164 Palestinian Talmud b. Ḥagigah y. Berakhot 12a 152 8:5 (12b) 175 19a 49 9:1 (12d) 80 26a 65, 66 y. Beṣah b. Horayot 1:11 (61a) 167 12a 114, 219, 222, 226–228 y. Ḥagigah b. Ḥullin 2:6 (78b) 49 35b 52 y. Horayot b. Keritot 3:3 (47c) 209 5b 219, 226–228 y. Makkot 28b 124 2:7 (32a) 209 b. Megillah y. Nazir 9b 114 7:2 (56b) 153 b. Menaḥot y. Shabbat 64b 114, 115, 116 1:3 (3c) 64 109b 115 2:6 (5b) 155 b. Nedarim y. Sheqalim 39b 75 6:1 (49c) 219, 222–223, b. Pesaḥim 227–229 49a–b 40 y. Soṭah 54a 75 8:3 (22c) 219, 222–223, 57a 124 227–229 64b 114 y. Taʿanit b. Qiddushin 2:1 (65a) 209 66a 114, 115, 121–126 3:9 (66d) 102–106 b. Sanhedrin 4:8 (68d) 284 19a 124 19a–b 115, 116 Babylonian Talmud 38a–b 153 b. ʿAbodah Zarah 82a–b 130 60b 323 b. Soṭah b. Baba Batra 22b 114, 115, 127–131 3b–4a 114, 115, 116 33a 114, 115 4a 124 49b 114, 115, 116 22a 164 b. Taʿanit b. Baba Meṣiʿa 18b 114 87a 64 22b–23a 114 b. Baba Qamma 23a 103–106 82b 114, 115, 116 29a 113 Index Of Ancient Texts 375 b. Yebamot Numbers Rabbah 4a 324 15:10 209–210 61a 124 b. Yoma Lamentations Rabbah 19b 114 2:2 284 21a 209 2:4 284 38b 113 4:2 113 52b 219, 222, 226–228 53b 219, 222, 223, Qohelet Rabbah 226–228 6:9 238 69a 114, 115 Song of Songs Rabbah Midrash Rabbah 8.9.3 209 Genesis Rabbah 1:3 78–84, 85, 87–88, 92 Other Texts 1:4 75, 90, 91 ʾAbot de-Rabbi Nathan A 3:4 167 18 164 5:5 160 31 87–89, 92 8:1 152 41 219 8:10 237–239 9:5 203 ʾAbot de-Rabbi Nathan B 11:2 174–177 7 113 11:9 71 14:1 155 ʾAggadat Bereshit 14:8 153 Gen 6:2 317 19:4 160 20:12 203 ʾAggadat Hamashiaḥ 297 21:3 152 21:9 137 Anonymous Piyyuṭim 23:1 123 Shivʾata on Gen 1:1–7 24:2 152 Vv. 3–4 73–76 When) אז באין כל 137 78:1 79:8 325–326 All Was Not) 76–77, 85–86, 88 94:9 203 Ll. 191–196 158, 160 162, 242 ʾEzel Moshe)) אזל משה 197 98:5 אז על אדירים Exodus Rabbah (Then, Upon the 1:26 12 Mighty Ones) 78 78 God of Old)) אלוהי קדם 148–147 15:22 אנהם ברב כח 162 21:6 22:3 336 (I Shall Loudly Cry) 77 את מי נועצת 137 25:2 (With Whom did Leviticus Rabbah You Take Counsel?) 78, 85 12:1 326 14:1 152 Baraita d’Melekhet 18:2 152 ha-Mishkan 217 20 328 7 219, 220–224 20:10 166 376 Index of Ancient Texts

Bereshit Rabbati 234, 268, 312 Ps 104 138 Gen 6:2 317 Ps 144 333 209 (on Gen 45:8) 179–207 passim 209:10–15 202–203 Midrash Petirat 209:13 203 Moshe Rabbeinu 317

Book of Elijah 281 Pesiqta Rabbati 20–24 (Midrash on the Divrei ha-Yamim shel Ten Commandments) Moshe Rabbeinu 11 174–176 34:2 317 Elazar berabbi Qillir 36 285–286 Qedushta 77 Qedushta for Shavuot 170 Pirqe de-Rabbi Eliezer 312 141 ,138 4 בימים ההם ובעת ההיא (In Those Days and at That Time) 11 153, 238–239 281 11–12 155 20 155 Hekhalot literature 75, 78, 80, 84–85 Prayers of R. Shimʿon b. Yoḥai Megillat Taʿanit 296–297, 298–300 Shevat 28 123 Pseudo-Rashi Megillat Taʿanit, Scholion 2 Chr 35:3 220 Kislev 21 115 Shevat 2 114 Radak (R. David Kimchi) Shevat 28 123 2 Chr 35:3 220

Mekhilta of Rabbi Ishmael Ramban Baḥodesh Commentary on Song 7 174–176 of Songs 209 Vayassa 2 327–328 Rashi Exod 25:18 336 Mekhilta of Rabbi Shimon bar Yoḥai b. Ḥagigah 26a 65 Vayassa b. Ḥullin 35b 52 38.1 327–328 b. Yoma 21a 209

Midrash Alpha-Bethot 166 Secrets of Rabbi Shimʿon b. Yoḥai 296–297, 299–300, Midrash Haggadol 302 Num 24:19 298–299, 302 Seder Olam Rabbah Midrash Ḥaserot 24:16–68 219–220, 224 ve-Yeterot 209 30 123

Midrash on Psalms Sefer Eldad ha-Dani 267–268 90:12 90–91 Ps 8:7 160 Index Of Ancient Texts 377

Sefer Yosippon 113, 278 Yalqut haMakhiri 35 7 210 Yalqut Shimʿoni Sefer Zerubbabel 295–296 62 (Genesis 18) 357 Isaiah 60 285 Shāhīn-i Shirāzi Bereshit–nāmah 268–269 Yose ben Yose 161 (ʾAz Be-Daʿat Ḥaqar) אז בדעת חקר Sifra 7 168 Beḥuqotai 24 161 (ʾAzkir Gevurot) אזכיר גברות 115 1:1 Zabim 15 166 1:1 52 16 168 38 159 (ʾAtta Konanta) אתה כוננתה Sifre Numbers 41 176–177 3 169 115 282 21–24 159–162 131 129, 323 157 (Matot) 339 Samaritan Literature Sifre Deuteronomy 348 325 Molad Mosheh 276

Sifre Zuta Tibat Marqe 178 Num 15:36 323–324 2.44 153–156 4.34 173 Tanḥuma 162, 170 Behaʿalotekha 6:11 209 Christian Literature Fragments on Gen 1 170, 238 Ḥayye Sarah Abbot Daniel 3 147 Pilgrimage 14, 18 Ki Teṣe 16 337 Abraham of Nathpar Lekh Lekha On the Kinds of Prayer 264–265 9 323 Qedoshim Abraham Zabāyā 7:2 88–90, 92 Life of Rabban Bar ʿIdtā 263 Vayeshev 4 72 Acts of Andrew 306

Tanna DeBe Eliyahu Acts of John 306 1 75, 91 31 75, 91 Acts of Thomas 306

Tuviah b. Eliezer Ambrose Midrash Lekaḥ Tov Hexameron Balak 58a 297 1.5.19 86 378 Index of Ancient Texts

Anania Katʿołikos Sermon Concerning MH 10th c.: Hospitality 355 255, 297–298 355 6 353 37 357 Anastasius of Sinai 47 355 Questions and Answers 48 353 80 258 53–54 354 55 352 Andrew of Crete Sermon on the People of Sodom Great Canon 6–39 passim, 20–21 351–352, 355 PG 97:1365D 25 Story of Father Abraham PG 97:1368A 22–23 351–352, 355 PG 97:1368CD 31 8 355 PG 97:1375B 23 Story of Terah and Father Abraham 351–354, 355 Aphrahat 6 353 Demonstrations 27 357 6.1 263 46 355 6.6 263 52–53 354 6.19 263 54 353 8.9 37 Story of the Holy Father 18.12 263 Abraham, and of Isaac, his Son, and of Mamrē, Apocalypse of John his Servant 355 37 21 6 359 29 357 Apocalypse of Sedrach 32 356–357 5.1–2 247–248 Synaxarium, Constantinople 1730 13 353, 354, 355–356 Apostolic Constitutions Ten Trials of Abraham 358 8.12.6–27 86–87 The Tree of Sabek and Melchizedek 351–354, 359 Ariston of Pella 284, 286 2 354 13–14 353 Armenian Abraham Saga 18 353 343–359 Biblical Paraphrases Athanasius on Abraham 351 Festal Letter 39 305–308, 310–311, 18 355 320–321 Genealogy of Abraham 358 21 305 Memorial of the Forefathers 351 Augustine 24 355 The City of God Poem on Abraham, Isaac, 11.9 72–73 Melchizedek,and Lot 350–354, 355 3–4 352 Barhebraeus 9 356 Ethicon 13 354 1.6.1 263 1.6.5 263 Index Of Ancient Texts 379

Lamp of the Sanctuary Ephrem the Syrian 7.1.1.1 265–266 Commentary on Genesis 2.32.1 250–251 Basil of Caesarea Commentary on the Diatesseron Divine Liturgy 20 16.22 263 679 (PG 31:1637D) 21–22 Hymns on Faith Hexameron 50.6.2 251 1.5 86 Hymns on Paradise 6.24 247 Basilides Hymns on the Church Epiphanius, 84–85 48.11.1 251 Pan. 24.1.8–9 Hymns on the Fast 1.1.1–4 251 Biblia Pauperum 5 Hymns on the Nativity 21.15.4 251, 263 Cave of Treasures 19, 231, 244–257 Nisibene Hymns passim, 265, 270, 299 68.3–4 251 2 245–247 68.7 251 2.4 160 2.8–11 245–246 Epiphanius 2.12–14 245 Panarion 304–321 2.13 160 1.3 312 2.15–19 245 3.4 312 3.1–4 246 8.6.1–4 306, 307 9.4.12 312 Chronicle of Zuqnin 300–302 20.2.2 304 24.1.8–9 (Basilides) 84–85 Chronicle up to the Year 1234 39.1.1–2 315 Chapter 1 253–254 39.2.1–3.5 314 39.4.2 314 Clement of Alexandria 164, 305, 311 39.5.1 304–305, 306–307 Stromata 39.5.4 314 2.36.2–4 (Valentinus) 160 39.5.5–8 314 39.5.9 314 Cyril of Jerusalem 39.6.1 307–308, 314–315 Letter to Constantius 279 39.6.2–4 315 39.6.6 315 Didymus 307, 311, 313–314 39.7.1–3 315 39.7.5 315 Dionysius bar Ṣalībī 39.8.1–5 315–316 Response to the Arabs 39.9.1 315 Chapter 26 261–262, 264 39.10.6 316 40.2.1 306 Diyarbakir Commentary 45.4.1 306 Gen 2:15 252–253 56.2.2 306 63.2.1 306 Encomium on Saint Michael the Archangel 67.1.5 306 248, 257 67.3.4 306 380 Index of Ancient Texts

Epistle of Peter to James Grigor Narekac‘i 1.2 319 Book of Lamentations 93.5 343 Eusebius 164 Canons 313 Hippolytus Chronicle 284 On the Antichrist Ecclesiastical History 17 243 4.6.2 284–286 53 243 4.8.4 284 Commentary on Daniel Preparation for the Gospel 4.12 243 9.39.5 (Eupolemos) 214 13.12.10–11 History of Moses 37 (Aristobulus) 163 Irenaeus George Cedrenus Against Heresies Compendium historiarum 1.30.6 152 1.6.1–4 8 1.9.10–13 9 Jacob of Edessa Letter to John the Stylite George Monachos 1 265–266 Chronicon 9, 33 Jacob of Sarug George Syncellus Homilies on the Six Days of Creation Chronography The First Day 85 3.4–18 9 The Sixth Day 251–252, 266 4.20–22 8 Poem on the Sleepers 9.6 27 of Ephesus 93, 106–107, 108 11.4–6 35 19.24–20.4 319 James de Voragine 27.11–12 7 Legenda Aurea 33.1–18 314 24 93–94, 107 105.6–108.11 10 107.22 10 Jerome 307, 311 123.22–25 9 Commentary on Matthew 16.13 13 Georgios Chumnos Letters Metrical Paraphrase of 78.20 313 Genesis and Exodus 32–33 Gallican Psalter Ps 103:4 135 Gregory II (Pseudo-Gregory) Liber de situ et nominibus 32 locorum Hebraicorum 195 13 Gregory of Nazianzus 6–7 In sanctum baptisma Joannes Phocas (Orat. 40) 7 Description of the Holy Land 11 18 Gregory of Tours Glory of the Martyrs 94 [95] 93, 106–108 Index Of Ancient Texts 381

John Chrysostom 6–7 Narsai 72, 253 Adversus oppugnatores On the Making of Adam and Eve, vitae monasticae and on the Transgression of the 3.17 16 Commandment Concerning the Consumption 4.101–125 249–251 of Wine 7 On the Making of Creatures In sanctum Romanum 1.221–240 249–251 Homily 1 16 Nicephorus Callistus John of Dalyatha Historia Ecclesiastica Letters 30 18 4.3–5 263 Opus imperfectum in Matthaeum John of Damascus 260 Homily 2 300–301

Joseph of Telkepe Oracle of Baalbek 280 On Revealed Truth Prologue 17–19 255 Origen 305, 311 Commentary on Matthew Julius Africanus Matt 27:3 4 Chronography 7, 313, 314, 319, 320 Contra Celsum Fragment 30 355 4.23 156, 157, 160–161 4.24 161 Justin Martyr 4.29 161 Dialogue with Trypho 4.30 161 19.4 19 4.74–99 156 33.2 19 4.76 158 First Apology 4.99 156–157 31.6 284 6.43–44 243 On First Principles Lactantius 1.5.4–5 243 Divine Institutes 3.2.1 4 2.15 317 Homilies on Numbers 7.15.6 283 12.3 243 7.19.2–5 280–285 Selecta on Numbers 7.24.6 281 PG 12:577b 37

Melito of Sardis Palaea Historica 1–39 passim, 353 Fragment 12 2 1–2 2 3.8 2 Michael Glycas 3.13 10 Annales 5:1–2 2 43.8–18 35 5.3 21 206.20–22 9 5.5 2 254.38–255.33 15 7.3–4 6 263.20–264.6 8–9 7.5 2 382 Index of Ancient Texts

8.2 20 58.19 2 9.3 20 59.1 3 11.1 26 62.9 26 14.4 20 67.11 2, 3, 21 14.5 27 69 11–12 16.1 11 70 11 16.12 20 75.10 4 16.41 21 76–84 24 20.4–6 28 91 2 20.5 10 92.4–93.5 29–30 21.10 21 97.8 21 22.2 38 98.8 30 22.9 38 99.3–8 30 22.11 21 99.10 21 23.2–9 20 100 30 23.3–6 21–22 100.10 21 23.14 11 101.1 5 25.2 13 103.6 21 26.2 10 103.8–11 2 26.7–9 10 103.11 21 26.8 10 104.4–12 30 27.66 2 105.13 30 28–29 17 105.15 21 30.2 2 106.3 21 31.6 18 106.5 21 31.7–8 17 107.3–11 23 32–38 3, 16–19 107.6 29 32 18 114.5–16 24 32.5 38 114.16 21 36.4 17 115.1–4 2 36.7 16 115.5 21 36.9 31 116.10–11 21 38.7 17 117.5 21 39.1 2 119.4 2 45.6 21 119.5 21 46 17 119.7 21 46.5 21 119.8 6, 20 46.7 2 120.1–4 20 48–49 4 121.1–8 36–37 52.8 21 121.6–8 4 53.9 6–7 122.2 2 53.10 6–7 123.6–7 13 53.11 26 125.8 21 54–55 14–16 127.10–12 20 55.3 13 128–129 3, 23 55.4 16 130.4 21 55.5 15 130.22 20 55.7 15 133.9 20 Index Of Ancient Texts 383

136.4–7 26 PG 28:740A 27 136.8 20, 26 PG 28.740BC 2 138.6 6, 20 138.9 270 Pseudo-Basil of Caesarea 139.1–5 22–23 Sermon 11 7 139.5 20 140 29–30 Pseudo-Clementines 140.21 20 Homilies 316–321 141.14 20 1.18 317 141.15 2 2.16–18 317 143–145 25 2.38 319 144.17–18 6.20 2.38–52 320 145–146 3 3.4–6 320 147–148 30–31 3.9–11 320 148.6–7 32 3.17–21 317, 320 148.9 20 3.18–19 319 151–161 26–27 3.23–25 317 156.14 21 3.37–51 320 159.5 24 3.47 319 160.10 20 7.8 318 161.4 3 8 308 161.6 26 8–11 316 161.16 21 8.10 317 161.17 5 8.11–15 317 162 31–32 8.11–20 317 162.5–6 32 8.12 317 163–167 6 8.15–16 318 168 14 8.17 318 169.2–8 3 8.18–19 318 169.9–10 29 8.18–20 318 9 308 Palaea Interpretata 1, 33 9.2–18 317 9.12–14 318 Peter Comestor 10.7–23 317 Historia Scholastica 5 11.29 319 16.9–14 320 Photius 16.19.4–7 162 Bibliotheca 108 18.12–13 320 18.18–22 320 Priest David Recognitions 316–319 passim Book of Knowledge 1.24–38 317 and Belief 343 1.29 317 4–6 316 Pseudo-Athanasius 4.26–27 317 History of Melchizedek PG 28:525–529 17 Pseudo-Dionysius the Quaestiones in scripturam sacram Areopagite 35 PG 28:737BC 26 384 Index of Ancient Texts

Pseudo-Germanus Theodoret Historia Ecclesiastica 34 Religious History Prologue 2–3 263 Pseudo-Hegesippus Questions on the Octateuch Hegesippi qui dicitur historiae libri V 70 4 5.44 278 262.2 37

Pseudo-Jacob of Sarug Theodore the Studite 6–7, 260 Homilies on the Six Days of Creation 265, 266 Theodosius De situ terrae sanctae 106 Pseudo-Jerome Questions on the Book of Samuel Theophilus of Antioch 2 Chr 35:3 220 Ad Autolycum 2.13 170, 171 Pseudo-Zachariah Rhetor Chronicle 106 Timothy I Epistle 34 258–259 Questions of Bartholomew 4.53–55 243, 247–249 Valentinus Clement, Samuel Kamrǰajorecʿi Strom. 2.36.2–4 160 MH 10th c.: 742–743 343 Vision of Enoch the Just 296

Story of Melchizedek 17–18, 349–354 Yoḥannān bar Penkāyē Ktābā d-rēš mellē Symeon Logothete Book 1 252 Chronicon 10.14–17 (8.4) 9 Yōḥannān bar Zōʿbī 25.11–12 (24.2) 8 Interpretation of the Mysteries Prologue 254–255 Symeon Metaphrastes Hypomnemata PG 115:437–445 108 Mandaean Literature

Tertullian 305 Right Ginza Against Praxeas 1.88 271 12 171–172 2.21 271 Against Marcion 5.11.11 243 5.17.8 243 Islamic Literature

Theodore Abū Qurrah Al-Ḥakīm al-Tirmidhī On the Veneration of the Holy Icons Kitāb ʿIlm al-awliyāʾ 263–264 259–261 ʿAlī b. ʿUthmān al-Huǧwīrī Theodore bar Koni 72 Kašf al-Maḥǧūb 263–264 Index Of Ancient Texts 385

Qurʾān 231, 234 Philolaus 2 (al-Baqarah) 261 Fragment 12 172 2:30–34 257 2:30–36 256 Phlegon of Tralles 95 7:10–19 256 15:26–35 256 Pliny 17:61–65 256 Natural History 18.8–25 111 7.175 96 18:50 256 20:116–117 256 Simplicius 38:71–78 (Ṣād) 256–257, 259–267 Commentary on Aristotle’s passim Physics 95, 107

Tacitus Classical Literature Histories 5.13.1 281 Aristotle Physics Theopompus 95 4.11 94–95, 107 Manuscripts Cited Cassius Dio Roman History Bayerische Staatsbibliothek, Cod. 69.11.4 293 Hebr. 222 Secrets of Rabbi Shimʿon b. Yoḥai Celsus 300 True Reason 156–157, 158, 160–161 Berlin Staatliche Museen P 10598 Diogenes Laertius Numbers fragment 202 Lives of Eminent Philosophers 1.109 94–96 Biblioteca Palatina Cod. Parm. 2342 Secrets of Rabbi Shimʿon b. Yoḥai Hermes Trismegistus 283 300 Hippocrates Biblioteca Palatina Cod. Parm. 3122 On Airs, Waters, and Places Secrets of Rabbi Shimʿon b. Yoḥai 5.6 318 300 Julian Bodleian Library Ms. Heb. 370.11 Against the Galileans Baraita d’Melekhet Ha-Mishkan 261E 299 219–221 Maximus of Tyre Bodleian Library Ms. Heb. 2740.14 Orations 78 God of Old)) אלוהי קדם 10.1 96 38.3 96 Bodleian Library Ms. Heb. d. 46/11 Secrets of Rabbi Shimʿon b. Yoḥai Pausanias 300 Description of Greece 1.14.4 96 386 Index of Ancient Texts

Bodleian Library Ms. Heb. d. 89 Paris, Bibliothèque nationale, Exod 2:23–25 202 cod. gr. 37 Palaea Historica 1, 20, 33, 38 British Library Or. 1481 Yemenite midrash 298 Paris, Bibliothèque nationale, cod. gr. 139 British Museum Add. Ms. 25875 Palaea Historica 27 Cave of Treasures 244 Paris, Bibliothèque nationale Cairo Genizah Ms. Hébreu 31 179–207 T–S NS 261.111 passim Apocalyptic text, Judeo–Arabic 297–298 Paris, Bibliothèque nationale, T.–S. NS 3.21 supp. gr. 928 Genesis fragment 202 Palaea Historica 35

Hekhalot text 75, 80, 91 Pierpont Morgan Library M. 638 Morgan Crusader Bible Galata 154 F. 1r 269–270 Genealogy of Abraham 358 Sassoon no. 368 (Farḥi Bible) Kaufmann A 50 48 Ff. 146, 403 179–207 passim Leiden Or. 4720, Palestinian Talmud Scorialensis Ψ.11.20 (gr. 455) y. Ber. 8:5 (12b) 175 Palaea Historica 34

London, Harl. 5508 Vatican 60 b. Yoma 209 Genesis Rabbah 174–175

Mingana Syr. 179 Vatican Ottob. gr. 205 F. 4v: Yoḥannān bar Penkāyē, Palaea Historica 1, 20 Ktābā d-rēš mellē 252 Venice, Biblioteca Nazionale Nash Papyrus 179 Marciana, gr. 501 Palaea Historica 1, 20, 33, 38–39 Oxyrhynchus Papyrus No 4365 319 Vienna, Cod. slav. Vindob. no. 158 Palaea Historica 33–34 Panaghia, cod gr. 68 Palaea Historica 1 Vienna, Vindob. hist. gr. 119 Palaea Historica 1 Index of Modern Authors

Aalen, S. 164, 167 Berti, V. 258 Adelman, R. 312 Beyer, H. W. 290, 291 Adler, W. 1, 7, 9, 10, 305, 307, 311, 313, 319, 353 Beyerle, S. 284, 294 Aḥituv, S. 340 Bianchi, U. 262 Albeck, C. 71, 180 Biliarsky, Ι. 33 Alexander, P. J. 280 Binggeli, A. 258 Alexander, P. S. 71 Birenboim, H. 41 Allison, D. C. 4, 275, 276, 302 Birley, A. R. 293 Alon, G. 125–126 Bishai, W. 257 Altmann, A. 167, 239 Bitton-Ashkelony, Β. 37 Alverny, M.-T. d’ 73, 81 Böck, Α. 18 Amara, D. 331 Bogaert, P. 100, 105, 214 Ameling, W. 110 Bohak, G. 75, 297–298 Andersen, F. I. 81, 192 Bolman, E. S. 19 Anderson, G. A. 162, 232, 236, 241, 247, 250 Bolton, J. D. P. 96 Aptowitzer, V. 27, 124, 185, 189, 211 Borgen, P. 157, 159, 160, 171 Aran, G. 128 Böttrich, C. 237 Ariel, D. T. 288 Boustan, R. S. 318 Arndt, T. 203, 204 Boyarin, D. 169, 239, 241 Auerbach, E. 344 Boyce, M. 302 Auffret, P. 331 Brakke, D. 305, 307 Aune, D. E. 262, 277 Brock, S. P. 107, 109, 343 Aus, R. D. 276 Broek, R. van den 239 Aveta, M. 263 Brooke, G. J. 274 Awn, P. J. 257 Brubaker, L. 27 Bruns, P. 252 Bacher, W. 176 Büchler, A. 40, 63, 104 Baer, Y. 170 Bultmann, R. 164 Baert, B. 14, 15, 16 Balberg, M. 53 Calasso, G. 257 Baneth, E. 50 Carriker, A. 284 Barag, D. 126 Cathcart, K. J. 273 Batsch, C. 128 Chadwick, H. 157 Bauckham, R. 36, 37, 171 Charles, R. H. 36, 144, 176–177, 214, 218, 321 Bauernfeind, O. 279 Charlesworth, J. H. 321 Baumgarten, A. I. 54, 63, 320 Choufrine, A. 164 Baumgarten, J. M. 42, 274 Clark, E. A. 109 Bayer, B. 333 Clements, R. A. 22, 129 Beatrice, P. F. 280, 283 Clines, D. J. A. 196 Beck, E. 256 Cohen, S. J. D. 112–115, 124, 125 Becker, A. H. 250 Cohon, S. S. 59 Ben-Dov, J. 81, 286 Collins, A. Yarbro 65, 163 Ben-Ḥayyim 153–154, 173 Collins, J. J. 273, 274, 275, 286, 294–295, 310 Berger, K. 36 Collins, M. F. 224 Bernstein, M. J. 129, 309 Colson, F. H. 171 388 Index of Modern Authors

Conzelmann, H. 164 Fleischer, E. 71 Cook, J. G. 299 Fletcher-Louis, C. H. T. 234 Cousland, J. R. C. 233 Florentin, M. 186 Crenshaw, J. L. 341 Flusser, D. 1, 2, 4–6, 12, 13, 17, 19, 28–29, 31, Culbertson, P. L. 35 32, 36, 38, 41, 158, 210, 277, 281–283 Fontanille, J.-P. 288 Dagron, G. 31, 32 Forsyth, N. 230 Dan, J. 80, 85 Fossum, J. E. 79, 80, 81, 241 Davies, W. D. 275, 276, 302 Fraade, S. D. 55 Davila, J. R. 5, 36, 101, 215 Fraenkel, Y. 322 Day, J. 208 Frank, S. 263 Denis, A.-M. 17, 100, 101, 307 Franklin, S. 319 Derenbourg, J. 123, 124 Freund, S. 281–282, 283 De Rossi, G. B. 185, 186, 187, 300 Frey, A. 232, 243 Deutsch, R. 292 Friedlander, I. 123–124, 125 Diels, H. 95, 97, 172 Friedmann (Ish-Shalom), M. 75, 217 Dimant, D. 311 Funk, W.-P. 304 Dochhorn, J. 17, 234, 237, 243, 350, 353 Furstenberg, Y. 40, 42, 47, 49, 55, 59, 63, 65, Dodds, E. R. 96, 97 66, 290 Doering, L. 142, 172, 173 Donner, H. 106 Gaca, K. L. 285 Dorival, G. 276, 299 García Martínez, F. 41, 274, 310, 331 Drijvers, H. J. W. 250, 288, 317 Geiger, A. 234 Drijvers, J. W. 279 Geller, M. J. 125, 126 Dudley, M. 357 Gelzer, H. 7, 8 Dunn, J. D. G. 128, 129 Giakalis, A. 260 Giannouli, Α. 16, 33, 39 Efron, J. 120, 121, 124 Gignoux, P. 249 Elbaum, J. 75 Gilders, W. K. 318 Elbogen, J. 78 Ginsburg, C. D. 196, 201 Eldridge, M. D. 247 Ginzberg, L. 12, 35, 71, 72, 151, 153, 156, 166, Elizur, S. 69, 71, 74, 77, 170 210–211, 219, 220, 234 Epstein, A. 180, 184, 185, 194, 197, 203, 204, Glessmer, U. 204 267, 268 Goldberg, A. 286 Epstein, J. N. 48, 285, 327 Goldschmidt, D. 282 Eshel, E. 164, 340 Goldstein, J. A. 212–213 Eshel, H. 292, 294 Goodblatt, D. 124, 125, 126, 292 Evans, E. 172 Goodenough, E. R. 288 Goodman, M. 42 Fabricius, J. A. 304, 307, 313, 321 Goshen-Gottstein, M. H. 189 Falk, D. K. 167–168 Gottlieb, I. B. 323, 324, 326 Fallon, N. 14 Gouillard, J. 32 Felder, S. 294 Graffin, F. 251 Feldman, L. H. 128 Granat, Y. 70, 71, 75, 77, 80, 90, 145, 169–170 Ficker, G. 2 Grant, R. M. 171 Fine, S. 292 Greatrex, G. 106 Firestone, R. 346 Greenberg, M. 183 Fishbane, M. 279 Greenfield, J. C. 164 Fitzmyer, J. A. 168 Grenet, F. 302 Index Of Modern Authors 389

Grierson, F. 36 Jonge, M. de 101, 231–232, 233, 247 Griffith, S. H. 260 Joseph, I. 271 Griffiths, A. H. 96 Jung, L. 234 Grypeou, E. 346 Guillaumont, A. 72, 86 Kaddari, M. Z. 337 Gundel, H. G. 288 Kaestli, J.-D. 232, 234–235, 243 Kahana, M. 174, 177, 221 Habermann, A. M. 196, 197 Kalimi, I. 213 Halpern-Amaru, B. 315 Kalmin, R. 115, 119 Hamerton-Kelly, R. G. 70 Kasher, M. M. 299 Hanauer, J. E. 15 Kelly, H. A. 230 Haran, M. 208, 218 Kennicott, B. 185, 186, 187 Harkavy, A. A. 184–185, 204 Kessel, G. M. 265 Harl, M. 153 Kim, Y. R. 306, 308 Harrington, D. J. 82, 83 King, K. L. 239 Harrington, H. K. 46, 55 Kirschner, R. 220 Hay, D. M. 352 Kister, M. 41, 70, 72, 74, 76, 77, 79, 80, 82, 83, Hayes, C. E. 46, 53 89, 132, 166, 168, 170, 173, 177, 218–219, Hegedus, T. 275 242–243, 257, 275, 279, 280, 282, 291, 299 Heinemann, I. 322, 328 Klawans, J. 41, 42, 46 Heinemann, J. 334 Klijn, A. F. J. 314 Heller, B. 110, 111 Kloner, A. 292 Hempel, C. 45 Koch, J. 94, 106, 107, 110, 111 Hengel, M. 178, 289 König, R. 96 Henze, M. 144 Kooten, G. H. van 170, 171, 273 Herrmann, K. 241 Kraeling, C. H. 288 Herzer, J. 97, 99, 105–106 Kranz, W. 95, 172 Hilhorst, A. 275 Kraus, H.-J. 103 Himmelfarb, M. 41, 181, 309, 310–311 Kreps, A. 311 Holladay, C. R. 163 Kronholm, T. 251 Honigmann, E. 106, 107, 108–109 Krueger, D. 20, 38 Horbury, W. 273, 274, 285–286 Küchler, M. 289 Horovitz, H. S. 328 Kugel (Kaduri), J. L. 4, 12, 27, 71, 73, 139, 142, Horowitz, W. 278 143, 147, 151, 225, 247, 273, 305, 309 Horst, P. van der 86 Kuhn, K. H. 101, 110, 214 Horton, F. L. 3 Kulik, A. 100, 166, 244, 357 Houghton, A. 288 Kutscher, E. Y. 183–184, 189, 191, 193–194, Huber, M. 95, 102, 103, 105, 106, 111 200–201 Hurst, T. R. 258 Kuyumdzhieva, Μ. 27

Ilan, T. 112–114 passim, 116, 124, 127 Laderman, S. 290 Irshai, O. 279 Lambert, R. 293 Issaverdens, J. 296 Lamoreaux, J. C. 259 Lampe, C. H. W. 16 Jackson, B. S. 334 Landau, B. C. 300–301 Jacobson, H. 24, 82, 83 Lange, A. 180, 184, 186, 194, 198, 204 James, M. R. 5 Lange, C. 251 Jansma, T. 72 Langer, G. 197 Jones, F. S. 316, 317, 318 Langermann, Y. T. 298 390 Index of Modern Authors

Laqueur, R. 114 Mildenberg, L. 284, 285, 287–289, 292 Lascaratos, J. 31 Milik, J. T. 146, 199 Layton, R. 313–314 Milikowsky, C. 123, 219 Le Déaut, R. 169, 280 Miller, S. J. 276 Leclercq, H. 94 Milson, D. 239–240 Leemans, J. 276 Mingana, A. 214, 252 Leeven, J. 191 Minov, S. 161, 163, 244 Leiman, S. Z. 185 Mirsky, A. 160 Leonhard, C. 244, 280 Mizrahi, N. 141 Levine, B. A. 273 Moehring, H. R. 173 Levine, L. I. 120, 291 Monferrer-Sala, J. P. 256 Levison, J. 156 Moorsel, P. van 18, 19 Licht, J. 44, 344 Morag, S. 268 Lichtenberger, H. 142 Moreen, V. B. 268–269 Lieberman, S. 1, 2, 12, 56, 58, 62–63, 67, 183, Muehlberger, E. 73, 263 200, 201, 203, 204, 212, 213 Mutius, H.-G. von 189, 190 Lietzmann, H. 290, 291 Loewinger, S. 185, 191, 193, 195, 197, 203–204 Najman, H. 309, 310 Loon, G. J. M. van 5, 19 Narkiss, B. 181 Loos, M. 3 Nau, F. 21, 265, 266 Lucchesi, E. 22, 34–35 Neubauer, A. 184–185 Lührmann, D. 313–314 Neusner, J. 63, 124 Luijendijk, A.-M. 319 Newman, H. I. 215, 293–294, 295 Lust, J. 274 Newman, J. 86 Newsom, C. 141, 168 Macaskill, G. 153, 235, 236 Nickelsburg, G. W. E. 81, 139, 176, 213, 233 Maccoby, H. 65 Nicklas, T. 275 Mack, H. 180 Niehoff, M. R. 239 Main, E. 125, 126 Niessen, F. 297 Mann, J. 212 Nir, R. 233 Marcovich, M. 171 Noam, V. 46, 69, 112–116 passim, 121–126, Marketos, S. 31 131–132 Markschies, C. 79, 80, 83, 160 Nocchi Macedo, G. 288 Marshall, F. H. 5, 33 Noga-Banai, G. 292 Marti, M. 232 Norton, J. D. H. 203 Maser, P. 291–292 Noth, M. 338 Mason, S. 119–120 Mathévossian, A. 343 Obolensky, D. 2 Mathews, E. G. 72, 85 Oppenheimer, A. 40, 59, 63 Mayerson, P. 11 Orbe, A. 171–172 McDonald, L. M. 307 Orlov, A. A. 28, 81, 152, 230, 236–237, 241, McNamee, K. 288 353–354 Mégas, G. A. 33 Ortlund, D. 128 Melamed, E. Z. 327 Outtier, B. 232, 243 Meshorer, Y. 287 Meyer, H. 293 Pagels, E. 230 Mézange, C. 280 Paramelle, J. 22, 34–35 Michalak, A. R. 5 Parry, K. 260 Michel, O. 279 Patrich, J. 292 Index Of Modern Authors 391

Peeters, P. 106 Russell, J. B. 230 Pettorelli, J. 232, 243 Rutgers, L. 289–290 Pietersma, A. 4, 24 Ruwet, J. 305 Pillinger, R. J. 93, 292 Pinggéra, K. 265 Saltman, A. 220 Piovanelli, P. 350–351 Samellas, A. 108 Poirier, P. H. 304 Samir, S. K. 257 Polotsky, H. J. 282 Sanders, J. A. 73, 330–332 Poorthuis, M. J. H. M. 241 Sarfatti, G. B. 104, 105 Pourkier, A. 306 Sassoon, D. S. 181 Price, R. 203 Satran, D. 353 Purvis, J. D. 213 Schäfer, P. 75, 79, 239–240, 284–285, 289 Schalit, A. 279 Qimron, E. 141, 193 Schaller, B. 97, 101 Quinn, E. 14 Schelbert, G. 334 Schiffman, L. H. 185 Rabbinovicz, R. N. 209 Schloessinger, M. 267 Rabin, C. 124, 189 Schlossberg, E. 298 Rabinowitz, L. I. 268 Schmidt, F. 277, 281 Rad, G. von 165 Schoeler, G. 257 Radtke, B. 264 Scholem, G. 167 Rajak, T. 5, 11 Schott, J. 306, 309, 312 Ramsey, J. T. 278, 293 Schremer, A. 60 Rand, M. 77, 78, 85 Schürer, E. 42, 212 Rasimus, T. 80 Schwartz, D. R. 114, 119–120, 124, 130, 213, 278 Ratzon, E. 81 Scopello, M. 304 Rea, J. R. 293 Scott, J. M. 311–312 Reed, A. Y. 8, 301, 305, 306, 308, 316, 317, 318, Sed-Rajna, G. 181 319, 321 Seeligmann, I. L. 164 Reeves, J. C. 296, 297, 299, 346 Segal, A. F. 80 Reinhart, J. 1, 33–34 Segal, M. 196, 309, 331 Reynolds, B. H. 286 Segal, M. H. 183, 191, 196, 197, 204 Reynolds, G. S. 257 Seleznyov, N. N. 254 Ri, S.-M. 244 Shahîd, I. 257 Riaud, J. 99, 172, 173 Shemesh, A. 42–45, 52 Rilliet, F. 106 Shinan, A. 325, 326 Roberts, B. J. 185 Shivtiel, A. 297 Robinson, S. E. 17, 19 Shulman, D. 344 Roggema, B. H. 258 Shuve, K. E. 320 Rohde, E. 95, 96, 107 Siegal, M. Bar-Asher 240 Romanoff, P. 289 Siegel, J. P. 179, 183–184, 185, 189–201 passim, Rönsch 307, 313, 317 203 Rösel, M. 274 Sievers, J. 125 Rosenstiehl, J.-M. 230, 244 Simpson, M. S. 269 Rosenzweig, F. 303 Sirat, C. 202 Rowell, G. 357 Smith, C. B. 80 Rubenstein, J. L. 40, 102 Smith, J. Z. 89 Ruiten, J. van 273 Smith, M. S. 165, 169 Runia, D. T. 155, 161 Sokoloff, M. 162, 242 392 Index of Modern Authors

Sollamo, R. 146 Tuval, M. 114 Sonne, I. 212 Tzaferis, V. 15 Sorlin, I. 33 Speller, E. 294 Urbach, E. E. 88, 167, 170, 238, 297 Sperber, A. 193, 197 Urowitz-Freudenstein, A. 238 Sperber, D. 295 Spurling, H. 346 VanderKam, J. C. 9, 72, 81, 82, 146, 176, 307, Stemberger, G. 180 309, 310 Stern, M. 114, 124, 281 Vassiliev, A. 1–2, 5, 7, 21, 36, 38 Steudel, A. 352 Vermes, G. 104, 164 Steyn, G. 135 Vollmann, B. K. 106 Stichel, R. 13, 27 Vööbus, A. 249, 263 Stökl Ben-Ezra, D. 237 Stone, M. E. 37, 81–82, 101, 138, 159, 162, 164, Wacholder, B. Z. 211, 214, 295, 310 222, 231–236, 340, 343–359 passim Wagner, H. 94–95 Strecker, G. 316, 317, 320 Wander, S. H. 27 Stroumsa, G. G. 314 Waszink, J. H. 95 Stroumsa, S. 163 Wegner, P. D. 203 Sundermann, W. 283 Weil, G. E. 200 Sussmann, Y. 68, 174 Weiss, D. H. 269 Swartz, M. D. 169 Weiss, Z. 288 Weitzman, S. 217–218 Takahashi, H. 265, 266 Weltecke, D. 253 Tal, A. 154, 186 Werman, C. 310 Talmon, S. 189 Wertheimer, S. A. 209 Ta-Shma, I. M. 181 Whittaker, G. H. 171 Tcherikover, V. 124 Wilkinson, J. 14, 18 Terian, A. 156, 157 Williams, F. 80 Teule, H. G. B. 254, 261, 265, 266–267 Williams, M. A. 239 Theodor, J. 71 Wilson, J. F. 293 Thompson, S. 102 Witztum, J. 257 Tigay, J. H. 183 Wolff, C. 215 Tigchelaar, E. J. C. 317, 318, 331 Wolfson, H. 80 Tobi, J. 299 Worthington, J. D. 168 Tobin, T. H. 171 Toepel, A. 249–250 Yahalom, Y. 162, 169, 242 Tottoli, R. 261 Yardeni, A. 191, 202 Tov, E. 183, 186, 189, 191, 192, 196, 197, 203 Yuval, I. J. 239 Treitl, E. 155 Tromp, J. 36, 231, 232, 247, 347 Zakovitch, Y. 325, 326, 333 Turdeanu, Ε. 32 Ziegler, J. 99 Turner, J. D. 314 Zimmermann, N. 290, 292 Tuschling, R. M. M. 263