The Caves of Qumran (Studies on the Texts of the Desert of Judah) 9004316493, 9789004316492

Citation preview

The Caves of Qumran

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 118

The titles published in this series are listed at brill.com/stdj

The Caves of Qumran Proceedings of the International Conference, Lugano 2014 Edited by

Marcello Fidanzio

LEIDEN | BOSTON

Library of Congress Cataloging-in-Publication Data Names: Fidanzio, Marcello, editor. Title: The caves of Qumran : proceedings of the international conference,  Lugano 2014 / edited by Marcello Fidanzio. Description: Leiden ; Boston : Brill, [2016] | Series: Studies on the texts of  the desert of Judah, ISSN 0169-9962 ; volume 118 |  Includes bibliographical references and index. Identifiers: LCCN 2016036866 (print) | LCCN 2016045137 (ebook) |  ISBN 9789004316492 (hardback : alk. paper) | ISBN 9789004316508 (E-book) Subjects: LCSH: Qumran Site (West Bank)—Congresses. | Qumran community. Classification: LCC DS110.Q8 C38 2016 (print) | LCC DS110.Q8 (ebook) |  DDC 933/.51—dc23 LC record available at https://lccn.loc.gov/2016036866

Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface. issn 0169-9962 isbn 978-90-04-31649-2 (hardback) isbn 978-90-04-31650-8 (e-book) Copyright 2017 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 and produced in a sustainable manner.

Contents Foreword IX George J. Brooke List of Maps, Figures, and Tables XI Abbreviations xIv Introduction 1 Marcello Fidanzio

Part 1 Topography 1 The Qumran Caves in their Regional Context: A Chronological Review with a Focus on Bar Kokhba Assemblages 9 Joan E. Taylor 2 Cacher et se cacher à Qumrân : grottes et refuges. Morphologie, fonctions, anthropologie 34 Jean-Baptiste Humbert

Part 2 Manuscripts 3 The Contents of the Manuscripts from the Caves of Qumran 67 Florentino García Martínez 4 The Profile and Character of Qumran Cave 4Q: The Community Rule Manuscripts as a Test Case 80 Charlotte Hempel 5 Scribal Characteristics of the Qumran Scrolls 87 Emanuel Tov 6 La paléographie des manuscrits de la mer Morte 96 Emile Puech

vi

Contents

Part 3 Other Finds 7

Terracotta Oil Lamps (Roland de Vaux’s Excavations of the Caves) 109 Jolanta Młynarczyk

8

The Unpublished Textiles from the Qumran Caves 123 Mireille Bélis

9

Miscellaneous Artefacts from the Qumran Caves: An Exploration of their Significance 137 Dennis Mizzi

10 The Distribution of Tefillin Finds among the Judean Desert Caves 161 Yonatan Adler

Part 4 Chronology, Functions, Connections 11 When and Why Were Caves Near Qumran and in the Judaean Desert Used? 177 Mladen Popović 12 The Connection between the Site of Qumran and the Scroll Caves in Light of the Ceramic Evidence 184 Jodi Magness 13 The Functions of the Caves and the Settlement of Qumran: Reflections on a New Chapter of Qumran Research 195 Jürgen K. Zangenberg

Part 5 Short Papers 14 The Inscriptional Evidence from Qumran and its Relationship to the Cave 4Q Documents 213 Sidnie White Crawford

Contents

15 The Coins of Khirbet Qumran from the Digs of Roland De Vaux: Returning to Henri Seyrig and Augustus Spijkerman 221 Bruno Callegher 16 Dating the Scroll Deposits of the Qumran Caves: A Question of Evidence 238 Gregory L. Doudna 17 History of the “Qumran Caves” in the Iron Age in the Light of the Pottery Evidence 247 Mariusz Burdajewicz

Appendix 18 Finds from the Qumran Caves: Roland de Vaux’s Inventory of the Excavations (1949–1956) 263 Marcello Fidanzio and Jean-Baptiste Humbert Bibliography 333 Index of Modern Authors 352 Index of Sites and Place Names 358

vii

Foreword This collection of essays is a most welcome addition to scholarship on the finds in the region of Khirbet Qumran. In fact, the volume as a whole represents a change of direction which should lead to a far more refined understanding of all the discoveries in and from the caves. That change of direction has five aspects to it which I briefly note here in appreciation of what the volume is moving towards as it attempts to contribute collectively to the construction of a new paradigm in the combination of archaeological and textual studies of the late Second Tempe period. First, the volume contains a set of studies that pay attention to details that have often previously been overlooked. Sometimes earlier publications have made very brief comments in their observations, but here the materials themselves have been revisited and the details given proper focus. In earlier stages of research many matters were not given their due weight in either description or interpretation, perhaps because scholars often were too eager to move towards the reconstruction of a larger picture or because they were thinking in categories of coherence and synthesis which were based on other data. It is overwhelmingly important that the evidence is scrutinised closely in its own right so that it can be used to interrogate received opinion. The essays here facilitate such interrogation through their respect for evidence and the permission they give the evidence to speak for itself. Second, the attention to detail has been made possible in large part through the availability of new technologies. In archaeology the range of tools that have become available during the previous generation have notably encouraged attention to and increased awareness of very small matters of evidence, whether a small collection of pollen, the structure of textile remains, the archaeometric profile of clay, or the chemical composition of tiny pieces of animal or vegetable remains. This focus on minutiae has required the learning of new skills, not all of which can be appreciated fully by any one person. Scholarship in the area of archaeological observation and interpretation has increasingly become a subject of teamwork to which specialists across several technical sub-disciplines contribute with fascinating precision. It is not surprising that it requires a volume like this to reflect the character of such team work as specialist studies are juxtaposed with one another in appropriate ways. The details are important, but their juxtaposition with other matters is equally so. Third, such attention to detail as offered by a new generation of specialists is not without a wider concern for

a bigger picture. Here the matter is one of contextualisation. In the first generation after the discoveries in the Qumran caves and at the site nearby, the excitement in much of the new data, especially textual data, meant that very many biblical scholars and others wanted to offer an opinion on what was made public. In a second generation, the enthusiasm waned: some texts remained unpublished, the significance of the data for matters already known seemed to be indirect at best, and the whole affair could be characterised as reflecting the life and times of a distinctive marginal group of Jews who had deliberately taken themselves away from larger society and who therefore could not be representative of anything. The current third generation of research began with the general release of all the fragmentary textual evidence in the early 1990s and fresh archaeological investigations since then too. In this third generation there is a new direction which the essays in this volume reflect and contribute to, namely that a much wider context needs to be taken into account for the better understanding of what is now observed in much closer detail. Although there might be several distinctive features in the profile of the caves at and near the Qumran site, such distinctiveness can only be properly recognised and appreciated when set in a larger context than the Qumran locale alone. The increase in the number of sites with information relating to late Second Temple period has serendipitously assisted in this wider awareness of context. Much remains to be done, but there are important indicators in several of the essays in this volume as to how scholarship might proceed. Fourth, readers will note that this volume of essays contains contributions from those who are either recognized specialists in the study of material remains or who are specialists in the reading and analysis of textual evidence. The collection together of these contributions is a very important step in the greater appreciation of how manuscripts are the material culture of textual data and how other kinds of material culture might be set alongside the manuscript remains to provide an integrated picture of the caves and their contents. In fact, with some obvious exceptions, it is surprising to see how often archaeologists in reporting on the caves or the site at Qumran have ignored the manuscripts as archaeological artefacts and equally interesting to see how commonly the texts are read out of context as if the date of composition of a text is somehow self-evidently more important than the social circumstances of the text’s copying, use, and storage as attested by a particular manuscript.

x Fifth, this book contains a wealth of evidence that is suggestive of various stages in the use and storage of objects. Thus, alongside very careful description which is sometimes presented in a somewhat synchronic fashion, in several of the contributions there are strong suggestions for the diachronic reading of the evidence. And what is even more important in this interaction of the synchronic and the diachronic is that not everything need or can be put into a single mould. There are strong indications of the need for the evidence from the caves, and of

Foreword

course from the wider site at Qumran, to be understood in a pluralist fashion as implying multiple fazes, variable uses, several purposes. Diversity or multiplicity in interpretative strategy is increasingly shown to be important. All this requires fresh nuance and insightful subtlety and the studies in this volume display those characteristics in abundance. George J. Brooke

List of Maps, Figures, and Tables Maps 1.1 1.2 1.3 1.7 1.8 7.12 17.1

The artificial caves cut in the marl terrace on the edge of the Wadi Qumran, within the occupation zone of the Qumran settlement 11 The map drawn by J. Ziegler of the 1952 caves expedition with de Vaux’s numbering system 12 The key wadis and sites of the western Dead Sea 15 Map of the Northern Dead Sea in the Second Temple Period showing settlements, roads and wadis 32 Amended version of Ziegler’s plan, showing locations of caves yielding remains 33 Distribution of lamp finds in the caves in the Qumran area according to their types 122 Location of caves with the Iron Age Pottery 256

Figures 1.4

The Cave of the Column, at the front of the rocky cliffs, with Cave 11Q to the north 17 1.5 Cave GQ37, located 1 km south of Qumran 20 1.6 The marl terrace artificial caves on the edge of the Wadi Qumran and Qumran plateau, Cave 4Qa, 4Qb, 5Q, 7Q, 8Q, 9Q and 10Q 30 2.1 Fouilles sauvages au sud de la grotte 11Q 34 2.2 Les environs au sud de la grotte 11Q en 1951 34 2.3 Excavations récentes au sud de la grotte 11Q 34 2.4 La grotte 11Q après la fouille de 1956 35 2.5 La grotte 11Q en 2013 35 2.6 Les anfractuosités autour de la grotte 2Q 43 2.7 L’entrée de la grotte 1Q en cours de dégagement ; les ouvriers étendent une poche de cendres sous l’éboulis 44 2.8 Enlèvement de l’éboulis devant l’entrée basse de la grotte 1Q 44 2.9a Le déblai moderne après dégagement de l’entrée de la grotte 3Q 45 2.9b Les ouvriers dégagent les gros blocs qui scel­lent l’entrée de 3Q 45 2.10 Entrée de la grotte 11Q 46 2.11a et b L’abri sous roche GQ15, avec la jarre prise dans l’éboulis et vue rapprochée 47 2.12 Les deux refuges dominent le Wadi Qumrân et dont séparés par la ravine qui borde à l’ouest la terrasse du site 48 2.13 Panorama des deux refuges occidental et oriental 49 2.14 Le Wadi Qumrân et les grottes artificielles creusées dans la berge abrupte 50 2.15 Façade de l’éperon oriental avec les grottes 7Q, 8Q 50 2.16 Les crevasses verticales marquent les chambres effondrées entre les grottes 4Q et 10Q 51 2.17 La découpe profonde de l’éperon des grottes 4Q 52 2.18 L’embrasure de la fenêtre de la chambre 4Qa, ouverte vers l’est 53 2.19 La façade orientale de l’éperon des grottes 4Q et 5Q 53 2.20a, b, c, d Les conduits internes de la grotte 4Q 54

xii 2.21 2.22a 2.22b 2.23 2.24 2.25 2.26 2.27 2.28 2.29 2.30 2.31 7.1 7.2

List of Maps, Figures, and Tables

Intérieur de la grotte 4Qa 55 Plans de 4Q 56 Essai de restitution du refuge 56 Puits d’entrée de la grotte 4Q 57 La grotte 5Q à la fin de la fouille 57 Relevé de la grotte 5Q 57 Accès escarpé à la grotte 4Q 57 Relevé schématique de la partie conservée du refuge du Qasr al-Abiadh 60 Relevé schématique du Qasr al-Tûba par A.-G. Barrois 60 Vestiges de cavités éboulées à l’ouest de 4Q 62 Tuyaux verticaux d’érosion et de dissolution 63 Effondrement des zones de conduits verticaux (les “pipes” des badlands) 63 Lamps Gr 1Q43 and Gr 1Q44 117 “Qumran family” of lamps. Qumran Types: 032 (KhQ 171), 033.1 (G 1Q-43), 033.2 (KhQ 2210, KhQ 1015), 034.1 (KhQ 2206), 034.2 (KhQ 5087), 035 (KhQ 1257) and 035-Prime (KhQ 2308) 118 7.3 Lamps of the “Qumran family” from Loc. 130 119 7.4 Fragmentary lamp Gr 1Q-56 119 7.5 Lamp GQ 29-1 119 7.6 Lamp GQ 8-12 120 7.7 Lamp GQ 9-1 120 7.8 Lamp Gr 10Q3 before and after restoration 120 7.9 Fragmentary lamp Gr 10Q3 120 7.10 Fragments of lamp Gr 8Q12 121 7.11 Lamp Gr 4Q22 121 8.1 The original boxes and drawer as found in the Rockefeller Museum containing the Ebaf textile collection.  124 8.2 A parasite in a 11Q textile under DinoLite 124 8.3 Jar GQ39-2 in the Amman Museum, with linen through its lacunae 126 8.4 Just empting of the jar 127 8.5 Cloth 1Q n°19 having a parallel, G2020 in the Amman collection 129 8.6 Fragments of a cloth G516, in the Amman collection 130 8.7 Violet from 8Q 132 8.8 Blue fibers from 11Q under DinoLite 132 15.1 Final synthesis of Augustus Spijkerman’s revision of the Khirbet Qumran Hoard, f. 30 (recto) 236 15.2 Small black-and-white photographs of the obverse and reverse of the silver coins (Khirbet Qumran Hoard), sent from the Archaeological Museum-Amman to father Augustus Spijkerman (November 1960) 237 17.2 Jars: 1–2 = Cave XII/49; 3–4 = Cave 2 in Wadi Murabba⁡ʿat 257 17.3 Jugs and juglets: 1 = Cave 2 in Wadi Murabba⁡ʿat; 2–3 = Cave of the Pool; 4-6 = XII/49; 7 = Cave of the Pool; 8 = Cave 2 in Murabba⁡ʿat; 9–10 = Cave of the Pool; 11–12 = Cave XII/56 257 17.4 Bowls: 1–4 = Cave XII/49; 5 = Cave of the Pool; 6–7 = Cave 13; 8–9 = Caves 1 and 2 in Wadi Murabba⁡ʿat; 10 = Cave XIII/13 258 17.5 Cooking pots: 1–2 = Cave of the Pool; 3-6 = Cave XII/49; 7–8 = Cave 13 259 17.6 Lamps: 1–2 = 11Q; 3 = 9Q; 4 = Cave of the Pool; 5 = Wadi Murabba⁡ʿat 259 18.1 Handwritten Card Index (HCI): Q6 264 18.2 Typewritten card index (TCI): GQ 3–2 268 18.3 Typewritten book (TB): p. 2 269

List of Maps, Figures, and Tables

18.4 Index of typewritten book 269 18.5a, b GQ 3-8 in the typewritten card index, and in the typewritten book 279 Handwritten Card Index (HCI) 281 Typewritten Card Index (TCI) 299 Tables 5.1 Comparison of Hebrew, Aramaic, Greek, Latin, and Nabatean-Aramaic Papyri and Leather Texts from the Judean Desert 91 9.1 Caves in the region of Qumran containing miscellaneous artefacts 155 12.1 Correlation of analyses performed on the same ceramic vessels by more than one team 192 14.1 Table of inscriptional remains from Khirbet Qumran 219 17.1 Juxtaposition of caves with the Iron Age pottery 255 18.1 Comparison of the data in de Vaux’s Handwritten Card Index (Cave 1Q) and the data published in RB 56 (1949) and DJD 1 266 18.2 Comparison of the data in de Vaux’s Typewritten Inventory of the caves (1952–1956) and the data published in RB 60 (1953), DJD 3, and DJD 6 272

xiii

Abbreviations AASOR Annual of the American Schools of Oriental Research ABD Anchor Bible Dictionary ADAJ Annual of the Department of Antiquities of Jordan ADPV Abhandlungen des Deutschen Palästina-Vereins AH De l’Archéologie à l’Histoire AJEC Ancient Judaism and Early Christianity AJN American Journal of Numismatics ANYAS Annals of the New York Academy of Sciences BA Biblical Archaeologist BAH Bibliothèque Archéologique et Historique de L’institut français d’archéologie de Beyrouth BAIAS Bulletin of the Anglo-Israel Archeological Society BAMA British Academy Monographs in Archaeology BAR Biblical Archaeology Review BAR.IS British Archaeological Reports International Series BASOR Bulletin of the American Schools of Oriental Research BIRPA Bulletin de l’Institut Royal du Patrimoine artistique BISNELC Bar Ilan Studies in Near Eastern Languages and Culture BJS Brown Judaic Studies BZAW Beihefte zur Zeitschrift für die Alttestamentliche Wissenschaft CHANE Culture and History of the Ancient Near East CQS Companion to the Qumran Scrolls CREJ Collection de la Revue des Études Juives CRINT Compendia Rerum Iudaicarum ad Novum Testamentum DJD Discoveries in the Judaean Desert DSD Dead Sea Discoveries EDEJ The Eerdmans Dictionary of Early Judaism EDSS Encyclopedia of the Dead Sea Scrolls EI Eretz-Israel EJL Early Judaism and its Literature EncJud Encyclopaedia Judaica, 2nd ed. ESI Excavations and Surveys in Israel FRLANT Forschung zur Religion und Literatur des Alten und Neuen Testaments HBM Hebrew Bible Monographs HSS Harvard Semitic Studies HUCA Hebrew Union College Annual IAA Israel Antiquities Authority IEJ Israel Exploration Journal INJ Israel Numismatic Journal ISACR Interdisciplinary Studies in Ancient Culture and Religion JAJSup Journal of Ancient Judaism Supplements JAN The Journal of Archaeological Numismatics JAOS Journal of the American Oriental Society JArS Journal of Archaeological Science JBL Journal of Biblical Literature JDS Judean Desert Studies JJS Journal of Jewish Studies JNES Journal of Near Eastern Studies JNG Jahrbuch für Numismatik und Geldgeschichte

Abbreviations

JQR JRA JRASS JSJ JSJSup JSOTSup JSP JSPSup LA LevantSup MEAH MSSMNIA NC NCirc NEAEHL NTOA NTOA.SA PAM PapyBrux PEQ PHeid PIA PIBRS QC QR RB RevQ RN RSN SAOC SBF SBF.CMa SBT ScrHier SFSHJ SJSJ SPB STDJ TSAJ TU VT VTSup WUNT ZDPV

Jewish Quarterly Review Journal of Roman Archaeology Journal of Roman Archaeology, Supplement Series Journal for the Study of Judaism Journal for the Study of Judaism, Supplement Series Journal for the Study of the Old Testament, Supplement Series Judea and Samaria Publications Journal for the Study of the Pseudepigrapha, Supplement Series Liber Annuus Levant Supplementary Series Miscelánea de Estudios Arabes y Hebraicos Monograph Series of the Sonia and Marco Nadler Institute of Archaeology Numismatic Chronicle Numismatic Circular The New Encyclopedia of Archaeological Excavations in the Holy Land Novum Testamentum et Orbis Antiquus Novum Testamentum et Orbis Antiquus, Series Archaeologica Palestine Archaeological Museum Papyrologica Bruxellensia Palestine Exploration Quarterly Veröffentlichungen aus der Heidelberger Papyrus-Sammlung Publications of the Institute of Archaeology, Tel Aviv University Publications of the Israel Bible Research Society Qumran Chronicle Qedem Reports Revue Biblique Revue de Qumran Revue Numismatique Revue Suisse de Numismatique Studies in Ancient Oriental Civilizations Studium Biblicum Franciscanum Studium Biblicum Franciscanum, Collectio Maior Studies in Biblical Theology Scripta Hierosolymitana South Florida Studies in the History of Judaism Supplements to the Journal for the Study of Judaism Studia Post-Biblica Studies on the Texts of the Desert of Judah Texte und Studien zum Antiken Judentum Texte und Untersuchungen zur Geschichte der altchristlichen Literatur Vetus Testamentum Vetus Testamentum, Supplements Wissenschaftliche Untersuchungen zum Neuen Testament Zeitschrift des Deutschen Palästina-Vereins

xv

Introduction Marcello Fidanzio This volume presents the proceedings of an international conference dedicated to the caves of Qumran, organized by the Istituto di cultura e archeologia delle terre bibliche (ISCAB) of the Facoltà di teologia di Lugano in collaboration with the É cole biblique et archéologique française de Jérusalem (EBAF). The conference was held in Lugano on 20–21 February 2014 and was attended by a group of thirtyfive Dead Sea Scrolls and Qumran scholars, representing two generations of Qumran research, together with thirty PhD students. The theme of the conference was based on a simple premise: the scrolls were found in caves. Therefore the caves need to be examined in order to better understand the complex reality that we call Qumran. With the exception of 11Q, the caves investigated by Roland de Vaux constitute the only part of the archaeological work that he carried out at Qumran for which he produced a final report in the DJD series. Here de Vaux published not only the findings on the so-called manuscript caves, but also on the entire series of holes, caves and crevices where significant archaeological material was unearthed. After de Vaux’s work, exploration of the cave has continued until recently, not only in the vicinity of Qumran but also throughout the entire western region of the Dead Sea, providing new useful textual and archaeological data (for the history of the explorations and excavations, with methodological observations see the papers by Joan E. Taylor and Jürgen K. Zangenberg). The attention of scholars has largely been focused on the manuscripts, while archaeological research has concentrated above all on the settlement. Thus the caves have generally been left in the background. The aim of this conference was to place the caves at the forefront of the research. Indeed, various participants have highlighted the novelty of the subject, and the present publication is intended as a step forward on a path that still has a long way to go, both in terms of documentation and in interpretation. We need to address the ambiguity for the commonly used designation “Qumran caves” as it denotes both a geographical situation and a conceptual one (i.e. the relationship with the settlement), which do not necessarily coincide (see the observations in the paper by Dennis Mizzi). This is just one small indication of the amount of work that is still required, even in as far as the definition of the phenomena.

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_002

There are two main problems that we face in the study of the Qumran caves. First, that the majority of them have been disturbed by natural and/or cultural post-depositional processes. Secondly, that for the majority of the caves, we only have cursory documentation that does not allow the stratigraphy to be reconstructed. These strongly limits the possibilities of reassembling the context (see the paper by Zangenberg). Nevertheless, the findings from the caves (and the caves themselves) are available for typological studies and distributional analysis. With this in mind and on the basis of their specific area of expertise, each of the speakers at the conference was asked to: – conduct a comparison between the various caves of Qumran, in order to verify their commonalities and assess their peculiarities; – compare the manuscripts and other archaeological materials found in the Qumran caves with the other finds in the Dead Sea region, in order to pinpoint continuity or discontinuity between the Qumran caves and their regional context. What can we say in terms of chronology, function, or sociology? Can we highlight differences between the Qumran caves? Can we argue that what we comprehensively call Qumran shows singular features in relation to their regional context? This renewed attention given to the caves is not related to a preconceived stand on the major themes that, for decades, have animated the debate on the origin of the scrolls and their relationship with the settlement. Instead, the aim was to gain a more detailed understanding of the caves themselves and reconsider them in their broader historical and geographical context, in the light of the data we currently possess. This volume is divided into five parts, in accordance with the sessions of the conference. The first part focuses on the topography. Joan E. Taylor (King’s College London), The Qumran Caves in their Regional Context: A Chronological Review with a Focus on Bar Kokhba Assemblages, gives an overview of the archaeological investigations that were carried out in the caves across region of the north-western Dead Sea, from En Gedi to Jericho. Taylor places an emphasis on the chronology of

2 the discoveries as well as on the typologies of the cave repertoires. The Bar Kokhba remains, which were first discovered in the caves of Wadi Murabba‘at and further to the south, are present throughout the region, including the Qumran area. According to the author, after the Iron Age II there was a widespread phenomenon of temporary occupation of caves only at the time of the Second Revolt. By contrast, during the First Revolt, the phenomenon was that of a reoccupation of some Iron Age II settlements along the route connecting Jericho to En Gedi. A comparative analysis of the material remains reveals the highly distinctive nature of the caves in the vicinity of Qumran. Furthermore, the state of preservation of the manuscripts found in Murabba‘at and the repeated use of the same caves over successive periods can be usefully compared with the discoveries in the Qumran caves. Jean-Baptiste Humbert (É cole Biblique et Archéologique Française de Jérusalem), Cacher et se cacher à Qumrân: Grottes et refuges. Morphologie, fonctions, anthropologie, considers that the isolation of the region of the Dead Sea, which was only sparsely populated and with only seasonal stays related to the limited farming opportunities, would have offered the right conditions to “hide (something) and hide oneself.” Humbert analyses the natural caves in the rocky cliff and the artificial caves excavated in the marl separately. He suggests that the use of the term “grottes” (caves) is inappropriate for natural cavities formed by tectonic movements. With a few exceptions, the natural caves in the vicinity of Qumran are too narrow and too low to have been inhabited. Instead, they may have been used as hiding places to conceal precious belongings. As regards the artificial caves, the author draws attention to the fact that erosion is the cause of important modifications. He suggests that different caves were connected, envisaging networks of caves linked with narrow corridors, similar to the hiding complexes present mainly in the Shephelah, but also in the region of the Dead Sea. The artificial caves were not suitable as hiding places nor as dwellings in the strictest sense, but rather the manuscripts were abandoned there when danger was imminent. Humbert’s text is accompanied by a note by Gérard Massonnat on the geology of the artificial caves. The second part focuses attention on the manuscripts. Florentino García Martínez (KULeuven), The Contents of the Manuscripts from the Caves of Qumran, opens his contribution with a useful review of the recent studies on the caves. He points out that the traditional opinion (i.e. that a single collection of manuscripts was hidden in the caves for safekeeping from the Romans) is now disputed by various hypotheses such as: storehouse for scrolls, burial

Fidanzio

deposits or genizot for manuscripts, multiple collections formed by different groups or by a single group, multiple deposits at different times all belonging to a single collection. The author then examines the contents of the manuscripts found in the individual caves. He highlights the commonalities, in particular between the manuscripts found in caves 1Q, 4Q and 11Q and more widely with those found in the other caves around Qumran. He does not fail to observe the peculiarities of the different caves. Nevertheless, the profile of each cave does not differ significantly from that of the overall corpus of manuscripts. The results of the analysis have steered the author in the direction of the traditional opinion. Based on their contents he believes that the corpus of the manuscripts is a collection of Jewish religious literature representing a particular stream of Judaism in the Second Temple period. Charlotte Hempel (University of Birmingham), The Profile and Character of Qumran Cave 4Q: The Community Rule Manuscripts as a Test Case, emphasizes that the texts from cave 4Q represent a learned and eclectic medley of materials and data that were, in all probability, reserved for the highest tiers of the community members. She outlines the distinctive features of cave 4Q: the presence of texts in cryptic script, the presence of technical calendrical materials, the prominence of the Maskil, the largest number of works in multiple copies, and the “workaday quality” of a number of texts in contrast with the evidence of refinement found in Cave 1Q. She focuses on the Serek tradition and highlights that the complex evidence of the S manuscripts from cave 4Q is compatible with several features that more widely span across the nature of cave 4Q. Furthermore, 4QSe (259) emerges as the most learned and “avant-garde” exemplar of the Community Rule. In several respects, 4QSe seems to be particularly representative of the distinctive aspects of the character of Cave 4Q. Emanuel Tov (Hebrew University of Jerusalem), Scribal Characteristics of the Qumran Scrolls, highlights the special scribal characteristics of the Qumran corpus as opposed to those found in other Judean Desert sites and those dating from other periods, namely: the occurrence of texts written in cryptic and Paleo-Hebrew scripts, the abundance of leather texts, the frequency of scribal signs, and the special writing given to divine names. The scribal signs and the special writing of divine names are found in specific texts written according to the Qumran scribal practice. The texts written in cryptic script could have been written by the same scribes as those who wrote other Qumran scrolls. Conversely, it is possible that the biblical texts written in Paleo-Hebrew script came from Sadducean circles. The prevalence of texts written on

Introduction

leather is due to the special nature of the Qumran collection, which is almost exclusively literary. According to the author, although the texts found at Qumran were not all written by the Qumran community, their nature reflects the features and preferences of that community. As regards the individual Qumran caves, Tov observes that, with the exception caves 6Q and 7Q, these are not homogeneous in scribal practices. He suggests that cave 11Q contains more sectarian texts than the other caves. Lastly he finds acceptable the theory of Stökl Ben Ezra, who distinguishes between “old caves” and “young caves.” Emile Puech (CNRS Paris and EBAF Jerusalem), La paléographie des manuscrits de la mer Morte, presents the contribution of the paleography for the study of the Scrolls. The analysis of the ductus offers a relative chronology for the manuscripts. The author describes different phases in the evolution of the shapes of the letters and remarks that the majority of the manuscripts from Qumran date to the Hasmonean and Herodian periods, while none relate to the post-Herodian period (70–135 CE). He also attests to the value of palaeography in the decipherment of fragments when identifying partially conserved letters and damaged texts. Lastly, the author supports the analysis of Ada Yardeni, who attributed an important quantity of the Qumran manuscripts to a single scribe: 57 + 37, an equivalent to 10% of the entire Qumran corpus. These manuscripts belong to different literary genres and were written mostly on leather, but also on papyrus, dating from the end of the first century BCE until slightly later. These were found in various caves: 1Q, 2Q, 3Q, 4Q, 6Q, (8Q), 11Q. Puech highlights Yardeni’s analysis in support of the unity of the collection and of one-time deposition. The third part focuses on the other material finds from the Qumran caves. Jolanta Młynarczyk (Institute of Archaeology, University of Warsaw), Terracotta Oil Lamps (Roland de Vaux’s Excavations of the Caves), presents a study of the typology, place of production and chronology of the oil lamps dating to the Greco-Roman period, which were found in the caves. With one exception, the oil lamps from the caves always have analogous types in the settlement. In particular, two of the lamps found in cave 1Q belong to the type of so-called “Qumran lamp family” that Młynarczyk considers to be specific to Qumran. Examining the assemblage of oil lamps she concludes that, regardless of their actual manufacturing place, nearly all of them must have come to the caves by way of the settlement of Qumran. As to the chronology, the three oil lamps found in cave 1Q date to different periods, which suggest that the place was visited more than once, probably at an interval of several decades. Four oil

3 lamps found in cave P13=X35 characterize the use of that cave as a place of habitation, most probably a Jewish hideout at the time of the First Revolt. Lastly, she highlights the presence of two oil lamps, which are identical in type and dating, and which were probably deposited during the Second Revolt and testify to the presence of different actors. The first lamp was found at the entrance to cave 4Q and was most likely left by a Roman soldier, searching the caves; the second was found in cave P24 with the disc broken, a sign of Judaic use. Mireille Bélis (chercheur associé É cole Biblique et Archéologique Française de Jérusalem), The Unpublished Textiles from the Caves of Qumran, gives a preliminary report on the textiles stored at the Citadel Museum in Amman and some of those kept in the Rockefeller Museum. The textiles now number over 300 fragments. They consist of a homogeneous corpus of linen textiles, scroll wrappers or jar covers. Some show blue lines in stripes or in welt yarn. An exception is represented by two fragments found in cave 8Q, which the author considers to have been dyed in purple or violet purple. In her opinion, these can be linked to a short-term stay of a single visitor to the cave. Bélis highlights the connection between the presence of jars and textiles in the caves and that of the manuscripts. She considers the possibility that the “empty” or “minor” caves were originally scroll caves which were later “emptied.” However, to develop this argument, it is necessary to ascertain the origin of many textiles found in Amman. As regards the deposition, according to the author the presence of the textiles supports a “slow-hiding scenario.” On the contrary, the absence of textiles or jars in other caves (or at least not present in sufficient quantities) suggest an alternative situation, such as a rapid response to a situation of emergency. With the term “miscellaneous artefacts” Dennis Mizzi (University of Malta), Miscellaneous Artefacts from the Qumran Caves: An Exploration of their Significance, refers to artefacts other than pottery and textiles. He offers for the first time an analysis of these objects. The miscellaneous artefacts give an unambiguous domestic dimension to a cultural assemblage. In the caves in the vicinity of Qumran, with very few exceptions, their presence is very limited in terms of both quantity and variety, unlike in other caves and subterranean complexes of Judea and the Judean Desert. This observation contrasts with the interpretation of the caves as permanent or continuous dwelling places, even though it does not necessarily provide us with conclusive answers. The author also notes that the caves around Qumran differ from the others in the region on account of their direct relationship to a settlement.

4 In view of this and within the framework of the Essenesectarian hypothesis, it is not implausible that some of then were used as long-term dwelling places. Some natural caves differ from the others in terms of the quantity or variety of the miscellaneous artefacts found there, or by the presence of certain distinctive objects indicative of a domestic setting or a different material culture. These are interpreted by the author as probably non-Qumranite, which thereby highlights the difference between “Qumran caves” intended as a geographical designation or as a conceptual category relating the caves to the Qumran settlement. The caves present different histories not only along a diachronic continuum but also along a synchronic one. Yonatan Adler (Ariel University), The Distribution of Tefillin Finds among the Judean Desert Caves, studies the characteristics and the distribution of the tefillin (phylacteries) that were found in the caves. The presence of these ritual objects in the scroll caves (1Q, 4Q, 5Q, 8Q, [11Q?]) contributes to the understanding of the nature of the textual deposit and the function of the caves. The author examines the main theories regarding the caves around Qumran in the light of the presence of tefillin and also comparing them with the finds in other Judean Desert caves. Moreover, a detailed analysis of the artefacts leads Adler to identify typological differences, both among the leather cases and among the tefillin slips. When examining the distribution of the various types of tefillin, he observes a plurality of practices at Qumran, whereas in the “Bar Kokhba caves,” the situation appears uniform. This may be interpreted as a diachronic development of the halakhic practice or as a testimony of contemporaneous differing practices used by different groups during the Second Temple period. Various types of Tefillin may even have been used within the same community of Qumran, where a pluralistic approach towards texts is also been detected. The fourth part focuses on the chronology and function of the caves, their relationship with the settlement and the regional context. The paper of Mladen Popović (Qumran Institute, University of Groningen), When and Why Were Caves Near Qumran and in the Judaean Desert Used?, deals with the chronology of the caves, the purpose for their use, and the connections and networks through which people ended up in the caves. Popović presents some evidence relating to the deposit of manuscripts in other caves in the Judean Desert and the Dead Sea region, which he compares with the finds in the caves around Qumran. The function of the caves of Qumran differs from that of other caves in the Judean Desert that bear traces of the presence of refugees. The caves around Qumran could have been used as a hiding-place for the manuscripts. The

Fidanzio

author does not interpret this difference in terms of isolation; on the contrary, studying examples of connections at regional level, he suggests the possibility that the manuscripts could have been sent to Qumran from elsewhere in Judea for safekeeping. This would be due to a combination of factors, through which Popović highlights the special character of the settlement. Jodi Magness (University of North Carolina at Chapel Hill), The Connection between the Site of Qumran and the Scroll Cave in Light of the Ceramic Evidence, updates the discussion on the pottery in the light of recent excavations and studies. Yitzhak Magen and Yuval Peleg have identified Qumran as a pottery production centre. In contrast, Magness demonstrates the inconsistency of this proposal and explains that local production must have been limited and related to the needs of its inhabitants. As regards the morphology of the cylindrical jars, this remains a clear indicator of the relationship between the caves and the settlement. The largest part of Magness’ paper is dedicated to the Instrumental Neutron Activation Analysis and Petrographic Analysis that have been applied to the Qumran pottery by different research teams. Analyses have identified different groups of pottery according to the clay and the researchers have indicated the source of the clay of each group. Further interpretation of these observations supports the idea that some of the Qumran pottery, including some of the Scroll Jars, came from elsewhere. A comparison between the analyses demonstrates contradictions, sometimes in the raw data, but especially in the interpretation of the clay sources. Magness explains that there is no need to assume that the pottery was produced anywhere else other than Qumran. The results of the different analyses within each of the groups testify to a connection between the pottery from different scroll caves as well as the caves and the Qumran settlement. The chemical composition of the pottery offers an additional argument by which to establish the link between caves and settlement. Jürgen K. Zangenberg (Universiteit Leiden), The Functions of the Caves and the Settlement of Qumran: Reflections on a New Chapter of Qumran Research, believes that putting the caves at the forefront of research—where investigations start from the caves and not from the settlement—allows new avenues to be explored. Zangenberg’s intention is to reflect on the proper criteria by which to approach the subject. He reviews the history of the explorations of the caves, describing the methods used and the results obtained. He also describes the technical limits facing this research and consequently advises caution when approaching interpretative hypotheses. The author analyses the regional context searching for connections

Introduction

rather than higlithing the (supposed) isolation and uniqueness of Qumran. He refers the contributions that illustrate the accessibility of the site and the analyses that consider the Qumran pottery as non-exclusive. According to Zangenberg, the site and the caves, with their facilities, attracted different groups, in addition to the inhabitants of the settlement. He sees the deposition of the manuscripts within this framework, in the context of the First Jewish Revolt. The fifth part presents a series of short papers. Sidnie White Crawford (University of Nebraska-Lincoln), The Inscriptional Evidence from Qumran and its Relationship to the Cave 4 Documents, examines the inscriptions found in the settlement and the caves. This includes the languages of the inscriptions, the find sites, the dates of the inscription, the results of the INAA on the clays of the ostraca and the genres of the inscriptions. She then investigates the connections between the inscriptions and the scroll collection from cave 4Q: the presence of scribal exercises, the documentary texts and accounts, and the onomasticon. According to the author the fact that small worthless exercises were found in cave 4Q indicate that they are part of the material from the settlement that was thrust helter-skelter into the cave in anticipation of the Roman attack. In general White Crawford makes the claim that there are modest written connections between the settlement and the caves. The coins found in Qumran (both the hoards found in Loc. 120, and the single finds) and the documentation on them (in particular the archives of Henri Seyrig and Augustus Spijkerman), which were believed to be partly lost or were unknown, are now available again for study. Bruno Callegher (Università degli Studi di Trieste), The coins of Khirbet Qumran from the digs of Roland de Vaux: returning to Henri Seyrig and Augustus Spijkerman, traces the history of this particular area of research, he emphasize the valuable contributions of its early protagonists and clarifies the shortcomings and obscure chapters of subsequent research. In the final part he makes a numismatic comparison between Khirbet Qumran and the caves. While the former shows signs of a thoroughly monetized economy at all levels, with an extensive and prolonged use of both silver and divisional currency, no coins have actually been found in the caves. Consequently, the author believes that the caves did not have a residential, and therefore functional use, unlike the khirbeh. Gregory L. Doudna, Dating the Scroll Deposits of the Qumran Caves: a Question of Evidence, presents a series of arguments against the traditional dating of the scroll deposits to the time of the First Revolt. The author has long argued the necessity to antedate the entire deposit of

5 the scrolls to the end of the first century BCE. In his view, the assumption of the First Revolt as the endpoint for the deposits, is neither plausible nor supported by evidence. He analyses the historical allusions contained in the manuscripts, the reasons for the dating of the deposit to the First Revolt, the dating of the pottery, the premises for the exact dating of the palaeographic script charts, the pluriformity of the biblical texts. He concludes that the scrolls in the caves of Qumran in their entirety sould occupy their rightful place as the remains of a lost textual world that existed prior the first century CE. Mariusz Burdajewicz (Institute of the Mediterranean and Oriental Cultures, Polish Academy of Science) A History of the “Qumran Caves” in the Iron Age in the Light of the Pottery Evidence, maps the the finds of the Iron Age II–III pottery in the caves around Qumran and in the other caves in the cliffs running along the littoral of the Dead Sea. He introduces the different pottery types, giving their place of origin, description, dating and parallels found at other sites. Then, on the basis of the pottery evidence, he presents some observations concerning the use of the caves during the Iron Age II–III. As for their function, he suggests that most of the caves probably served as temporary shelter, used in the event of a threat or as hiding-places for all manner of “outlaws.” As regards the artificial caves in the marl terrace, he considers it plausible that these caves were hewn out and used as dwelling places by the first settlers in the Iron Age II. Lastly, taking into consideration the Iron Age pattern of cave-occupation in the cliffs along the littoral of the Dead Sea, he observes that a marked cluster of caves exists in close proximity to Khirbet Qumran, which is not the case in the vicinity of the other large settlement at En Gedi. The volume ends with the inventory of the materials excavated from the caves, prepared by the team led by Roland de Vaux. This is being published for the first time and is preceded by a comprehensive introduction prepared by Marcello Fidanzio (ISCAB, Facoltà di teologia di Lugano) and Jean-Baptiste Humbert (EBAF), Finds from the Qumran Caves: Roland de Vaux’s Inventory of the Excavations (1949–1956).1 Taken all together, one can notice the convergence of different contributions on some specific aspects, thereby inviting future studies to test these observations and to 1  Two other papers were presented at the conference but are not published in this volume. Orit Shamir and Naama Sukenik, The Differences in the Textiles from the Qumran Caves Compared to Those Found in the Other Judea Caves; Stephen J. Pfann, The Ancient Library or Libraries of Qumran? The source and character of the manuscript caches in the Judean Wilderness.

6 expand upon them. Numerous caves in the vicinity of Qumran were involved, during the late Second Temple period, in a specific phenomenon that has no parallels in the regional context. Analysing the contents of the individual caves, one notices that many elements point to a unitary phenomenon. Nonetheless differences are attested between various caves, primarily between natural and artificial caves but also between individual caves. The variety increases when all the caves’ finds, from the different periods, are considered. The traditional view that sees a connection between the caves and the settlement is confirmed on various levels. In turn, some papers highlight the connectivity of the caves and the khirbeh with the regional context. The organization of this international conference and the publication of its proceedings have been possible thanks to the generous help of various individuals and institutions. It gives me great pleasure to thank the Istituto di cultura e archeologia delle terre bibliche of the Facoltà di teologia di Lugano and in particular its president, Giorgio Paximadi, who supported the project, gave me valuable advice and allowed me to spend the necessary time in Jerusalem to work on it. The É cole biblique et

Fidanzio

archéologique française de Jérusalem offered the collaboration of the director Marcel Sigrist, of Emile Puech, and a close cooperation with Jean-Baptiste Humbert. The Department of education, culture and sport of the Canton Ticino, the Facoltà di teologia di Lugano, and the Fondazione Girolamo di Stridone per gli studi biblici and gave an essential financial contribution, together with those of the Swiss associations of the Equestrian Order of the Holy Sepulchre and the Sovereign Military Order of Malta. Mireille Bélis helped me shape the content and identify the contributors to the conference. Florentino García Martínez and Dennis Mizzi were always available and gave me valuable suggestions and advice. Sara Della Ricca assisted me with great commitment and dedication in both the preparation of the program and in the organization of the conference. The students of the Facoltà di teologia di Lugano, together with some grant-holders offered their help in the practical running of the event under the guidance of Laura Re. In the editing of the proceedings I was helped by Federica Argiolas, Giulia Viani, Marco De Pietri, and Vladimir Olivero. I am grateful to George Brooke for accepting this volume for publication in this prestigious Series.

Part 1 Topography

∵

CHAPTER 1

The Qumran Caves in their Regional Context: A Chronological Review with a Focus on Bar Kokhba Assemblages Joan E. Taylor The caves in which the manuscripts of the Dead Sea Scrolls were discovered are among the most famous in the world, but they exist in a context of many caves formed naturally in the rocky hills west of the Dead Sea, along with some artificial ones. The discovery and exploration of these caves, and the excavation of their contents, has led to other interesting finds, and has also enabled a greater awareness of the chronology of the occupations. When the Dead Sea Scrolls were first discovered, the caves of the region as a whole were not known. In terms of understanding the Qumran manuscripts and their situation in the caves, as archaeological objects, it pays then to think again in terms of the history of the excavations of the caves in this area of the Dead Sea and review the evidence as it now stands. In this discussion, I will briefly chronicle the Qumran caves as they were brought to light, and situate their discovery and further exploration in the context of cave surveys and excavations in the region. I will reflect on what a broader knowledge of the caves of the region of the northwestern Dead Sea might add to our understanding of the caves around Qumran. Within this volume, the study overlaps in part with that of Mladen Popović and Dennis Mizzi in approaching the Qumran caves within their regional context, with the emphasis here being placed on the chronology of the excavations (which accounts for certain theories along the way) and also on the character of cave assemblages or repertoires as they can now be viewed. In this discussion I am not concerned with texts alone: a deeply informed survey of the discovery of the Qumran texts in the light of other texts found in the Judaean desert was done for Charlotte Hempel’s The Dead Sea Scrolls: Texts and Context volume by the late Hanan Eshel, and recently too a comparative review of the manuscript collections of the Judaean desert has been presented by Mladen Popović.1 I am interested more expansively in 1  Hanan Eshel, “Gleaning of Scrolls from the Judean Desert,” in The Dead Sea Scrolls: Texts and Contexts (STDJ 90; ed. C. Hempel; Leiden: Brill, 2010), 49–87; Mladen Popović, “Qumran as Scroll Storehouse in Times of Crisis? A Comparative Perspective on Judaean Desert Manuscript Collections,” JSJ 43 (2012): 551–94.

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_003

different dates of occupation and different repertoires of objects. At the conclusion of this study, I will also reflect a little on how all this may impact on the understanding of the marl terrace caves 4Qa and 4Qb. 1

A Review of Explorations and Excavations

De Vaux and Harding: The Qumran Caves and Murabbaʿat, 1946–1958 The fascination with the caves of the north-western Dead Sea of course began with the discovery of the cave now known as 1Q. In the winter of 1946 or spring of 1947, ancient Hebrew scrolls came to light in this cave 1 km north of the site of Kh. Qumran, when Bedouin herders Mohammed ed-Dhib and his associates found jars with ancient scrolls tightly wrapped in linen. As the world learnt of this sensational discovery, the Bedouin of the Ta‘amireh tribe continued to find more manuscript fragments in the cave, and released them into the antiquities market.2 In January 1949 the Arab legion, under Captain Akkash el Zebn, undertook a careful search and discovered its location, whereupon Gerald Lankester Harding, chief inspector of the Department of Antiquities of Jordan, invited Father Roland de Vaux of the École Biblique et Archéologique Français in Jerusalem to excavate this cave with him, along with the Palestine Archaeological Museum. Excavators Harding and de Vaux noted that the cave had been very thoroughly gone over by Bedouin and others searching for anything of value. Nevertheless, they spent 15 days excavating, and found that in the half a metre of earth built up in the cave there were 600 fragments identified later as coming from some 70 different 1.1

2  However, it should be noted here that in the 1990s Weston Fields was told by a Bedou, Abu Daoud, that there was another original cave where manuscripts were found, possibly the back entrance to the “Cave of the Column” later excavated by Vendyl Jones, see Weston W. Fields, The Dead Sea Scrolls: A Full History. Vol. 1: 1947– 1960 (Leiden: Brill, 2009), 110–13. However, no manuscript fragments or broken cylindrical jars were later found in this other cave, and the testimony to it is late, singular and suspect.

10 Hebrew m ­ anuscripts, along with a great deal of broken pottery, mainly coming from cylindrical jars and lids, though there were also bowls, a single small cooking pot, a single juglet and remains of four lamps. Along with the pottery there very many pieces of linen scroll wrappers and packing. They even found one decomposed scroll still inside its linen wrapper, stuck to the broken neck of a jar.3 They initially identified the pottery—jars and bowls—as coming from the end of the 2nd century BCE to the beginning of the 1st century CE, though curiously there were potsherds they identified as being dated to the 2nd and 3rd centuries CE,4 but they revised this in the official publication to a date range up to 70 CE.5 This was called the “Ain Feshkha cave,” since it was defined by the nearest well-known locality on the northwestern Dead Sea: an oasis irrigated by natural cold springs, comprising large pools, a great variety of plants and trees, and a rich bird and freshwater fish life. The cave was not connected to the tumble of ruins at Khirbet Qumran, because up until that time the general view was that these ruins represented a Roman fort, dated to the 3rd or 4th century. The spring and pool of Ain Feshkha seemed a more appropriate landmark, since Christian travellers used to stop there en route between the Jordan River and Mar Saba monastery, further inland on the edge of the Kidron ravine, because of the potable water. Such a water resource was important to the Bedouin who held this territory, and the ruins of Qumran themselves could be called Khirbet (“ruin of”) Feshkha by Bedouin.6 Because of this old assumption that Qumran was a Roman fort, de Vaux and Harding initially rejected the site as relevant to the scrolls cave,7 though they did a surface inspection and dug up two tombs.8 In 1951 they went back, 3  See the photograph in Dominique Barthélemy and Józef T. Milik, eds., Qumran Cave 1 (DJD 1; Oxford: Clarendon Press, 1955), Pl. I: 8–10, and see p. 12. This item is preserved in the IAA collection and classified as 990e. I am grateful to Pnina Shor for allowing me to see it during a research visit in July 2014. 4  Gerald L. Harding, “The Dead Sea Scrolls,” PEQ 81 (1949): 112–16; Raymond J. Tournay, “Les anciens manuscrits hébreux récemment découverts,” RB 56 (1949): 204–33; Roland de Vaux, “Post-Scriptum: La cachette des manuscrits hébreux,” RB 56 (1949): 234–37; idem, “La grotte des manuscrits hébreux,” RB 56 (1949): 586–609. 5  Roland de Vaux, “La poterie,” in DJD 1:11. 6  See Joan E. Taylor, “Khirbet Qumran in the Nineteenth Century and the Name of the Site,” PEQ 134 (2002): 144–64 (148, 156). 7  In 1949, de Vaux wrote of Qumran: “il suffira de dire ici qu’aucun indice archéologique ne met cette installation humaine en relation avec la grotte où furent cachés les manuscrits.” De Vaux, “La grotte des manuscrits hébreux,” 586, n. 2. 8  Roland de Vaux, “Fouille au Khirbet Qumrân: Rapport Préliminaire,” RB 60 (1953): 83–106 (89); idem, Archaeology and the Dead Sea

Taylor

digging quickly, from 24 November to 12th December, with 15 workers, for three weeks, working out a sequence of use from the late 2nd century to the later 1st century, on an Iron Age foundation. Most importantly, they found embedded in the floor of locus 2 a cylindrical jar identical to a type of jar in the cave, and a coin dated to 10 CE (KhQ 17).9 Other pottery coming to light was of the same type as that found in the cave.10 Therefore, the association between the cave and the site of Qumran was established chronologically and typologically. Since de Vaux already believed that the cache of manuscripts belonged to the Essenes—largely on the basis of the “Manual of Discipline” (1QS) and the writings of Pliny, Josephus and Philo on the Essenes—Qumran too was understood to have been an Essene site.11 After Bedouin had made an initial discovery in October 1951, de Vaux and Harding progressed from Qumran to work in the Wadi Murabbaʿat (Na�al Darga), 18 km south of Qumran, in January 1952 (see Figure 1.1).12 Here they found four caves with well-preserved remains from the Chalcolithic, Middle Bronze and Iron Age II periods, as well as from the Roman period. Later a fifth cave was found. Three of the caves, 1Mur, 2Mur and 5Mur, yielded manuscripts. In the first cave, 1Mur, there were actually just two texts: Mur 2 (parts of the book of Deuteronomy),13 and Mur 78 (a very short double-lettered Abecedary).14 In the last cave, 5Mur, there was only Mur 88 (fragments of a scroll of the twelve Prophets),15 so in fact it is only one cave, 2Mur, that has yielded the 170 remaining Greek, Aramaic and Hebrew texts, including ostraca, phylactery texts, legal and administrative documents and letters from Bar Kokhba, leader of the Second Revolt.16 In addition,  Scrolls (The Schweich Lectures 1959; rev. ed.; trans. David Bourke; London: Oxford University Press for the British Academy, 1973), vii. 9  Jean-Baptiste Humbert and Alain Chambon, eds., Fouilles de Khirbet Qumrân et de Aïn Feshkha. Vol. 1: Album de photographies; Répertoire du fonds photographiques; Synthèse des notes de chantier du Père Roland de Vaux, (NTOA.SA 1; Fribourg/ Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 1994), 292. 10  De Vaux, “Fouille au Khirbet Qumrân,” 83–106. 11  De Vaux, Archaeology, 126–38. 12  Gerald L. Harding, “Khirbet Qumran and Wady Murabbaʿat: Fresh Light on the Dead Sea Scrolls and the New Manuscript Discoveries in Jordan,” PEQ 84 (1952): 104–9. 13  Józef T. Milik, “Textes hébreux et araméens,” in Les grottes de Murabbaʿât (DJD 2; ed. P. Benoit, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1961), 78–79; Pl. XXI. 14  Milik, in DJD 2:178; Pl. LIV. 15  Milik, in DJD 2:181–205; Pls. LVI–LXXIII. 16  Joseph A. Fitzmyer, A Guide to the Dead Sea Scrolls and Related Literature (Grand Rapids: Eerdmans, 2008), 136–39.

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

Figure 1.1 The artificial caves cut in the marl terrace on the edge of the Wadi Qumran, within the occupation zone of the Qumran settlement. See also Figure 1.6. © Joan E. Taylor.

there is a palimpsest of a papyrus letter of the 8th century BCE rewritten with a list of names in the 2nd century CE, both of which are in paleo-Hebrew (Mur 17).17 Importantly, there was a distinctively different repertoire of artefacts in this cave in comparison to what had been excavated in the “Ain Feshkha Cave.” There were no jars, lids or scroll wrappers, and the assemblage was rich and diverse. There were numerous ropes and cables, a stone cup, “Herodian” lamps, storage jars, cups and bowls, arrow heads, rings from armour, wooden plates, nails, a hook, knives, medical items, a die, wooden spoons, combs and earrings, woollen textiles, basketry, leather (including sandals with hobnails, but also two children’s sandals), a hairnet, hair needles, spindle whorls, cosmetic equipment, military equipment, a key, and tools. All of this indicated that people, male and female, had come here for refuge, with their domestic and precious objects. The textiles are particularly beautiful, including a piece of blue linen with a tapestry band (no. 110) and another piece of linen with silk (no. 113). In terms of the date, 20 coins dating from procurators under Nero, one from Ascalon (84–5 CE), one from Tiberias struck under Hadrian 119–20 CE, and nine

17  Fitzmyer, A Guide to the Dead Sea Scrolls, 137.

11 from the Second Revolt (132–5), all relate to the time of this Revolt, while two 2nd-century bronze coins with the stamp “Legio X Fretensis” indicate a period afterwards.18 Indeed, the repertoire of the cave cannot be associated entirely with Bar Kokhba rebels and refugees, as there is clear evidence of further use of the cave that continued on through to about 190 CE, illustrated by datable documents. A Greek papyrus, Mur 114, is written “under the consulship of Stativius Severus” (c. 171 CE), with the name of Emperor Severus (180–92 CE), and a Latin document, Mur 117, begins: “C. Julius . . .” Making the picture even more complex, there were also fragments of 9th–10th-century Arabic writing with magical texts (Mur 169–73). The post-Second Revolt Roman use of the cave is also evidenced by the employment of fine scroll parchment for accounts and lists of cereals, written in Greek (Mur 90–94, 96–102).19 With good reason, any connection of Murabbaʿat with Qumran was completely ruled out by de Vaux. Bar Kokhba and the Qumran Essenes were totally separated both by ideology and by chronology, and Murabbaʿat was therefore not discussed in de Vaux’s British Academy lectures.20 Back at Qumran, in February 1952, Bedouin reportedly found another cave with manuscripts, 100m south of the original cave, which led to the classification system we now follow: the “Ain Feshkha Cave” became 1Q, and the new one was designated 2Q. A similar repertoire was discovered: manuscript fragments and pieces from over six scroll jars, a lid and three bowls. Following this, around 300 caves around Qumran were searched during March 1952 in a survey jointly undertaken by the American School of Oriental Research in Jerusalem, the École Biblique and the Palestine Archaeological Museum, under the direction of de Vaux, which focused on caves from two kilometres north of cave 1Q to the region between Qumran and Ain Feshkha, as shown in the published map drawn by J. Ziegler (Figure 1.2).21 It seems that this range was picked as one that would yield results that could be linked with the site of Qumran, and indeed cave 3Q was discovered, 18   Józef T. Milik and Henri Seyrig, “Trésor monétaire de Murabbaʿât,” RN 6/1 (1958): 11–26. 19  Mur 8 would fit this pattern also, but it is in very rough Aramaic writing, perhaps indicating a Roman auxiliary soldier who has had basic literacy training within a Jewish or Samaritan context. For this observation of continued Roman occupancy see also Popović, “Qumran as Scroll Storehouse,” 558. 20  De Vaux, Archaeology, ix. 21  William L. Reed, “The Qumran Caves Expedition of March, 1952,” BASOR 135 (1954): 8–13.

12

Figure 1.2 The map drawn by J. Ziegler of the 1952 caves expedition with de Vaux’s numbering system. Note that GQ37, placed above Ain Feshkha, is wrongly located; the actual GQ37 is to be placed about half way between GQ32 and GQ33, see Figure 8.

Taylor

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

which provided fragments of parchment and papyrus, over 30 broken jars, over 20 lids, two jugs and a lamp, as well as the Copper Scroll.22 It should be noted that the “Q” is important in identifying caves, since this is the system of indicating manuscript caves only, not the sequence of caves in general, which were assigned a different numeration: the natural caves 1Q, 2Q, 3Q, and 6Q were given cave numbers 14, 19, 8 and 26, respectively, often thereafter prefixed by the letters “GQ” (“Grottes de Qumran”). However, two caves discovered in 1956, after the completion of this survey, close to cave 3Q, were simply called “Cave A” and “Cave B.” There remains to my knowledge no universal system of numbering all the caves of the western Dead Sea, even in the Qumran area.23 It was determined that 40 of the caves contained ­pottery and other objects, and 26 of these had pottery ­corresponding to the distinctive type and repertoire found in Cave 1Q: jars, lids, bowls, occasional small storage pots and lamps (de Vaux 1973: 51). “Cave A” and “Cave B” were investigated in 1956 and noted as having Qumran types of pottery. Despite the categorical separation of the Bar Kokhba material of Murabbaʿat as having nothing to do with Qumran, the cave evidence that came to light brought Bar Kokhba rebels and refugees quite close to the Qumran ruins. Pottery from the “later” Roman period was identified in caves GQ33 and 34, south of Qumran, and in caves GQ36 and 38,24 though no significant excavation was done in them. Furthermore, the same kind of chronological range of occupation as Murabbaʿat came to light in a number of other caves, with Chalcolithic as well as Early or Middle Bronze material in caves GQ3, 15, 37, and “Cave B,” and Iron Age II remains in caves GQ6, 13, 27, 39, “Cave A” and “Cave B.” In addition, Byzantine material was found in GQ23 and 38, and “Arab” material in GQ4, 6, 23, and 38. In GQ17, just south of 1Q, five wooden poles identified as tent poles were found along with pottery otherwise of the Qumran type.25 Then, famously, the Bedouin managed to discover caves 4Qa and 4Qb beside the ruins, just four months 22  Roland de Vaux, “Exploration de la région de Qumrân: Rapport préliminaire,” RB 60 (1953): 540–61; idem, “Archéologie,” in Les ‘petites grottes’ de Qumrân: Explorations de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 3–36. 23  See Frederick F. Bruce and Stephen J. Pfann, “Qumran,” EncJud 16:771–72. 24  De Vaux, in DJD 3:16. 25  De Vaux, in DJD 3:Pl. VII:3; de Vaux, Archaeology, 57.

13 after the official caves survey was completed.26 While these are actually two separate caves, they were called “a” and “b” because the resulting manuscripts originally deriving from these, found by Bedouin, could not be definitively assigned to one or the other, though in fact all save a few apparently came from 4Qa. Excavations in September 1952 jointly undertaken by the Department of Antiquities of Jordan, the École Biblique, and the Palestine Archaeological Museum uncovered in 4Qa 12 almost complete storage jars, and other pottery including small pots, lids, bowls, carinated bowls, jugs and one lamp.27 Along with this pottery there were manuscript fragments, leather, some textiles, scroll tags and ties, phylactery cases and some wood. A further cave was noticed, 5Q, also excavated in September 1952; it contained some manuscript fragments but no pottery. De Vaux still did not look in other possible caves here, even as he excavated at Qumran. He searched for collapsed caves only three years later, in 1955, after he found the remains of a rock-hewn staircase leading from the plateau to the artificial caves to the south of the buildings, and thus caves 7Q–10Q came to light, though meanwhile the Bedouin discovered cave 6Q, lying close to the Wadi Qumran itself, with manuscript fragments, a jar, and bowl of “Qumran type.”28 Cave 7Q, like 4Qa, had manuscript fragments, as well as parts of two jars, one lid, a large bowl and a lamp, matching types in existence at Qumran. In cave 8Q, which had two openings, there were fragments of manuscripts, phylactery texts, a prayer and mezuzah, with 100 leather strips used for scroll fastening. There were three storage jars, four lids, a handled jar and a lamp with a pared nozzle. In cave 9Q, there was a small fragment of Hebrew or Aramaic writing,29 dates and date-stones, three pieces of cord and string, and also the side of an Iron Age II bowl.30 Cave 10Q had a woven mat, parts of a lamp with a pared nozzle, an ostracon with Hebrew letters, dates and further sherds.31

26   De Vaux had thought there were some water-eroded cavities here that would be archaeologically barren, so de Vaux, Archaeology, 52. 27   Roland de Vaux, “Archéologie,” in Qumrân Grotte 4.II: I. Archéologie; II. Tefillin, Mezuzot et Targums (4Q128–4Q157) (DJD 6; ed. R. de Vaux and J. T. Milik; Oxford: Clarendon Press, 1977), 9–20; cf. idem, Archaeology, 52. 28  De Vaux, in DJD 3:10, 26; idem, Archaeology, 51. See Humbert and Chambon, Fouilles de Khirbet Qumrân, 343–44. 29  Maurice Baillet, “Textes des grottes 2Q, 3Q, 6Q, 7Q, à 10Q,” in DJD 3:163, Pl. xxxv. 30  De Vaux, in DJD 3:31. 31  De Vaux, in DJD 3:124, 164; de Vaux, Archaeology, 53.

14 In 1956 Bedouin found cave 11Q, just south of 3Q, and it was subsequently excavated by de Vaux’s team, who found a similar range of objects as elsewhere, in terms of material dating from the Chalcolithic, Iron Age II and Roman periods.32 The pottery repertoire included material of the Qumran type (broken cylindrical jars, lamps, juglet, lids), but artefacts included linen, cords and basketry, an assemblage slightly wider than is usual in the “Qumran type” caves. There were also iron objects: a knife, a small pick-axe, an arrow-head, two keys and a chisel, along with “leather” items, a copper buckle and a bead,33 the latter perhaps indicating a woman’s presence. De Vaux explored two further caves in front of 3Q, both containing Qumran types of pottery and linen fragments, along with Iron II sherds, and one had Chalcolithic sherds and a flint blade, and another a bit of parchment without writing.34 As for dating, de Vaux himself was not very precise regarding the caves, but he thought in terms of periods he had discerned in the Qumran excavations as Period Ib (late 2nd cent. BCE to 31 BCE) and II (c. 10 BCE–68 CE), with correlating pottery defined as the “Qumran type.” Interestingly, however, de Vaux noted himself that “forms characteristic of both periods have been found in the same cave, and no cave can be positively stated to have been used only during Period Ib. . . . The greater part of the materials which have survived probably belongs to Period II.”35 This chronological range of pottery is something to note; the cave pottery does not belong only to the end of Period II. De Vaux’s close associate J. T. Milik, who worked on the Qumran excavations and the texts, also engaged in some independent work and provided an overview of the cave discoveries in a book published in 1957, with an updated English edition, translated by John Strugnell, published in 1959. In this updated edition, he noted a new discovery: “Another small cave, with a terrace in front of it, was discovered above Ain Feshkha by myself in the spring of 1958. It contained two distinct strata, the lower with Iron II sherds, and the upper with pottery of the Qumran type mixed with ashes, and, in a corner a neat pile of date stones.”36 The terrace was large enough to accommodate an “Essene hut.” This was not noted in de Vaux’s publica32  Roland de Vaux, “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e et 5e campagnes,” RB 63 (1956): 573–74; idem, Archaeology, 51. 33  See Humbert and Chambon, Fouilles de Khirbet Qumrân, 344. 34  De Vaux, “Fouilles de Khirbet Qumrân,” 574. 35  De Vaux, Archaeology, 54. 36  Józef T. Milik, Ten Years of Discovery in the Wilderness of Judaea (SBT 26; London: SCM Press, 1959), 151.

Taylor

tion of the caves survey, but it seems to have been identified as GQ37 in Ziegler’s map (Figure 1.2). However, the actual GQ37 as described by de Vaux37 is further north, as will be explained below. For our purposes we will call it “Milik’s Cave.” Southern Explorations: The Judean Desert Caves Expedition (1952–65) Further south of Wadi Murabbaʿat, in the area assigned to Israel in 1948, there was a parallel exploration of caves beside the Dead Sea. In this investigation, the dominant interest was in the rebels and refugees of the Second Revolt. In 1952 a young officer showed archaeologist Yohanan Aharoni a small fragment of an ancient scroll he had found in caves near En Gedi. At the same time, Aharoni learnt that Bedouin were searching in this area, conscious of the lucrative sales of Dead Sea Scroll manuscripts, and curators of the Palestine Archaeological Museum (now the Rockefeller Museum) purchased material from Bedouin supposedly found in Wadi Seiyal, located in Israel, though they are now thought to have come from Na�al Ḥever (= Wadi Habra) (see Figure 1.3). In 1953 Aharoni organised a small expedition that began investigations in the area between En Gedi and Na�al Ḥever, and found eight caves, including the Cave of Letters, and the remains of Roman siege camps 100m above caves 5/6 and 8. Later 20 cave openings were noted in the Na�al Arugot, Na�al En Gedi and Na�al David, including the “Pool Cave,” which had fragments of pottery vessels, palm mats and food.38 In March 1960 the expedition for the study of the Judaean Desert Caves, organised by Joseph Aviram of the Israel Exploration Society (in association with the Hebrew University, the Israel Department of Antiquities and assisted by the Israel Defence Forces), began work here. Nahman Avigad surveyed the Na�al David (= Wadi Sdeir), investigating tombs and the Pool Cave, used by Bar Kokhba refugees, though documents (a fragment of a scroll of Genesis, a promissory note and unidentified Greek texts) had been found already by Bedouin searchers (see DJD 38; Sdeir 1, 2, 3, 4). Yigael Yadin worked in Na�al Ḥever in the “Cave of Letters” (1960–1961), originally named Ḥever 5/6 because it has two openings. Here there were also the skeletal remains of eight women, six children and three men 1.2

37  De Vaux, in DJD 3:12. 38  Shimon Dar, “The Search for Scrolls in the Judaean Desert Caves in the Years 1950–1960—An Archaeological Memoir,” BAIAS 28 (2010): 141–57.

15

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

Figure 1.3 The key wadis and sites of the western Dead Sea, using a 1972 NASA image as a base.

that had been wrapped in cloth and placed in baskets covered over with mats. Yadin found Greek, Hebrew and Aramaic papyri, some which are letters from Bar Kokhba to Yehohanan, son of Be‛aya of En Gedi.39 In a leather bag

were 35 letters wrapped in cloth, belonging to Babatha, daughter of Shimon, dating from 93 to 132 CE, and another collection of Eleazar son of Shemuel totalled six ­documents.40 In addition, there were marriage and other

39  Yigael Yadin, The Finds from the Bar Kokhba Period in the Cave of Letters (Jerusalem: Israel Exploration Society, 1963), 38–40.

40  Yigael Yadin, Bar Kokhba: The Rediscovery of the Legendary Hero of the Last Revolt against Imperial Rome (Jerusalem: Israel Exploration Society, 1971), 222–53.

16 contracts (published in DJD 38 and the JDS series).41 The cave repertoire was rich, like that of Murabbaʿat, with an especially important collection of women’s artefacts: mirrors, cosmetics, bright textiles, hairnets, wool and unspun wool for spinning, spindle whorls, and perfume vials. There were also metal vessels including incense shovels, glass plates, palm mats, sandals, basketry, keys, tools and other personal artefacts. Yohanan Aharoni worked in the Na�al Ḥever in the “Cave of Horror” (cave 8), located on the southern side of the Na�al Ḥever, where the skeletal remains of 40 men, women and children were found, and also in the Wadi Seiyal (=Na�al Ze’elim), finding fragments of a Greek scroll of Twelve Minor Prophets (8Hev1; published in DJD 8), fragments of a Hebrew prayer scroll, a Greek papyrus fragment, contracts and ostraca (see DJD 27, 38 and 39).42 Meanwhile, Bar Adon, in Na�al Mishmar, found a hoard of 429 Chalcolithic vessels—and thus this cave was called the “Cave of Treasure”43—along with some Bar Kokhba fragmentary texts and two ostraca (1Mish 1–8). Surveys of caves in the region of En Gedi have generally yielded material dated to the time of the Second Revolt, sometimes confirmed with coinage. The documents indicate thriving Jewish communities in the southern part of the Dead Sea and elsewhere in Judaea, providing important evidence of continuing Jewish life after the First Revolt. There was then the curious phenomenon of material in caves north of Murabbaʿat especially around Qumran, being largely considered to come from a time prior to the First Revolt, and most of the material from Murabbaʿat south being associated with the Second Revolt. Unlike in the southern caves, however, around Qumran there was no corresponding documentary material or coins to date the caves, and pottery was the chief dating marker. Everything then hinged—and still hinges—on how accurately pottery typology could be dated in terms of age ranges. 1.3 Lapp: The Wadi ed-Daliyeh (1962–64) In 1962, Bedouin brought to light papyri from the Wadi edDaliyeh, north of Wadi Qelt and Jericho, and the cave in which these were found was excavated in 1963–64 by Paul 41  See Hanan Eshel and David Amit, eds., Refuge Caves of the Bar Kokhba Revolt (Tel Aviv: Israel Exploration Society, 1998), Vol. 1 [Hebrew]; Hanan Eshel, “Excavations in the Judean Desert and at Qumran under Israeli Jurisdiction,” in The Dead Sea Scrolls in Scholarly Perspective (STDJ 99; ed. D. Dimant; Leiden: Brill, 2012), 381–99 for further extensive bibliography. 42  Yohanan Aharoni, “Expedition B—The Cave of Horror,” IEJ 12 (1962): 186–99. 43  Pesach Bar Adon, “Expedition C—The Cave of the Treasure,” IEJ 12 (1962): 215–26.

Taylor

Lapp, for the American Schools of Oriental Research. Three hundred skeletons from the 4th century BCE were found, with fragments of some 38 Aramaic legal documents of this time (WDAP 1–38), along with clay ­bullae.44 However, cave 2 in the Wadi ed-Daliyeh provided a rich repertoire of the same kind of objects found in Murabbaʿat and in the wadis around En Gedi, dated to the time of Bar Kokhba. As with Murabbaʿat there was a wide range of storage jars, jugs and juglets, pots, lamps and personal items, including sandals, basketry, keys, a wooden comb and coins, on top of a previous Middle Bronze occupation. This brought the Bar Kokhba evidence to the north of Qumran though some 30 km away. 1.4 Allegro: The “Christmas Cave” (1962–63) In 1960, John Allegro led a team looking for the locations of the treasure of the Copper Scroll, and excavated in 2 caves above Ain Feshkha. In 1961 he found a cave with remains in the Wadi en-Nar (old Israel grid 189887/121095). It was called the “Christmas Cave,” because Allegro discovered it on Christmas Day, after being alerted to the existence of the cave by a Bedouin man showing him some ancient kohl in a wooden bottle, a wooden toilet box inlaid with ivory pieces—one of them in the shape of a bird—and other wooden items including a spoon. There were also “whorls of wool, stone and bone, a spindle, a large tooth sharpened at one end and still polished from picking up the threads of wool and linen on the loom.”45 The assemblage corresponds to what we would expect of a Bar Kokhba cave. With co-excavator Howard Stutchbury Allegro then found Bar Kokhba coin here, a Roman lamp, bone needle, adze shaft, potsherds, leather and numerous textiles46 which were removed and remained in boxes labelled “QCC” in the Rockefeller archives, a case we will return to presently.47 Allegro also explored some caves near 11Q, which he called “Ibrahim’s Caves” after the illegal excava44  Paul W. Lapp, “Wâdī ed-Dâliyeh,” RB 72 (1965): 405–9. Paul W. Lapp and Nancy L. Lapp, Discoveries in the Wâdī ed-Dâliyeh (AASOR 41; Cambridge, MA: American Schools of Oriental Research, 1974); Mary J. W. Leith, Wadi Daliyeh I: The Wadi Daliyeh Seal Impressions (DJD 24; Oxford: Clarendon Press, 1997); Douglas M. Gropp, Wadi Daliyeh II: The Samaria Papyri from Wadi Daliyeh; Moshe Bernstein et al., eds., Miscellanea, Part 2 (DJD 28; Oxford: Clarendon Press, 2001). 45  John M. Allegro, Search in the Desert (London: W. H. Allen, 1964), 7. 46  Judith Anne Brown, John Marco Allegro: Maverick of the Dead Sea Scrolls (Grand Rapids: Eerdmans, 2005), 118–45. 47  Roi Porat, Hanan Eshel, and Amos Frumkin, “The Christmas Cave in the Lower Part of Na�al Kidron,” in Refuge Caves of the Bar Kokhba Revolt (ed. H. Eshel and R. Porat; Jerusalem: Israel Exploration Society, 2009), 2:31–52 [Hebrew].

17

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

tor. These contained Chalcolithic and Roman remains, and here Allegro was presented with a wooden bowl.48 The “Christmas Cave” discovery brought rich (though now dispersed) Bar Kokhba material to a site much closer to Qumran than Murabbaʿat, and provided a link to the Bar Kokhba material found in caves between Qumran and Ain Feshkha. 1.5 Bar Adon: Surveys and Excavations (1968–82) With the change in the circumstances of the region after the Six Day War of 1967, the area of the north-western Dead Sea opened up to Israeli researchers. In 1968, Pesach Bar Adon undertook survey work in the north-western Dead Sea. This was mainly concerned with the ruins of the Dead Sea coast.49 Bar Adon noted a cave west of Kh. Mazin with some Roman pottery sherds,50 and another 2 km south of Kh. Mazin (old grid 19031185) with Early Bronze, Hellenistic, Roman and Byzantine pottery and a coin from Year 2 of the First Revolt, hence the “Coin Cave,”

Me‘arat Hamatbe‘a.51 Caves west of Ain et-Turaba had sherds dated to the Chalcolithic, Early Bronze and Iron II,52 with Iron II remains also in a cave above Ain el-Ghuweir. Bar Adon also identified in the Wadi Murabbaʿat some structures along the stream and a plastered water cistern with a channel in the hillside, and Hellenistic, Roman and Byzantine sherds along the stream bed.53 All this added to the impression of key periods of occupation. Bar Adon then progressed to a site near cave 11Q consisting of two caves side by side, separated what might appear to be a rock pillar, which he identified as the “Cave of the Column” of the Copper Scroll (3Q6:1) (Figure 1.4). He named it “Twin Cave,” Me‘arat Hate’omot (old grid 19371293), and noted a pattern of Chalcolithic, Early Bronze, Hellenistic, Roman, Byzantine and Early Arab remains.54 Later, he dug the cave in three excavations in 1971, 1977 and 1982,55 discovering some further pottery and glass from the Hellenistic and Roman periods, with his excavations being continued by Vendyl Jones (see below).

Figure 1.4 The Cave of the Column, at the front of the rocky cliffs, with Cave 11Q to the north. © Joan E. Taylor.

48  Allegro, Search in the Desert, 162. 49  Pesach Bar Adon, “The Judaean Desert and Plain of Jericho,” in Judaea, Samaria and the Golan: Archaeological Survey 1967– 1968 (ed. M. Kochavi; Jerusalem: Carta, 1972) [Hebrew]; idem, “Excavations in the Judaean Desert,” ʿAtiqot 9 (1989): 3–14 [Hebrew]. 50  Bar Adon, “Excavations,” 28–29.

51  Bar Adon, “Excavations,” 30–32. 52  Bar Adon, “Excavations,” 41–49. 53  Bar Adon, “Excavations,” 88. 54  Bar Adon, “Excavations,” 15–17. 55  Eshel, “Excavations in the Judean Desert,” 388.

18 1.6 The Israel Cave Research Centre (1979–93) The Israel Cave Research Center (ICRC) was founded by the Society for the Protection of Nature in Israel in 1979, headed by Amos Frumkin, and five new caves were discovered between 1979 and 1993. The most interesting cave was located opposite the Qarantal Monastery west of Jericho, south of Wadi el-Mafjar, in an area now named Ketef Jericho. This main cave was called the “Abior Cave,” which was excavated in 1985, 1986 and 1993. In 1986 Hanan Eshel found the remains of 38 refugees of the Bar Kokhba Revolt, as well as six fragments of papyrus scrolls, one in Aramaic from the 4th century BCE, mentioning one Abior, hence the name of the cave.56 There were also two documents in Aramaic and three in Greek dated to the 1st–2nd centuries,57 along with pottery, textiles, ropes, sandals and other remains dating to the Bar Kokhba Revolt. In 1993, under the Israel Antiquities Authority, Hanan Eshel and Boaz Zissu found—in an excavation of material thrown outside the cave in later disturbance58—a total of 15 papyrus documents, ten in Aramaic and five in Greek, along with a Roman sandal at a cave they then called “Cave of the Sandal” (with a humorous nod to Monty Python’s Life of Brian), a cave which yielded a further two fragments of a Greek document and 25 Roman coins, cosmetic objects and jewellery. The texts are published in DJD 38. The ICRC explored a cave named Araq el-Na‛asaneh in the Wadi ed-Daliyeh in 1980. Further excavation brought to light 16 silver denarii of Trajan and Hadrian. Continuing excavations under the auspices of the ICRC uncovered caves in the upper part of the Wadi Darajeh, Wadi Makkukh and el-Matsia. In the Wadi Ḥever, the “Cave of the Tetradrachm” revealed Bar Kokhba remains,

56  Hanan Eshel and Haggai Misgav, “A Document from the Fourth Century B.C.E. from a Cave in Ketef Jericho,” Tarbiz 56 (1987): 461–77 [Hebrew]; Michael Maggen, “The Conservation of Papyri Documents dated to the 4th Century BCE,” Restaurator 18/3 (1997): 153–61. 57  Hanan Eshel and Esther Eshel, “Fragments of Two Aramaic Documents which were Brought to the Abi‘or Cave during the Bar Kokhba Revolt,” in EI 23: Archaeological, Historical and Geographical Studies; Avraham Biran Volume (ed. E. Stern and T. Levi; Jerusalem: Israel Exploration Society/Hebrew Union College, 1992), 276–85 [Hebrew]. 58  Eshel and Zissu have noted that disturbance of cave material occurred in the Abior cave, with activity in the Mamluk period, 1948 and 1967, which resulted in an inverted stratigraphy: a list of loans with names in Aramaic, dating to the later 4th century (Jer 1), was lying in a pile of soil on top of Roman cooking pot sherds, and a similarly-dated document (Jer 6 1–4) was found on top of Roman period texts underneath the cave entrance.

Taylor

as well as material from the Chalcolithic, Iron Age II, and Roman periods.59 This work north of Qumran indicated clearly that Bar Kokhba rebels and refugees—fleeing from Jericho and surrounding settlements—occupied the caves west of Jericho in the same way as they occupied caves from the Christmas Cave to En Gedi. Their remains were typified by a distinctive assemblage of artefacts, often including personal items. Still, however, the immediate region of Qumran itself seemed less rich in Bar Kokhba material. 1.7 Patrich: Surveys and Excavations (1984–91) Overlapping with the work of the ICRC, Yigael Yadin and Joseph Patrich undertook surveys of caves north of Qumran and two south of the site in the years 1984–5, work that was continued by Patrich through to the year 1991, on behalf of the Institute of Archaeology of the Hebrew University of Jerusalem and the Zinman Institute of Archaeology of the University of Haifa. From 1986 to 1988 the search and excavation proceeded in four caves in the region of Qumran on behalf of the Institute of Archaeology of the Hebrew University of Jerusalem and Institute for Judaic-Christian Research (Arlington, Texas), directed by Joseph Patrich, in association with Vendyl Jones, who excavated at Bar Adon’s Cave of the Column, near cave 11Q, independently to 1992. In 1984–5 the work focused over a range 4 km north of Qumran to 4 km south. New numbers were assigned to the caves that were identified.60 Out of 57 caves marked on aerial photographs north of Wadi Qumran, 15 yielded archaeological remains (as opposed to 27 caves in the 1952 survey), but only two caves with finds were determined south of Qumran (as opposed to 13 in 1952), probably because potsherds had been already collected by the earlier surveyors. An unknown cave, identified as cave 24 in the new survey,61 just 50m to the north of 11Q, yielded 59  David Amit and Hanan Eshel, “The Bar Kokhba Period Finds from the Tetradrachm Cave,” in Eshel and Amit, Refuge Caves of the Bar Kokhba Revolt, 189–204 [Hebrew]. 60  Joseph Patrich, “Caves of Refuge and Jewish Inscriptions on the Cliffs of Na�al Michmas,” EI 18 (1985): 153–66 [Hebrew]; idem, “Hideouts in the Judean Wilderness: Jewish Revolutionaries and Christian Ascetics Sought Shelter and Protection in Cliffside Caves,” BAR 15/5 (1989): 32–42; idem, “Khirbet Qumran in Light of New Archaeological Explorations in the Qumran Caves,” in Methods of Investigation of the Dead Sea Scrolls and the Khirbet Qumran Site: Present Realities and Future Prospects (ANYAS 722; ed. M. O. Wise et al.; New York: New York Academy of Sciences, 1994), 73–95. 61  In Fig. 1.8 this cave and others found by Patrich are labelled “PQ” in line with Stephen Pfann’s classification, as shown

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

a rich collection of pottery, including a bag-shaped jar with the Hebrew letter alef drawn in black charcoal on its shoulder and a lamp in which a pagan motif had been mutilated, both of which indicate that the occupants were Jewish. Patrich identified the pottery as coming from the later 1st century onwards.62 There were at least 10 jars, six cooking pots and five juglets and remains of fireplaces, which suggest people took food, cooked and ate it. A bronze ear or nose ring indicates that among the occupants there was at least one woman, and a bronze needle and some cloth would also be suggestive of this. A layer of goat dung radiocarbon-dated to the Roman period indicates that certain occupants had herds here at this time. Patrich identified the occupants as refugees, since there is no permanent oven installation and no effort was made to narrow or block the opening. Still, perhaps a goat dung layer might also suggest that actually shepherds sheltered flocks here before refugees used it. In a location 2.8 km north of Qumran and 150m west of a track leading from Kibbutz Qalia and Kibbutz Almog, there was a cave probably identified as no. 2 in the 1952 survey, tentatively identified as cave 13 in the new survey, lying north of cave 3Q.63 This cave contained Chalcolithic, Iron Age II and Roman sherds, and one pot—“cooking pot no. 9”—was identified as being “more characteristic of the end of the third quarter of the 1st century CE. . . . This type continued to be used up to the beginning of the second century.”64 The repertoire included a local imitation of red Pompeian ware in the form of a frying pan (no. 6), comparable to a bowl from Herodium.65 There was also a “typical Herodian lamp” and a juglet wrapped in palm fibres, deliberately hidden between rocks, which they thought contained balsam oil.66 The pottery repertoire (frying pan, bowl, plate, cooking pot, juglet, storage jar, lamp) indicates that people were cooking and eating in this cave. One complete jar in which dates were stored was found, along with five jar-stoppers, indicating that other storage items had been brought in here. in his map found at http://www.uhl.ac/en/resources/blog/ 56-years-ago-qumran-cliffs-survey/ 62  Patrich, “Khirbet Qumran,” 90. 63  Joseph Patrich and Benny Arubas, “A Juglet containing Balsam Oil (?) from a Cave near Qumran,” IEJ 39 (1989): 43–59 (43). 64  Patrich and Arubas, “A Juglet containing Balsam Oil,” 48. 65  Rachel Bar Nathan, “Pottery and Stone Vessels of the Herodian Period,” in Greater Herodium (Qedem 13; ed. E. Netzer; Jerusalem: Hebrew University, 1981), 112; Pl. 1:17. 66  However, Vendyl Jones claims that his team found the juglet in Cave 24, see http://www.texasmonthly.com/story/masqueraderlost-ark/page/0/5, and that it is the Shemen Afarsemon of Ps 133, the holy anointing oil of the Temple.

19 Caves 3Q and 11Q were re-excavated in 1986 and 1988 respectively, and Chalcolithic material came to light in 3Q.67 Most importantly it was noted that the cave had suffered collapse even before anything was placed in it. In 11Q the team found small Roman potsherds and a further two pieces of cloth.68 South of Qumran the new survey looked again at cave GQ37 of de Vaux’s 1952 survey,69 which they noted as being wrongly marked as lying above Ain Feshkha. In fact, it was just 1 km south of Qumran (see Figure 1.8). They discovered on the terrace a bronze coin of Antioch with the bust of the emperor deliberated erased—typical of the Second Revolt—and a wooden plate. An iron nail and two iron arrowheads were found, one of which is a trilobate form which is paralleled in Masada, Na�al Ḥever and Gamla. Also on the terrace in front of the cave, there were fragments of four glass plates, and remains of at least five jars, 2 stone vessels, 3+ cooking pots, 12+ juglets, 3+ bowls and a “Herodian” oil lamp. The terrace floor was made with mud mixed with palm fronds; Patrich noted that in his cave 13 a similar terrace floor was dated to Iron Age II, though in this case it may be better understood as Roman.70 It adds to the caves with Bar Kokhba remains previously identified by de Vaux in the area near Qumran. Patrich also reported finding important Iron Age material from this cave, including two Iron Age II burials, one a child, beams possibly from a ploughshare, a juglet, and an Iron Age arrow shaft with decorative ringlets and feather indications.71 GQ37 is actually a cave complex, entered via a terrace measuring 10 × 5m, with a large first chamber and a western tunnel leading to a second chamber. In the first chamber were two rooms high-placed to the right and left. Additionally, there were two small caves near the terrace on which pottery had been strewn, still used by shepherds at the time of de Vaux, though there was no pottery found by de Vaux in the main chamber,72 and Patrich noted that “all survey finds came from the terrace.”73 However, regarding the Iron Age burials, in the caption of the article by Patrich and Eisenman in the BAR article of 1989, the arrowhead and reed shaft with black and white paint 67  Patrich, “Khirbet Qumran,” 77. 68  Patrich, “Khirbet Qumran,” 90. 69  They call it FQ37. 70  Patrich, “Khirbet Qumran,” 91–3, with Figs. 15–17 on pp. 88–89. 71  Robert Eisenman, “The 1988–1992 California State University Dead Sea Walking Cave Survey and Radar Groundscan of the Qumran Cliffs,” QC 9 (2000): 123–30 (123); Patrich, “Hideouts in the Judean Wilderness,” 34. 72  De Vaux, in DJD 3:12. 73  Patrich, “Khirbet Qumran,” 92.

20

Taylor

Cave GQ37

Figure 1.5 Cave GQ37, located 1 km south of Qumran. © Joan E. Taylor.

traces is defined as being found in a “Judean wilderness cave” in which there was “extraordinarily low humidity” and a “protective environment,” with the arrow head found “in the floor nearby” the cave wall it bounced off.74 Patrich has confirmed (personal communication) that excavations were on the terrace and also in the cave itself. While the burials and “clay vessels” are Iron Age, the arrowhead and shaft has now been reclassified by Guy Stiebel as

74  Caption by Patrich and Eisenman in Patrich, “Hideouts in the Judean Wilderness,” 34.

Roman,75 and thus it is to be placed within the category of the Bar Kokhba remains, showing that the Roman military shot into this cave, or that the rebels reused a Roman arrowhead. GQ37, just 1 km south of Qumran, brings then the Bar Kokhba remains—both of the refugees and the military 75  Guy Stiebel, “Armis et Litteris—The Military Equipment of Early Roman Palestine in Light of the Archaeological and Historical Sources” (Ph.D. diss., University College London, 2007) Pl. III.18. 4–5. I am grateful to Joseph Patrich for this information supplied by emails in July 2014.

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

that followed them—very close to the Qumran ruins indeed. In terms of its placement, it seems likely that Ziegler confused it with Milik’s Cave76 just above Ain Feshkha and placed it incorrectly at this southern location. Milik’s Cave also had a terrace, and a mixture of Iron Age II and Roman finds, but it was a small cave, not a cave complex like GQ37. Patrich also surveyed 42 almost inaccessible rock-cut caves in the northern cliffs of Na�al Michmash (= Wadi Suweinit), Wadi el-Habibi and Wadi Fara, west of Jericho.77 Of these, he noted that 17 of the 42 caves consisted of a single room, but there were larger and more complex arrangements, the most complex were multi-roomed, and had water installations inside (others had access to cisterns outside—obviously, hide-outs had to be in striking distance of water). In the Wadi Suweinit, at el-‘Aleiliyat, a site later occupied by Byzantine hermits, there were four Jewish Aramaic inscriptions, depictions of two menorot and a pentagram drawn in charcoal on the plaster of a bell-shaped cistern. These were identified then as being originally occupied by Jewish refugees. Inscription no. 2 reads in Aramaic: ‫יועזר התעקר עלו (ות) מטרן‬: “Joezer was uprooted. The guards entered.” A western complex of caves at the site yielded jar rims of the mid-2nd century CE. Overall, finds throughout the large number of caves and the sophistication of some of the internal arrangements within them led Patrich and his team still to associate the habitations with the First Revolt, with the caves re-occupied in the Second Revolt. Patrich went to Josephus to identify those who lived here, noting that in 68–69 CE Josephus had Simon ben Giora, rebel leader of Jerusalem, in a ravine named Pheretae—identified as the Wadi Fara—in which he used caves, some of which he widened, for storage and for army habitations (War 4: 511–513). However, one has to say that one does not have Josephus for the Second Revolt, and the archaeological profile again would fit this time quite closely; the occupants of this time may have intermixed their material with that of some 65 years earlier. In terms of Qumran, Patrich has argued that the associated caves provide evidence of Jewish refugees during both the First and Second Revolts. His results indicate an overall pattern of cave occupation stretching from the territory north of Jericho to Na�al Ḥever which indicates that refugees tried to find some kind of temporary sanctuary: Qumran then is embedded in an area used by refugees. 76  See above, p. 14. 77  Joseph Patrich, Benny Arubas and Eyal Naor, “Jewish Caves of Refuge in the Cliffs of Na�al Mikhmas,” Qadmoniot 19 (1986): 45–50 [Hebrew].

21 Partly overlapping with this, and partly in conjunction with Patrich (for the first year), the University of California Long Beach also undertook three cave surveys in the region from 1988 to 1992, from 5 kilometres north of Qumran to 20 km south. They entered some 485 caves in total and identified 25 caves between the limit of de Vaux’s 1952 survey and the Wadi Murabbaʿat showing signs of human habitation, one cave having a “dressed stone entrance” with lintel and doorposts, indicating long-term use, located about 25 km south of Qumran.78 However, the photograph of this does not show dressed stone, even though clearly there is a proper doorway built up here, with a long stone used as a lintel and a stone for a threshold. Along with these, they found water-channels, ancient grave areas, and other traces of ruins that indicated human use of the vicinity. Iron Age II potsherds were found in their cave 11, and there were also Chalcolithic, Early Bronze, Byzantine and “Islamic” remains.79 The most striking cave was an apparent look-out post 200m above the Kidron consisting of curved walls facing the rock cliff, with pottery remains identified somewhat loosely as “Second Temple.”80 This has a large wall four metres high and above it a large cavity three metres broad with a walled entrance and exit. Another 15m further along this cliff face is a smaller cavity, with finds dating to the Second Revolt.81 In the spring of 1992 Vendyl Jones continued excavations at the Cave of the Column, and a back entrance was found—entered by a small hole on a ledge—in which he uncovered a lot of fill and red material that he identified as Temple incense, along with a coin of the First Revolt.82 However, Zohar Amar later very plausibly analysed the installations and materials (ash and red earth) from the cave as indicating a zone for the production of soap from the potassium-rich plants (Chenopodicae) which grow wild in this area.83 This would then not be a refugee cave, but rather a site connected with Qumran involving industry. In fact, in the excavations of Magen and Peleg at 78  Eisenman, “California State University,” 126, Fig. 2. 79  Eisenman, “California State University,” 126. 80  Eisenman, “California State University,” 127–28. He also suggests this may be identified as site 115 in Bar Adon’s survey (p. 126). 81  Eisenman, “California State University,” 127–28. 82  Note that no fragments of pottery or manuscripts were found here, which throws doubt on the suggestion that this was the original scrolls cave (pace Fields, The Dead Sea Scrolls, 110–13). For a description of the find, see http://www.torah-voice.org/ cave%20excavation.htm [accessed 10 September 2015]. 83  Zohar Amar, “The Ash and the Red Material from Qumran,” DSD 5 (1998): 1–15; idem, “The Balsam Wood in the Temple Incense,” Tehumin 17 (1997): 473–79 [Hebrew].

22 Qumran, an ostracon was discovered that reads “Eleazar the son of Yeshua the soap maker.”84 The position of the Cave of the Column close to caves 1Q, 3Q and 11Q also makes this link with Qumran highly probable, and ties in with the goat-dung layer in Patrich’s cave 24. It is then possible to see this area as being utilised by the residents of Qumran for soap-making and herding and, after Qumran was abandoned, it was re-used by Bar Kokhba refugees and rebels. The apparent First Revolt remains here may then not be from refugees but from occupants of Qumran engaged in work. 1.8 IAA “Operation Scroll” (1993–94) The Israel Antiquities Authority surveyed the area in an intense project dubbed “Operation Scroll,” from November 1993 to January 1994, led by Amir Drori and Yizhak Magen, in which they searched for any further scroll fragments between Qumran and Jericho.85 Reports of this operation were published in Atiqot 41 in 2002. The name of the expedition is quite telling, and clearly sums up the underlying concerns. Further pottery from the Second Temple Period, up to the Bar Kokhba Revolt, was recovered. In Patrich’s cave 13 two oil lamps and a coin from the second year of the First Revolt were found.86 The most interesting result of this project was related to the Ketef Jericho cave complex close to the Qarantal monastery (old grid 1912 1419). Operation Scroll subjected the cave complex to further excavation within four areas A, B, F and G. During work in area A, located on a terrace at the foot of the Abior Cave, further papyrus documents that had apparently fallen out of Abior Cave were also brought to light. Eshel and Zissu wrote about these excavations in numerous publications, and these were also quickly published in DJD 38.87 Notably also, in the “Cave of the Warrior,” a 5000-year old skeleton, wrapped in shrouds

84  Yizhak Magen and Yuval Peleg, The Qumran Excavations 1993– 2004: Preliminary Report (JSP 6; Jerusalem: Staff Officer of Archaeology, Civil Administration of Judea and Samaria/Israel Antiquities Authority, 2007), 21–22; Hanan Eshel, “Excavations in the Judean Desert,” 391. 85  Amir Drori et al., “Operation Scroll,” in Twentieth Archaeological Conference in Israel: Abstracts (Jerusalem: Israel Exploration Society, 1994), 12–17 [Hebrew]. 86  Michel Itah, Yoni Kam, and Ronny Ben Haim, “Survey and Excavations of Caves in the Fault Escarpment South of Almog Junction and West of Qalya,” ʿAtiqot 41 (2002): 169–76 [Hebrew]. 87  Hanan Eshel and Boaz Zissu, “Ketef Jericho: Archaeological Introduction,” in Miscellaneous Texts from the Judaean Desert (DJD 38; ed. J. Charlesworth et al.; Oxford: Clarendon Press, 2000), 3–20.

Taylor

and associated with grave goods, was discovered in Wadi el-Makkukh.88 Broshi and Eshel: Surveys and Excavations (1994–6) During the winter of 1994 Magen Broshi and Hanan Eshel considered again the caves of the marl terrace on which the ruins of Kh. Qumran are located. Four other caves were discovered.89 They noted stairs leading to a further marl cave which is now eroded.90 Therefore, there are now known to be ten artificial marl caves on the wadi: 4Qa, 4Qb, 5Q, 7Q, 8Q, 9Q and 10Q, and the four collapsed caves. Broshi and Eshel also surveyed an area north of Qumran off an ancient track, where they located a further six marl caves, labelled A-F. In Cave A they found the rim of a bowl. Cave C had two openings with three quarried niches (cf. the niches made in the marl cave 4Qa). There were remains of 280 sherds, including storage jars, cooking pots, bowls and part of a lamp. Cave F had collapsed, and the debris yielded 110 sherds, mainly storage jars and bowls.91 Some 100m north of the site Broshi and Eshel found pottery almost flush with the ground: three bowls in one place (between a point identified as 2X and Cave E), a goblet in another (in a gully north-west of point 2X) and elsewhere (1X) a bowl with 25 bowl rims and fragments of at least 25 jars. The complete bowls are identified as relating to the Period II type at Qumran. At point 2X were 3 coins of the 1st century: one of the procurator Festus (59 CE), one of Agrippa I (41/42 CE), and a third of Year Two of the First Revolt (67/68 CE), which suggests that the occupation took place during or after this time. At point 1X they also found four iron hobnails of the same kind discovered along the path north, an iron peg that would have held a tent rope, and part of a stone vessel. It may be noted that these caves lie along a pathway that soldiers would have trod, and iron nails are distinctive 1.9

88  Tamar Schick. The Cave of the Warrior: A Fourth Millennium Burial in the Judean Desert (IAA Reports 5; Jerusalem: Israel Antiquities Authority, 1998). 89  Hanan Eshel and Magen Broshi, “The Archaeological Remains on the Marl Terrace around Qumran,” Qadmoniot 30 (1997): 129– 33 [Hebrew]; idem, “Three Seasons of Excavations at Qumran,” JRA 17 (2004): 321–32; Magen Broshi and Hanan Eshel, “How and Where did the Qumranites Live?” in The Provo International Conference on the Dead Sea Scrolls: Technological Innovations, New Texts, and Reformulated Issues (STDJ 30; ed. D. W. Parry and E. C. Ulrich; Leiden: Brill, 1999), 267–73; idem, “Residential Caves at Qumran,” DSD 6 (1999): 328–48. 90  Broshi and Eshel, “Residential Caves,” 334. 91  Broshi and Eshel, “Residential Caves,” 328–30.

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

on shoes of the military, rather than civilians.92 Eshel has also noted that these types of shoes were used by Jewish fighters in the Bar Kokhba Revolt.93 However, Broshi and Eshel identified the pottery in terms of its relationship with types found at Qumran in Periods Ib and II,94 indicating perhaps that there was a long period of human settlement in these caves, from the 1st century BCE to most of the 1st century CE. I have previously noted that the repertoire of shoe nails and a tent peg would fit with a short occupation by a small unit of auxiliary Roman soldiers, who came to Qumran after it was burnt in 68 CE and it was renovated in a defensive way.95 While they renovated the site these soldiers needed shelter. They may have walked up and down these paths in nailed shoes, erected tents, and used available pottery. The date of the latest coin indicates exactly this time.96 However, both scenarios may be equally true, with the military briefly occupying the caves previously used by the occupants of Qumran, hence the loss of their sandal nails. Porat, Eshel and Frumkin: Further Investigations around En Gedi, 2001–4 Further explorations around En Gedi proceeded from 2001 in survey work sponsored by the Department of Land of Israel and Archaeology of Bar Ilan University and the Cave Research Center of the Hebrew University of Jerusalem, concentrating on the dispersion of Bar Kokhba refuge caves, basically to explore what refuge caves existed between the area of En Gedi and the Wadi Murabbaʿat. Seven further caves were brought to light.97 In one just north of the Na�al David, the Har Yishai cave, Chalcolithic and Roman remains were found, including papyri, coins, a dozen arrowheads, arrow shafts, pottery and glass vessels, a stone bowl, textiles and leather, basketry, ropes, as 1.10

92  Joan E. Taylor, “Kh. Qumran in Period III,” in Qumran, the Site of the Dead Sea Scrolls: Archaeological Interpretations and Debates; Proceedings of a Conference Held at Brown University, November 17–19, 2002 (STDJ 57; ed. K. Galor, J.-B. Humbert, and J. Zangenberg; Leiden: Brill, 2006), 133–46. 93  Hanan Eshel, “Nailed Sandals in the Jewish Sources in Light of the Finds of the Refuge Caves,” in Eshel and Amit, Refuge Caves of the Bar Kokhba Revolt, 225–32 [Hebrew]. 94  Broshi and Eshel, “Residential Caves,” 340–43. 95  Taylor, “Qumran in Period III.” 96  However, there were two Hasmonean coins dropped along the pathway dating to Alexander Jannaeus: Broshi and Eshel, “Residential Caves,” 344. 97  Roi Porat, Hanan Eshel, and Amos Frumkin, “The ‘Caves of the Spear’: Refuge Caves from the Bar Kokhba Revolt North of ‘En-Gedi,” IEJ 59 (2009): 21–46.

23 well as other organic remains.98 Thus the repertoire of personal and utilitarian items is that of the Bar Kokhba period. The nearby “Sabar Cave” contained a coin hoard in a disintegrated cloth pouch, with coins ranging from the reign of Nero to the Bar Kokhba Revolt.99 The survey continued along the cliff face from the “Sabar Cave” and found a scythe blade, a knife, pottery and glass vessels, a piece of cloth, as well as date and olive pits. They discovered another group of five caves, called “Caves of the Spear,” because of the unique find of a spear head, in caves along the Wadi Marrazah (old grid 18653/10495), about 5 km south of Wadi Murabbaʿat, though they had been “extensively looted.”100 The finds came from the Chalcolithic and Early Roman (Bar Kokhba) period, and included leather, a key ring of a jewellery box, part of a sandal, olive and date pits, pomegranate peel, cooking pots, storage jars, jugs, a weaving needle, lamps, arrowheads and bone objects, pieces of textiles and straw rope, basketry fragments, glassware, and four coins, including two coins from the third year of the Second Revolt. This rich personal and utilitarian repertoire of the assemblage is again typical of refuge caves of the Bar Kokhba Revolt. Bedouin then found four fragments of a Hebrew scroll in a small cave on the Na�al Arugot, which turned out to be of Lev. 23 and 24. It was excavated and found to yield pottery dating to the Bar Kokhba Revolt.101 1.11 The “Christmas Cave” Revisited John Allegro’s “Christmas Cave” had a period of obscurity, until it was rediscovered in 2007 by Roi Porat. Thereafter it was re-surveyed by Porat with Hanan Eshel and Amos Frumkin.102 The contents and the textiles were presented in papers at a COST conference in Jerusalem in 2010, which are now available in the electronic publication of Jan Gunneweg and C. Greenblatt.103 The cave is located on the Kidron (Wadi en-Nar), and comprises a 60m long passage 98  Porat, Eshel, and Frumkin, “The ‘Caves of the Spear’,” 38. 99  Porat, Eshel, and Frumkin, “The ‘Caves of the Spear’,” 44. 100  Porat, Eshel, and Frumkin, “The ‘Caves of the Spear’,” 23. 101  Hanan Eshel, Yosi Baruchi, and Roi Porat, “Fragments of a Leviticus Scroll (ArugLev) Found in the Judean Desert in 2004,” DSD 13 (2006): 55–60. 102  Porat, Eshel, and Frumkin, “The Christmas Cave.” 103  Jan Gunneweg and Charles Greenblatt, eds., Outdoor Qumran and the Dead Sea: Its Impact on the Indoor Bio- and Material Cultures at Qumran and the Judean Desert Manuscripts; Proceedings of the Joint Hebrew University and COST Action D-42 Cultural Heritage Workshop held at the Hebrew University of Jerusalem in May 25–26, 2010 (Jerusalem: Hebrew University, 2012). Free Electronic Book in http://micro5.mscc.huji.ac.il/~msjan/qumranproceed ings2010.html.

24

Taylor

between two entrances, a northern low one and a southern high one. There were 255 textiles, 184 from the Roman Period, 71 from the Chalcolithic Period, and 5 Medieval, largely woollen, but also made of linen, and some of goat hair, many of the Roman-period woollen textiles being colourfully dyed, and there was also a linen hairnet, indicating a woman’s presence. These remnants were clearly from clothing, as worn by those who took their belongings to the cave, so the conclusion that the people of the “Christmas Cave” were entirely distinct from Qumran on the basis of bright woollen clothing,104 may be moot in that we do not have personal clothing among the Qumran textile remains of scroll wrappers and packing, though this is sometimes made from selected garments. Other archaeological material came from the Chalcolithic era, Early Bronze Age, Hasmonean period, First and Second Revolts. There were potsherds, two bronze coins, several combs, metal objects and food remains. 2

Key Issues and Considerations

There are many implications resulting from looking at the caves of the western Dead Sea holistically, and in the chronological sequence of their discovery. It is now clear from the surveys as a whole that there was a very extensive use of the caves by refugees during the Second Revolt. Other periods of cave use include the Iron Age II, at which point Qumran was also occupied, along with a string of small settlements between En Gedi and Jericho, and also in the Buqei’a. We can now view the Dead Sea caves broadly to consider how they functioned at different times, in order to further refine our awareness of the nature and dating of the Qumran caves in which scrolls and/or scroll (cylindrical) jars and lids have been discovered, and consider the small finds additional to the pottery in a more refined way.105 Bar Kokhba Rebels and Refugees: A Distinctive Repertoire in Assemblages Overall, the initial partitions of areas occupied by Bar Kokhba rebels have now broken down. As we consider now what all the cave surveys have revealed, we can see a clear picture in the second century of Bar Kokhba activity in caves not only in the area at Murabbaʿat and to the south 2.1

104  Orit Shamir and Naama Sukenik, “Qumran Textiles and the Garments of Qumran’s Inhabitants,” DSD 18 (2011): 206–25. 105  See the paper by Dennis Mizzi in this volume.

of it, but also near Ain et-Turaba and Ain el-­Ghuweir, at the Christmas Cave, in a cluster of caves near Ain Feshkha, in cave GQ37 just 1 km south of Qumran, probably also in the caves close to 3Q and 11Q, and north to Ketef Jericho. These Bar Kokhba rebels and refugees ranged all over the north-western Dead Sea and Jericho area. Additionally, it is sometimes hard to separate what might come from the First Revolt or earlier from this widely attested picture of Bar Kokhba activity. Mixed in are the remains of the Roman military, since they encamped near places where rebels and refugees were located and engaged in a systematic operation to either flush them out, thereafter at times occupying the hideouts themselves, as we see at Murabbaʿat. This puts into perspective the kind of early intrusion we are likely to have had in the Qumran scroll caves, and leads to some further considerations. The repertoire of material found in Bar Kokhba caves is, overall, quite distinctive. While the pottery does continue forms that are quite similar to those of 60–70 years earlier, meaning it is hard to date them on pottery alone, the caves of this time have a much greater range of artefacts than the Qumran caves where cylindrical jars and lids were found, even those without existing scroll fragments. The Bar Kokhba caves held personal items (some of great value), including also women’s artefacts, as well as papyrus documents, arrow heads, wooden plates, ropes, sandals, basketry, glassware, metal objects, coins and so on. Along with these are utilitarian items for cooking, storage and also tools and keys. Therefore, when one finds a subset of these kinds of objects within a cave setting, even without other datable artefacts, there is some reason to suggest that they should be considered refugee types of objects. The assemblage indicates what we may call a “Bar Kokhba repertoire.” The wooden artefacts presented to Allegro conform to such a repertoire, including the wooden plate apparently discovered in “Ibrahim’s cave” close to cave 11Q. The richness of the artefact assemblage of the refuge caves serves to highlight the very limited range of artefacts we have in a cave like 1Q. As Jodi Magness has observed,106 nowhere else in the Judaean Desert caves are there the distinctive cylindrical jars and lids, or linen scroll wrappers, we find in the Qumran caves associated with scrolls. In cave 11Q, 106  Jodi Magness, “Why Scroll Jars?” in Religion and Society in Roman Palestine: Old Questions, New Approaches (ed. D. R. Edwards; London: Routledge, 2004), 146–61, and see Dennis Mizzi, The Archaeology of Khirbet Qumran: A Comparative Approach (Ph.D. Diss., Oxford University, 2009), 120–24.

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

however, there is comparatively more material than elsewhere, including items that do overlap with the kind of artefacts we have from refugee caves, such as an iron key and iron arrowhead, which actually leaves a question about whether Bar Kokhba refugees might have used this cave; the issue really is where the objects were found exactly in the cave relative to the jars, and we may never know. As Popović asks: Did someone drop the key before or at the time the scrolls were deposited there? Did that person perhaps also stay there for a while? Might it indicate clear and present danger—belonging to Romans who sought the place out, or to Jewish defenders— or was its loss in the cave not related to any violent circumstances?107 The question of whether the distinctive repertoires containing personal objects come from the 1st century or Second Revolt or both may not always be possible to determine, but the possibility of repeated use of certain caves is particularly interesting. In the case of visits close in time it may be impossible to distinguish whether objects were deposited together or separately. The case of the Murabbaʿat caves is particularly relevant in terms of this model of repeated use of the same vicinity, not only at points close in time (during and after the Second Revolt), but in multiple different eras. As to the dating of refugee activity as a whole, the vast majority of datable artefacts found in all excavations and surveys indicate that this was an area used by people at the end of the Bar Kokhba Revolt—and subsequently. However, material possibly dated to the First Revolt in the cluster of caves close to 11Q and 3Q might be also be ascribed to the occupants of Qumran in Period II or even Period III. The soap-making distinguished in the “Cave of the Column” could be connected with Qumran industries, meaning that the occupation area of Qumran should be linked with this site. A problem is that both coins and pottery types of 68 CE could still be widely employed in 132–35 CE: in Judaea this was a period in which little innovation took place, though there are some important exceptions for diagnostic purposes (see below). The artefact repertoire of cave 1Q, and other caves in which cylindrical “scroll” jars and lids have been found, is

107  Popović, “Qumran as Scroll Storehouse,” 562.

25 sparse.108 This repertoire, as well as the very limited range of small finds (and no coins) found in the Qumran caves, is nothing like the caves of refuge used by Bar Kokhba rebels and refugees. In addition, the rich material from refugees has to be placed on one side if we are to understand the Qumran caves in the light of a chronologically appropriate context. If it is included, it confuses the picture. By this I mean that we have to consider that, at the time these Qumran caves were used (co-inciding with Qumran’s habitation), the last significant use of caves, both natural and artificial, was in Iron Age II. The caves were, to those who lived at Qumran in the 1st century BCE to 68 CE, secure and quite remote. No one was taking refuge in them. Overall, the range of the natural “jar caves” spans an area in which Qumran is roughly at the centre, and Qumran is a site with exactly the same kinds of wide-topped, cylindrical hole-mouthed jars found distinctively only here in any significant numbers. The highly distinctive nature of the caves proximate to Qumran in terms of their pottery and linen artefacts from the 1st century BCE to later 1st century CE stands out as unique in the context of the northwestern Dead Sea at this time. Bar Kokhba cave material is personal, utilitarian, various, diverse, and assemblages completely lack cylindrical jars, lids and scroll wrappers, though they do contain documents and manuscripts people took with them to hide away, decomposed in the same way as we find with the manuscripts in the marl caves of Qumran. Repertoires of object assemblages as a whole are key in defining the character and dating of cave occupancy, and there is a stark difference between what we find in the Qumran manuscript and/or jar caves and the Bar Kokhba caves; the only exception to this is cave 11Q, which perhaps may then have been used by refugees some time after the manuscript deposits. 2.2 Wadi Murabbaʿat and Qumran One of the most remarkable features of Murabbaʿat is that so many documents of such a diverse nature were found in one cave, 2Mur. Since it is not usual for a single cave to have any document from even a single time, here we have documents apparently deposited at different times: in the Second Revolt, in the later 2nd century and, finally, in the 10th-century. In addition, there is the

108  See Joan E. Taylor, The Essenes, the Scrolls, and the Dead Sea (Oxford: Oxford University Press, 2012), 277–81.

26 o­ riginal letter from the 8th century BCE (Mur 17) perhaps found in the cave and re-used in the 2nd century.109 It is also interesting that in two of the caves only small fragments of biblical manuscripts were found. They are in extremely damaged condition. This kind of damage to parchment or lightly tanned leather—as opposed to papyrus—is because it is very attractive to rats, as de Vaux himself noted, and he even photographed pieces of manuscripts in a rat nest.110 The scroll manuscript fragments are precisely what we would expect to remain from parchment manuscripts left on the ground and not preserved in jars. There are no scroll jars in these caves, and it is likely that the manuscripts that were brought here were considered precious possessions of the community attested in the documents, taken to safety at a time of crisis. Murabbaʿat then is important for understanding the Qumran scrolls as artefacts because it gives a test case of scrolls that really were not carefully bound up tightly, wrapped and placed in jars to protect them over the ­centuries.111 In other words, we can be sure that the well-preserved scrolls from caves 1Q and 11Q were found preserved in jars because they would not otherwise have survived so intact; the jars that were smashed open in past centuries in these caves would have resulted in the accompanying fragments, since outside the jars the scrolls were no longer protected. The profile of Murabbaʿat manuscripts is very like what we find in Cave 4Qa, where manuscripts were not preserved in jars, which suggests strongly that manuscripts were not placed in jars in Murabbaʿat 109  It is unlikely that both hands come from the 8th century BCE, see Fitzmyer, A Guide to the Dead Sea Scrolls, 137, who dates the second hand to the 2nd century CE. 110   D JD 2: Pl. II: c. 111  For an argument that the scrolls were specially preserved in jars for long-term storage, see Joan E. Taylor “Buried Manuscripts and Empty Tombs: The Qumran Genizah Theory Revisited,” in ‘Go out and study the Land’ ( Judges 18:2): Archaeological, Historical and Textual Studies in Honor of Hanan Eshel (JSJSup 148; ed. A. Maeir, J. Magness, and L. H. Schiffman; Leiden: Brill, 2011), 269–316; eadem, The Essenes, the Scrolls, and the Dead Sea, 272–303. It should be noted that this argument does not presuppose that the scrolls are a genizah as defined by later times, or that they were assembled by using the criteria of later genizah collection. In addition, I now think it unlikely that there was a bitumen preparation used on linen wrappers, since I have found no extant items to confirm the reports, but I remain convinced that the careful rolling, wrapping and placing of scrolls in heavy jars, and the laborious removal of these jars to various barely accessible caves, required time and planning, and could not be done in a hurry, and that the pottery repertoire of the caves indicates a longer period than one just around the year 68 CE, cf. Popović, “Qumran as Scroll Storehouse,” 583–84.

Taylor

either. One might also recommend that proper soil-sifting excavations be done in any cave in which smashed cylindrical scroll jars and lids were found, as the possibility of there being manuscript fragments in these caves is, I think, quite high. In my view the jars in the natural caves were invariably scroll containers, though they may have had other utilitarian uses prior to their deposit. In addition, Murabbaʿat raises the issue of the activity pattern of settlement and refuge in the Second Revolt. At the time of the First Revolt, there was a string of settlements along the western coast of the Dead Sea, from Jericho all the way to En Gedi, of which Qumran was one; these were redeveloped from Iron Age II sites and lay along the road between Jericho and En Gedi.112 However, on the higher level of the plain at the top of the hills, known as the Buqei’a, it does not appear that the Iron Age II sites here were occupied in the period before 68 CE. The Iron II fortified farms had been abandoned in the wake of the Babylonian invasion in the 6th century BCE, but had been taken over only briefly again as camping grounds for the Hasmonean army in the 1st century BCE. The key site in this vicinity was Hyrcania. However, the settlement picture is different in the years following the destruction of the Jerusalem Temple in 70 CE, as refugees dispersed to outlying regions. The Iron Age II fortified farms, which had sophisticated dams used to harness flood rains,113 were reoccupied, at Khirbet Abu-Tabaq, at Khirbet es-Samrah, Ain Feshkha, Kh. el-Maqari and elsewhere. The archaeological examinations done in the Buqei’a by Frank Moore Cross and J. T. Milik114 identified that these sites here were inhabited at a time corresponding to de Vaux’s Period III (Humbert’s Level 4) at Qumran; Cross and Milik noted that a 1st-century Roman ribbed cooking pot and Roman lamp were identical with those at Qumran in this period. This cluster of settlements thus continued to the end of the Second Revolt. In fact, Murabbaʿat documents attest continuing farm buying and settlement in the region nearby (e.g. Mur 23, 24, 25, 26, 27, 28, 30, 31).

112   See the discussion in Joan E. Taylor and Shimon Gibson, “Qumran Connected: The Paths and Passes of the North-Western Dead Sea,” in Qumran und die Archäologie: Texte und Kontexte (WUNT 278; ed. J. Frey, C. Claussen, and N. Kessler; Tübingen: Mohr Siebeck, 2011), 1–51. 113  Lawrence E. Stager, “Ancient Agriculture in the Judean Desert: A Case Study,” (Ph.D. Diss., Harvard University, 1975); idem, “Farming in the Judean Desert during the Iron Age,” BASOR 221 (1976): 145–58 (145). 114  Frank M. Cross, “El-Buqei’a,” NEAEHL 1:267–69; Frank M. Cross and Józef T. Milik, “Explorations in the Judaean Buqê’ah,” BASOR 141 (1956): 3–17.

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

As I have explored on the ground with Shimon Gibson,115 all these sites were directly connected to Qumran, by means of re-used Iron Age II tracks. The pass travels up from Qumran in sharp bends, just north of the wadi, then joins with a route towards the Buqei’a plain. This is not a significant road, or trade route, but it connects a number of small sites (Figure 1.4). In the period from 70 to 135 CE Herodium seems to have been a main administrative centre, as mentioned in Mur 13 and confirmed by recent archaeology, and En Gedi was under the control of the Judaean rebels (Mur 46). Milik himself identified a site in Mur 45 (line 6), Mesad Hasidim, with Qumran, which in this case would probably have been an old name still in use.116 With this distribution pattern of settlement prior to and during the Bar Kokhba Revolt, we would also expect the occupants of the Buqei’a, faced with a Roman advance, to hide in the caves towards the Dead Sea. The track from Kh. Abu Tabaq leads directly to Qumran. Kh. Abu Tabaq is located at the intersection of tracks leading north-south between Jericho and Bethlehem, and eastwest from Qumran and Ain Feshkha. From Jerusalem this north-south track crosses the Kidron Valley, and then goes to the Wadi Ta‘amireh, which begins in Bethlehem. The caves of Wadi Murabbaʿat are a short detour off this valley, which continues along the Na�al Darga (Wadi Darajeh) and ends today at Kibbutz Mitzpe Shalem. The caves here are actually two hours walk inland from the cliffs of the Dead Sea,117 but not such a long walk from the higher ground of the Buqei’a. Further down from the cliffs there are flood-fed pools in the Na�al Tekoa that would have provided winter drinking water for refugees, and later for the Roman military.118 Given the route connectivity, the settlement of the Buqei’a in the period 70–135 CE raises again the issue of how late Qumran was occupied. While de Vaux’s close associate Milik always assumed that Period III stretched to the end of the first century, and Qumran was then briefly re-occupied by Bar Kokhba rebels,119 de Vaux dated Period III (Humbert’s Level 4), only to 73 CE, and saw the site as definitively a Roman fort at this time, perhaps 115  See Taylor and Gibson, “Qumran Connected.” 116  Milik, in DJD 2:163–64; Pl. XLVII. 117  Fitzmyer, A Guide to the Dead Sea Scrolls, 309. 118  For a good account of walking in this area see http://www.jpost .com/Travel/Around-Israel/Dare-to-delve-into-Nahal-Darga. The cistern and water system found by Bar Adon near the caves was most likely built by Roman soldiers. Their presence is also attested by a Roman road built in the Na�al Darga, identified on the British Mandate map of Palestine, and the documents of 2 Mur itself. 119  Milik, Ten Years of Discovery, 55–56.

27 influenced by earlier identifications of the visible ruins, though he too saw it as re-occupied in the Bar Kokhba Revolt: the buildings “were used either as a hiding place or as a centre of resistance by the insurgents of the Second Jewish Revolt in AD 132–5.”120 In terms of this chronology of Period III, or Level 4, I have previously argued that the occupation is likely to have been longer than 73 CE.121 Milik’s reasoning based on coins seems correct: from de Vaux’s excavations there was a coin of Agrippa II, from 87 CE, found outside the buildings.122 There were three Judaea Capta coins (71–79/81 CE) found in loci 24, 32 and 43, and two city coins (end of 1st cent.) found in loci 13 and 34, along with one of Trajan (98–117 CE) found in locus 41.123 Magen and Peleg found a coin of Titus, Caesarea issue, of 71–81 CE.124 Jolanta Młynarczyk also identifies several lamps as coming from the latter part of the 1st century or 2nd century found in Qumran: KhQ 5101 (type 046) and KhQ 3136, the latter being an imported lamp from the last third of the 1st century (type 047.1), also attested in Masada and Murabbaʿat.125 On the basis of these coins and lamps, and the sheer wealth of Period III remains, it seems likely that the occupation may have stretched at least through the reign of Domitian to Trajan. Qumran was clearly abandoned after earthquake damage that created north-south cracks across the site, which were mistakenly assigned to the earthquake of 31 BCE by de Vaux, as Humbert has noted,126 though this earthquake did damage the site quite significantly. I have also suggested that the most likely earthquake to account for the cracks is that attested in 115 CE.127 This very strong earthquake of a possible magnitude of 7.4 was felt from Caesarea to Petra (Dio Cassius, Hist. 68).128 Whether there is any significant subsequent construction by Bar Kokhba rebels and refugees seems to me 120  De Vaux, Archaeology, 45. 121  Taylor, “Qumran in Period III.” 122  De Vaux, Archaeology, 44. 123  Roland de Vaux, Die Ausgrabungen von Qumran und En Feschcha. Vol 1A: Die Grabungstagebücher (NTOA.SA 1A; trans. and notes F. Rohrhirsch and B. Hofmeir; Fribourg/Göttingen: Universitätsverlag/Vandenhoeck & Ruprecht, 1996), 126–27. 124  Magen and Peleg, The Qumran Excavations, 23, n. 19. 125  Jolanta Młynarczyk, “Terracotta Oil Lamps from Qumran: The Typology,” RB 120–21 (2013): 99–133 (123). 126  Jean-Baptiste Humbert, “Some Remarks on the Archaeology of Qumran,” in Galor, Humbert, and Zangenberg, Qumran, the Site of the Dead Sea Scrolls, 19–39. 127  Taylor, The Essenes, the Scrolls and the Dead Sea, 263–65. 128  Kenneth W. Russell, “The Earthquake Chronology of Palestine and Northwest Arabia from the 2nd through the Mid-8th Century A. D.,” BASOR 260 (1985): 37–59 (40–41).

28 unlikely. The wall built on top of cistern 58, attributed to Bar Kokhba rebels by de Vaux, may rather be a Byzantine wall, built after the bottom of cistern 58 was filled by debris from Period III/Level 4 collapse, and also after a “natural deposit” had formed on top of it.129 While the initial collapse could be from earthquake damage from 115 CE, the natural deposit should indicate the passing of centuries. Since a Byzantine wall may seem perplexing to some, it should be remembered that Byzantine activity at Qumran is attested. There are a number of 4th-century and later Byzantine coins that have been found at Qumran, including two from the reign of Theodosius (379–95 CE) in loci 34 and 152, with six other coins of the 4th century in loci 7, 68, 88, 91, 96, 119. Two later Byzantine coins were found in loci 42 and 76.130 The excavations by Randall Price also uncovered a coin of Constans dated to the 4th century.131 The full extent of Byzantine employment of Qumran is not the subject of this discussion, but it is important that linen (KhQ 3649) from Tomb 1 of the small cemetery south of the Wadi Qumran at the foot of the marl terrace, where there were some 30 graves of varying orientations,132 was radiocarbon dated to 565–635 CE (QUM–524 B003) at 1 standard deviation.133 This grave contained the remains of an adult female and 30 beads (KhQ 2670), and either two earrings or an earring and ring (KhQ 2671 and KhQ 3651). A Byzantine date coheres with the typological jewellery study made by Christa Clamer,134 who noted that the earrings from Tomb 33 in the southern arm of the main cemetery and Tomb 1 in the southern cemetery as well as a crumbled glass bead from Tomb 32 should be all be dated to sometime between the 4th and 8th/9th centuries. All this refutes the proposals made by Zias, that 129  De Vaux, “Fouilles de Khirbet Qumrân,” 548; idem, Archaeology, 45. 130  De Vaux, Die Ausgrabungen von Qumran, 127–8. 131  Yoav Farhi and Randall Price, “The Numismatic Finds from the Qumran Plateau Excavations 2004–2006, and 2008 Seasons,” DSD 17 (2010): 210–25 (220). 132  De Vaux, Archaeology, 58; Humbert and Chambon, Fouilles de Khirbet Qumrân, 352. 133  Kaare L. Rasmussen et al., “Cleaning and Radiocarbon Dating of Material from Khirbet Qumran,” in Bio- and Material Cultures at Qumran: Papers from a COST Action G8 Working Group Meeting held in Jerusalem, Israel on 22–23 May 2005 (ed. J. Gunneweg, C. Greenblatt, and A. Adriaens; Stuttgart: Fraunhofer IRB Verlag, 2006), 139–64. 134  Christa Clamer, “Jewellery Finds from the Cemetery,” in Khirbet Qumrân et ‘Aïn Feshkha. Vol. 2: Études d’anthropologie, de physique et de chimie (NTOA.SA 3; ed. J.-B. Humbert and J. Gunneweg; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 171–84 (171–76).

Taylor

some Qumran graves are from modern Bedouin.135 De Vaux himself noted Byzantine sherds in cave GQ23, which is located just above Qumran near the aqueduct route.136 At Ain Feshkha there was occupation in the north-east corner of the stable (locus 20),137 and a similar utilisation of a disused cistern for an anchoritic dwelling at Qumran would be consistent with this evidence. The supposition by Hirschfeld, that the Dead Sea level was too high for occupation of Ain Feshkha in the Byzantine period,138 is not supported by recent studies.139 That hermits occupied the ruins at certain times is possible. We should remember also the discoveries of scrolls in this region already in the year 217 CE,140 and coins shortly after this time have been found in Qumran locus 104 (2054), 91 (1583) and 135 (2315).141 While there was then no Bar Kokhba building at Qumran, Bar Kokhba rebels did clearly visit Qumran and made use of the structures, because de Vaux found ten coins stashed underground in a bowl in locus 29, at the bottom of the tower: five of these coins were from 135   Joseph E. Zias, “The Cemeteries of Qumran and Celibacy: Confusion Laid to Rest?” DSD 7 (2000): 220–53. 136  De Vaux, in DJD 3:23. 137  Roland de Vaux, “Fouilles de Feshkha: Rapport Préliminaire,” RB 66 (1959): 225–55 (253–54). 138  Yizhar Hirschfeld, Qumran in Context: Reassessing the Archaeological Evidence (Peabody, MA: Hendrickson, 2004), 189–91. 139  Yehouda Enzel et al., “Late Holocene Climates of the Near East deduced from Dead Sea Level Variations and Modern Regional Rainfall,” Quarternary Research 60 (2003): 263–73 (265); Fig. 2A. 140   As is also well-known, the scholar Origen (who wrote his Hexapla between the years 228 and 254), stated that he had the use of a (Greek) version of the Psalms that was found “in a pithos (earthenware jar) near Jericho at the time of Antoninus the son of Severus” (the emperor Caracalla, 211–27; Eusebius, Hist. Eccles., 6: 16: 3). The finding of the manuscripts in pithoi (plural) is reported by Pseudo-Athanasius in his Synopsis and by Epiphanius, who writes of the discovery being “in the seventh year of Antoninus, son of Severus” (217 CE). Both Athanasius and Epiphanius specify that pithoi contained “manuscripts of the Septuagint as well as other Hebrew and Greek writings” (Epiphanius, De Mens. et Pond., 17–18; PG 43, cols. 265–68; Pseudo-Athanasius, Synopsis, PG 28: col. 432). 141  De Vaux, Die Ausgrabungen von Qumran, 127. The discussion of the silver coins up until this time by Kenneth Lönnqvist, The Report of the Amman Lots of the Qumran Silver Coin Hoards: New Chronological Aspects of the Silver Coin Hoard Evidence from Khirbet Qumran at the Dead Sea (Amman: National Press, 2007): 32–34, 51–53, Nos. 347–54, on the basis of the holdings of the Amman Archaeological Museum, is difficult to assess as it may be there are intrusive examples.

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

the Second Revolt (490–493, 495), and from this era also was one coin in locus 41 (714), with a further three coins of this date in Ain Feshkha (AF 137, 140, 141). The coins of the locus 29 hoard usefully include two from the First Revolt (483, 484), the one from the reign of Vespasian (486) and three from the time of Trajan (487–489), giving an indicator of the range of coins in circulation at this time.142 The fact that the coins were hidden under a floor in the tower may indicate that Qumran at this time was an abandoned site. In addition, Magen and Peleg state that in their excavations the Qumran pottery “dates from the Iron Age and from the 1st century BCE to the destruction of the site during the Great Revolt and to the Bar Kokhba Revolt,”143 though careful analysis of later Qumran pottery in the light of the repertoire of surely datable Bar Kokhba caves has yet to be done comprehensively. There was no destruction layer to seal material beneath it when Qumran Period III/Level 4 was left to whether the ages, apart from the falling debris caused by earthquake damage, and the occupants of the site would have taken away anything useful. The buildings were left in ruins, but they were not necessarily reduced to rubble. Many walls and even roofs could have been intact. Bar Kokhba rebels might yet have used such structures, with some clearance and sweeping out, just 20 years after they were abandoned. The paths through to the Buqei’a still existed, along with the Qumran pass. The aqueduct would still have filled some of the surviving cisterns on the west side, since the earthquake damage was concentrated in the water system of the east. Clearly, from the widespread evidence now found, Bar Kokhba rebels and refugees were in this area, looking for caves anywhere in this region that could be used as temporary shelters and hide-outs. Additionally, there would have been reason to occupy Qumran strategically for a short time as a look out post. A ruined site at a strategic location, with caves nearby, would have been perfect for temporary occupation, though it was too exposed to be a refuge. All this has ramifications for the Qumran caves, because this activity was happening when the manuscripts had already been placed inside them, and it raises issues about the integrity of these caves and the possibility that they were utilised by Bar Kokhba refugees, as noted in regard to cave 11Q. However, in terms of the use of the caves at Qumran itself, by which I mean the marl caves, there is no repertoire of objects that would indicate Bar Kokhba occupation, or coins. However, at the ancient entrance to 142  De Vaux, Die Ausgrabungen von Qumran, 127. 143  Magen and Peleg, The Qumran Excavations, 15.

29 Cave 4Qa de Vaux discovered a circular lamp (Gr4Q 22), recently analysed by Jolanta Młynarczyk and classified as type 047.2, dated to between the late 1st and early 3rd century.144 Lamp Gr4Q 22 then indicates that these caves were visited in Period III or in the Bar Kokhba period. This kind of “discus lamp” was found in the Caves of the Spear,145 the Cave of the Sandal,146 and the Cave of the Tetradrachm.147 It was also found in Cave 24 of Patrich’s survey, 50m to the north of cave 11Q, which would date occupation of this cave to the later 1st century (post-70 CE) or the Bar Kokhba Revolt.148 2.3 The Marl Caves Can we determine for sure whether the lamp Gr4Q 22 should be associated with the Period III or the Bar Kokhba period? Perhaps its placement at the old entrance to the cave is telling: it is the last in the pottery sequence. The marl caves are very close to the structures of Qumran, accessed by rock-cut stairways and ladders, with 7Q–9Q only accessible from the plateau on which the buildings stand. It is possible that here manuscripts at the site were quickly dumped ahead of the attack in 68 CE, though I am more persuaded that remains in 7Q–9Q were collected manuscripts in the process of being conserved in the proper preservation-burials in the natural caves.149 The state of the manuscripts in cave 4Qa at the time they were deposited there and in the period following their deposit remains unknown. Hannah Cotton and Erik Larson have contested de Vaux’s theory that manuscripts here were torn up in antiquity by Roman troops: for example 4Q174 and 4Q376 show breaks along the straight edges of drylines.150 But they point out that an opisthograph, 4Q460/4Q350, indicates that someone used one of

144  De Vaux, DJD 6, 17–18; Fig. 6:10; Pl. III: c; Młynarczyk, “Terracotta Oil Lamps,” 123–24. 145  Porat, Eshel, and Frumkin, “The ‘Caves of the Spear’,” 21–46 (30): Fig. 9.2. 146  Hanan Eshel and Boaz Zissu, “Finds from the Bar Kokhba Period in the Caves at Ketef Jericho,” in Eshel and Amit, Refuge Caves of the Bar Kokhba Revolt, 126 no. 8. 147  David Amit and Hanan Eshel, “The Bar Kokhba Period Finds from the Tetradrachm Cave,” in idem, Refuge Caves of the Bar Kokhba Revolt, 198 no. 64. 148  Patrich, “Khirbet Qumran,” Fig. 10. 149  Taylor, The Essenes, the Scrolls, and the Dead Sea, 292–95. 150  Hannah Cotton and Erik Larson, “4Q460/4Q350 and Tampering with Qumran Texts in Antiquity,” in Emanuel: Studies in Hebrew Bible, Septuagint, and Dead Sea Scrolls in Honor of Emanuel Tov (VTSup 94; ed. S. M. Paul et al.; Leiden: Brill, 2003), 113–25.

30

Taylor

Cave 4Qb Cave 4Qa Cave 9Q Cave 8Q

Figure 1.6 The marl terrace artificial caves on the edge of the Wadi Qumran and Qumran plateau, Cave 4Qa, 4Qb, 5Q, 7Q, 8Q, 9Q and 10Q. © Joan E. Taylor.

the manuscripts for writing a cereal list in Greek.151 The same cereal lists, and accounts, are found in papyrus and parchment fragments 4Q351–358 in Hebrew or Aramaic, and are very like the cereal lists written in cave 2Mur, which include one written badly by someone in Aramaic (Mur 8). As noted by Popović, 2Mur gives us a scenario of ­multiple deposits by different people using the cave for different purposes.152 Cotton and Larson argue against Ada Yardeni’s speculations that such Cave 4Q writings exhibit such an evident “late” scribal script that they must actually originate from caves further south,153 and suggest that a Roman auxiliary unit who occupied Qumran post-68 CE utilised the 151  Cotton and Larson, “4Q460/4Q350.” 152  Popović, “Qumran as Scroll Storehouse,” 558–59. 153  Ada Yardeni, “Introduction to the ‘Qumran Cave 4’ Documentary Texts,” in Aramaic, Hebrew and Greek Documentary Texts from Naḥal Ḥever and Other Sites: With an Appendix Containing

marl caves and used the manuscripts lying here. Cave 4Qa would have been a good storage place for cereals, since it could be kept rodent free. Romans returned to the area in 135 CE, but they appear not to have set up camp at this time. We know they were here, because they shot an arrow into cave GQ37, 1 km south of Qumran, where Bar Kokhba refugees had taken shelter, and never retrieved it. It would be useful to understand the stratigraphy of Cave 4Qa, though details are lacking. Importantly, according to de Vaux’s account of what the Bedouin told him, when an old man first located Cave 4Qa (thanks to a ­partridge flying in to a small opening), he saw pottery pieces and a lamp. This gave the Bedouin the idea to try digging, and they dug a tunnel into the floor of the cave. Inside, as Milik stated, they “had already turned over several cubic metres Alleged Qumran Texts (The Seiyâl Collection II) (DJD 27; ed. H. M. Cotton and A. Yardeni; Oxford: Clarendon Press, 1997).

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

of earth when, suddenly, their hands came upon a compact layer of thousands of manuscript fragments.”154 In other words, there was ancient pottery on an upper surface layer, seen from the small opening above (the original entrance), but the Bedouin found the manuscripts buried underneath this layer at some distance down.155 The testimony may not be entirely ­accurate, but it remains credited by de Vaux and Milik, and unless the Bedouin had seen pottery on the surface they would not have spent time digging here; there are hundreds of barren caves. Since lamp Gr4Q 22 is associated with the entrance, and the Bedouin report seeing a lamp, it is tempting to suggest this was indeed what the Bedouin saw, or one like it. By the time of its deposit, then, the manuscripts had been covered by sediment fall. As such, its association with Bar Kokhba remains elsewhere in the Dead Sea area might indicate rebels visited the cave. Sediment had already covered the manuscripts (including the opisthograph), probably as a result of the earthquake of 115 CE. The manuscripts discovered by de Vaux were largely found in the deposit of the principal chamber of Cave 4Qa but not on the actual cave floor.156 Rather, they were found throughout the chamber of 4Qa at different levels; and, as de Vaux noted, there was already a disturbance before all the sediment fell that covered the manuscripts.157 De Vaux also notes of the actual floor level that it was uneven and had troughs: there was a rectangular one in front of the eastern recess (window), measuring 1m × 0.65m. De Vaux does not note in his report all the windows, and discusses the terrace of the south-facing window as a “chambre secondaire,” where no manuscripts were found. Now the southern window opens to a ledge at a higher level than the cave floor. De Vaux considered this ledge to be a recess that has partly collapsed, so that the southern window is actually not original but is the recess exposed by cave erosion.158 154  Milik, Ten Years of Discovery, 17. 155  De Vaux, in DJD 6:3. 156  De Vaux, in DJD 6:3. Note that cave 4Qa and b were “excavated” by the Bedouin who were only interested in texts, and therefore many items of pottery and small finds may have been thrown out down the wad and de Vaux only excavated what remains. Even with the proper excavation of de Vaux, he himself did not do sherd counts and was only concerned with diagnostic pieces. 157  De Vaux, in DJD 6:21. 158  It seems doubtful to me that this was constructed simply as a recess without a window because the edge of the window ledge is quite clearly defined with rocks, see https://www .flickr.com/photos/palestineexplorationfund/5904630712/in/ set-72157627120966518/.

31 De Vaux’s team found another lower trough cut into the ground, where they retrieved wood and manuscript fragments in the fill, with a jug that had slid down into the trough from above (4Q17).159 This wood is in fact highly significant but, as far as I know, it has not been analysed; might it represent a wooden box of manuscripts deliberately placed in the recess? Many phylactery cases with a bit of pottery and some written fragments were found in what de Vaux thought was an annex in Cave 4Qa cut into the south-western angle of the principal chamber at a slightly lower level. This leads to the “window” that now looks out to Cave 4Qb.160 De Vaux thought then of this as a passage that led to a missing cave that had crumbled down the wadi, a passage which then led through also to Cave 4Qb. In the paper discussing this by Humbert, in this volume, he has reconstructed an original missing cave in between 4Qa and 4Qb, so that what is often seen as a lower recess and window is actually a passage through to the collapsed cave, though why it is created at such a low level, making access difficult, is hard to understand. Because of the natural form of the hill with water run off I am not even sure that a cave could have been located here. However, the point that what we see as a window might have originally been a recess is significant, and would make better sense in terms of this being a place where materials were deposited. I note all this simply to explore the situation of the manuscripts in the cave prior to its major collapse. The buried manuscripts subsequently were damaged by decomposition over time and by burrowing ants, and were buried even further by cave collapse in later centuries. One final thing to note is that there were no Byzantine remains in Cave 4Qa, despite the presence of anchorites in the area. Did those who left the lamp at the entrance also seal it, for protection, ­perhaps recognising some biblical manuscripts that still lay around the edges of cave collapse? If there were a detailed stratigraphy of de Vaux’s excavations in Cave 4Qa perhaps further light could be shed on this matter.

159  De Vaux, in DJD 6:9, Fig. 6:9. Note that 4Qa also contained “thongs” and reinforcing tabs as also in Cave 8Q, see DJD 6, Pl. iv and v. The phylacteries are shown on Pl. vi–xxiv. Fastenings from the marl caves are described by J. Carswell, Appendix 1, 23–28 and comprise 200 items, 100 of these coming from cave 4Q, the rest coming from cave 8Q, with some also bought back from the Bedouin. 160  De Vaux, in DJD 6:12, Fig. 3.

32

Taylor

3 Conclusions Progressive explorations of the region of the western Dead Sea from En Gedi to Jericho have illuminated the occupations of this area from the Chalcolithic era to the Bar Kokhba Revolt, and on to Byzantine and later periods. There is an overall pattern of rich finds testifying to temporary cave occupation during the Bar Kokhba Revolt in particular. These show a distinctive repertoire in assemblages, with personal and utilitarian objects, as well as occasionally some manuscripts and documents considered valuable. Refugees (men, women and children) fled to the caves from settlements of the Judaean Desert, En Gedi and the southern Dead Sea, and from Jericho and the areas north of the Dead Sea, around the year 135 CE, and they were closely followed by the Roman military who flushed them out and/or killed them, and could use their caves. An interesting case is that of Murabbaʿat, where there are four phases of document deposit, most importantly from Bar Kokhba rebels and refugees followed by Roman soldiers, as well as an indication of what happens to parchment manuscripts left exposed on the ground. It thus provides a model for a quick hiding scenario for precious scrolls. By the time of the Second Revolt, Qumran was lying abandoned. The site and nearby caves were nevertheless used for a short time by Bar Kokhba rebels and refugees, as elsewhere, and their remains can be contextualised now with what we know of these people in the region. What they made of caves where manuscripts were lying remains a matter for conjecture, but the indications of caves being used at different times by different people needs always to be borne in mind. The evidence of re-used parchment in 4Qa for a profane list would be consistent with the marl caves being used for storage by the Roman military, encamped in the area, post-68 CE. A lamp Gr4Q 22, indicates the latest use of the cave. This was, I suggest, deposited at the entrance after cave collapse (in 115 CE) covered the manuscripts. However, there is no distinctive

Figure 1.7 Map of the Northern Dead Sea in the Second Temple Period showing settlements, roads and wadis. © Joan E. Taylor.

Bar Kokhba repertoire in the assemblage here, at least as excavated, meaning the cave was not a hiding place for refugees. In terms of the caves in close connection with Qumran during its period of occupation from the 1st century BCE to the early 2nd century CE, prior to the Bar Kokhba Revolt, the definitive repertoire of the artefacts in the caves is extremely limited, with a small range of pottery types and an overwhelming number of cylindrical “scroll” jars and lids, not evidenced in any other caves of the region. The exception is cave 11Q, which has a repertoire of objects that may indicate a later use by refugees.

THE QUMRAN CAVES IN THEIR REGIONAL CONTEXT

Figure 1.8 Amended version of Ziegler’s plan, showing locations of caves yielding remains. Q indicates a scroll cave; GQ indicates the 1952 Cave Survey and PQ is used of Patrich’s survey caves. The placements are largely those of Stephen Pfann, as shown in his own modification of Ziegler’s plan found at http://www.uhl.ac/en/resources/blog/56-years-ago-qumran-cliffs-survey/. The red circles indicate the presence of Bar Kokhba period remains.

33

CHAPTER 2

Cacher et se cacher à Qumrân : grottes et refuges. Morphologie, fonctions, anthropologie Jean-Baptiste Humbert Les “Grottes de Qumrân” tirent leur célébrité de ce qu’elles ont livré les Dead Sea Scrolls. Dans l’imaginaire collectif, l’antre suscite la fable et le merveilleux. Qumrân n’échappe pas à la règle et ses grottes furent l’écrin fabuleux d’un patrimoine où la surenchère paraît sans fin. Elles ont excité curiosité et convoitise. Une investigation hâtive ou désordonnée, des menées plus ou moins clandestines, qui sait ? Ont vandalisé ici et là des cavités ou leur approche (Fig. 2.1). Qu’espérait-on ? Trouver d’autres cachettes ou

Figure 2.2 Les environs au sud de la grotte 11Q en 1951.

Figure 2.1 Fouilles sauvages au sud de la grotte 11Q.

poursuivre une fouille qui n’aurait pas été complète ? Les environs de 11Q ont été bouleversés (Figs. 2.2, 2.3), et le sol de la grotte a été creusé et sur-creusé (Figs. 2.4 et 2.5). Outre le respect dû à un site archéologique qu’il fallait préserver à tout prix, la perte d’informations est dommageable pour la science et entraîne un discrédit pour la recherche quand on sait que le bénéfice de de telles entreprises est presque nul.

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_004

Figure 2.3 Excavations récentes au sud de la grotte 11Q.

Cacher et se cacher à Qumrân : grottes et refuges

35

Figure 2.4 La grotte 11Q après la fouille de 1956.

Figure 2.5 La grotte 11Q en 2013.

En Judée, des gens se sont cachés dans des souterrains. Qui, quand et pourquoi ? Des manuscrits ont été cachés dans les environs immédiats de Qumrân. Par qui, quand et comment ? Autant de questions simples et quand chacune recueille une réponse simple, leur assemblage suscite une nouvelle question. L’événement est là, évident, que Qumrân a été élu dans l’Antiquité comme lieu de convergence de plusieurs initiatives heureuses et d’impératifs contrariés. Cacher y fut une entreprise de dissimulation de dépôts dans des abris banalisés. Les dépôts sont

des jarres cachées dans des grottes, contenant des textes cachés dans des toiles ; fut aussi caché le rouleau de cuivre, et ce sont des faits. Les parchemins et la plaque de cuivre, qui relate à son tour l’entreprise d’une cache1, sont des textes. Qumrân est à notre point de départ un lieu crypté en abyme. Ont convergé des forces religieuses, politiques, historiques. Le fait appartient à l’archéologie et les textes à 1  Nous distinguons les termes cache et cachette. La cache est l’action de cacher et la cachette est le lieu de la cache.

36 l’histoire. Or le fait ressort aussi de la petite histoire et les manuscrits sont malgré tout des objets archéologiques. La recherche de cinquante ans tente en zigzags de formuler une interprétation acceptable. On a espéré que les grottes en cachaient la clé. La tâche s’annonce confuse. La recherche avance parfois grevée d’idées reçues, jaillies d’une approche ignorante ou approximative de l’écologie du lieu, d’une anthropologie contemporaine plus ou moins fondée, gênée par des poncifs archéologiques. Pour éviter une fuite en avant virtuelle, le mieux est encore de se plier aux règles d’une investigation tour à tour prospective et rétrospective. Évitons le rebond sur des idées biaisées, parfois fausses, dont la répétition en accroît le crédit. Dans le dossier des grottes, des foyers de campement ne font pas une période d’occupation, dix ou vingt poteries ne font pas une habitation. On ne peut prendre la partie pour le tout, et l’on ne confondra pas un épisode archéologique avec ce qui n’en est que l’accident. L’accident ne signifie pas et son sens, s’il porte, n’est qu’une indication. Nous pouvons considérer les trois étapes de la démarche dont chacune se plie à sa propre méthode. 1 – Il y a encore à voir : le matériau brut. Ce qui reste du corps d’avant la dissection, en avant et en arrrière : le site et les grottes. Les dissocier équivaut à une décollation. Le site muet dans sa modestie dérange par l’étrangeté de quelques dispositifs. L’observateur hésite entre une anthropologie religieuse ou profane, et bute sur les raisons d’une occupation humaine aux vestiges composites, inclassables, sans parallèle. La complexité du matériau excuse les hésitations et légitime les interprétations divergentes. L’étude n’est pas si bien engagée quand n’est plus intact le potentiel des grottes telles qu’elles sont aujourd’hui, d’abord fouillées et vides pour la plupart, parfois saccagées. 2 – Les archives documentaires sont ce qui a été appris avec ce qui est consigné, des données polyvalentes sur le matériau brut : apparition des manuscrits, découverte des grottes, méthode de la fouille de la khirbeh. Ce qui a été couché par écrit et ce qui a été collecté, des relevés, un journal de fouilles qui est déjà une interprétation, un mobilier enfin. Tout ce qui a été tiré ou appris des bédouins, ce qui n’est pas des moindres. Et en écho, on appelle ce qui a été oublié, négligé, jeté ou perdu. 3 – Le dossier Qumrân : autopsie des archives documentaires. Le débat académique de l’interprétation, au risque de se perdre dans les sables. Ce qu’est la recherche avec le surgissement aléatoire des

Humbert

hypothèses, les aléas de l’archéométrie qui croit en la vérité des nombres, auxquels s’ajoutent les impératifs d’une histoire qui s’impose et qu’accompagne l’autorité des manuscrits devenue monopole. 1

La recherche sur la documentation

Revenir de la documentation toujours en mouvement au matériau brut et inerte est un passage obligé. Nous recevons sans a priori ce qui a été dit et écrit des grottes. R. de Vaux en a publié une archéologie rapide mais correcte, que nous devons respecter comme le document le plus proche des découvertes, presque pris sur le vif. Revenus à la publication, nous acceptons que l’inventaire des mobiliers ne peut pas être modifié puisque de première main. Il reste que l’interprétation anthropologique dans le contexte qumrânien semble bien ne pas toujours convenir et qu’il nous faut revoir les mobiles pour reformuler la proposition. Quand il fut connu que les manuscrits venaient de plusieurs grottes, s’imposa la nécessité d’une exploration systématique de la falaise au sud et au nord de Qumrân. Elle fut menée entre 1952 et 1956c : des centaines de cavités visitées, un certain nombre décrites ; quelques initiatives isolées, certaines non officielles comme celle d’Allegro en 1962/64 ; avec l’occupation des Territoires palestiniens, d’autres explorations ont été tentées sous la houlette de P. Bar Adon (1968), J. Patrich (1984), M. Broshi et H. Eshel (1994), etc. La plus ambitieuse, plus systématique et de plus grande extension, fut celle menée à partir de 1993 par les forces d’occupation, sous la direction de A. Drori et Y. Magen : plusieurs milliers de cavités et de trous ont été visités, un certain nombre décrits. À part les trouvailles publiées, les bons résultats ont déçu puisque les nouveaux manuscrits escomptés dans la proximité de Qumrân ont manqué. Malgré quelques apports fournis par les explorations plus récentes, et qui ont eu l’avantage d’élargir le sujet et de mieux documenter une archéologie régionale, l’étude de de Vaux demeure le document fondateur. De Vaux avait établi que 43 grottes ou simples fentes de rocher offraient une information utile. Il a conçu son interprétation des grottes2 dans le moule de sa version 2  Roland de Vaux, “Archéologie”, in Les ‘petites grottes’ de Qumrân : Exploration de la falaise ; Les grottes 2Q, 3Q, 6Q, 7Q à 10Q ; Le rouleau de cuivre (DJD 3 ; ed. M. Baillet, J. T. Milik, et R. de Vaux ; Oxford : Clarendon Press, 1962), 3-36 ; idem, “Archéologie”, in Qumrân Grotte 4.II : I. Archéologie ; II. Tefillin, Mezuzot et Targums (4Q128-4Q157) (DJD 6 ; ed. R. de Vaux et J. T. Milik ; Oxford : Clarendon Press, 1977), 3-22 avec de bons plans-coupes ; Gerald L. Harding, “Introductory,

37

Cacher et se cacher à Qumrân : grottes et refuges

d’une communauté qumrânienne sectaire, selon ce qu’il avait appris des sources et de la Règle, dans un cadre préétabli essénien et communautaire. La ruine de Qumrân fut pour lui monastique, et ses alentours cénobitiques. Il s’y est tenu sans revenir sur ses intuitions, parce l’interprétation justifiait une bibliothèque de centaines de manuscrits sur le site même, au cœur d’une communauté nombreuse. Aujourd’hui, une telle version n’est pas crédible. Des doutes ont été émis, il est improbable que le site ait été communautaire, et l’idée s’insinue que l’ensemble des rouleaux est un regroupement. Il a fallu tout relire et aller voir. La fréquentation répétée des lieux nous en a restitué le cadre étendu et escarpé. Il apparaît nécessaire de reconsidérer le profil écologique des lieux que de Vaux avait évalués, de la palmeraie à la falaise avec le site archéologique en charnière entre elles. Les grottes sont aujourd’hui dépouillées mais leur morphologie demeure. Archives et nouvelles visites fournissent assez de données pour dresser le portrait des grottes à placer à côté de celui de la ruine, deux portraits debout, posés sur la table encadrant le livre fermé des manuscrits. Nous voici revenus sur le contexte des cachettes puis de l’entreprise qu’elles supposent, menée par une société que l’on peine à saisir. Quel était le peuplement autour de la mer Morte ? Quelles étaient les conditions climatiques et économiques de la région qui, au Ier siècle, avaient permis ce peuplement ? Quels étaient les possesseurs des rouleaux et les acteurs des caches ? Qui furent les témoins de connivence ? Nous voudrions montrer que tout fut possible dans le contexte d’un peuplement dispersé et d’un lieu si peu fréquenté que les garanties de discrétion et d’oubli y étaient les meilleures à ce moment-là ; que le site archéologique fut surévalué, la conception d’une “bibliothèque” abritée en ce lieu est anachronique si on la conçoit comme celle de l’établissement. Quand on considérera notre proposition infra avec la probabilité que la série des “petites grottes” a contenu une grande quantité de manuscrits perdus, l’accumulation des rouleaux est si exceptionnelle, dans la modeste thébaïde que fut Qumrân, qu’une logique historique fait défaut. Reste un fait massif difficilement récusable : Qumrân aura tenu le rôle central dans l’entreprise des caches.

the Discovery, the Excavation, Minor Finds”, in Qumran Cave 1 (DJD 1 ; ed. D. Barthélemy et J. T. Milik ; Oxford : Clarendon Press, 1955), 3-7 ; Roland de Vaux, “La poterie”, in DJD 1 :8-17.

2

Le peuplement de la mer Morte

Il nous a semblé utile de décrire le peuplement de la mer Morte pour mieux estimer le rôle des acteurs dans les deux siècles avant et après le tournant de l’ère. Des manuscrits ont été cachés en un endroit circonscrit et le moment de la cache s’est déroulé dans un cadre social et topographique spécifique. Il nous a fallu fréquenter la mer Morte pendant quarante ans pour pressentir ce que le rapport de l’homme à sa géographie y eut de particulier. La mer Morte fut un monde clos, séparé et dans une certaine mesure “étranger”, en marge de la Judée. Qumrân est au fond de la fosse de la mer Morte où contrastent climat et paysage. On ne cessera de souligner que le lac Asphaltite forme une région à part, isolée par le désert de Judée, peu aisée à traverser au risque de s’y perdre ou de mourir de soif, et protégée à son bord par des falaises abruptes dont il fallait connaître les passes. L’accès naturel à Qumrân se fait depuis Jéricho, par le nord, sans accident. Le bassin lui-même reçoit moins de 100 mm de précipitations par an : il est semi-désertique avec un haut relief rapproché où les terres arables sont rares au débouché des wadi, et cultivables quand des cours d’eau pérennes permettaient l’irrigation. La mer Morte rebute en même temps qu’elle offre une retraite paisible, et les deux temps ont commandé le va-et-vient des hommes. Le dattier acclimaté depuis les temps les plus reculés, s’accommodant du sol salin, formait au fond des baies des îlots de végétation isolés les uns des autres. Plus que d’oasis, il vaudrait mieux parler de palmeraies, et les palmiers chers à Pline y sont en place. En dehors d’elles, les sols sont rocheux ou marneux, et stériles par manque d’eau. Les rives de la mer Morte n’ont donc fourni qu’une agriculture parsemée et ingrate. Leur réputation non usurpée d’avoir produit des baumes ne peut être exagérée, le baumier n’y a pas atteint une production à grande échelle. La récolte des bitumes, que l’on sait sous tutelle d’un monopole, n’était pas quotidienne. Il serait bien étonnant que la commercialisation du luxe ait eu des retombées locales sensibles. La mer Morte n’a jamais été qu’une région agricole pauvre parce que sèche et, puisque pauvre, peu et pauvrement peuplée. En raison de cette austérité naturelle, deux questions se posent : “quelles étaient les raisons d’habiter la mer Morte ?” et “qui habitait la mer Morte ?” Les palmeraies isolées les unes des autres par des barrières rocheuses purent favoriser des formes d’autonomie sociale3 qui communiquaient entre elles par mer. Les plus 3  Les palmeraies devaient constituer des domaines organiquement liés aux forteresses qui les dominaient. Elles ne jouissaient pas d’une totale indépendance économique sous la tutelle de “propriétaires

38 vastes, à cause de l’abondance des sources, surplombent la rive orientale comme ez-Zara-Callirhoé, ou s’étendent en retrait de la Lisan, au débouché des torrents des Wadi Kerak et al-Hasa avec les longues terres cultivables de Mazraa, Ghor es-Safi, et tout au sud Zoora, l’ancienne Zoara4. Sur la rive judéenne beaucoup plus avare en eaux, seules les palmeraies d’Aïn Boqeq (Umm al-Baghaq), d’Engaddi et de Qumrân-Aïn Feshkha jouirent de quelque développement5. Il faut entendre de tout cela que la rive transjordanienne, mieux dotée en ressources vivrières, était mieux peuplée que la rive occidentale. Qumrân ne pouvait ignorer l’autre rive et en premier ez-Zara, le domaine de Machéronte à quelques quinze kilomètres de navigation, où le magnifique port d’Hérode était relié à la Judée et Jérusalem par la cale royale de Khirbet Mazen6. Il est probable que la forte cohérence géographique du bassin de la mer Morte a entraîné, pour les populations qui y vivaient retirées, une singularité culturelle, peut-être même une identité commune aux deux rives, en dépit des tutelles politiques iduméenne à l’est et judéenne à l’ouest. On sait que les esséniens vivaient dans la région qui n’était certainement pas peuplée que d’esséniens. En revanche, restreindre l’implantation des esséniens à Qumrân est un non-sens et en contradiction avec les sources. À ce jour, il n’y a pas en Transjordanie de témoignage d’une présence essénienne démontrée. Cependant, on en trouve une indication dans Dion Chrysostome,

aristocratiques” qui profitaient de leurs productions agricoles. La résidence hasmonéenne qu’il faut voir à l’origine de Qumrân aurait été la forteresse-villégiature d’hiver de l’Hyrcanion Le chemin qui, dans la gorge, franchit la falaise a été l’objet d’aménagements maçonnés en pierres sèches ou pavés qui témoignent du soin que l’on avait pris pour relier les deux places. La passe reliait l’Hyrcanion dans son environnement désolé aux jardins de Qumrân et de Aïn Feshkha. Le modèle de “la forteresse et son jardin” se répète, en symétrie sur le littoral oriental de la mer, à ez-Zara, domaine de Machéronte. Des résidences aristocratiques hérodiennes y ont été fouillées. La distance entre la place forte et son jardin est la même dans les deux cas. Ces domaines offraient une économie certes modeste mais non négligeable. De plus, les deux places étaient reliées par voie maritime. Cf. Jean-Baptiste Humbert, “L’espace sacré à Qumrân : Propositions pour l’archéologie”, RB 101 (1994) : 161-214. 4  Qui fut aussi le domaine agricole du gros site du Bronze ancien de Bâb edh-Dhrah. 5  Aïn Ghuweir entre Engaddi et Aïn Feshkha possède quelques terres mais moins d’eau. 6  Yizhar Hirschfeld pense avoir retrouvé le monnayage du péage déjà sous Alexandre Jannée : Yizhar Hirschfeld et Donald Ariel, “A Coin Assemblage from the Reign of Alexander Jannaeus Found on the Shore of the Dead Sea”, IEJ 55 (2005) : 66-89.

Humbert

vivant au premier siècle et contemporain de Qumrân, cité au IVes. par Synesius à propos d’une note sur les esséniens. Dion les nomme une polis, c’est-à-dire une entité politique ou religieuse établie au sud de la mer Morte7. R. North peut avoir vu juste quand il propose d’étendre vers le sud le Pays de Damas, avec référence aux tentes, jusqu’à la rive orientale de la mer Morte8. L’auteur recense les faits historiques et archéologiques pour réévaluer la place et le rôle des Transjordaniens dans les premiers siècles avant et après le tournant de l’ère. Au début du second siècle, les Romains avaient nommé l’Outre-Jourdain Province d’Arabie puisque arabe désignait une population sous tente. Aujourd’hui quelques rares nomades résistent encore à la sédentarisation. Le contrôle économique et politique nabatéen s’étendait au moins jusqu’au nord de Bostra dans le pays de Cham (Damas). L’essénisme fut sans doute un mouvement géographiquement dispersé ; on en retrouve un écho dans la polis de Dion quand sont conjuguées les rares données de la démographie antique de la mer Morte. Qumrân a pu s’imposer comme un lieu de rassemblement occasionnel des adeptes des deux rives, parmi d’autres mais le mieux désigné au plus proche de Jérusalem9. 3

Vers un habitat saisonnier

Au soleil brûlant de la canicule succèdent le tempéré des demi-saisons et la douceur des hivers. Suivant les modes hellénistiques, quand la société hasmonéenne eut pris goût de la “campagne”, des villégiatures d’hiver plus ou moins élaborées surgirent autour de la mer Morte. L’archéologie les a reconnues sur la côte occidentale où s’échelonnent Umm al-Baghaq, Engaddi, Aïn Turabi, Aïn Ghuweir, Aïn Feshkha – Qumrân enfin. Bien qu’écarté, le séjour de Jéricho y tient une place de choix puisque royale ; l’on est alors amené à attribuer ces domaines à l’aristocratie hasmonéenne de Jérusalem. À l’est, l’unique villégiature de ez-Zara-Callirhoé dépendait de la place forte de Machéronte, elle aussi dans l’orbite hasmonéenne 7   Cité in Jean-Baptiste Humbert, “Reconsideration of the Archaeological Interpretation”, in Khirbet Qumrân et ‘Aïn Feshkha. Vol. 2 : Études d’anthropologie, de physique et de chimie (NTOA.SA 3 ; ed. J.-B. Humbert et J. Gunneweg ; Fribourg/Göttingen : Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 427. 8  Robert North, “The Damascus of Qumran Geography”, PEQ 87 (1954) : 34-48. 9  Humbert, “Reconsideration of the Archaeological Interpretation”, 434.

Cacher et se cacher à Qumrân : grottes et refuges

de la Judée, poste d’Hérode avancé en Nabatène10. Les villégiatures n’étaient fréquentées qu’en hiver ou dans des temps d’insécurité. Le mode de vie alterné été-hiver y semble non seulement adapté mais salutaire aux populations. À l’été trop chaud, le mouvement séculaire des hommes est encore aujourd’hui, pour les derniers nomades transjordaniens, de regagner les hauteurs11. Il ne serait pas étonnant que dans l’Antiquité, les récoltes maraîchères achevées, la majorité de la population ait quitté la région du fossé. Sur place, une maintenance et un gardiennage étaient assurés. On redescendait à l’automne, au temps de récolter les dattes. Le schéma épisodique convient alors à un Qumrân hasmonéen qui accueillait une villégiature aristocratique qui gérait ses jardins. Le statut de résidence, que l’on y veut temporaire comme dans les autres places, est compatible avec une exploitation agricole : la grande maison qui dominait la palmeraie de son escarpe, ne pouvant en aucun cas lui être étrangère, devait logiquement en être la propriétaire. Il y eut toujours à Qumrân une présence, parfois réduite, aussi bien dans la résidence aristocratique qu’au temps d’un un établissement sectaire, voire e­ ssénien : une société qui se voulait séparée pouvait accepter d’y vivre temporairement ; nous avons renoncé à y établir une communauté stable. L’aristocratie ne faisant que passer, restent ceux qui cultivent. Depuis cinquante ans, l’archéologie dans la région de la mer Morte a mis au jour à répétition des édifices qui sont pour la plupart des architectures hasmonéo-hérodiennes : Burj el-Bahar, Qumrân, Kh. Mazen, Aïn Turabi, Engaddi, Masada, Umm al-Baghaq, ez-Zara. On doit ajouter l’Hyrcanion et l’Hérodion. Paradoxalement les installations domestiques sont rares et on ne peut pas dire qu’on ne les a pas encore découvertes, les explorations s’étant multipliées depuis vingt ans. Si l’on excepte l’agglomération d’Engaddi qui fut un vrai village, les quelques exemples d’installations de peu d’étendue ou 10  Machéronte fondée par Alexandre Jannée. Excepté les trésors, la majorité des monnaies recueillies à Qumrân ont été frappées sous ce prince ; les deux places ont été contemporaines et l’on doit supposer qu’elles étaient liées. Détruite par Gabinius, elle a été reconstruite par Hérode en 30 av. J.-C. Le trajet maritime de Machéronte aboutit à la cale de Kh. Mazen puisque le Cédron est l’itinéraire le moins accidenté pour rejoindre Jérusalem. Cependant il dut exister un débarcadère desservant Aïn Feshkha qui n’était pas nécessairement en dur. 11  Ainsi que font les quelques bédouins qui campent sur la rive orientale. La palmeraie de ez-Zara est un pâturage d’hiver pour de petits troupeaux de chèvres. La pâture n’est guère adaptée au mouton.

39 sommaires sont à Aïn Ghuweir, au sud de Qumrân12, les huttes d’ouvriers agricoles sur une corniche à Engaddi13, et sur la crête, le groupe des maisons de Kh. Samra qui domine l’oasis14. Constatant la précarité de l’habitat domestique, nous relevons le paradoxe qu’en revanche, les cimetières jouxtent les terres les plus généreuses, au sud-est de la mer Morte : cinq mille tombes à Qazone15 au bord des exploitations modernes de Mazraa, des milliers à Zoora (Zoara) au bord des anciennes cultures; on peut ajouter Feinan, à quelques dizaines de kilomètres plus au sud, où des milliers de sépultures jouxtent les berges fertiles au débouché du Wadi Dana16. La modicité des installations domestiques, en contraste avec la taille des cimetières, renforce l’hypothèse d’un habitat saisonnier. On mourait loin de chez soi. Le peuplement de la mer Morte, peu dense ne fut qu’agricole. La proposition de voir à Qumrân une industrie potière régionale surprend, le marché fermé fut si réduit qu’il aurait été des moins ­rentables17. Il est toujours possible de cuire n’importe quelle terre mais sans produire une céramique de la qualité requise. Il n’est pas encore prouvé que les alluvions recueillies dans les débouchés des wadi aient fourni le matériau adapté à une industrie potière. La terra rossa, l’argile de Moza, qui convient, gît sur les crêtes de la Judée ou en symétrie sur le plateau transjordanien. L’essentiel de la vaisselle des habitants des bords de la mer Morte venait des montagnes. 12  Pesach Bar Adon, “Another Settlement of the Judean Desert Sect at ‘En el-Ghuweir on the Shores of the Dead Sea”, BASOR 227 (1977) : 1-25. 13  Yizhar Hirschfeld, “A Settlement of Hermits above ‘En Gedi”, Tel Aviv 27 (2008) : 103-55. 14  Pesach Bar Adon, “Excavations in the Judean Desert”, ʿAtiqot 9 (1989) : 83-85 [Hébreu]. 15   Le cimetière de Qazone est en continuité de l’importante nécropole de Bab edh-Dhrah (Bronze ancien et moyen). Pour le cimetière : Konstantinos D. Politis, “The Discovery and Excavations of the Khirbet Qazone Cemetery and its Significance Relative to Qumran”, in Qumran, the Site of the Dead Sea Scrolls: Archaeological Interpretations and Debates; Proceedings of a Conference held at Brown University, November 17-19, 2002 (STDJ 57 ; ed. K. Galor, J.-B. Humbert, et J. Zangenberg; Leiden : Brill, 2006), 213-21. Politis souligne le caractère transjordanien du contexte archéologique. 16  George Findlater et al., “The Wadi Faynan Project : The South Cemetery Excavations, Jordan 1996 ; A Preliminary Report”, Levant 30 (1998) : 69-83 (et 71, remarque sur les tombes nabatéennes). 17  Yizhak Magen et Yuval Peleg, “Back to Qumran: Ten Years of Excavations and Research, 1993-2004”, in Galor, Humbert, et Zangenberg, Qumran, the Site of the Dead Sea Scrolls, 92-94.

40

Humbert

Que la région ait été un endroit à part et peu fréquenté offre l’avantage d’un refuge à qui était menacé par les polices, on dirait aujourd’hui aux maquisards et par exemple, tels furent en leur temps les zélotes ; l’avantage aussi d’un retrait à qui désirait se couper du monde urbanisé, ainsi que firent des esséniens et plus tard les anachorètes ; à qui, enfin, voulait mettre à l’abri un patrimoine. Il n’était pas utile d’errer ailleurs dans le désert de Judée. Autrement il n’y avait rien à y faire. Quand les circonstances l’imposaient, les habitants des palmeraies gagnaient les villes et les villages des montagnes, à l’est comme à l’ouest, par des cheminements balisés18 : Engaddi montait à la Tombe de Lot à Beni Naïm (vers Hébron), Qumrân à Jéricho par le Ghor et Jérusalem par le Cédron, ez-Zara montait à Machéronte. On s’y croisait exceptionnellement car depuis toujours l’homme va partout, mais rarement puisque telles n’étaient ni l’utilité ni l’habitude dans un désert, hostile par nature. Pour les sociétés isolées les unes des autres, il était beaucoup plus facile et rapide de se rencontrer par la navigation que par le franchissement des promontoires. Au premier siècle, le niveau de la mer était beaucoup plus haut qu’aujourd’hui et noyait le pied des falaises. La circulation terrestre était limitée et les trajets fractionnés. Les falaises n’étaient pas fréquentées parce qu’escarpées, stériles, inutiles. La dizaine de grottes entre Qumrân et Engaddi et au sud d’Engaddi, où furent dégagés des sédiments archéologiques substantiels, n’infirment pas le propos mais le confirment, puisque ceux qui les ont élues ont cherché l’inaccessible et témoignent de la volonté de se soustraire à la vue. Que des familles y aient vécu relève de l’exception. 4

Habiter Qumrân et les alentours

Maintenant, peut-on savoir enfin qui habitait Qumrân ? Pour l’archéologue, la khirbeh dans son ensemble présente tous les aspects d’un établissement compact, encombré et fermé sur lui-même, sans la planification préalable d’une communauté organisée qui l’aurait programmée comme de Vaux en avait forgé l’idée. Nous avons proposé d’y voir une grande maison hasmonéenne aux espaces largement conçus, désaffectée puis réinvestie par un groupe au mode de vie radicalement différent. Les espaces ont alors été redistribués, les chambres compartimentées ou réaménagées, les accès déplacés et rétrécis, les façades aveuglées. Des ajouts périphériques en ont rompu la symétrie primitive ; leurs fonctions restent imprécises. Les surfaces 18  Il faudra attendre le monachisme aux V-VIe s. pour y rencontrer une population organisée.

couvertes ne sont pas si nombreuses et l’établissement demeure modeste. L’encombrement de l’espace vital correspond à cette période d’occupation et la plupart des installations furent des ateliers. La circulation interne, contrainte par maints obstacles, ne convient pas à un grand nombre, le logement pour les habitants manque, l’établissement n’a rien de communautaire. Nombreux sont ceux qui résistent à abandonner la version communautaire comme si l’oublier mettait en danger l’édifice d’une interprétation reçue, dictée du pupitre des manuscrits, refusant de “jeter le bébé avec l’eau du bain”. La dimension communautaire de la khirbeh est exigée parce qu’elle répercute en écho le nombre des manuscrits. Sans une communauté au pied des grottes, les manuscrits seraient-ils orphelins ? Le nombre des rouleaux suppose un soutien communautaire quand le site ne l’est pas. Leur nombre laisse pressentir celui des acteurs, artisans pour produire les rouleaux, copistes, bibliothécaires, lecteurs, sauveteurs des lots, que le site trop petit ne peut pas contenir. L’incapacité de Qumrân à accueillir du monde et le mode de vie épisodique de la population locale achèvent d’affaiblir l’interprétation d’une société pluridisciplinaire qui jusqu’ici avait prévalu. La filiation des textes s’estompe. Il faut chercher ailleurs leur parenté et si l’on accepte que la communauté fut une polis diffusée en Judée et sur les rives de la mer Morte, l’identité des acteurs se précise et la succession des étapes s’organise. La cohérence tisse, dans les grottes, le lien entre les manuscrits et le site. Qumrân a joué un rôle dans l’entreprise des caches, mais lequel ? La présence de jarres de forme non encore attestée ­ailleurs19, identiques dans les grottes et sur le site, le prouve. La communauté élargie, politique et religieuse, à laquelle appartenaient les manuscrits est alors présente dans les grottes par procuration et les grottes appartiennent au site. Il vaut mieux renverser la proposition initiale et dire que Qumrân appartient à la communauté élargie, à une secte essénienne ou non. De Vaux, qui avait besoin pour sa démonstration d’une communauté recentrée à Qumrân, a campé sur les pentes et couché dans les grottes ceux des adeptes que le site ne pouvait abriter ou ceux qui, ayant enfreint la Règle, étaient exclus temporairement. La proposition d’y loger une partie de la supposée communauté de Qumrân n’est pas conforme aux conditions de la géographie physique et humaine. Rendons-lui la justice que, l’ayant proposée, il n’a pas insisté plus qu’il ne fallait. Le méplat caillouteux

19  En dépit de quelques propositions, les jarres tubulaires dites “à manuscrits” n’ont été trouvées qu’à Qumrân.

Cacher et se cacher à Qumrân : grottes et refuges

au pied des falaises aurait été la place du campement20 et les grottes auraient été un habitat favorisé. La proposition reflète une concentration sectaire centripète sur le lieu de Qumrân qui reste indémontrable, et l’essénisme ne peut se réduire à un centre dit communautaire. Il faut mieux regarder. Un campement de deux siècles laisse des traces au sol et rien de tel ne se voit : empreinte effacée des tentes, avec autour d’elles les cheminements et les rejets d’une occupation longue, tessons, débris de faune, matériau de rebut, etc. De plus, nous n’accepterons pas l’éventualité qu’une société ait habité deux siècles sous des tentes sans construire en dur dans un paysage où la pierre est partout. Habiter, ne serait-ce que quelques décennies, de façon permanente dans un abri troglodytique, entraîne normalement des aménagements domestiques, même sommaires – abaissement des sols sous un plafond trop bas, rétrécissement des ouvertures et aménagement des abords – et rien de tel ne se voit nulle part. À la longue, les extérieurs auraient été redessinés par le piétinement : les sentiers soulignés, l’avant des porches plus ou moins nivelé, entouré par les rejets et les déblais. Rien de tel n’a été remarqué devant aucune des cavités des alentours, pas de dépôts, rareté des tessons, pas de déblais. La falaise n’a révélé aucun abri domestique à long terme et il faut renoncer à croire qu’elle fut habitée. De Vaux a surexploité le caractère d’habitation des grottes. La relation des explorations successives mentionne quelques murs de protection, aménagements sommaires de fugitifs ou de bergers, dont l’édification ne peut être datée. Les cendres accumulées, en deux endroits épaisses, ne trahissent qu’une fréquentation aléatoire pendant des siècles, ou seulement répétée dans certains cas : abri, maquis, refuge, hivernage. Les foyers qu’on y voit sont des feux plus ou moins anciens et les gens qui passèrent ne furent pas nombreux. Rien n’indique une longue gestion de l’habitation, même sommaire, d’une communauté à Qumrân pendant un siècle ou deux. Si les grottes n’ont pas été habitées au sens strict, elles n’en sont pas pour autant toutes stériles. Comme ailleurs dans le bassin de la mer Morte, de rares traces chalcolithiques et du Fer ont été détectées dans les abris qu’il faut se garder de surévaluer. De Vaux, qui les a soigneusement notées par ailleurs, compte dans l’enregistrement 20  Il est possible que l’idée du campement lui soit venue du Document de Damas où il est dit que les exilés dans le Pays de Damas habitaient sous des tentes. Il est vrai que l’armée romaine avait vécu sous tente dans les camps de siège de Masada et sur les crêtes dominant Engaddi, mais le temps d’une campagne militaire dans des conditions d’urgence stratégique et non à long terme.

41 du mobilier : 156 jarres environ de différents types, dont 82 dites “à manuscrits”, sans compter les autres vases, et 116 couvercles. En revanche, l’interprétation de cachettes s’affirme quand l’habitat s’estompe et il n’est guère permis d’hésiter sur la fonction des “grottes”. Certains avancent comme preuve d’une véritable occupation les objets retrouvés en formes complètes – quatre petites cruches, une marmite, deux gobelets, trois lampes, moins de dix bols et une assiette pour 40 sites ; ce rare mobilier éparpillé dans les abris ou au porche des grottes ne suffit pas pour les croire habitées de façon permanente à un moment ou à un autre. Que les explorations plus récentes aient augmenté l’inventaire ne change pas la donne. Il faut y voir au contraire les témoins de passages épisodiques, et il semble raisonnable d’y reconnaître un mobilier abandonné par des réfugiés. Depuis l’époque chalcolithique, les cavités de la rive occidentale de la mer Morte ont été des refuges, toujours temporaires ; on ne vit pas dans des falaises sinon pour s’y retrancher. Les grottes n’ont pas été des habitations puisque, si l’on veut s’installer dans le voisinage, en dehors de la khirbeh, le mieux est encore de vivre dans les jardins où les eaux sont pérennes21, sous les palmiers ainsi que Pline l’a dit. 21  Le source d’Aïn Feshkha pose-t-elle problème pour faire de la palmeraie de Qumrân un lieu habitable ? Son débit assez fort allait d’abord à l’irrigation des jardins. Elle était prisée par les juifs religieux comme bain rituel, toujours disponible. Son eau est-elle potable ? Elle n’est certes pas nocive, nous avons vu des troupeaux de chèvres y boire, nous en avons bu nous-même sans répulsion quand elle fait un thé amer. Les oasis de Bahariyah, Farafra, par exemple dans l’ouest égyptien, ont aussi des sources chaudes à l’amertume variable qui n’ont pas empêché de vivre et d’y prospérer. Et lorsque le niveau de la mer Morte était encore haut, nous avons vérifié tout au long du rivage de Feshkha, vers le sud, que des points d’eau (d’émission 40 litres/minute pour ne pas parler des ruissellements) donnaient des eaux aux températures et goûts variés et parfois ni salées ni soufrées. La maison hasmonéenne originale a été transformée en zone industrielle. M. Bélis y voit une fabrique d’indigo, cf. Mireille Bélis, “Des textiles : catalogue et commentaire”, in Humbert et Gunneweg, Khirbet Qumrân et ‘Aïn Feshkha, 247-50. De Vaux avait proposé une tannerie pour les peaux et y avait renoncé quand les analyses de résidus avaient montré l’absence de tanin. Mais l’eau de la mer Morte peut traiter les peaux sans tannin par son pouvoir stérilisateur de vermines. Une telle industrie n’est pas incompatible avec l’hypothèse de la production d’indigo. Le traitement des peaux a pu s’effectuer ailleurs dans la palmeraie qui n’a pas été explorée dans le détail. Le dossier est peut-être à reprendre. Il reste qu’un artisanat de tannerie se concevait dans l’hypothèse d’une grande bibliothèque à Qumrân même ; ce qui n’est guère tenable aujourd’hui.

42 5

Humbert

Les cavités naturelles

5.1 Morphologie géologique Les “Grottes de Qumrân” ne sont pas des grottes à proprement parler. Dans la grande cassure du Rift, la falaise occidentale résulte d’une bascule sismique, récente. Elle ne recèle pas les cavernes hydrologiques des vastes réseaux souterrains des reliefs karstiques, dotés d’habitats troglodytiques, souvent profonds, fréquentés de la préhistoire à l’époque moderne. À Qumrân le mot “grotte” est inapproprié, mais il est reçu. Les creux sont des accidents produits par une mécanique architectonique, brutale : simples fissures, anfractuosités, failles de fractures, abris sous roche, poches de décollement, d’affouillement, couloirs de suffosion. À de rares exceptions, toutes ces cavités manquent d’amplitude, sont peu aménageables et, au mieux, n’offrent que des abris. 5.2 Des refuges et des cachettes Qumrân a été une palmeraie tranquille, naturellement défendue. La falaise de Qumrân est une passe difficile à franchir, escarpée, qu’il vaut mieux connaître pour ne pas trop hésiter ou se perdre. La route principale, vers Jéricho au nord et Aïn Feshkha au sud, évitait le pied de la falaise découpé par les wadi, et longeait le rivage en traversant les jardins. La falaise, où personne n’allait, était l’endroit idéal pour une entreprise que l’on voulait mener en toute discrétion : un endroit ni habité ni fréquenté, aux alentours arides, en un mot inutile même aux quelques chevriers, trop loin des eaux avec des rochers qui n’ont rien d’un pâturage. La chèvre perdue que l’on cherche dans un trou est un conte bédouin pour tromper les étrangers. Une dizaine de fausses grottes furent des refuges à qui voulait disparaître ; dans une solitude hostile, le millier d’anfractuosités peu spacieuses, plus ou moins visibles, inaccessibles sans escalade ou difficiles d’accès, convenait exactement pour cacher son bien22. Les agriculteurs peu nombreux de la palmeraie, en contrebas, n’étaient pas vraiment en vue de qui circulait dans les rochers. On ne le voyait pas non plus. Toutefois si l’endroit présente toutes les garanties pour mettre en sécurité ce que l’on cherche à soustraire ou à mettre au 22  Les bédouins de l’époque moderne n’ont pas fait autrement, dissimulant dans une fente de rocher qui un manteau, qui un outil, qui verres et théière enroulés dans un chiffon. Dans le Sinaï, on a vu, pendues aux arbres, des tentes roulées en attente du prochain campement. La découverte de la grotte 1Q n’aurait-elle pas été faite par un bédouin qui ne cherchait rien mais qui voulait cacher quelque chose ? M. Bélis suppose plutôt que le caillou jeté était destiné à sonder la profondeur de la fissure.

secret, de Qumrân même on ne pouvait l’ignorer. Les résidents furent partie prenante de l’entreprise de la cache, ce qu’avec la distribution des jarres nous savions déjà. Que des cachettes soient restées inviolées jusqu’au XXe s. prouve que le choix de Qumrân était le bon. L’isolement et la complexité du relief, combinés avec la dispersion des dépôts, furent la garantie du secret. Les milliers de trous, cavités et crevasses, fissures, fentes et failles, devaient décourager ou retarder une éventuelle perquisition, comme “chercher une aiguille dans une botte de foin”. Les difficultés physiques inévitables, les efforts déployés et le temps passé aux explorations modernes le montrent. La dispersion topographique des cachettes augmente l’efficacité du secret. La roche est souvent friable, les ressauts et quelques à-pics sont dangereux à franchir, les pierriers sont raides. La lenteur de tout déplacement, dans un relief tout en verticales séparées par des éboulis instables, oblige d’être observé ou au moins vu de Qumrân. À plus forte raison quand, avec une crête abrupte, on ne peut accéder à la falaise que par le bas. D’en haut, sans être vu, on voit l’ennemi qui vient. Sécurité de conservation, aussi. Nul n’ignorait dans l’Antiquité que la rareté de la faune et la grande sécheresse évitaient la corruption des matières organiques à moyen ou long terme. La falaise, face à l’est, est protégée des vents froids et violents de l’ouest et de la pluie quand parfois elle vient. Les “grottes à manuscrits”, choisies sèches, à l’écart des ravines et des ruissellements, donc saines, ont rempli toutes les conditions d’un entrepôt sûr et stable. 5.3 Les modalités de la cache En plus des quatre grottes de la falaise 1Q, 2Q, 3Q, 6Q, 11Q qui ont livré des manuscrits, il faut mentionner les 18 sites susceptibles d’avoir été des asiles pour des dépôts : GQ1, GQ2, GQ3, GQ7, GQ12, GQ15, GQ17, GQ19, GQ21, GQ22, GQ28, GQ29, GQ30, GQ31, GQ32, GQ37, GQ39. Douze d’entre elles (GQ1, GQ2, GQ3, GQ7, GQ12, GQ17, GQ19, GQ28, GQ29, GQ31, GQ32, GQ39) recélaient en tout 32 jarres cylindriques dites “à manuscrits” et 66 couvercles23. La probabilité que celles-là aient contenu des textes reste forte : le type spécifique de la jarre “à manuscrits” leur a été associé avec une quasi certitude, sans toutefois qu’on puisse affirmer qu’elles ont abrité des rouleaux. L’évaluation perd de sa précision quand on tient compte que nombre de grottes ont été pillées ou vidées de leur mobilier dès l’Antiquité ou par les bédouins24. Le nombre exact de mises à l’abri de manuscrits est donc impossible à établir. Ose-t-on estimer leur nombre à plus de vingt 23  De Vaux, in DJD 6 :14-15. 24  Bélis, “Des textiles”, 238-41.

43

Cacher et se cacher à Qumrân : grottes et refuges

Figure 2.6 Les anfractuosités autour de la grotte 2Q.

sites ? Le chiffre n’est pas invraisemblable et il nous faut accepter que le nombre des rouleaux cachés fut probablement plus du double de ce qui a été recouvré. L’ampleur de la cache s’en trouve accrue. Une cachette est d’autant plus efficace qu’elle est discrète. Celles qui ont été choisies, en général hors de la vue de qui longeait la falaise, offraient de préférence assez de profondeur en arrière d’une ouverture resserrée, type 1Q, ou une position perchée, type 2Q (Fig. 2.6). Celles dont l’orifice était trop visible ont dû bénéficier d’un camouflage qui banalisait leur approche. Il semble logique que les ouvertures aient été dissimulées, maquillées par des éboulis de blocs et de pierraille comme 3Q. On suppute que les plus vulnérables ont été pourvues en hâte ou en catastrophe, au dernier moment, et mal protégées. Il est techniquement possible que d’autres niches inconnues

soient encore scellées par des éboulis. La configuration des abords de la grotte 1Q en est un bon exemple : l’éboulis qui dissimulait l’entrée ne semble pas une formation naturelle et seule la lucarne en contre-haut, mal praticable, serait restée ouverte. De Vaux a noté : “Une recherche a été faite pour déterminer le contour exact de l’entrée primitive”25. En place en 1949 lors de la découverte, le remblai a été enlevé pour restituer l’entrée primitive et faciliter l’accès, les photographies en rendent compte. Sur le document 1Qa, des ouvriers entament un éboulis sous lequel gît une poche de cendres (Figs. 2.7 et 2.8).

25  De Vaux, in DJD 3 :9.

44

Figure 2.7 L’entrée de la grotte 1Q en cours de dégagement ; les ouvriers étendent une poche de cendres sous l’éboulis.

Figure 2.8 Enlèvement de l’éboulis devant l’entrée basse de la grotte 1Q.

Humbert

45

Cacher et se cacher à Qumrân : grottes et refuges

La valeur des cachettes varie. La grotte 3Q n’est qu’à un jet de pierre du cheminement au pied de la falaise, heureusement en contre-haut et dans un creux du rocher. L’ouverture, telle qu’on la voit aujourd’hui, a été largement

exposée. L’accès avait été empêché par de gros blocs et des gravats qu’il a fallu écarter. Quand les notes en disent peu les photos le montrent (Figs. 2.9a et 2.9b).

Figure 2.9a

Le déblai moderne après dégagement de l’entrée de la grotte 3Q.

Figure 2.9b

Les ouvriers dégagent les gros blocs qui scel­lent l’entrée de 3Q.

46

Figure 2.10

Humbert

Entrée de la grotte 11Q.

Assez vaste poche de décollement, elle s’élargissait sans hauteur sous plafond : “La grotte était très basse et lorsque les jarres étaient intactes, elles devaient presque toucher le plafond”26. L’ouverture, basse, était facile à occulter et les jarres auraient pu être remisées tout au fond, dans une petite chambre et un diverticule qui ont livré fragments de parchemins et fragments de cuir, et fort peu de tessons. Cependant, devant les petites chambres : “Une grande quantité de jarres et de couvercles brisés, mêlés aux débris du plafond et sans remplissage de terre”27. Les jarres ont pu être extraites du diverticule ou de la petite chambre et cassées là où l’espace était plus praticable. L’absence de terre dans les interstices des fragments du plafond tombé indique que la manutention des jarres fut récente et bédouine. La cache antique avait été repoussée au fond. Le rouleau de cuivre avait été posé simplement à l’entrée de la chambre du fond : “Juste à l’angle nord de la chambre, deux rouleaux de cuivre inscrits étaient disposés l’un au-dessus de l’autre, contre la paroi rocheuse”28. La cachette était riche, de Vaux a enregistré trente jarres et compté vingtet-un couvercles. Il s’agit certainement de la cachette de Qumrân la plus abondante, avec les grottes 1Q et GQ29, et si elle a pu rester inconnue jusqu’à la période moderne, nous devons admettre sa fréquentation dans l’Antiquité. La grotte 11Q s’ouvre sur le plateau littoral, exposée en pleine vue du chemin en contrebas (Fig. 2.10). Nous noterons l’approche aisée et le paradoxe d’aucune trace de fréquentation après 68, la cachette était sûre. Ici aussi 26  De Vaux, in DJD 3 :7. 27  De Vaux, in DJD 3 :7. 28  De Vaux, in DJD 3 :7-8.

l’entrée avait été obturée. Qu’on y ait pourtant découvert des fragments de manuscrits dans les interstices des blocs de l’entrée suppose que l’ouverture avait d’abord été dégagée par les bédouins. De Vaux a noté29 : “On enlève. . . . de gros quartiers de roc et de poudingue. . . . pour améliorer l’entrée et faciliter le travail” (Journal, 11Q, 26/2/1956). Dans la cavité, la plus vaste des grottes à manuscrits (50 m2 où l’on peut se ternir debout), les jarres n’avaient pas été enfouies mais posées à l’écart. Les photographies montrent la fouille du sol de la première salle. Puis : “On fouille dans la grotte intérieure à droite de l’entrée . . .” (Journal, 13/3/56) qui n’a livré que des tessons chalcolithiques et du Fer. Aujourd’hui tout le sol a été retourné par les fouilles suivantes. Au dire des bédouins, “C’est dans cette première salle que les grandes trouvailles ont été faites . . .” (Journal, 1/03/56). Les jarres pouvaient avoir été dissimulées dans une anfractuosité dans la pénombre du fond. Cependant, c’est à l’intérieur de la grotte que gisaient “Surtout des éclats de vases cassés par les bédouins” (Journal, 28/2/56). En 1956, les bédouins connaissant la valeur marchande des jarres, il semble improbable qu’ils les aient cassées pour en extraire les rouleaux. Les jarres ont été retrouvées cassées et les tessons furent abandonnés. Le cas de la Grotte GQ15 retient l’attention par la position aberrante d’une jarre dont témoigne une photographie (Figs. 2.11a et b). L’anfractuosité s’ouvre largement 29   Jean-Baptiste Humbert et Alain Chambon, eds., Fouilles de Khirbet Qumrân et de Aïn Feshkha. Vol. 1 : Album de photographies ; Répertoire du fonds photographique ; Synthèse des notes de chantier du Père Roland de Vaux (NTOA.SA 1 ; Fribourg/Göttingen : Éditions Universitaires/Vandenhoeck & Ruprecht, 1994), 344.

47

Cacher et se cacher à Qumrân : grottes et refuges

Figure 2.11a et b

L’abri sous roche GQ15, avec la jarre prise dans l’éboulis et vue rapprochée.

sans profondeur. La cachette semble mal choisie. Le cliché pris en cours de fouille montre le moment où une jarre est apparue, prise dans un amoncellement de pierraille. La jarre est renversée et il est difficile d’accepter qu’elle soit tombée avec la masse des pierres sans se casser. D’ailleurs, d’où viendrait naturellement une telle quantité de cailloux dans un porche abrité ? La jarre est dans une position verticale et semble avoir été mise ainsi dans la position où elle a été trouvée : elle a été enfouie avec soin dans l’amoncellement. Comme les notes ne disent rien de son contenu, nous supposerons qu’elle était vide. Son contenu périssable aurait disparu avec le temps ; la question se pose alors que des jarres auraient servi à conserver autre chose que des manuscrits. 5.4 La topographie des caches Les cachettes répertoriées sont circonscrites dans un périmètre assez bien délimité, moins d’un km2 pour la falaise au nord de Qumrân, et un km2 pour la falaise au sud. Le

secteur choisi pour mettre en sécurité un patrimoine avait donc été déterminé, jalonné, éventuellement garanti par la proximité de Qumrân. Les différents dépôts avaient été guidés et dans ce cas, les résidants de Qumrân ont été impliqués dans les opérations. La décision de circonscrire le périmètre des cachettes suggère qu’on avait planifié l’opération, que l’on pouvait surveiller depuis Qumrân même, et qu’on aurait entrepris de repérer des différentes localisations, en vue du recouvrement. Le secteur choisi reste à proximité du chemin qui, par la gorge de Qumrân, vient de la Judée montagneuse et distribue la falaise sud. L’aboutissement de la route venant de Jéricho distribue la falaise nord. On supposera donc que les dépôts peuvent avoir été apportés du nord. Si l’on accepte que des groupes esséniens ou simplement juifs vivaient sur la rive orientale de la mer Morte, rien n’empêche que certains rouleaux en soient venus. Qumrân a pu être le centre de ralliement d’une entreprise de sauvetage d’un patrimoine largement distribué.

48 6

Humbert

Des cavités artificielles

Les cavités (“grottes”) dites 4Q, 5Q, 7Q, 8Q, 9Q, 10Q présentent d’abord le caractère indubitable d’avoir été creusées de main d’homme dans le banc marneux de la Lisân, relativement facile à excaver. Sur la photographie (Fig. 2.12) on vérifie que la terrasse de Qumrân (au milieu) sépare les deux passifs érodés. Nous insistons sur le fait que l’érosion naturelle a considérablement modifié la distribution originale des souterrains. Nous devons tenir compte de leur état de conservation fortement dégradé : “Toutes ces grottes, creusées au flanc de la terrasse, ont été presque complètement emportées par des effondrements ou des érosions”30. Le constat nous permet de réinterpréter dispositif et fonction, et de renoncer à leur fonction originale d’habitation ou de cachette. Premier paradoxe : les grottes artificielles n’ont pas été creusées avec une fonction de cache. Un rapide examen convainc que l’exécution d’un travail aussi considérable et dans des conditions escarpées n’était pas nécessaire quand tant de cavités naturelles étaient à portée de main pour sauvegarder des biens. À preuve, une dizaine de niches naturelles ont mieux sauvé les manuscrits et les grottes artificielles ont été saccagées dès l’Antiquité. En revanche, nous pouvons faire valoir que les grottes 7Q, 8Q, 9Q, inscrites dans le périmètre de l’établissement de Qumrân, s’imposaient dans sa dépendance et son contrôle, en raison de leur proximité et du lien que tend le “long mur” au bord oriental du promontoire. Revenant un instant dans le jeu interprétatif de de Vaux, nous pouvions concevoir que, la menace venue, le projet de mettre à l’abri la bibliothèque de Qumrân aurait envisagé de creuser des souterrains sécurisés par l’établissement, d’en faire une espèce de chambre-forte dans un abord vertigineux quasi inexpugnable, et d’y déposer son patrimoine le temps venu31. Pourtant de Vaux ne le dit pas et il était sur la bonne voie en pensant à d’anciennes habitations souterraines pour les membres de sa communauté qu’il logeait en dispersion : “L’utilisation des grottes n’est pas seulement contemporaine de Khirbet Qumrân, elle lui est liée organiquement. Il est déjà significatif qu’elle commence et qu’elle s’achève en même temps que l’occupation principale du Khirbeh . . .”32 Si l’occupation s’est achevée en 68, rien n’indique le moment de leur creusement. Ensuite elles ne sont pas des habitations au sens strict. Vivre à

30  De Vaux, in DJD 3 :27. 31  De Vaux, in DJD 6 :21-22. 32  De Vaux, in DJD 3 :32.

Figure 2.12

Les deux refuges dominent le Wadi Qumrân et dont séparés par la ravine qui borde à l’ouest la terrasse du site.

long terme dans un souterrain obscur n’est ni confortable ni dans l’usage. Il vaut mieux croire que les grottes artificielles sont plus anciennes que l’occupation de Qumrân et qu’elles ont été utilisées par les nouveaux venus. De Vaux put croire qu’elles avaient abrité la bibliothèque qu’il voyait à Qumrân puisque des manuscrits y avaient été cachés. Seule la proximité de l’établissement pouvait conforter la proposition. Toutefois une bibliothèque considérable à Qumrân n’a guère de sens, et la cachette dans les grottes creusées aurait été mauvaise parce que peu sûre. Qu’elles aient été vandalisées par l’armée romaine, comme il le suggère avec de bonnes raisons, en fournit la preuve. Second paradoxe, le mode de cache serait sans cohérence. Les grottes 4Qa et b s’exposent à la vue avec de grandes baies ouvertes à l’est (4Qa) et à l’ouest (4Qb) qui, ne pouvant se dissimuler, attirent l’attention. De Vaux savait que les grottes s’étaient effondrées mais il pensait probablement que la grotte 4Q était presque en l’état. Par ailleurs, était-ce bien nécessaire de disposer de petites cavités, 7Q, 8Q, 9Q plus ou moins juxtaposées, quand à partir d’une ouverture discrète il aurait été plus utile et pas plus difficile de poursuivre l’excavation dans la continuité, comme fut conduite celle de 4Qa et b ? Enfin, il tombe sous le sens que les trois “petites grottes” artificielles 7Q à 9Q auraient été suffisantes pour cacher plus d’un millier de manuscrits, à plus forte raison quand la seule 4Qa, elle aussi, aurait probablement tout contenu.

49

Cacher et se cacher à Qumrân : grottes et refuges

Doit-on envisager que les lots aient été volontairement dispersés ? Quel est l’intérêt de deux ensembles de dépôts aussi distincts, d’un côté les grottes creusées dans l’abrupt de la terrasse, et de l’autre côté à l’ouest du ravin, creusées dans les “doigts,” vulnérables et moins contrôlables depuis l’établissement ? Notons encore que le creusement des six grottes artificielles a demandé du temps et requis beaucoup d’énergie, et nous devons savoir pourquoi. Il est évident que la tâche n’a pas été menée dans l’urgence.

Figure 2.13

Panorama des deux refuges occidental et oriental.

Considérons séparément les deux séries : occidentale 4Qa et b, 5Q et 10Q, et orientale 7Q, 8Q, 9Q (Fig. 2.13). Les six cavités ont été présentées en deux séries, l’une à l’est l’autre à l’ouest, de chacune trois chambres sans aucune relation entre elles : le ravin occidental, ici fort profond et qui borde le promontoire de Qumrân, les isole l’une de l’autre. Les occupants respectifs ne pouvaient communiquer qu’à la voix. Qu’elles aient été des habitations n’a pas échappé à de Vaux et il distingue encore les deux séries par le mobilier recueilli. La série orientale a livré avec parcimonie des documents écrits mais de la poterie en quantité raisonnable, avec des restes de fruits séchés, pour rappeler qu’on y avait vécu ; des lampes dont l’une est de bonne fabrique pour rappeler qu’on y avait dormi ; des poteries distinguées par des inscriptions peintes pour signaler le rang des usagers ; des débris de lanières et de ­sandale en cuir, des cordelettes, des phylactères avec leur étui pour identifier sans équivoque la société qui l’avait habitée. L’ensemble désigne le groupe humain au caractère juif marqué qui l’avait occupé. En revanche, la série occidentale, qui recélait aussi un lot de poteries domestiques, une natte de branches de palmier et quelques fragments de cuir et de textiles, possédait un lot de parchemins dont le nombre est estimé à plus de 600 rouleaux ; c’est le même groupe qui l’a fréquentée.

6.1 Morphologie des grottes artificielles (Fig. 2.14) Les deux séries de cavités ont été creusées dans le haut banc marneux de la rive nord du Wadi Qumrân. Grâce à la géologie il faut comprendre que, contrairement à la rive sud du wadi déjà régularisée, le ressaut de la rive nord est d’autant plus fragile qu’il est abrupt, et qu’il a été, sur le long terme, violemment taraudé par les eaux. Le profil de l’escarpe a été modifié au cours des deux millénaires : l’affouillement au niveau du lit en a creusé la base et a entraîné des ruptures de crête, puis la chute des fronts, et c’est pourquoi les rapports de fouilles ont, à bon escient, répété la dégradation des façades et l’effondrement des plafonds. De Vaux n’a pas été insensible au bouleversement géologique de la croupe marneuse mais son interprétation n’en a pas tiré toutes les conséquences. Dans le cas de 7Q–9Q, il a noté que les chambres sont séparées les unes des autres et que “Toutes ces grottes . . . ont été presque entièrement emportées par des effondrements ou des érosions” (Fig. 2.15). Dans le cas de 9Q, il note : “Chambre ronde, les 2/3 emportés”33. Autour des grottes 4Q et 5Q, il a constaté des creusements qu’il n’a pas cherché à interpréter : “Une grotte ovale certainement creusée par l’homme . . . le plafond bas était effondré . . . avant que le creusement fut achevé. Le travail fut interrompu et la grotte ne servit jamais”. Et ailleurs : “Au nord de la grotte 5Q . . . une petite plateforme . . . fragments d’une jarre ovoïde . . . On ne peut mettre cette trouvaille en relation avec une grotte”34. Il aura confondu interruption et dégradation. En fait, le schéma topographique de cachettes dispersées dans la falaise, isolées les unes des autres, a été appliqué aux grottes artificielles. Il était pourtant compréhensible que les cavités creusées dans l’épaisseur de la croupe rocheuse formaient tout un ensemble auquel nous pouvons trouver une cohérence. Quand on aura accepté qu’une bonne partie de la falaise s’est effondrée, il est inéluctable que du mobilier entreposé ou abandonné dans les cavités ait été emporté dans la chute (Fig. 2.16). Il n’est pas impossible qu’une fouille profonde au pied et à l’aplomb des grottes puisse en retrouver des débris. La fouille serait difficile en tranchées profondes sans être assuré de retrouver un matériel utilisable. De Vaux y avait pensé35.

33  De Vaux, in DJD 3 :28. 34  De Vaux, in DJD 3 :28. 35  Communication orale de James Charlesworth.

50

Humbert

Figure 2.14

Le Wadi Qumrân et les grottes artificielles creusées dans la berge abrupte.

Figure 2.15

Façade de l’éperon oriental avec les grottes 7Q, 8Q.

51

Cacher et se cacher à Qumrân : grottes et refuges

Figure 2.16

Les crevasses verticales marquent les chambres effondrées entre les grottes 4Q et 10Q.

En dépit des précautions prises par de Vaux, les grottes artificielles ont été considérées comme telles, en l’état. S’impose pourtant le fait qu’elles ne conviennent pas comme des cachettes. Il n’empêche que mieux regarder les lieux aide à modifier l’idée reçue. À cause du bouleversement géologique, retrouver la disposition topographique primitive, exacte, des cavités est une tâche vouée à l’échec. En revanche, restituer leur fonction est possible. La morphologie de la croupe marneuse où furent creusées 4Q, 5Q, 10Q est un paysage devenu si familier qu’il paraît naturel. Or il ne l’est pas. A-t-on jamais observé que le contour de la croupe, comme une griffe de lion ou les doigts d’une main à plat vue d’avion, n’a pas un profil d’érosion naturel et qu’on ne l’observe nulle part ailleurs ? La photographie (Fig. 2.17) l’illustre déjà au regard averti. Les creux entre les doigts correspondent à des cavités effondrées, que le ravinement de deux millénaires a aggravées. A-t-on jamais pensé que creuser des cavités dans un relief accusé en respectant sa découpe était une mauvaise initiative ? Il faut estimer que le profil géologique du Wadi Qumrân a plus ou moins reculé de plusieurs mètres ou l’érosion en a ensuite émoussé les reliefs. L’organisation des chambres

n’est plus lisible. Un événement décisif est venu fragiliser les falaises : un ou plusieurs tremblements de terre de forte amplitude ont secoué la région. Nous avons déjà montré ailleurs36 que la terrasse sur laquelle la khirbeh repose a été fracturée par trois failles principales zigzaguant dans les maçonneries et au long desquelles aucune restauration n’a été faite ; et celles qui ont affecté l’aqueduc sont décisives à la rectification de la chronologie. Le séisme et ses conséquences postérieurs à 68, ont affecté gravement les à-pics qui entourent le site et endommagé le relief des grottes 4 et 5Q et 7 à 9Q. Il faut accepter que le mobilier et les manuscrits ont été déposés dans les cavités avant les séismes. Les grandes fenêtres qui ouvrent 4Qa (Fig. 2.18) à l’est et à l’ouest puis 4Qb à l’ouest sont aberrantes dès lors qu’on interprèterait les chambres comme des cachettes ; elles attirent le regard aujourd’hui comme elles l’auraient attiré dans l’Antiquité. Creusées de plein-pied depuis les chambres, les grandes ouvertures donnaient sur l’abîme

36  Humbert, “Reconsideration of the Archaeological Interpretation”, 436, 443 (plan).

52

Figure 2.17

Humbert

La découpe profonde de l’éperon des grottes 4Q.

et auraient été dangereuses. De plus, ouvertes à tous les vents et aux oiseaux, elles auraient mal protégé le mobilier dont les parchemins. Choisir de tels souterrains pour soustraire à la vue et p ­ rotéger des centaines de rouleaux aurait été le pire des choix que les auteurs de la cache n’ont pas fait. Il ne s’agit pas de fenêtres, et une observation même rapide démontre que les baies sont ce qui reste en section des chambres coupées par l’effondrement de la crête. Les ouvertures nécessaires à la ventilation devaient être des soupiraux discrets. 6.2 Description du Refuge occidental (4Q, 5Q, 10Q) Il est le mieux conservé. Un géologue a réexaminé l’endroit car la configuration du relief de la croupe, si parti-

culière avec sa forme des doigts d’une main, ne paraît pas résulter d’une érosion naturelle, mais de l’adoucissement d’un massif évidé qui s’est effondré (cf. note en annexe) (Fig. 2.19). Le cône d’effondrement, dont la base tombe dans le lit du wadi, peu régularisé et donc récent, pourrait le confirmer s’il était mieux revu. À propos de la morphologie des cavités, de Vaux n’a pas exploité toutes les possibilités qui se présentaient. Une meilleure observation du site convainc qu’il n’est pas faux, loin de là, de faire se rejoindre 4Qa et b, au contraire de ce qui a été affirmé, et d’y ajouter probablement 10Q. Nous sommes ici en présence d’une espèce de refuge souterrain en une succession de chambres reliées entre elles par des boyaux ou étroitures (Figs. 2.20a, b, c, d).

Cacher et se cacher à Qumrân : grottes et refuges

Figure 2.18

L’embrasure de la fenêtre de la chambre 4Qa, ouverte vers l’est.

Figure 2.19

La façade orientale de l’éperon des grottes 4Q et 5Q.

53

54

Figure 2.20a, b, c, d

Humbert

Les conduits internes de la grotte 4Q.

Cacher et se cacher à Qumrân : grottes et refuges

Figure 2.21

55

Intérieur de la grotte 4Qa.

La “fenêtre” de 4Qa vers l’est s’évase, et avec une lumière frisante on perçoit qu’une embrasure extérieure est taillée (Fig. 2.21). Il convient donc de restituer en avant une chambre disparue. Immédiatement à sa gauche et en contre-haut quand on regarde vers l’ouest, se devine un creux avec des angles encore marqués qui pourrait être ce qui reste d’une autre petite chambre. Le “doigt” nord de 4Q est percé et pourrait être l’élément d’un diverticule ou l’ultime témoin d’un boyau vers 5Q. La fenêtre de 4Qa vers l’ouest possède aussi une embrasure, et l’ouverture bée vers 4Qb. Il convient donc de rétablir un passage entre 4Qa et b que la crevasse aurait détruit. On peut se fier au relevé de Vaux-Coüasnon, qui ne donne rien de précis, mais restitue utilement la découpe générale des reliefs. La forme à peu près orthogonale de la crevasse qui sépare a de b restitue assez bien l’épure d’une chambre emportée ; une poche représentée à l’ouest, qui n’a pas une forme naturelle, serait une niche épargnée par l’effondrement. En tenant compte des découpes du relief, nous proposons une reconstitution du refuge 4Qa, b, c, d (Fig. 2.22a et 2.22b). Il n’est pas impossible de poursuivre par un passage jusqu’en 10Q qui, isolée, trop petite mais dans l’axe de 4Qb, n’a guère de sens ; le relevé exact de sa position manque. Selon le plan qui nous est parvenu, 4Qa (Fig. 2.19) et 4Qb possèdent chacune leur puits (Fig. 2.23) et tunnel d’accès : un refuge est plus sûr avec deux ouvertures. Il en est de même pour 5Q (Fig. 2.24) qui a fait l’objet d’un relevé à main levée par de Vaux (Fig. 2.25). Pourtant, des photographies en ont été prises. Absente des plans, elle pourrait être le vestige d’un autre ensemble détruit. L’entonnoir érodé aux alentours de 5Q signale un effondrement massif de la falaise (Fig. 2.19). Le refuge 5Q a dû être plus déve-

loppé que celui de 4Q. Des évidements fort dégradés en contre-haut et en face de 5Q le laissent croire. De Vaux a noté dans la proximité une chambre inachevée. Il est probable qu’un conduit ait pu rejoindre 4Qa ; tout dépend du volume de la croupe effondrée. Deux souterrains indépendants sont cependant possibles. La petite chambre ovale en contre-haut de 5Q et le fond d’une chambre avec une jarre devaient être en lien avec 5Q. En conséquence, il s’agirait d’un système troglodytique cohérent, une espèce de refuge où les renfoncements, les niches et les cavités intérieures, aveugles, étaient prévus pour engranger des provisions en un temps d’insécurité. Des creux plus profonds ont pu abriter de l’eau. Deux accès étaient préférables pour éviter le cul-de-sac. Enfin, quelle peut être la qualité de refuge de ce genre de souterrain ? Efficace certainement mais en dernier recours car les entrées, bien que dissimulées dans des dépressions, étaient visibles à qui inspectait le plateau. Un assaillant engagé dans une telle cavité à l’accès accidenté (un ressaut d’un mètre cinquante), menant à des chambres séparées par des étroitures en forte déclivité, n’aurait aucune chance d’en sortir vivant. La pratique est universelle d’enfumer les asiles enterrés pour expulser les retranchés, mais les étroitures peuvent avoir été des sas successifs, faciles à obturer. Il est probable qu’une fuite nocturne, par le wadi, était possible à l’aide de cordes (Fig. 2.26). 6.3 La série orientale L’enchaînement 7Q, 8Q, 9Q peut être restitué comme un autre ensemble tout aussi détruit, qui n’a pas été précisément localisé, et on le regrettera. Les trois cavités étaient logées dans l’important effondrement du sud-ouest de la terrasse de Qumrân. L’accès s’y pratiquait de l’est par

56

Figure 2.22a Plans de 4Q publié dans DJD 6.

Figure 2.22b Essai de restitution du refuge.

Humbert

57

Cacher et se cacher à Qumrân : grottes et refuges

Figure 2.23

Puits d’entrée de la grotte 4Q.

Figure 2.24

La grotte 5Q à la fin de la fouille.

Figure 2.25

Relevé de la grotte 5Q (de Vaux, carnet de fouilles).

Figure 2.26

Accès escarpé à la grotte 4Q.

58 7Q : “On y accédait par un escalier partant du bord de la plateforme, au nord-ouest de la chambre ; les marches inférieures de l’escalier sont seules conservées”37. 9Q est aux deux tiers détruite et 8Q est la plus vaste. Les trois chambres sont assez rapprochées les unes des autres pour s’enchaîner, les corridors qui les reliaient auraient disparu. La présence de cordes en 9Q suggère que les réfugiés ont pu prévoir de s’enfuir en escalade de rappel de la hauteur des douze mètres de la falaise dans le Wadi Qumrân. Une fuite précipitée expliquerait l’abandon d’objets aussi précieux que des livres et des phylactères. 6.4 Fonction et chronologie Nous avons là les restes d’un ensemble de refuges que les habitants de Qumrân ont connu et utilisé. S’il est douteux qu’il faille les leur attribuer, le contraire n’a pas été démontré. Il faudrait déterminer la date de leur creusement et quels en furent les auteurs, mais la morphologie des cavités et le mode de taille ne nous enseignent rien. Qu’on ait voulu y voir des tombes du Fer n’a strictement aucun fondement, car les tombes de l’époque n’en n’ont ni la forme ni surtout la disposition en loculi contigus. La présence d’un tesson du Fer recueilli par de Vaux dans la grotte 9Q38 constitue une faible preuve pour qu’on puisse renvoyer avec assurance les souterrains aux VI-VIIe s. av. J.-C. Elle demeure cependant valable. Restent l’archéologie et mieux encore l’histoire de la région. Les associer à l’établissement hasmonéen n’est pas interdit, mais ce genre de refuge, nécessaire en cas de conflit ou d’instabilité politique, ne cadre pas avec la retraite à la campagne de la famille aristocratique qui séjourna à Qumrân. De Vaux n’a pas hésité à voir une habitation dans 4Qa et à en faire remonter l’occupation au moins jusque dans sa période Ib (de 150 à 31 av. J.-C.), au témoignage de sa lecture de la poterie de la khirbeh. Aujourd’hui, la typologie de sa période Ib est de toute façon remise en question. Il faut rappeler que de Vaux n’exclut pas une occupation jusqu’en 68, date de l’abandon du site, et la proposition est juste. On ne conçoit guère que deux siècles de fréquentation d’un local aussi réduit aient produit si peu de mobilier, même dans le cas d’une fréquentation épisodique. Et sur le sol rocheux des chambres, de Vaux n’a pas noté une sédimentation en épaisseur suffisante qui témoignerait d’un long terme. La poterie retrouvée en 4Qa renvoie à l’ultime occupation des lieux, elle est contemporaine du dépôt des manuscrits et ne renseigne pas sur l’occupation dans la durée. Le dépôt contredit une occupation en des temps pacifiques. La poterie est celle 37  De Vaux, in DJD 3 :29. 38  De Vaux, in DJD 3 :31.

Humbert

de réfugiés qui n’avaient pas eu d’autre choix que de se cacher dans des souterrains, qui avaient avec eux les rouleaux et qui les y ont laissés. Les rouleaux et la poterie ont pu être déposés en 68. Nous avons souligné l’opposition entre les modes de cache dans les grottes de la falaise et dans les grottes artificielles qui suppose des urgences différentes, et si les mobiles ont été les mêmes dans les deux cas, la cache est soignée dans l’un et pas dans l’autre. Il faut aussi tenir compte de la présence à Qumrân de sept monnaies de la seconde Révolte ; sans être considérable, le fait ne peut être minoré. Il est possible que la poterie recueillie dans les grottes artificielles puisse en décider. À condition que la typologie subisse un changement radical après 68, ce qui reste à démontrer. Peut-on tirer argument de la conservation de poteries communes pendant soixante ans ? Les refuges n’ont pas été creusés comme des cachettes mais sont devenus des “grottes à manuscrits” faute de mieux. Que des centaines de rouleaux y aient été acheminés en vrac sans la protection de jarres sous-entend une démarche précipitée, menée en catastrophe. Quelle put en être alors l’occasion précise et qui en furent les acteurs ? Nous laissons aux historiens de Qumrân le soin de répondre. 6.5 Les refuges de la région La Palestine méridionale paraît criblée de cavités creusées dans les massifs montagneux et interprétées à juste titre comme des refuges. Elles ont fait l’objet de publications détaillées. La Shephelah, en a fourni un nombre impressionnant dont certaines ont été attribuées au premier millénaire avant notre ère39. Dans d’autres régions, de nombreux systèmes souterrains ont été explorés, tant en Galilée que dans la montagne d’Hébron. La Shephelah a produit les plus spectaculaires ; ils y sont fort complexes, profonds, étendus, ramifiés pour les rendre plus sûrs. Les plans qui en ont été dressés prouvent qu’ils ont été agrandis par étapes et que l’habitation troglodytique s’est développée, ici et là, avec le temps. Aux boyaux étroits et sinueux menant à de modestes abris de forme ovoïde, succèdent des ensembles de plus en plus vastes et de mieux en mieux organisés. Les plus récents recèlent des habitations élaborées, aux chambres à plan orthogonal. 39   Parmi d’autres : Amos Kloner et Yigael Tepper, The Hiding Complexes in the Judean Shephelah (Tel Aviv : Hakibbutz Hameuchad Publishing House, 1987) [hébreu] ; et dans une perspective historique Yuval Shahar, “The Underground Hideouts in Galilee and Their Historical Meaning”, in The Bar Kokhba War Reconsidered : New Perspectives on the Second Jewish Revolt against Rome (TSAJ 100; ed. P. Schäfer ; Tübingen : Mohr Siebeck, 2003).

Cacher et se cacher à Qumrân : grottes et refuges

59

Certains étaient devenus de véritables habitats permanents. L’impression demeure que des abris précaires ont été réhabilités en grand à une époque ultérieure. Enfin, il n’est pas impossible de voir dans la grotte 4Q deux étapes de creusement. La cavité frappe par ses dimensions et la largeur des chambres pourrait correspondre à l’élargissement d’entrées, de boyaux et d’alvéoles oblongs primitifs. L’hypothèse milite en faveur d’un creusement initial antérieur à l’occupation sectaire de Qumrân. Les plus anciens asiles remonteraient aux Macchabées, peut-être plus haut encore. Autant il est difficile de dater leur premier creusement, autant l’essai de les comparer dans le temps à ceux de Qumrân n’est pas sûr d’aboutir. Le tesson du Fer de la grotte 9Q demeure une interrogation. Quant à leur abandon, des tessons mamelouks témoignent dans quelques cas d’une fréquentation jusqu’au MoyenÂge. La plupart des complexes ont cependant livré une poterie romaine datée, signant le dernier passage, que les auteurs n’ont pas hésité à mettre en relation avec la Révolte de 135. Les maquisards de 135 se seront glissés dans les refuges élus par ceux de 68. Comparer la morphologie de nos grottes artificielles avec les retraites complexes de la Judée est contestable. Grottes, cavités, refuges troglodytiques ne manquent pourtant pas dans les falaises autour du Lac Asphaltite et des liens sont possibles. La Grotte des Lettres et la Grotte de l’Horreur aux développements profonds de plusieurs dizaines de mètres, sont uniques dans le désert de Judée. Les deux Révoltes y avaient choisi les mêmes retraites qui furent des asiles si inaccessibles qu’elles n’étaient pas habitables ; elles semblaient inexpugnables. Fréquentées à maintes reprises elles ont abrité des fugitifs mais la Grotte des Lettres a servi de recel pour les biens de valeur d’une société aristocratique. Que les petits camps romains qui les dominent en aient fait le siège, prouve la haute qualité des réfugiés. Le cas est exceptionnel40. Le repaire dans la gorge retirée de Murabba‘ât est glorieux par la geste de Bar Kokhba. La lettre Mur45 reçue à Murabba‘ât avait été envoyée de la “forteresse des pieux”, laquelle peut être Qumrân41. Les deux refuges étaient assez proches pour avoir des liens. Ce que l’on sait de la Christmas Cave, long corridor sur une haute escarpe au débouché du Cédron, avec deux entrées, la donne comme une retraite de qualité. Le mobilier exhumé répète, en plus riche, celui de 4Q. Masada fut l’ultime repli en conclusion de la première

Révolte, même quand des installations et du mobilier doivent être attribués à la seconde. Les caches de Qumrân sont plus modestes et leur nature fut autre. Aux quelques parallèles répertoriés dans la région de la mer Morte, nous sommes amenés à ajouter des installations attribuées à des laures byzantines, creusées au cœur de la presqu’île de la Lisân. Jugeons utile de les réexaminer. Rien ne prouve qu’elles ne soient pas plus anciennes et qu’elles aient été réoccupées par des moines byzantins, comme le fait est attesté à maintes reprises en Judée. La Lisân est transjordanienne avec une présence juive qui n’est pas à démontrer ; au sud, Zoara est proche avec son cimetière, et celui de Qazone, qui compte des tombes que l’on soupçonne juives, s’étend à l’est, à moins de deux kilomètres. Masada domine en face. Rappelons que Dion Chrysostome place ses esséniens dans la proximité. Des ermitages byzantins ont été explorés dans la Lisân par une expédition suédoise dès les années 1990, et il est utile de les mettre en parallèle avec nos grottes artificielles42. Quatre ermitages ont été visités : Qasr al-Tûba, Dayr al-Qattar, Qasr al-Abiadh, le Eastern Hermitage. Un relevé schématique de Qasr al-Abiadh que nous avons redessiné (Fig. 2.27) nous a retenu car les similitudes avec 4Q a et b nous frappent. Mêmes dimensions, même implantation découpée dans le banc marneux avec chambres longues ou adjacentes munies de niches, et diverticules. Les auteurs signalent deux entrées dont l’une est en boyau étroit. Ils notent avec insistance les ravages de l’érosion et la forte dégradation des lieux avec plafonds effondrés et chambres disparues. La description morphologique et l’état des grottes artificielles de Qumrân valent pour la Lisân. Que les souterrains aient été occupés et parfois modifiés par les anachorètes byzantins est évident, et nous pouvons montrer qu’ils ont investi un site plus ancien. Il est surprenant que des graffitis hébreux, ou du moins ce qui en reste, aient été relevés sur les parois de la grande chambre A. On ne voit guère qu’il faille les attribuer aux ermites qui ont aussi gravé en grec, et dont les textes mieux conservés seraient plus tardifs. Une inscription hébraïque gravée sur le mur occidental, dont la paléographie doit être précisée, n’a conservé que : “David père de . . .” ; d’une autre sur le mur oriental, on a pu lire : “Appartient à celui qui a fui . . .”43 La mention d’un fugitif nous renvoie à un temps de crise. La fin de la première Révolte convient. Ce qui n’est pas encore une preuve constitue une sérieuse indication.

40  Yigael Yadin, The finds from the Bar Kokhba Period in the Cave of Letters (Jerusalem : Israel Exploration Society, 1963). 41  Józef T. Milik, “Textes hébreux et araméens”, in Les grottes de Murabba‘ât (DJD 2 ; ed. P. Benoit, J. T. Milik, et R. de Vaux ; Oxford : Clarendon Press, 1961), 163-64.

42  Richard Holmgren, Anders Kaliff, et Jan Svennson, “The Hermit Life on al-Lisân Peninsula – Results of the Swedish Dead Sea Expedition: a Preliminary Report ”, ADAJ 49 (2005) : 167-88. 43  Holmgren, Kaliff, et Svennson, “The Hermit Life”, 174.

60

Figure 2.27

Humbert

Relevé schématique de la partie conservée du refuge du Qasr al-Abiadh, d’après ADAJ 49 (2005) : 169.

Le Qasr al-Tûba, plus au nord, tient sa sécurité d’un sévère isolement : il est caché dans un wadi sous la forme d’une grosse tour ronde, détachée du banc marneux par l’érosion. Sans position saillante, le mamelon ne put avoir une fonction stratégique ou d’observation. La morphologie de la cavité diffère sensiblement de celles que nous cherchons à comparer. Le Qasr a été exploré par l’École biblique en 1928 et A.-G. Barrois en a dressé un plan s­ chématique44 (Fig. 2.28). Un corridor traverse le mamelon de part en part, où deux cellules peuvent avoir été habitées. Le site ne semble pas avoir souffert d’effondrements et les chambres ouvertes dans les façades purent être fermées ; elles l’ont été par les cénobites byzantins et il en restait en 1928 des vestiges de terre crue. Ici encore deux issues rappellent la disposition décrite de 4Q et de Qasr al-Abiadh. Une fouille de 1996 a reconnu des chambres adjacentes et une citerne inférieure. Un petit fragment de papyrus qui semble rédigé en grec ne constitue pas un repère chronologique. Les ermitages de la Lisân ont été des refuges juifs qu’il est rationnel d’attribuer aux Révoltes et de comparer au potentiel qumrânien. Le scénario dans les deux cas présente des similitudes qui, de part et d’autre de Masada, désenclave géographiquement Qumrân d’une inutile singularité. Sur l’autre rive on ne peut oublier la forteresse juive de Machéronte qui a subi les hauts et les bas Figure 2.28

Relevé schématique du Qasr al-Tûba par A.-G. Barrois, RB 38 (1929) : 250.

44  Félix-Marie Abel, “Notes complémentaires sur la mer Morte”, RB 38 (1929) : 237-60 et 250 (le plan).

61

Cacher et se cacher à Qumrân : grottes et refuges

de l’effervescence politique du milieu du siècle. La place a été ruinée en 36 par Arétas IV. Quel qu’ait pu être son état, elle a été réinvestie en 66 par les rebelles45. Elle est tombée en 71. Les retranchés ont correspondu. Les ostraca en araméen, recueillis dans les fouilles, ont été attribués aux cinq années que les rebelles ont tenu la place. Ils sont interchangeables, comme les parties d’un lot, avec ceux de Qumrân et de Masada ; dans les trois lieux, nous n’avonslà que des bribes mais d’une même histoire passée dans un temps court46. S’il convient de formuler une conclusion, nous envisageons que les grottes de la falaise et les cavités creusées de main d’homme constituent deux dossiers différents. Les grottes pour les hommes, les caches pour les livres, les refuges pour les rebelles évoquent les événements d’une crise exceptionnelle dont Qumrân n’est qu’un épisode. Elle a dû se dérouler en deux temps qui ne sont pas forcément consécutifs. D’abord un groupe ou des groupes ont entrepris et planifié, dans des jarres fermées, la mise en sûreté des rouleaux dans les anfractuosités de la falaise. Ensuite, en grande urgence, les derniers fugitifs, retranchés en hâte dans les souterrains sous la terrasse de Qumrân, auront abandonné en vrac les parchemins avec les phylactères, un patrimoine qu’ils cherchaient à sauver. Les refuges autour du site furent de bien piètres cachettes. L’entrepôt de centaines de rouleaux y était risqué et ne devait être, somme toute, que l’étape d’une errance dont nous ne savons rien. Ceux qui les avaient entassés avaient prévu de venir les reprendre. Il n’en fut rien. De Vaux a noté que les manuscrits de 4Q, au contraire des grottes de la falaise, avaient été saccagés et que le vandalisme avait été perpétré peu après l’abandon47. Il y a vu le fait de la botte romaine. Il avait probablement raison.

45  Flavius Josèphe, Ant. Jud., XVIII 5, 1-2. 46  Haggai Misgav, “The Ostraca”, in Machaerus. Vol. 1 : History, Archaeology and Architecture of the Fortified Herodian Royal Palace and City Overlooking the Dead Sea in Transjordan ; Final Report of the Excavations and Surveys, 1807-2012 (ed. G. Vörös; Milano : Edizioni Terra Santa, 2013), 259-72. 47  De Vaux, in DJD 6 :21-22.

Géologie des grottes artificielles de qumran Gérard Massonnat Les grottes de Qumrân peuvent être regroupées en deux catégories principales : des cavités naturelles formées par dissolution des calcaires crétacés dans les falaises qui dominent le site, et des grottes creusées de main humaine dans les terrasses marneuses du Quaternaire récent. Ces dernières sont accessibles dans une falaise escarpée de 50 m de hauteur qui domine le Wadi Qumrân, une gorge au fond de laquelle le lit d’une rivière descend vers la mer Morte. Le sommet de la falaise a été aplani au cours du temps par des pluies violentes qui y ont déposé environ 1 m de galets grossiers et de conglomérats. Les grottes artificielles, taillées dans les marnes de Lisân, constituent une formation géologique Pléistocène déposée en discordance sur les séries carbonatées du Cénomanien. Localement, la formation Lisân est constituée de lits minces de marne blanche, tendre et de texture fine avec des cristaux de gypse et d’aragonite. L’hétérogénéité verticale du contenu argileux conduit à un litage très fin que souligne le profil de dureté et d’érosion de la falaise. L’origine sédimentaire des dépôts de la Lisân Fm. doit être recherchée dans un environnement sédimentaire lacustre, à une époque où la mer Morte, avec un plus haut niveau, était beaucoup plus étendue qu’à présent. Les faciès indiquent des conditions de dépôt marins pélagiques, probablement dans une zone située en position distale ou latérale par rapport aux apports détritiques fluviatiles dans le lac salé. L’âge récent des sédiments, donc peu enfouis, explique leur faible consolidation et l’aspect friable qui les caractérise. Peu de fractures affectent les marnes qui accommodent certainement la déformation de manière ductile plutôt que fragile. Au dessus du site de Qumrân, quelques fractures en dalles démontrent que les directions principales NW-SE and NNW-SSE sont cohérentes avec les directions du paléo stress post-Pléistocène de la région de la mer Rouge et de la zone du Rift. Les marnes de la formation Lisân sont affectées par une intense dissolution naturelle : à petite échelle, les nombreux joints de stratification entre les lits marneux montrent de gros trous qui constituent un réseau de vides bien connectés dans la roche. Les effets d’érosion différentielle des marnes, en relation avec le contenu variable en composant calcaire, déterminent le profil de la falaise. À plus grande échelle, les rares fractures qui affectent la

62

Figure 2.29

Humbert

Vestiges de cavités éboulées à l’ouest de 4Q.

série permettent le développement de tuyaux d’érosion et de dissolution verticaux qui s’agrègent spatialement. La zone où sont creusées les grottes apparait avec une morphologie particulière, en relief à la fois par rapport aux niveaux sous-jacents riches en argile, mais surtout par rapport aux mêmes horizons latéraux. La zone des cavités creusées est marquée ainsi par un relief proéminant qui se diverticule en plusieurs “doigts” séparés par des conduits verticaux. La question est de savoir si le relief caractéristique de la zone des grottes est préalable au creusement ou la conséquence de l’excavation humaine. Dans le premier cas, la morphologie issue d’une érosion sévère des marnes locales pourrait suggérer que les dépôts naturels de la zone des grottes sont constitués par des sédiments plus consolidés et moins friables que leurs équivalents latéraux. Les grottes auraient été creusées à cet emplacement en raison de la plus grande dureté du sédiment pour meilleure sécurité de l’ouvrage. L’explication est contredite par l’observation des lits marneux dont la composition est exactement continue latéralement dans la zone examinée. D’autre part, la morphologie de la falaise n’est pas exclusive à la zone des grottes creusées, mais se retrouve ailleurs dans le wadi, lorsqu’existent des cavités naturelles. Dans la seconde hypothèse, l’activité de creusement aurait déstabilisé le matériau et favorisé une plus forte érosion par concentration du ruissellement. Fortement fragilisés, affectés par une dissolution intense à petite échelle, les sédiments n’ont certainement pas résisté à l’existence d’un

réseau creusé en parallèle à la falaise. Des effondrements ont rapidement eu lieu, ainsi que le prouvent d’autres grottes effondrées, dans les marnes à proximité (Fig. 2.29). L’explication la plus pertinente de la morphologie de la falaise aux grottes artificielles est donc la suivante : – Le creusement des grottes dans la falaise en face du site archéologique a certainement déstabilisé des sédiments marneux peu consolidés, déjà partiellement dissous et friables. – Au bout d’un certain temps, certaines parties de ce réseau de cavités se sont effondrées, entraînant une concentration du ruissellement, une érosion préférentielle et le développement de conduits verticaux de dissolution (Fig. 2.30). – L’effondrement du réseau de vides et la dissolution préférentielle dans ces zones ont probablement conduit à des effondrements partiels de la falaise, d’abord au-dessous de la voûte des conduits verticaux, puis jusqu’au sommet par propagation (Fig. 2.31). Ce processus a été décrit pour des zones de badlands similaires et déjà évoqué comme possible à Qumran par des auteurs spécialistes de ces morphologies. A la place du réseau initial de vides, ne subsisteraient aujourd’hui plus que quelques grottes isolées séparées par les zones d’érosion préférentielle. La morphologie actuelle de la falaise résulterait donc en partie de l’effondrement des zones de conduits verticaux (les “pipes” des badlands).

Cacher et se cacher à Qumrân : grottes et refuges

Figure 2.30

Tuyaux verticaux d’érosion et de dissolution.

Figure 2.31

Effondrement des zones de conduits verticaux (les “pipes” des badlands).

63

Part 2 Manuscripts

∵

CHAPTER 3

The Contents of the Manuscripts from the Caves of Qumran1 Florentino García Martínez As it is clear from the conference program, my own presentation and that of Charlotte Hempel are supposed to be centered on the contents of the manuscripts found in the Qumran caves. I think that in the minds of the organizers, a profile or profiles of the manuscript contents would help to complement, sharpen, and/or correct the conclusions provided by the presentation of the archaeological and material elements of the caves (the main focus of the conference). Charlotte Hempel’s presentation will give a specific profile of Cave 4, while I will offer, if possible, a general profile of the whole collection or collections. In order to achieve this goal, it is necessary to look at both the commonalities of the manuscript contents from the different caves, and the peculiarities of the contents of the different caves. Before embarking on the profile of each individual cave, it will be helpful to list the conclusions of the most relevant studies dealing with the caves, not from the point of view of the contents, but from the perspective of material culture, paleography, regional context, etc., because this will hopefully reveal some of the problems we need to take into consideration when looking at the contents of the manuscripts. I will list the conclusions of these studies in chronological order, which, in my opinion, are the most relevant in the last fifteen years for the present topic. The 1999 article and excavation report by Magen Broshi and Hanan Eshel “Residential Caves at Qumran”2 established firmly the need to distinguish between the residential caves in the marl terrace (4a, 4b, 5, 7, 8, 9 and 10, and the two excavated by Broshi and Eshel) and the natural caves in the cliffs (1, 2, 3, 6 and 11), which were considered unsuitable for sustained habitation, thereby disagreeing with de Vaux’s3 assertation that Cave 11 could have 1  I want to thank Eibert Tigchelaar and Mladen Popović, who have read and commented on a previous version of the paper. They are in no way responsible for the errors which may have remained. I also want to thank Jeremy Penner, who has polished my English. 2  Magen Broshi and Hanan Eshel, “Residential Caves at Qumran,” DSD 6 (1999): 328–48. 3  Roland de Vaux, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973), 51: “Cave 11 was inhabited in the Chalcolitic period, in Iron Age II, and finally at the same period as Khirbet Qumran, as the pottery found there (but rare elsewhere) attests.”

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_005

been inhabited.4 This basic division has been generally accepted and it will allow us to consider separately the contents of the two groups of caves. In the very important 2007 article “Old Caves and Young Caves,”5 and in his study in 2011, “Wie viele Bibliotheken gab es in Qumran?,”6 Daniel Stökl Ben Ezra established that: the average age of the dated scrolls from Cave 4 and from Cave 1 differs to such an extent from that of the manuscripts of Caves 2, 3, 5, 6, and 11 that the possibility that they are all randomly chosen samples of the same “population,” the same library, becomes improbable. In other words, it can be shown statistically to be highly unlikely that the manuscripts from Caves 1 and 4 are random samples coming from the same collection of manuscripts as those from Caves 2, 3, 5, 6, and 11, hidden in an emergency just before 68 CE. Instead, I propose that the manuscripts from Cave 1 and Cave 4 represent an older form of the Qumran manuscript collection than those from Caves 2, 3, 5, 6, and 11. The latter caves represent the manuscript collection of the same community but at a later stage.7 His conclusion that we are dealing with two temporally different collections, has found a wide consensus, but also opposition, both by those who propose a different paleographical dating of the manuscripts8 and by those 4  Broshi and Eshel, “Residential Caves,” 332: “Cave 11 is relatively big, but it does not have a leveled floor as expected from a place fit for habitation.” 5  Daniel Stökl Ben Ezra, “Old Caves and Young Caves: A Statistical Reevaluation of a Qumran Consensus,” DSD 14 (2007): 313–33. 6  Daniel Stökl Ben Ezra, “Wie viele Bibliotheken gab es in Qumran?” in Qumran und die Archäologie: Texte und Kontexte (WUNT 278; ed. J. Frey, C. Claussen, and N. Kessler; Tübingen: Mohr Siebeck, 2011), 327–46. 7  Stökl Ben Ezra, “Old Caves and Young Caves,” 315. 8  For example, Florentino García Martínez, “Reconsidering the Cave 1 Texts Sixty Years After Their Discovery: An Overview,” in Qumran Cave 1 Revisited: Proceedings of the Sixth Meeting of the IOQS, Ljubliana 2007 (STDJ 91; ed. D. Falk, S. Metso, and E. J. C. Tigchelaar; Leiden: Brill, 2010), 1–13 (8–9). But see Stökl Ben Ezra’s answer in

68 who find unconvincing the historical explanation of Stökl Ben Ezra (two chronologically different deposits, between 9 and 4 BCE for the old caves and 68 CE for the young caves)9 and the alternative explanation of Hanan Eshel10 (only one deposit in 68 CE, but the manuscripts from Cave 1 are a selection of the manuscripts from Cave 4 which served as stacks for aging manuscripts). Despite these reservations, Stökl Ben Ezra’s article has at the very least made scholars aware of the necessity of determining if the different caches of scrolls represent a single collection or multiple collections. The third article I will review that focuses on the material aspects of the caves was written by Joan Taylor and published in 2012: “Buried Manuscripts and Empty Tombs: The Qumran Genizah Theory Revisited.”11 As the title suggests, Joan Taylor revisits the genizah theory, originally formulated for Cave 1 by Sukenik12 and more recently defended by George Brooke for the same cave.13 Taylor’s article contains many tantalizing suggestions (e.g., on the time-span in which the scrolls may have been hidden, on the size of the collection and its origin), but her primary aim is to prove that the caves were selected as internment locations for worn out scrolls in need of preservation for a later burial deposit, not only in Cave 1 but in all the natural caves on the cliff (and possibly also the “empty” tombs of the cemetery). The scrolls found in the marl terrace caves, particularly Cave 4, could be a real genizah or temporary storehouse, “for manuscripts prior to processing and burial.” To quote her conclusion: “The Dead Sea Scrolls that we possess are the result of preservation-burials of manuscripts after processing at Qumran, during which time the temporary store area (genizah)-with workshops-was in the marl “Further Reflections on Caves 1 and 11: A Response to Florentino García Martínez,” in The Dead Sea Scrolls: Texts and Context (STDJ 90, ed. C. Hempel; Leiden: Brill, 2010), 211–23. 9  For example, Mladen Popović, “Qumran as Scroll Storehouse in Times of Crisis? A Comparative Perspective on Judaean Desert Manuscripts,” JSJ 43 (2012): 551–94 (580–81). 10  As reported by Stökl Ben Ezra, “Old Caves and Young Caves,” 331. 11  Joan E. Taylor, “Buried Manuscripts and Empty Tombs: The Qumran Genizah Theory Revisited,” in ‘Go Out and Study the Land’ ( Judges 18:2): Archaeological, Historical and Textual Studies in Honor of Hanan Eshel (JSJSup 148; ed. A. M. Meir, J. Magness, and L. H. Schiffman; Leiden: Brill, 2011), 269–315. 12  Already in the title of his two Hebrew volumes of 1948–49: Eliezer L. Sukenik, Megilot genuzot mitokh genizah keduma shenimtse’ah bemidbar Yehudah (Jerusalem: Bialik Institute, 1948–49), [Hebrew]. 13  George J. Brooke, Qumran and the Jewish Jesus: Reading the New Testament in the Light of the Scrolls (Cambridge: Grove Biblical Series 35, 2005), 9.

García Martínez

caves.”14 If I have understood Taylor correctly, she argues that the deposition of scrolls in the caves took place over an extended period of time (burial preparation was a long process): “a date in the reign of Herod at the earliest, continuing at least to the year 68 CE;”15 regarding the question of whether the scrolls found in the caves formed a single or multiple collection/s, she writes: “In my view, the vastness of the scroll preservation-burials (which indicates a correspondingly vast originating holding) can only be explained by thinking of the entire Essene school, over a period of time, not narrowly in terms of one single library hidden at a single point in time.”16 One article, which also deals with the genizah theory and takes into consideration the material elements and the paleography of the manuscripts, but puts the greatest emphasis on their contents, was published by Stephen Pfann in 2006: “Reassessing the Judean Desert Caves: Libraries, Archives, Genizas and Hiding Places.”17 In this study Stephen Pfann explores the possibility that “each individual cave might represent a single coherent library,” and concludes by arguing that the caves whose profile can be established (excluding thus Caves 7, 8, 9 and 10 which have an “elusive character”18) represent a different and distinct library: Cave 1 represents a priestly Essene group and Cave 6 a lay Essene group;19 Caves 4 and 5 were genizahs for both priestly and lay Essenes during the occupational phases of the site. Cave 11 represents a priestly zealot’s library and Cave 3 a lay zealot’s library, both brought from Jerusalem, possibly by Yehuda ben Yair;20 and finally Cave 2 would be a cave “potentially connected with Simon bar Giora.”21 I will finish this review with two studies published in 2012 that synthesize the relevant data on the material culture, palaeography, and regional context of Qumran caves, but which also pay attention to the contents of the manuscripts in their respective deposits. In “Qumran: Caves, Scrolls, and Buildings,” Sidnie White Crawford22 critically 14  Taylor, “Buried Manuscripts and Empty Tombs,” 305. 15  Taylor, “Buried Manuscripts and Empty Tombs,” 297. 16  Taylor, “Buried Manuscripts and Empty Tombs,” 303. 17   Stephen J. Pfann, “Reassessing the Judean Desert Caves: Libraries, Archives, Genizas and Hiding Places,” BAIAS 25 (2007): 147–70. 18  Pfann, “Reassessing the Judean Desert Caves,” 166. 19  Pfann, “Reassessing the Judean Desert Caves,” 156–58. 20  Pfann, “Reassessing the Judean Desert Caves,” 158–61. 21  Pfann, “Reassessing the Judean Desert Caves,” 167. 22  Sidnie White Crawford, “Qumran: Caves, Scrolls, and Buildings,” in A Teacher for All Generations: Essays in Honor of James C. VanderKam (JSJSup 153; ed. E. F. Mason et al.; Leiden: Brill, 2012), 1:253–73.

The Contents Of The Manuscripts From The Caves Of Qumran

discusses all the works that I have mentioned, and after looking into the contents of the different caves she concludes that Caves 5, 7 and 8 are collections for the private use of the Qumran inhabitants of these caves,23 that the manuscripts found in the caves of the cliff all have the same sectarian profile, and that the manuscripts from Cave 4 tie “all other collections to itself and to each other.”24 For Crawford the eleven caves form a single collection or corpus and, while she seems not to exclude completely the possibility “that some older manuscripts were stored away in certain caves before the end of the settlement,”25 she finally opts for the “quick hiding” scenario of the standard traditional interpretation put forward by de Vaux. To quote her conclusion: “All of these facts create a strong chain of evidence that it was the inhabitants of Qumran who owned the scroll collection and who hid the majority of the Scrolls, perhaps first in the relatively inaccessible caves in the limestone cliffs, but then finally and quickly in the large, conveniently nearby Cave 4 in 68 C.E.”26 The most lucid, incisive and complete study of the topic I know, is the long article by Mladen Popović, published also in 2012: “Qumran as Scroll Storehouse in Times of Crisis? A Comparative Perspective on Judaean Desert Manuscript Collections.”27 The article compares the collection of manuscripts from Qumran, comprised almost entirely of literary texts, with all other collections of manuscripts from the Judaean desert, collections made up primarily of documentary material. Popović thus underlines the distinctive features of the collection from Qumran as a scholarly, school-like collection. He also analyses the historical context surrounding the deposition, characterized by violence and conflict; he notes the absence of coins in the Qumran caves and concludes that refugees did not hide in them. This implies a somewhat different deposition scenario, perhaps reflecting the anticipation of ­violence.28 Broadly, the lived context of the Judaean desert collections other than the one at Qumran points to the deposition of many personal collections; the manuscripts from Qumran, however, point to a communal, school-like context,29 and after discussing if at Qumran we have single or multiple collections, Popović concludes:

23  White Crawford, “Qumran: Caves, Scrolls, and Buildings,” 265. 24  White Crawford, “Qumran: Caves, Scrolls, and Buildings,” 266. 25  White Crawford, “Qumran: Caves, Scrolls, and Buildings,” 271. 26  White Crawford, “Qumran: Caves, Scrolls, and Buildings,” 273. 27  Popović, “Qumran as Scroll Storehouse,” 551–94. 28  Popović, “Qumran as Scroll Storehouse,” 254–57. 29  Popović, “Qumran as Scroll Storehouse,” 578.

69

The movement behind the Dead Sea Scrolls can be characterized as a textual community, reflecting a milieu of Jewish intellectuals who were engaged on various levels with their ancestral traditions. The collection of texts attracted people and shaped their thinking, while at the same time people shaped the collection, producing and gathering more texts. In this sense, the site of Qumran and its surrounding caves functioned like a storehouse for scrolls.30 This quick review of the main studies dealing with the material culture, paleography, and comparative regional context of the caves shows the great diversity of opinions among scholars today. For many years the general opinion was that the scrolls found within the caves formed part of a single collection, the library of the Qumran community, hidden in the caves for safe keeping from the Romans.31 The questions now asked are whether the archaeological remains represent a storehouse for scrolls (Popović), burial deposits or genizahs for manuscripts (Taylor), multiple collections formed by different groups, each represented by one of the caves (Pfann), multiple deposits at different times all belonging to a single collection (Stökl Ben Ezra), or multiple collections from a single group coming from different places (Schofield32). What can a profile of the contents of each single cave add to this polyphony, or shall we say cacophony? Let us now look at the contents of each individual cave (Cave 9 and 10 can be safely ignored, as only one papyrus fragment with a few letters was found in the former, and only an ostracon with two letters in the latter) with the help not only of the indices of DJD 39,33 but also of the programmatic article by Eibert Tigchelaar.34 I can be short with the contents of Cave 1, since in 2010 I published an article dedicated to “revisiting” this cave, and in 30  As is formulated in the article abstract on p. 551, and developed on pp. 585–94. 31  See, for example, the classic hypothesis formulated by Józef T. Milik, Ten Years of Discovery in the Wilderness of Judaea (SBT 26; trans. J. Strugnell; London: SCM Press, 1959) or by Frank M. Cross, The Ancient Library of Qumran (3rd. ed.; Sheffield: Sheffield Academic Press, 1995, first edition from 1958). 32  Alison Schofield, From Qumran to the Yaḥad: A New Paradigm of Textual Development for the Community Rule (STDJ 77; Leiden: Brill, 2009). See also John J. Collins, Beyond the Qumran Community: The Sectarian Movement of the Dead Sea Scrolls (Grand Rapids, MI: Eerdmans, 2010). 33  Emanuel Tov, ed., The Texts from the Judaean Desert: Indices and an Introduction to the Discoveries in the Judaean Desert Series (DJD 39; Oxford: Clarendon Press, 2000). 34  Eibert J. C. Tigchelaar, “The Dead Sea Scrolls,” EDEJ 163–80.

70

García Martínez

which, after discussing the opinions of other colleagues, I reached a rather conservative conclusion: All things considered, the traditional opinion, which sees Cave 1 as the repository of part of the treasures of the library of Qumran in order to hide and protect them from impending danger, when presented in an orderly and thoughtful manner, seems still the best explanation. If we take seriously the high number of jars, already broken in antiquity, and the high number of linen textiles found in the Cave, we may conclude that the orderly hiding of the manuscripts was interrupted and never completed, or that Cave 1 was emptied of part of its treasures before modern times, as Stegemann concluded for Cave 3. We will never know. What we do know is what we have: a few well-preserved manuscripts and many more small remains of other compositions. And when we consider all of them, we have a perfect sample of the library of which the holdings of Cave 1 were once a part of, a cross section, as it were, of the Qumran collection as a whole. In the article I mentioned at the beginning35 where I compared the contents of Cave 1 with all the other known materials from Qumran, I concluded that the most interesting element brought forth by the completion of the publication was the change offered in the proportions between biblical, para-biblical, and sectarian manuscripts, and the increased importance of non-sectarian para-biblical materials as compared with the two other categories. And when one takes into account not only the seven big manuscripts published outside DJD but the forty manuscripts included in DJD 1 (leaving out of consideration the thirty other manuscripts non classified or identified reproduced on plates XXXIII– XXXVII) the profile of the contents of Cave 1 is rather similar to the profile of the collection as a whole: 15 “biblical” manuscripts, 9 “sectarian” compositions, and 22 “para-biblical” non sectarian compositions.36

35  I was referring to an article published in Spanish in 2006, Florentino García Martínez, “Qumrán en el siglo XXI: Cambios y perspectivas después de 50 años de estudios,” MEAH 55 (2006): 309–34, available also on the website of the Asociación Española de Estudios Hebreos y Judios: http://www.aeehj.net. 36  García Martínez, “Reconsidering Cave 1 Texts Sixty Years After Their Discovery,” 13.

I was, of course, operating at the time with the tripartite division of the manuscripts put forward in DJD 137 (biblical or canonical, apocryphal and/or pseudepigraphal, and sectarian or Essene; the problematic apocryphal and pseudepigraphical category was subsumed in the “parabiblical” category used later in the DJD series38). But, as you know, I have since then abandoned completely these classifications (not only “biblical” and “non biblical” but also “sectarian” and “non sectarian”) because they are an anachronism and do not help us to advance our understanding of these texts.39 I am on record for having defended the idea that we should consider all the texts (I mean all of them) by what they are in their historical context: a collection or collections of religious texts. Or, to express this idea with the words of Eibert Tigchelaar: “The most characteristic feature of both the entire corpus and the collections of the individual caves is the fact that virtually all the manuscripts contain texts of a religious nature or touching upon religious issues. Only a few, badly preserved fragments are the remnants of nonliterary texts such as letters, accounts or deeds, and it cannot be excluded that some of those actually stem from Naḥal Ḥever.”40 My conclusion for the profile of the collection from Cave 1 and its statistics may look thus a little outdated by my present thinking. Besides, we all know that we have no real idea of the original holdings of the cave. But in order to profile the cave we need to use what we have, which are the 46 useful manuscripts. If we apply the categories proposed by Armin Lange in DJD 39 for the non-biblical texts,41 or the thematic grouping of the manuscripts proposed by Tigchelaar42 and we start looking at the 46 useful 37   According to a proposal by Milik: “Ouvrages canoniques, Ouvrages non canoniques (Commentaires et Apocryphes), ouvrages nouveaux de la ‘bibliothèque Essénienne’,” Dominique Barthélemy and Jóseph T. Milik, “Les Textes”, in Qumran Cave 1 (DJD 1; ed. D. Barthélemy and J. T. Milik; Oxford: Clarendon Press, 1955), 46. 38   D JD 13, DJD 19, DJD 22, and DJD 30. 39   Florentino García Martínez, “¿Sectario, no-sectario, o qué? Problemas de una taxonomcorrecta de los textos qumránicos,” RevQ 91 (2008): 383–94. 40  Tigchelaar, “The Dead Sea Scrolls,” 164. 41  Armin Lange and Ulrike Mittmann-Richert, “Annotated List of the Texts from the Judaean Desert Classified by Content and Genre,” in DJD 39:115–64. These categories are: “parabiblical,” “exegetical,” “concerned with religious law,” “calendrical,” “poetical,” “liturgical,” “sapiential,” “historical,” “apocalyptic” and “eschatological.” 42   Tigchelaar, “The Dead Sea Scrolls,” 163–80: “Authoritative Scriptures and the Formation of the Bible,” “Extending Scriptures by Interpretative Rewriting,” “Expanding Scripture

The Contents Of The Manuscripts From The Caves Of Qumran

manuscripts from Cave 1 using these categories, we end up reaching the same conclusion I reached in my 2010 article. In Cave 1 we find texts written in Hebrew and in Aramaic,43 both on leather and papyrus,44 several of them copied by the same scribe responsible for manuscripts found in other caves45 (according to Ada Yardeni as many as 5346), and at least one manuscript (1QHa) was copied at Qumran according to the analysis of the ink used.47 One of the advantages of using the Study Edition48 is that we have noted for each manuscript the copies found in other caves. Regarding all the so-called “biblical texts” from Cave 1, copies have been found in other caves, as well as all of the following: Hodayot: 1QHa, 1Q35, 4Q427–432 Serekh Texts: 1QS, 4Q255–264, 5QS Serekh ha-‘Edah: 1QSa, 4Q24949 by Ascribing Traditions to Foundational Figures,” “Expounding Scripture in Commentaries and Pesharim,” “Interpreting the Law in Legal Works,” “Harmonizing Times and Festivals: Calendrical Documents and Annals,” “Performing Scripture: Liturgical and Poetical Manuscripts,” “Understanding All There Is: Sapiential Texts,” “Envisioning the End: Apocalypses and Eschatological Texts,” and “Returning to the Law: Community Rules and Related Texts.” 43  1Q19, 20, 21, 23, 24, 32, the non-characterized and non-identified texts 1Q63–66, and the non-classified text 1Q67 are in Aramaic; all the rest of the materials are in Hebrew. 44  Although papyrus is poorly represented, used only for 1Q70 and 1Q70bis. 45  For a discussion of the hand that copied both 1QpHab and 11Q20, see Florentino García Martínez, Eibert J. C. Tigchelaar, and Adam S. van der Woude, eds., Qumran Cave 11.II: 11Q2–18, 11Q20– 30 (DJD 23; Oxford: Clarendon Press, 1998), 364. For a discussion of the scribe that copied 1QS and several other manuscripts from Cave 4, see Eibert J. C. Tigchelaar, “In Search of the Scribe of 1QS,” in Emanuel: Studies in Hebrew Bible, Septuagint, and Dead Sea Scrolls in Honor of Emanuel Tov (VTSup 94; ed. S. M. Paul et al.; Leiden: Brill, 2003), 439–52. 46  Ada Yardeni, “A Note on a Qumran Scribe,” in New Seals and Inscriptions, Hebrew, Idumean and Cuneiform (HBM 8; ed. M. Lubetski; Sheffield: Phoenix Press, 2007), 287–98. 47  Ira Rabin et al., “On the Origins of the Ink of the Thanksgiving Scroll (1QHodayota),” DSD 16 (2009): 97–106, and Ira Rabin, “Archaeometry of the Dead Sea Scrolls,” DSD 20 (2013): 124–42 (139–40). 48  Florentino García Martínez and Eibert J. C. Tigchelaar, The Dead Sea Scrolls Study Edition (Leiden/Grand Rapids, MI: Brill/ Eerdmans, 2000). 49  S. J. Pfann has identified no less than 8 copies of the Serekh ha’Edah among what was considered a single papyri manuscript in cryptic writing from Cave 4 (4Q249) and edited them in Stephen J. Pfann, ed., Qumran Cave 4.XXVI: Cryptic Texts;

71

Aramaic Levi: 1Q21, 4Q213–214, 4Q540–541 Dibre Moshe: 1Q22, 4QDM Enoch-Giants: 1Q23–24, 2Q26, 4Q203, 4Q530–533, 4Q556, 6Q8 Instruction: 1Q26, 4Q415–418a, 4Q423 Mysteries: 1Q27, 4Q299–301? Apocryphon of Moses?: 1Q29, 4Q408, 4Q375–376 New Jerusalem: 1Q32, 2Q24, 4Q232, 4Q554–555, 5Q15, 11Q18 War Scroll: 1Q33, 4Q285, 4Q491–496, 11Q14? Liturgical Prayers: 1Q34, 4Q507–509 Jubilees: 1Q17–18, 2Q19–20, 3Q5, 4Q176a, 4Q216–24, 4Q482–483, 11Q12 The better represented works in other caves are Jubilees (manuscripts found in 5 caves), New Jerusalem (also found in 5 caves), and Enoch-Giants (found in 4 caves). This fact proves that Cave 1 has a similar profile with Cave 4 and 11 and with the corpus as a whole, and also that this corpus as a whole cannot be defined simply as Jewish religious literature, but rather Jewish religious literature representing a particular stream of Judaism because of the peculiar halakhic or ideological ideas appearing in some of the compositions. It is true that Hodayot, Serekh and War Scroll copies point to a peculiar group that many would identify as Essenes, but the other compositions like Jubilees or Enoch do not point in this direction, and they are the best represented within the corpus. Cave 250 has provided us with 26 useful fragments of manuscripts, all of them copied on leather, of which 17 represent “biblical” books and the rest contain diverse compositions, among which is the only preserved copy of Sirach found at Qumran,51 two copies of Jubilees,52 an Apocryphon of Moses,53 and an Apocryphon of David?;54 also attested in Cave 4 on two or three copies,55 and two Philip S. Alexander et al., eds., Miscellanea, Part 1 (DJD 36; Oxford: Clarendon Press, 2000), 547–71. 50  Edited by Maurice Baillet, “Textes des grottes 2Q, 3Q, 6Q, 7Q, à 10Q,” in Les ‘petites grottes’ de Qumrân: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 45–93. 51  2Q18, Baillet, in DJD 3:75–77. 52  2Q19 and 2Q20, Baillet, in DJD 3:77–79. 53  2Q21, Baillet, in DJD 3:79–81. 54  2Q22, Baillet, in DJD 3:81–82. 55  4Q372 and 4Q373 were edited and published under the title “4QNarrative and Poetic Compositionb and c” by Eileen Schuller and Moshe Bernstein, in Wady Daliyeh II: The Samaria Papyri from Wady Daliyeh, ed. D. M. Gropp; Qumran Cave 4.XXVIII: Miscellanea, Part 2, ed. M. Bernstein et al. (DJD 28; Oxford:

72 Aramaic texts also known from other caves: the New Jerusalem56 and a copy of the Book of Giants.57 The copy of Leviticus (2Q5) was written in palaeo-Hebrew characters, all the other manuscripts are in the later Hebrew alphabet, although a copy of Exodus (2Q3) has the tetragrammaton written also in palaeo-Hebrew as is done in some manuscripts from Caves 1, 3, 4 and 11. Cave 2 contains a couple of interesting characteristics when compared with the other caves: it holds the only copy of Sirach, a sapiential writing clearly different from the other sapiential writings from Caves 1 and 4, and it contains not a single yaḥad text (neither Serehk, nor CD, nor Milḥamah texts are represented in the holdings). But as for the rest, the profile is rather similar to that of Cave 1, even if the palaeographical dating of the manuscripts is later than that from Cave 1. According to the estimates of Baillet, the Cave 2 fragments range from the early to late Herodian period. Also according to Baillet, the editor of the fragments in DJD 3, the most noteworthy feature of the Cave 2 fragments are their occasionally sharp edges (e.g., 2Q1; 6; 11; 21 2; 24 8) which may be related to their destruction by enemies: “Leur destruction est d’ailleurs imputable a de multiples causes. La première et la plus ancienne semble avoir été l’intervention maligne des mains ennemies qui découpèrent certains rouleaux au moyen d’instruments tranchants.”58 This feature would favor a quick hiding scenario and would bring close the holdings of Cave 2 to the holdings of Cave 4 (of which the same has been asserted already by de Vaux;59 Cotton and Larson only accept as such the case of 4Q460/4Q35060). But after a thorough discussion of the matter, Popović concludes: “In light of the foregoing, I conclude that there is no evidence that clearly indicates Roman or auxiliary soldiers damaging manuClarendon Press, 2001), 165–97, 199–204. Both manuscripts overlap with 2Q22 and 4Q371 overlaps with 4Q373. E. Tigchelaar has published another copy, 4Q373a, “On the Unidentified Fragments of DJD XXXIII and PAM 43.680: A New Manuscript of 4QNarrative and Poetic Composition, and Fragments of 4Q13, 4Q269, 4Q525 and 4QSb,” RevQ 21 (2004): 477–85. 56  2Q24, Baillet, in DJD 3:84–89. Attested in 1Q32, 4Q554, 4Q554a, 4Q555, 5Q15 and 11Q18. 57  2Q26, Baillet, in DJD 3:90–91. Attested in 1Q23, 1Q24, 4Q203, 4Q530, 4Q531, 4Q532, 4Q533 and 6Q8. 58  Baillet, in DJD 3:45–46. 59  De Vaux, Archaeology and the Dead Sea Scrolls, 100, n. 3; Roland de Vaux, “Archeologie,” in Qumrân Grotte 4.II: I. Archéologie; II. Tefillin, Mezuzot et Targums (4Q128–4Q157) (DJD 6; ed. R. de Vaux and J. T. Milik; Oxford: Clarendon Press, 1977), 21–22. 60  Hannah Cotton and Erik Larson, “4Q460/4Q350 and Tampering with Qumran Texts in Antiquity?” in Paul, Emanuel, 113–25.

García Martínez

scripts from Cave 4 or any of the other Qumran caves.”61 I think we can conclude that the holdings from Cave 2 do not give us any reason to suppose that they represent a collection different from the better preserved manuscripts from Cave 1, and the multiple copies of Exodus, Numbers, Deuteronomy, Ruth or Jubilees cast doubt on the suggestion that we should envision the Cave 2 scrolls as those of a personal library. The extant holdings of Cave 362 are even more meager than those of Cave 2: extant are the remains of 6 fragmentary manuscripts identified by the editor, 3 texts qualified by Baillet as “Textes de charactère mal definie,” (3Q7–9), 2 groups of Hebrew fragments (3Q10–11) and 2 groups of Aramaic fragments (3Q12–13) non-identified, as well as some isolated fragments in Hebrew and Aramaic. And of course the Copper Scroll (3Q15).63 Baillet identified the fragments of Ezekiel (3Q1) and Psalms (3Q2) (one fragment each with only a few extant words), Lamentations (3Q3; two fragments with the Tetragrammaton in palaeo-Hebrew writing), one larger fragment (3Q4) with the remains of Isaiah 1 intersected with non-biblical words published as “commentaire d’Isaïe,” and from which Baillet says: “Ce que l’on saisit du commentaire donne l’impression qu’il était plus littérale que les pesharim connus.”64 3Q5 was published as “Une prophetie apocryphe”65 but the fragments were directly and independently identified by Rofé66 and by Deichgräber67 as a copy of Jubilees. Among the “Textes de Charactère mal defini” we have the 6 fragments of 3Q7, published as “Un apocryphe mentionant l’ange de la presence.”68 Baillet notes with some caution a possible relationship with the Testament of Judah, and in his edition of 4Q484 Testament de Juda (?) he is a little more affirmative: “Toute identification est impossible, mais les ff. 1, 7 et 19 pourraient suggérer une parenté avec le Testament de Juda . . . surtout si l’on se rappelle que 3Q7 61  Mladen Popović, “Roman Book Destruction in Qumran Cave 4 and the Roman Destruction of Khirbet Qumran Revisited,” in Qumran und die Archäologie: Texte und Kontexte (WUNT 278; ed. J. Frey, C. Claussen, and N. Kessler; Tübingen: Mohr Siebeck, 2011), 239–91 (250). 62  Baillet, in DJD 3:94–103. 63  Edited by Józef T. Milik, “Le rouleau de cuivre provenant de la grotte 3Q (3Q15),” in DJD 3:211–302. 64  Baillet, in DJD 3:96. 65  Baillet, in DJD 3:96–98. 66  Alexander Rofé, “Further Manuscript Fragments of the Jubilees in the Third Cave of Qumran,” Tarbiz 34 (1965): 333–36 [Hebrew]. 67  Reinhard Deichgräber, “Fragmente einer Jubiläen-Handschrift aus Höhle 3 von Qumran,” RevQ 19 (1965): 415–22. 68  Baillet, in DJD 3:99.

The Contents Of The Manuscripts From The Caves Of Qumran

pourrait bien représenter des restes du même passage.”69 Further, Cave 3 contained two very small fragments (3Q8) published as “Un texte mentionant un ange de paix (?),”70 and the three fragments of 3Q9, published as “Un texte de la secte”71—an identification without doubt because of the use of ‫ ועדתנו‬on the last line of frag. 3 together with the words ‫ אשמת פשע‬in line 2, a phrase also used in 1QS 9:4. The Copper Scroll (3Q15) has drawn a wide variety of interpretations, both in terms of its real meaning (a legend from folklore about imaginary treasures72 or a catalogue of hiding places for real treasures73), its origin (the Qumran community,74 the Temple of Jerusalem,75 or from the war of Bar Kokhba76), and its deposition date in the cave (before 70,77 around 100,78 later on79). As far as I can see there are three essential factors to decide upon regarding these matters: the date of the script, the language of the composition, and the place of deposition within the cave. As for the date of the script, according to the palaeographical study of Cross,80 the script “is to be placed in the second half of the Herodian era, that is, within the broad limits of AD 25–75;”81 Milik, however, prefers a date between 30 and 130 CE: “Au terme de l’enquête paléographique, on dispose d’un marge d’un siècle, env. 30–130 après J. C. pour situer l’écriture du catalogue; quelques indices semblent pourtant être en faveur de la

69  Maurice Baillet, Qumrân Grotte 4.III (4Q4Q482–4Q520) (DJD 7; Oxford : Clarendon Press, 1982), 3. 70  Baillet, in DJD 3:100. 71  Baillet, in DJD 3:100–1. 72  Milik, in DJD 3:211–302. 73  John M. Allegro, The Treasure of the Copper Scroll (London: Routledge & Kegan Paul, 1960). 74  Stephen Goranson, “Sectarianism, Geography, and the Copper Scroll,” JJS 43 (1992): 282–87 and most recently Emile Puech, The Copper Scroll Revisited (STDJ 112; Leiden: Brill, 2015). 75  As argued by the majority of scholars; see Judah K. Lefkovits, The Copper Scroll—3Q15: A Reevaluation (STDJ 25; Leiden: Brill, 1999). 76   Ernest-Marie Laperrousaz, “La grotte 3 de Qoumrân et le ‘Rouleau de cuivre’: L’origine et la nature du ‘Rouleau de cuivre’,” in Qoumrân et les manuscrits de la mer Morte: Un cinquantenaire (ed. E.-M. Laperrousaz; Paris: Éditions du Cerf, 1997), 207–13. 77   This was the common opinion in the 1950s and recently defended by Puech. 78  This is the opinion of Milik in the official edition of the scroll in DJD 3. 79  Ben Zion Luria, The Copper Scroll from the Judaean Desert (PIBRS 14; Jerusalem: Kiryat Sepher, 1963) [Hebrew]. 80  Frank M. Cross, “Écriture et chiffres,” in DJD 3:215–21. 81  Cross, in DJD 3:217.

73

seconde moitié de cette période.”82 No unanimity is thus among the scholars. The language of the composition is certainly closer to the defining features of Mishnaic Hebrew than Biblical Hebrew. But here also there is no unanimity. For Milik it is a spoken dialectal Hebrew,83 but for others it represents an earlier form of Mishnaic Hebrew and is thus related to the Hebrew of some other older compositions such as 4QMMT.84 As for the place of deposition within the cave, everybody agrees, already since it was first discovered by H. de Contenson, that the scroll was not found alongside the others in the cave, but somewhere else. But the interpretation of this fact is also disputed, since according to de Vaux, “Ils étaient un peu à l’écart de la masse des jarres et des couvercles brisés et on n’a recueilli dans leur voisinage aucun fragment écrit sur peau ou sur papyrus. Ces indices archéologiques ne suffisent pas à prouver que les rouleaux ont été déposés après la poterie et les autres textes mais ils ne s’opposent pas à une telle conclusion.”85 And Milik emphasized that “Le fait d’avoir trouvé dans la même grotte les textes esséniens (plut haut, 3Q1–14) et les rouleaux de cuivre n’est pas du tout décisif pour leur origine commune.”86 Because the Copper Scroll appears to be such an anomaly some scholars go so far as to exclude the contents of Cave 3 from the rest of the Qumran corpus. Daniel Stökl Ben Ezra in a forthcoming book is quite explicit: Die letzte Möglichkeit ist daher, zu erwägen, die Kupferrolle sei von Priestern in Höhle 3 verborgen worden und hänge nicht mit den anderen Rollen zusammen. Dies wiederum öffnet die Büchse der Pandora, dass nicht nur der Jachad als Bewohner Qumrans, sondern auch andere Gruppen in den Höhlen um Qumran Texte verborgen haben können. Besteht ein Unterschied zwischen den 15 Schriftrollen aus Höhle 3 und den Schriftrollenensembles in den anderen Qumranhöhlen? Zwar scheinen sie paläographisch aus der gleichen Zeit zu stammen, doch erzwingt der Inhalt keinesfalls eine Verbindung zu den anderen Qumranrollen. Kein einziges Fragment gibt den Anlass zur Vermutung 82  Milik, in DJD 3:283. 83  Milik, in DJD 3:227. 84   Albert Wolters, “The Copper Scroll and the Vocabulary of Mishnaic Hebrew,” RevQ 55 (1990): 483–95; Francisco Jiménez Bedman, El misterio del Rollo de Cobre de Qumrán: Análisis Lingüístico (Biblioteca Midrásica 25; Estella: Verbo Divino, 2002). 85  De Vaux, in DJD 3:201. 86  Milik, in DJD 3:276.

74

García Martínez

jachadischer Autorenschaft. Die Inventarliste gibt einen Pescher zu Jesaja an (3Q4)—dessen winziges Fragment allerdings nichts enthält, dass diese Identifikation berechtigen würde. Das gleiche gilt für 3Q9 „Sectarian Text“. Auch die anderen Fragmentgruppen (Ezekiel, Psalmen, Klagelieder, Jubiläenbuch, Hymne, Testament Judas, Text mentioning the Angel of Peace, nicht identifizierte hebräische und aramäische Fragmente) erzwingen keinen Schluss auf Zugehörigkeit dieses Korpus zu Qumran. Schließlich ist Höhle 3 von allen Höhlen mit Schriftrollen die nördlichste und am weitesten von Qumran entfernte Höhle. Von den fünf Krügen aus Höhle 3, die mit Neutronenaktivationsanalyse auf ihre Herkunft geprüft wurden, sollen alle aus Jerusalemton gefertigt worden sein. Angesichts des kumulativen Gewichts dieser Argumente ist die überzeugendste Lösung, nicht nur die Kupferrolle sondern auch die anderen Fragmente aus Höhle 3 aus dem Korpus der Qumranrollen auszuscheiden und sie einer anderen Gruppe zuzuordnen, die dem Tempel nahestand (Pfann).87 In spite of the difficulties surrounding the Copper Scroll, I see no hard evidence that proves without doubt it was deposited in Cave 3 at a time later than the other documents, and that we should therefore separate the deposit of the contents of this cave in two different periods: the initial hiding of the manuscripts 3Q1–14 followed by a later deposition of the Copper Scroll (3Q15) (something that otherwise would have been nothing abnormal in light of the regional context in which the caves served as places of refuge on different occasions). But, in my opinion, if we look to what can be ascertained from the scant remains of 3Q1–14 we need to conclude that the profile of Cave 3 is no different than the profiles of Caves 1 and 2: note the presence in Cave 3, next to the three “biblical” manuscripts, of a copy of Jubilees, of a pesher-like text, a possible copy of the Testament of Judah, and an ‘edah manuscript. A difference between the profile of Cave 3 and the profiles of Cave 1 and Cave 2 is that in Cave 3 we do not have multiple copies of the same composition. Since the Copper Scroll is, in my view, more of a documentary text than a literary one, it would not be completely dissonant with other documentary texts, lists, anales, etc. of the corpus found mostly in Cave 4, but 87  Daniel Stökl Ben Ezra, Lehrbuch Qumran (forthcoming). I thank Daniel Stökl Ben Ezra for sending me a copy of the manuscript of his forthcoming book and for allowing me to quote from it.

also in Cave 6. Much has been done by those wishing to exclude the Copper Scroll from the rest of the collection by examining the Greek letters, and particularly the Greek loanwords of the document, comparing them with the Greek and Latin loanwords in rabbinic literature, and then contrasting them with the lack of Greek loanwords in other documents (where do we find Persian and particularly Aramaic loanwords). My study of these loanwords, however, has shown that they are restricted to the architectural terminology and as such they are no different from the Greek loanwords of musical terminology we find in the much older Book of Daniel.88 Charlotte Hempel has recently published a thorough article examining the profile of Cave 4;89 after looking at what Cave 4 has in common with the other caves, particularly with Caves 1 and 11, and at the distinctive elements of Cave 4, she concludes: “A good case can be made that Cave 4 comprises the most eclectic and scholarly corner of the collection.”90 Indeed, a look at the “Annotated List of the Text of the Judaean Desert Classified by Content and Genre”91 shows that the Cave 4 contents are very closely connected to the contents of Cave 1, and that they are also related to the contents of Cave 11. But the same list also shows some of the peculiarities of the contents of Cave 4: Calendar texts are almost exclusively present in Cave 4,92 the same is evident with Sapiential texts, texts which Cave 4 has in common only with Cave 1,93 and, of course, with the texts written in Cryptic A.94 Calendar and Cryptic texts are generally considered to be “sectarian,” which would accentuate the commonalities of Cave 4 with the other caves, but not so with the Sapiential texts. As Tigchelaar notes, Sapiential texts “illustrate the ­merging 88  Florentino García Martínez, “Greek Loanwords in the Copper Scroll,” in Jerusalem, Alexandria, Roma: Studies in Ancient Cultural Interaction in Honour of A. Hilhorst (JSJSup 82; ed. F. G. Martínez and G. P. Luttikhuizen; Leiden: Brill, 2004), 119–45. 89   Charlotte Hempel, “ ‘Haskalah’ at Qumran: The Eclectic Character of Qumran Cave 4,” in eadem, The Qumran Rule Texts in Context: Collected Studies (TSAJ 154; Tübingen: Mohr Siebeck, 2013), 303–37. This chapter is found within the eigth part of the book titled: “Does 4Q Equal Qumran? The Character of Cave 4 Reconsidered.” It is published for the first time in her book. 90  Hempel, “ ‘Haskalah’ at Qumran,” 336. 91  Lange with Mittmann-Richert, “Annotated list of the Text of the Judaean Desert,” 115–64. 92  The only exception is 6Q17 published by Baillet in DJD 3:132– 33. The Calendar Texts from Cave 4 are analyzed by Hempel, “ ‘Haskalah’ at Qumran,” 319–29. 93  The only exceptions are 1Q26 and 1Q27 published by Milik in DJD 1:101–7. 94   The only exception is 11Q23, published in DJD 23:419–20. Hempel, “ ‘Haskalah’ at Qumran,” 312–17.

The Contents Of The Manuscripts From The Caves Of Qumran

together of many different kinds of knowledge, including the appropriation of non-Jewish scientific concepts, as well as the fusion of diverse literary forms and genres. This in turn suggests the rise of a new kind of Jewish scholar, who tries to integrate all available disciplines and fields of knowledge.”95 This sapiential interest would also explain “the presence of various scientific texts in the corpus, such as 4QHoroscope (4Q186), 4QPhysiognomy (4Q561), 4QZodiology and Brontology (4Q318), and perhaps also the various astronomical texts, such as the Astronomical Enoch and 4QLunisolar Calendar (4Q317).”96 Neither the texts classified by Lange as “Historical Texts and Tales” are considered “sectarian,” nor the “Documentary texts,” nor the “Letters” and the “Scribal exercises,” all of which are attested only in Cave 4.97 It is true that these texts are a clear minority within the holdings of the cave, and some of them are of an uncertain precedence, but they need to be explained within the general profile.98 Cave 4 contains nevertheless some oddities on its opistographs: for example, on the 4Q460/4Q350 scroll, the documentary text 4Q350, a Greek language ledger which ticks off quantities of cereals, was written on the verso of 4Q460 (4QNarrative Work and Prayer), or 4Q338 (4QGenealogical list) was written on the verso of 4Q201 (4QEnocha), 4Q414 (RitPur A), considered to be “sectarian,” was written on the verso of the “non-sectarian” 4Q415 (4QInstructiona), or 4Q433a, a Hodayot-like text written on the verso of 4Q255 (4QSa).99 Cave 5100 has yielded a limited number of manuscripts, 15 identified and 10 unclassified. The cave is located on the marl plateau, very close to the settlement. The cave that was discovered and excavated by archaeologists (as were Cave 3 and Cave 8), giving us the opportunity of considerer their holdings without the disturbance of the Bedouin’s previous excavation and dispersion. For de Vaux: “En dehors des débris de manuscrits et de quelques os de gros animal mêlés aux déblais, la grotte n’a livré aucun objet et pas un seul tesson de poterie. Malgré cette absence de mobilier, la grotte paraît avoir été creusée pour être une habitation.”101 De Vaux seems to refer to permanent habitation, in spite of the absence of objects and pottery, and 95  Tigchelaar, “The Dead Sea Scrolls,” 176. 96  Tigchelaar, “The Dead Sea Scrolls,” 176. 97  With the only exception of 6Q26, an Aramaic papyrus with remains of a contract, published by Baillet in DJD 3:138–39. 98  This seems to me perfectly possible within Hempel’s hypothesis. 99  These opistographs seem to me more difficult to explain within the same hypothesis. 100  Edited by Józef T. Milik, “Textes de la Grotte 5Q,” in DJD 3:167–97. 101  Roland de Vaux, “Les Grottes 5Q et 6Q,” in DJD 3:26.

75

the same assumption led Milik to define the holdings of Cave 5 as those of a private library: “Ils nous livrent les restes de la bibliothèque privée d’un anachorète essénien du Ier siècle de notre ère.”102 Milik dates, in fact, most of the manuscripts to the first century,103 or defines them as “écriture tardive” which “doit s’entendre: de la fin du Ier siècle avant J. C., ou des soixante-dix premières années du Ier siècle après J. C.”104 When we look at the contents of the 15 identified fragments we count 7 “biblical” scrolls, including two copies of Lamentations (5Q1 Deuteronomy, 5Q2 Kings, 5Q3 Isaiah, 5Q4 Amos, 5Q5 Psalms, 5Q6–7 Lamentations), a Phylactery in its case (5Q8), a composition published as “Ouvrage avec des toponymes,” which is a copy of the Apocryphon of Joshua, attested in Cave 4105 (5Q9), a composition “Ecrit avec des citations de Malachie” interpreted by Carmignac as a possible Malachie pesher (5Q10),106 a copy of the Serekh (5Q11), a copy of CD (5Q12), a composition entitled “Un règle de la secte” (5Q13), in which frag. 4 certainly quotes 1QS III 4–5, another composition with curses (5Q14), and a copy of the Aramaic New Jerusalem. This brief description of the contents makes clear that the profile of Cave 5 is rather similar to the profile of Caves 1 and 2. If Milik’s assertion were true that the contents of this cave are the remains of a private library, we would have a different profile. Milik is very specific in his assertions: some manuscripts “ont pu être apportés par notre ermite du dehors (nos 1 et 2), d’autres auront été recopies par lui-même dans son ermitage (l’un des deux exemplaires des Lamentations, nos. 6 et 7), mais la plupart doivent provenir de la bibliothèque ‘centrale’ du monastère.”107 Milik’s proposal seems logical. The number of manuscripts is not that high, although it is comparable to the amount found in Cave 6 and 11, and it offers a very similar profile to the holdings of Cave 1 and 4; but this has no external support, and does not seem compatible with the lack of ceramic and other material which would attest continuous habitation. My conclusion in the case of Cave 102  Milik, in DJD 3:167. 103  With the exception, of course, of 5Q1, a “écriture pré-hasmonéenne” dated to the first half of the second century BCE, and 5Q2, a “écriture hasmonéenne, postérieure à celle de 5QDeut, mais probablement relevant de la même école de scribes.” Milik, in DJD 3:169, 172. 104  Milik, in DJD 3:167. 105  4Q378 and 4Q389, edited by Carol Newsom in Qumran Cave 4.XVII: Parabiblical Texts, Part 3 (DJD 22; ed. G. Brooke et al.; Oxford: Clarendon Press, 1996), 241–62, 263–88. 106  Jean Carmignac, “Vestiges d’un Pesher de Malachie (?),” RevQ 13 (1964): 97–100. 107  Milik, in DJD 3:167.

76 5 is that its ­holdings are also a cross section of the corpus as a whole, and that the cave has the same profile as the holdings of the other caves examined until now. Cave 6 is situated on the cliff close to the water system; it was emptied by the Bedouins who took the manuscripts and also “une jarre et un bol,” although de Vaux considers the possibility of habitation doubtful.108 The holdings of Cave 6 comprise fragments of 31 scrolls, 18 of which have been identified by the editor who also proposed a tentative identification for the other 3.109 An interesting characteristic of this cave’s profile is that the high proportion of the texts found there is written on papyrus (more than 50%, if we count the unclassified fragments). Among the manuscripts copied on leather we find “biblical” texts: 6Q1 (Genesis), 6Q2 (Leviticus) (both written in palaeo-Hebrew) and 6Q6 (Canticles). We also find “non-biblical” compositions: 6Q11 (Allegory of the Vine), 6Q12 (Apocryphal Prophecy), 6Q13 (Priestly Prophecy), 6Q14 (Apocalypse Aramaic), 6Q15 (a copy of the Damascus Document with the divine name ‫ אל‬written in Palaeo-Hebrew; of the five fragments preserved, frag. 5 does not have correspondence within CD but with 4Q270 2 ii), 6Q17 (Calendrical Document) and the 3 tentatively identified fragments 6Q19–21. Among the manuscripts copied on papyrus we also find “biblical” texts: 6Q3 (Deuteronomy), 6Q4 (Kings), 6Q5 (Psalm 78?), 6Q7 (Daniel), and “non-biblical” compositions: 6Q8 (papEnGiants ar), 6Q9 (Apocryphon Sam-Kings), 6Q10 (Prophecy), 6Q16 (Benediction), 6Q18 (Hymn with divine name ‫אל‬ written in Palaeo-Hebrew) and all the unclassified fragments of 6Q22–31. Among these unclassified fragments Milik identified 6Q23110 as a copy of the Aramaic composition Words of Michael edited now by Puech as 4Q529,111 and Eshel has identified 6Q30 as a copy of Proverbs,112 6Q26 contains the extant remains of an Aramaic contract or of an account and is noteworthy because it provides a 108  Roland de Vaux, “Archéologie,” in DJD 3:10: “Douteux.” 109  Baillet, in DJD 3:105–41. The tentatively identified texts are: 6Q19, a “Texte en rapport avec la Genèse (?),” a single leather fragment with the remnants of 3 lines written in Aramaic; 6Q20, a “Texte en rapport avec le Deutéronome (?),” another single leather fragment written in Hebrew which may have come from the first column of the scroll; and 6Q21, a “Fragment prophétique (?),” also a single leather fragment with the remains of 3 lines from the top of a column. 110  Józef T. Milik, The Books of Enoch: Aramaic Fragments of Qumran Cave 4 (Oxford: Clarendon Press, 1976), 91. 111  Emile Puech, Qumrân Grotte 4.XXII: Textes araméens; Première partie 4Q529–549 (DJD 31; Oxford: Clarendon Press, 2001), 1–8. 112  Hanan Eshel, “6Q30, a Cursive Sin and Proverbs 11,” JBL 122 (2003): 544–46.

García Martínez

parallel to similar documentary texts from Cave 4 whose provenance is disputed. Michael Wise has raised the possibility that the manuscripts from Cave 6 represent the remains of a collection of private study copies, since a good number are written on papyrus and many in a cursive or semicursive hand.113 This has been rightly contested by Daniel Stökl Ben Ezra who has pointed out that the manuscripts were written by different persons and date to different eras.114 Thus is it unlikely that one hand would have copied them and that the characteristic assumed by Wise would provide us with a specific profile for this cave. The dubious possibility of habitation would certainly favor considering it as an emergency hiding place where scrolls and pottery were deposited at the time of the Roman attack, as in the other caves on the cliff. Pfann connects Cave 1 and Cave 6 as Essene caves, because of the presence of the Serekh, Milhamah, and Hodayot manuscripts in Cave 1, and the presence of CD in Cave 6. He suggests for this reason, and also because phylacteries and phylactery cases were found in Cave 1, that Cave 1 represents an Essene priestly library and Cave 6 an Essene lay library,115 a conclusion that seems to be contradicted by the fact that in Cave 4, both Serekh and CD texts were found, as well as phylacteries. After comparing the palaeographical dates of the documents of Cave 1 and Cave 6 Pfann concludes: “The final decline of collecting ends before AD25 for cave 1Q and about AD50 for cave 6Q, possibly indicating separate termini at which each group was forced to abandon the Qumran site.”116 Pfann thus assumes multiple collections, belonging to multiple groups, deposited at multiple times; but a careful look at the contents of the 18 identifiable scrolls do not seems to confirm these conclusions. Cave 6 does not have multiple copies of the compositions preserved unlike Cave 1. However, the use of the divine name ‫ אל‬written in Palaeo-Hebrew in 6Q15 and 6Q18 does correspond to the same use in 1Q14, 1Q27, and 1Q35, and the contents of the 18 identified manuscripts of Cave 6 show the same profile that we have found for the other caves with the same mix of religious texts, as indicated by the presence of 6Q15 (CD) and 6Q17 (a calendrical document), and thus

113  Michael O. Wise, “Accidents or Accidence: A Scribal View of Linguistic Dating of the Aramaic Scrolls from Qumran,” in idem, Thunder in Gemini and Other Essays on the History, Language and Literature of Second Temple Palestine (JSPSup 15; Sheffield: Sheffield Academic Press, 1994), 103–51 (120–22). 114  Stökl Ben Ezra, “Old Caves and Young Caves,” 323–24. 115  Pfann, “Reassessing the Judean Desert Caves,” 156–58. 116  Pfann, “Reassessing the Judean Desert Caves,” 158–59.

The Contents Of The Manuscripts From The Caves Of Qumran

b­ elonging to the same peculiar religious group that we have found in the other caves. Most scholars consider Cave 7 to be the deposit of a Qumran community member particularly interested in Greek, or the hiding place of the Greek works of the library, as the 18 papyri fragments and the papyri imprint on the clay found in the cave are all in Greek. Stökl Ben Ezra has formulated what seems to be a sort of communis opinio in this way: Some of the minor caves do indeed reflect a choice by the librarian or the reader. The most palpable case is Cave 7, which contains exclusively Greek documents and is the only cave to have revealed a considerable amount of Greek texts. Somebody with a particular taste for Greek readings probably dwelled in this cave at the time of the attack. The scrolls may reflect his private collection or borrowings from the main library. Alternatively, if the manuscripts were hidden in Cave 7 in an emergency, Cave 7 could reflect the Greek section of the library being secured in this cave just as Roman libraries usually had two sections, one for the Greek, one for the Latin books.117 But this “considerable amount of Greek texts” would be less considerable if 7Q4, 8, 11–14 were part of the same manuscript, as Emile Puech has proposed.118 In this case we do have only 3 identified manuscripts (7Q1 LXX Exodus, 7Q2 Epistle of Jeremiah, 7Q4, 8, 11–14 Enoch) and 9 unidentified fragments. Stökl Ben Ezra did not take into consideration Cave 7 in his analysis of the “Old Caves and Young Caves” (because the other caves with less that 10 dated manuscripts would be “statistically less reliable”119), but the 5 manuscripts dated by Roberts120 (frags. 1–4 are dated around 100 BCE and frag. 5 between 50 BCE and 50 CE) would make Cave 7 an “old cave,” which would be difficult to combine with his understanding of the cave as “a private collection or borrowings from the main library.”121 And Mladen Popović has astutely noted “There is an inscription (in Cave 7) of the name of Rom’a in Aramaic characters on a large jar, and that in a preliminary publication de Vaux refers to a small fragment in Hebrew from Cave 7, which was either a mistaken attribution or this 117  Stökl Ben Ezra, “Old Caves and Young Caves,” 323. 118  Emile Puech, “Sept fragments grecs de la lettre d’Hénoch (1 Hén 100, 103, 105) dans la grotte 7 de Qumrân (= 7QHén gr),” RevQ 70 (1997): 313–23. 119  Stökl Ben Ezra, “Old Caves and Young Caves,” 317. 120  As reported by Baillet in DJD 3:142. 121  Stökl Ben Ezra, “Old Caves and Young Caves,” 323.

77

fragment has since been overlooked.” He also signals that, “In most other Qumran caves we find Aramaic texts alongside Hebrew and, of course, some Greek manuscripts were also found on Cave 4. This does not suggest a linguistic division within the collection or collections of scrolls.” And concludes “This data, although meager—and this is also the case for the evidence concerning a single inhabitant with an interest in Greek texts only—corrects the impression of exclusively Greek writings from Cave 7.”122 If we consider the contents of the three identified manuscripts (Exodus, Epistle of Jeremiah, and 1 Enoch) we should conclude that the profile of Cave 7 is not essentially different from the profile of the other cave we have looked at up until now. Cave 8,123 located near Cave 7 and equally within the enclosure wall of the Khirbet, has preserved only 5 manuscripts: a copy of Genesis (8Q1), a copy of Psalms (8Q2), a Phylactery (8Q3), a Mezuzah (8Q4), and two fragments of a Hebrew manuscript, published by Baillet as a “passage hymnique,” and defined as “Le texte est de caractère hymnique, mais il faut sans doute le replacer dans le cadre d’un récit.”124 In my opinion, the text belongs rather to the exorcism genre or the like, as clearly indicated by the phrase ‫ בשמכה גנור אני מירא‬in 1 1 and the reference to ‫ הרוחות כול‬in 2 6. Taking into consideration all the material remains found in the cave, one is tempted to agree with de Vaux’s formulation: Si l’on considère seulement les grottes qui contenaient des documents écrits, la présence de ceux-ci s’explique de différentes façons. Ces textes peuvent être ceux qu’un membre ou un petit groupe de la communauté avaient à leur usage et qu’ils ont abandonnes dans la grotte qu’ils habitaient (grottes 5Q, 7 à 9Q, 11Q) ou qu’ils ont entreposés ou cachés, avec leur vaisselle, dans une cavité voisine de leur lieu de campement (grottes 2Q, 3Q, 6Q).125 Aside from the pottery, de Vaux describes the material remains of Cave 8 as follows: “Deux étuis à phylactère à quatre compartiments et un étui à phylactère à une seule case; les deux modèles s’étaient déjà rencontrés dans la grotte 1Q. Un morceau de semelle de sandale, une datte avec sa peau, une figue, plusieurs noyaux de dattes, un noyau d’olive. Beaucoup de fines lanières et des languettes 122  Popović, “Qumran as Scroll Storehouse,” 571. 123  Baillet, in DJD 3:147–62. 124  Baillet, in DJD 3:161. 125  De Vaux, in DJD 3:34.

78 de cuir. Restes d’étoffe et de ficelles.”126 According Joan Taylor the “fines lanières” were about 100 leather strips used for binding scrolls. For Taylor, these strips are evidence that the scrolls were processed for their burial in this cave, as in Cave 7 and 9.127 But, as Popović has noted, “the number of leather reinforcing tabs and thongs from Cave 8 is matched by those from Cave 4”128 and they can simply indicate that they were hidden with the scrolls. And in my opinion, olives, dates and the like point more to habitation than to a working place. But I am supposed to look to the contents of the manuscripts, and although more numerous than those of Caves 9 and 10, three compositions are too meager a basis to establish a profile, as even Pfann recognizes.129 Daniel Stökl Ben Ezra completely agrees: “The three scrolls from Cave 8 and the single manuscript from Cave 9, which may also reflect the private reading interests of their inhabitants or private property, are too few to sustain any argument.”130 The only thing I can add is that these 3 manuscripts are Jewish religious literature. Concerning the contents of the manuscripts from Cave 11, I can be equally short as with the contents of Cave 1, as I have previously written an article on the topic in 2010.131 In this article I tried to ascertain: What is peculiar to Cave 11? And, how do the materials from Cave 11 relate to the Qumran collection as a whole, now that all the manuscripts have been published? I also discussed: 1) Emanuel Tov’s proposition, which considers the contents of this cave more homogeneous and more sectarian than the contents of other caves132 (this, in my opinion, may be true for the scribal characteristics, but certainly not for the contents of the compositions), 2) the opinion of Daniel Stökl Ben Ezra who considers Cave 11 as a young cave133 (which to me seems correct since Herodian and late-Herodian manuscripts make up the majority of its contents, and does not require different deposition sce-

126  De Vaux, in DJD 3:31. 127  Taylor, “Buried Manuscripts and Empty Tombs,” 292. 128  Popović, “Qumran as Scroll Storehouse,” 572. 129  Pfann, “Reassessing the Judean Desert Caves,” 166. 130  Stökl Ben Ezra, “Old Caves and Young Caves,” 323. 131  Florentino García Martínez, “Cave 11 in Context,” in Hempel, The Dead Sea Scrolls: Texts and Context, 199–210. 132  Emanuel Tov, “The Special Character of the Texts Found in Qumran Cave 11,” in Things Revealed: Studies in Early Jewish and Christian Literature in Honor of Michael E. Stone (JSJSup 89; ed. E. Chazon, D. Satran, and R. Clements; Leiden: Brill, 2004), 187–96. 133  Stökl Ben Ezra, “Old Caves and Young Caves,” 313–33.

García Martínez

narios), and 3) de Vaux’s opinion, which considers the deposits of Caves 5, 7–9, and 11 as habitation caves and the that manuscripts as abandoned by the inhabitants.134 I also examined de Vaux’s notes on Cave 11 in his preliminary report135 and in his final statement in the Schweich Lectures,136 and concluded that the most likely scenario was an orderly hiding of some of the holdings of the collection of Qumran for safe keeping, such as in Cave 1: The location of Cave 11 is at a considerable distance from the Khirbet, the presence of the same jars and linen, attesting to the same way of preservation and transport of the manuscripts as in Cave 1, and even the fact that the entrance to Cave 11 was concealed in antiquity, would be consonant with this interpretation.137 Concerning the relationship of the holdings of Cave 11 with the holdings of the other caves I concluded: In my opinion the most significant observation to make in the wake of the full publication of the Scrolls in the DJD Series concerns the proportions of the categories of manuscripts which formed the collection as a whole. We now have some idea of the full spectrum of preserved material and are no longer dependent on best preserved manuscripts from Cave 1 which were published relatively speedily. Looking at the collection as a whole we notice a significant shift in the proportions of manuscripts that have been classified as “biblical,” “para-biblical,” and “sectarian.” In particular, the increased importance of “non-sectarian” “para-biblical” material compared with the other two categories is noteworthy. It is now possible to state without exaggeration that these sorts of materials constitute the majority of the collection outnumbering both the “biblical” and the “sectarian” manuscripts together. If we recall the overview over the contents of Cave 11 spelt out above (9 “biblical” texts, 3 or 4 “sectarian” compositions, and 8 or 9 “para-biblical” texts) we may safely conclude that the general profile of Cave 11 is very similar and practically identical to the profile of the 134  De Vaux, in DJD 3:34. 135  Roland de Vaux, “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e et 5e campagnes,” RB 63 (1956): 533–77. 136  Quoted in n. 3, but which apparently can only refer to temporary habitation. 137  García Martínez, “Cave 11 in Context,” 209.

The Contents Of The Manuscripts From The Caves Of Qumran

collection as a whole as it emerges today. Like the contents of Cave 1, the materials from Cave 11 form a perfect sample of the library of which the holdings of Cave 11 once formed a part and thus represent a cross-section of the Qumran collection as a whole.138

79

inadequate our historical data. This gives all the weight to the wise words of Eibert Tigchelaar: There is no evidence that all the manuscripts of the corpus were ever, at any time, together in one collection. Nor can one know, for that matter, whether all those manuscripts that were together at a certain time at the same place, were actively read and studied, or merely deposited. Even the status of Cave 4, as a library, repository, temporary place of concealment, or perhaps even a genizah, is unclear. We may not know the precise historical events which eventually resulted in the deposit of manuscripts in different caves at and near Qumran; we may conclude, however, from the contents of those manuscripts, that the corpus is not a random reflection of all kinds of available Jewish texts of the time, but is part of a specific current in Early Judaism.141

As in my review of Cave 1, I was operating here with the usual tripartite categories, and also as in the case of Cave 1, applying the categories of DJD 39139 and/or of Tigchelaar’s article140 would not make any difference. Upon completing this overview, it is time to make some conclusions. The review of the main articles on the topic showed us a polyphony of voices very far away from what for many years was the general opinion. Against the hypothesis of a single collection hidden in the caves for safe keeping from the Romans, the current debate now includes discussion of storehouses for scrolls (which is somehow compatible with the scenario of storing the scrolls for safekeeping against the Romans, even within a single deposit), the burial of manuscripts, genizahs, and multiple collections and multiple deposits (which is clearly incompatible with the traditional opinion). My review of the contents of all the caves leads me more in the direction of the traditional opinion, a single collection and a single deposit. But it is clear that the evidence preserved allows no conclusive argument to prove the point. We do have some good manuscripts in Cave 1 and Cave 11, which in my view are closely related, even if in Cave 11 we do not have the most characteristic yaḥad compositions. But for the rest of the caves, except for Cave 4, we have only meagre remains which can point in a certain direction, but which are too scanty to be considered as definitive proof. Besides, through the reports of other discoveries in antiquity we are now very much aware of how partial and accidental our evidence really is and how

We are not yet able to say exactly what this specific current in early Judaism was, nor if this current formed a large single group or if it was composed of by interrelated small groups which shared the same approach to Scripture and certain peculiar legal traditions (both would provide an explanation for the multiple copies of the same compositions). Therefore we are not able to prove definitively or disproof the traditional hypothesis or the new interpretations of the evidence we have. My analysis of the contents suggests that a single collection and a single deposit is a reasonable interpretation (but not hard proof) of the evidence we have. And since I do not believe anymore that the tripartite division of categories I have used in the past are helpful to understand the collection or collections of the caves, I will simply define it as a collection of religious Jewish literature from the Second Temple era.

138  García Martínez, “Cave 11 in Context,” 210. 139  Quoted in n. 41. 140  Quoted in n. 42.

141  Tigchelaar, “The Dead Sea Scrolls.”

CHAPTER 4

The Profile and Character of Qumran Cave 4Q: The Community Rule Manuscripts as a Test Case Charlotte Hempel 1 Introduction The Qumran marl Cave 4Q was discovered by the Bedouin in 1952 and revealed the lion’s share of texts to have emerged from the caves in the vicinity of Khirbet Qumran. Scholars estimate Cave 4Q to have contained almost 700 fragmentary manuscripts.1 Several scholars have suggested Cave 4Q is a library2 or the “stacks” of a library.3 Thus, Schiffman has interpreted the holes in the 1  Roland de Vaux, “Archéologie,” in Les ‘petites grottes’ de Qumrân: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 9–22; George J. Brooke, Qumran and the Jewish Jesus: Reading the New Testament in the Light of the Scrolls (Cambridge: Grove Books, 2005), 9; Devorah Dimant, “The Qumran Manuscripts: Contents and Significance,” in Time to Prepare the Way in the Wilderness: Papers on the Qumran Scrolls by Fellows of the Institute for Advanced Studies of the Hebrew University, Jerusalem, 1989–1990 (STDJ 16; ed. D. Dimant and L. H. Schiffman; Leiden: Brill, 1995), 23–58; Józef T. Milik, Ten Years of Discovery in the Wilderness of Judaea (SBT 26; trans. J. Strugnell; London: SCM Press, 1959), 16–18, 20; Mladen Popović, “The Manuscript Collections: An Overview,” in T&T Clark Companion to the Dead Sea Scrolls (ed. G. J. Brooke and C. Hempel; London: T&T Clark, forthcoming); Lawrence H. Schiffman, Reclaiming the Dead Sea Scrolls: The History of Judaism, the Background to Christianity, the Lost Library of Qumran (Philadelphia, PA: Jewish Publication Society, 1994), 54–56; Roland de Vaux, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973), 52; Sidnie White Crawford, “Qumran: Caves, Scrolls, and Buildings,” in A Teacher for All Generations: Essays in Honor of James C. VanderKam (JSJSup 153; ed. E. F. Mason et al.; Leiden: Brill, 2012), 1:253–73 (266–67). 2  Dimant, “The Qumran Manuscripts,” 36; Armin Lange, “The Qumran Dead Sea Scrolls—Library or Manuscript Corpus?” in From 4QMMT to Resurrection: Mélanges qumraniens en hommage à Émile Puech (STDJ 61; ed. F. García Martínez, A. Steudel, and E. J. C. Tigchelaar; Leiden: Brill, 2006), 177–93 (191); Schiffman, Reclaiming the Dead Sea Scrolls, 54–56. 3  Hartmut Stegemann, The Library of Qumran: On the Essenes, Qumran, John the Baptist, and Jesus (Grand Rapids, MI: Eerdmans, 1998), 74; Daniel Stökl Ben Ezra, “Old Caves and Young Caves: A Statistical Reevaluation of a Qumran Consensus,” DSD 14 (2007): 313–33 (327–29).

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_006

walls of the main chamber of Cave 4Q (technically to be divided into two sub-caves) as evidence of fixtures of ancient “shelves”—the wooden parts of which would not have survived the ravages of time.4 Jean-Baptiste Humbert envisages a carefully planned concealed library complex hollowed out in the marl terrace.5 Joan Taylor, by contrast, proposes that Cave 4Q was used in order to offer temporary storage for manuscripts ultimately destined for what she calls “preservation-burial.”6 Different explanations have been put forward to explain the many fragments that were found scattered across the floor of Cave 4Q. Taylor presupposes ancient disruption.7 Frank Moore Cross, by contrast, suggested the messy scenario went back to the circumstances associated with a hasty deposit.8 Reservations about the oft repeated suggestion that the Roman army entered Cave 4Q and caused considerable damage to its contents have recently been raised by Mladen Popović.9 Ever since we have been able to get a sense of the scope and nature of the contents of all eleven scroll caves from Qumran scholars have been able to fathom and describe the profile of the collection—or perhaps collections—for the first time. A number of scholars have started to conceive of a plurality of perhaps inter-related collections at Qumran raising the possibility that some 4  Schiffman, Reclaiming the Dead Sea Scrolls, 56. 5  Jean-Baptiste Humbert, “L’espace sacré à Qumrân: Propositions pour l’archéologie,” RB 101 (1994): 161–214 (194–95). 6   Joan E. Taylor, “Buried Manuscripts and Empty Tombs: The Qumran Genizah Theory Revisited,” in ‘Go Out and Study the Land’ ( Judges 18:2): Archaeological, Historical and Textual Studies in Honor of Hanan Eshel (JSJSup 148; ed. A. M. Maeir, J. Magness, and L. H. Schiffman; Leiden: Brill, 2011), 269–315 (294–95). 7  Taylor, “Buried Manuscripts and Empty Tombs,” 299; see also de Vaux, in DJD 3:21–22 and Milik, Ten Years of Discovery, 20. 8  Frank M. Cross, The Ancient Library of Qumran and Modern Biblical Studies (Grand Rapids, MI: Baker, 1961), 27 followed by White Crawford, “Qumran: Caves, Scrolls, and Buildings,” 272. 9  Mladen Popović, “Roman Book Destruction in Qumran Cave 4 and the Roman Destruction of Khirbet Qumran Revisited,” in Qumran und die Archäologie: Texte und Kontexte (WUNT 278; ed. J. Frey, C. Claussen, and N. Kessler; Tübingen: Mohr Siebeck, 2011), 239–91.

The Profile And Character Of Qumran Cave 4q

caves reflect a specific sub-section of a larger library or a reader’s preferences.10 A similar trend has for some time characterised recent work on the communities and movement behind the collection. Earlier scholarship was fairly confident we are dealing with a single community resident at Qumran alongside a wider camp movement spread across the region. More recently a number of scholars, myself included, propose to read the Rule texts as reflecting a number of inter-related communities.11 My own previous work has argued for evidence in the Rule texts ranging from small fellowship groups that met in a variety of places to eat, pray and exchange counsel (1QS 6:1c–3a // 4QSd 2: 6b–7a // 4QSg 2a-c 1–2a // 4QSi 2b–3) to more ­complex communities such as those being described in the most developed portions of the text as presented by the regulations for sessions of the many (1QS 10   Stephen J. Pfann, “Reassessing the Judean Desert Caves: Libraries, Archives, Genizas and Hiding Places,” BAIAS 25 (2007): 147–70; Popović, “The Manuscript Collections;” Stökl Ben Ezra, “Old Caves and Young Caves,” 316, 322–23; see also idem, “Wie viele Bibliotheken gab es in Qumran?” in Frey, Claussen and Kessler, Qumran und die Archäologie, 327–46 (333). 11  George J. Brooke, “From Jesus to the Early Christian Communities: Modes of Sectarianism in the Light of the Dead Sea Scrolls,” in The Dead Sea Scrolls and Contemporary Culture: Proceedings of the International Conference held at the Israel Museum, Jerusalem ( July 6–8, 2008) (STDJ 93; ed. A. Roitman, L. H. Schiffman, and S. Tzoref; Leiden: Brill, 2011), 413–34; John J. Collins, Beyond the Qumran Community: The Sectarian Movement of the Dead Sea Scrolls (Grand Rapids, MI: Eerdmans, 2010); Philip R. Davies, “The ‘Damascus’ Sect and Judaism,” in Pursuing the Text: Studies in Honor of Ben Zion Wacholder on the Occasion of his Seventieth Birthday (JSOTSup 184; ed. J. Reeves and J. Kampen; Sheffield: Sheffield Academic Press, 1994), 70–84, reprinted in idem, Sects and Scrolls: Essays on Qumran and Related Topics (SFSHJ 134; Atlanta, GA: Scholars Press, 1996), 163–77; Florentino García Martínez, “¿Sectario, no-sectario, o qué? Problemas de una taxonomía correcta de los textos qumránicos,” RevQ 91 (2008): 383–94; Charlotte Hempel, The Qumran Rule Texts in Context: Collected Studies (TSAJ 154; Tübingen: Mohr Siebeck, 2013), 47–62, 79–105, 293–98; eadem, “Qumran Communities: Beyond the Fringes of Second Temple Society,” in The Scrolls and the Scriptures: Qumran Fifty Years After (JSPSup 26; ed. S. Porter and C. Evans; Sheffield: Sheffield Academic Press, 1997), 43–53; Sarianna Metso, “Whom does the Term Ya�ad Identify?” in Biblical Traditions in Transmission: Essays in Honour of Michael A. Knibb (JSJSup 111; ed. C. Hempel and J. M. Lieu; Leiden: Brill, 2006), 213–35; Eyal Regev, “Between Two Sects: Differentiating the Ya�ad and the Damascus Covenant,” in The Dead Sea Scrolls: Texts and Context (STDJ 90; ed. C. Hempel; Leiden: Brill, 2010), 431–49; Alison Schofield, From Qumran to the Yaḥad: A New Paradigm of Textual Development for the Community Rule (STDJ 77; Leiden: Brill, 2009); Eibert J. C. Tigchelaar, “The Dead Sea Scrolls,” EDEJ 163–80.

81 6:8b–13a // 4QSb 11:5–8 // 4QSd 3:1–3), the complex admission process (1QS 6:13b–23 // 4QSb 11:8,11–13 // 4QSg 3:1), and the penal code (cf. 1QS 6:24–7:25 // 4QSd 5:1 // 4QSe 1:4–15; 2:3–9 // 4QSg 3:2–4; 4a-b:1–6; 5a-c:1–9; 6a-e:1–5).12 The complexity of the evidence is further enhanced by the presence of a large number of at times divergent copies of the Community Rule in Cave 4Q. This has provoked a range of interpretations. Alison Schofield and John Collins take the evidence to be indicative of a number of interrelated groups who each promulgated their own versions of the Rule and endeavoured to live by it.13 By contrast, I have argued elsewhere that the textual plurality of the Rules at Qumran is analogous to the textual plurality that so surprised us in the study of the “biblical” manuscripts from Qumran.14 I therefore argued that the movement behind the Scrolls—and perhaps Second Temple Judaism more widely—was comfortable with producing and preserving several versions of a text side by side. Ground-breaking work was pursued by Devorah Dimant with the publication of a pioneering and important paper on the profile of the library as a whole in 1995.15 Dimant’s particular concern was an initial comprehensive analysis of the corpus from a literary perspective, and she outlines her aims to meet “the need for a comprehensive literary description of the Qumran collection.”16 The issue of provenance was at the forefront of her attention, and she proposed to distinguish between non-biblical texts that attest “Community Terminology” (CT) and those “without Community Terminology” (NCT).17 In 2007 Daniel Stökl Ben Ezra offered another ­wide-ranging proposal on the profile of the contents of the Qumran scroll caves by way of an investigation of the average age of the scrolls found in a given cave which he has developed in a number of studies since. His results led him to distinguish between “old caves” (Caves 1Q and 4Q) and “young caves” (Caves 2Q, 3Q, 5Q, 6Q, and 11Q). In order to account for the differences he suggested two deposits in 9/8 BCE and 68 CE respectively.18 Stökl Ben Ezra is here 12  For bibliographical details of my own contributions see n. 11 above. 13  Collins, Beyond the Qumran Community and Schofield, From Qumran to the Yaḥad. 14  Hempel, The Qumran Rule Texts in Context, 271–84. 15  Dimant, “The Qumran Manuscripts.” 16  Dimant, “The Qumran Manuscripts,” 25. 17   Dimant, “The Qumran Manuscripts,” 26 passim; and more recently eadem, “The Vocabulary of the Qumran Sectarian Texts,” in Frey, Claussen and Kessler, Qumran und die Archäologie, 347–95. 18  Stökl Ben Ezra, “Old Caves and Young Caves;” see also Florentino García Martínez, “Reconsidering the Cave 1 Texts Sixty Years

82 building on the argument by Jodi Magness that the site of Qumran was briefly abandoned in 9/8 BCE as a result of an enemy attack.19 The topic of the profile of the caves was further investigated in a series of further studies.20 One conclusion that has been reached by a number of scholars—even those that otherwise disagree with one another—is some kind of a connection between Caves 1Q and 4Q.21 Particularly tangible evidence in support of this is the fact that the same scribe apparently copied 4QTestimonia (4Q175), 1QS, 1QSa, 1QSb, 4QSamc and corrected 1QIsaa.22 In addition, Ada Yardeni assigned a small number of fragments from Cave 1Q (chiefly New Jerusalem and perhaps Jubilees) and as many as 47 “apparent” and a further 27 “possible” manuscripts to the work of the same prolific individual scribe.23 After Their Discovery: An Overview,” in Qumran Cave 1 Revisited: Proceedings of the Sixth Meeting of the IOQS, Ljubljana 2007 (STDJ 91; ed. D. Falk, S. Metso, and E. J. C. Tigchelaar; Leiden: Brill, 2010), 1–13; idem, “Cave 11 in Context,” in Hempel, The Dead Sea Scrolls: Texts and Context, 199–209; Daniel Stökl Ben Ezra, “Further Reflections on Caves 1 and 11: A Response to Florentino García Martínez,” in Hempel, The Dead Sea Scrolls: Texts and Context, 211–23; and idem, “Wie viele Bibliotheken.” 19  Jodi Magness, The Archaeology of Qumran and the Dead Sea Scrolls (Grand Rapids, MI: Eerdmans, 2002), 67–68. 20   See especially Brooke, Qumran and the Jewish Jesus, 8–10; Devorah Dimant, “The Composite Character of the Qumran Sectarian Literature as an Indication of Its Date and Provenance,” RevQ 88 (2006): 615–30; García Martínez, “Cave 11 in Context;” idem, “Reconsidering the Cave 1;” Lange, “The Qumran Dead Sea Scrolls;” Frederick F. Bruce and Stephen J. Pfann, “Qumran,” EncJud 16:768–75; idem, “Reassessing the Judean Desert Caves;” Popović, “The Manuscript Collections;” Stegemann, Library of Qumran, 58–79; Taylor, “Buried Manuscripts;” Tigchelaar, “Dead Sea Scrolls;” Emanuel Tov, “The Special Character of the Texts Found in Qumran Cave 11,” in Things Revealed: Studies in Early Jewish and Christian Literature in Honor of Michael E. Stone (JSJSup 89; ed. E. G. Chazon, D. Satran, and R. A. Clements; Leiden: Brill, 2004), 187–96; and White Crawford, “Qumran: Caves, Scrolls, and Buildings,” 265–73. 21  Dimant, “The Composite Character of the Qumran Sectarian Literature;” Stökl Ben Ezra, “Old Caves and Young Caves;” see also García Martínez, “Cave 11 in Context.” 22  See Jonathan Campbell, The Exegetical Texts (CQS 4; London: T&T Clark, 2004), 89; Emanuel Tov, Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (STDJ 54; Leiden: Brill, 2004), 23; Eibert J. C. Tigchelaar, “In Search of the Scribe of 1QS,” in Emanuel: Studies in Hebrew Bible, Septuagint, and the Dead Sea Scrolls in Honor of Emanuel Tov (VTSup 94; ed. S. M. Paul et al.; Leiden: Brill, 2003), 439–52. 23  Ada Yardeni, “A Note on a Qumran Scribe,” in New Seals and Inscriptions, Hebrew, Idumean and Cuneiform (HBM 8; ed. M. Lubetski; Sheffield: Phoenix Press, 2007), 287–98; see also White Crawford, “Qumran: Caves, Scrolls, and Buildings,” 267.

Hempel

As has been clearly shown by the landmark study by Devorah Dimant mentioned above Qumran Cave 4Q gives the impression of lying at the very heart of the Qumran collection with many texts that are found in this cave also attested elsewhere.24 The centrality of Cave 4Q was already mooted in some pioneering studies, and both De Vaux and Milik held that the contents of Cave 4Q belonged to the main library once found on the settlement and subsequently hastily concealed in the marl cave to protect them from the consequences of an imminent Roman attack.25 2

Distinctive Features of Cave 4Q

Dimant’s initial assessment of the central position of Cave 4Q has withstood the test of time. My own recent contribution to this debate emphasized that alongside important connections to the contents of Cave 1Q in particular, the texts from Cave 4Q display some important distinctive features which are best accounted for by considering them to represent a learned and eclectic medley of materials and data that were in all probability reserved for a more restricted readership than the contents of the remainder of the library.26 My proposal that the contents of Cave 4Q testify to a scholarly and eclectic sub-collection accessible to the highest tiers of community members is based on the following features which characterise Cave 4Q:

• All but one of the cryptic texts hail from Cave 4Q—the single exception being the unidentified text 11Q23.27 • The majority of references to the Maskil, a figure also

associated with the cryptic material (cf. 4Q298 Words of the Maskil to the Sons of Dawn, which after an unencrypted title continues in the cryptic script), are attested in Cave 4Q texts:

24  Dimant, “The Qumran Manuscripts.” 25  Milik, Ten Years of Discovery, 20; de Vaux, Archaeology and the Dead Sea Scrolls, 105. 26  Hempel, The Qumran Rule Texts in Context, 303–37. Talmon suggested different readers for various calendrical materials ranging from general membership to a priestly audience, see Shemaryahu Talmon, “Calendrical Documents and Mishmarot, in Qumran Cave 4.XVI: Calendrical Texts (DJD 21; ed. S. Talmon, J. Ben Dov, and U. Glessmer; Oxford: Clarendon Press, 2001), 14. 27  For the three small fragments of 11Q23 preserving seven letters see Florentino García Martínez, Eibert J. C. Tigchelaar, and Adam S. van der Woude, eds., Qumran Cave 11.II: 11Q2–18, 11Q20–31 (DJD 23; Oxford: Clarendon Press, 1997), 419–20 and Pl. 48.

83

The Profile And Character Of Qumran Cave 4q

ing divine scheme that guides time and history.31 Most important for my own argument in favour of the eclectic character of Cave 4Q is the important conclusion to have emerged from the full publication of the calendar texts of a lack of a unified “calendar position.” In place of a clear party line that privileges the solar calendar, Cave 4Q contained an array of technical calendrical learning.32

The Maskil in CD and at Qumran CD

Cave 1Q

Cave 4Q

Cave 11Q

X2

X11

X26

x1

It is important to stress that the prominence of the Maskil in Cave 4Q goes beyond the number of attestations since the figure is linked not only to the significant block of cryptic texts found almost exclusively in this cave, but also to calendrical learning as I will argue below. Cave 4Q is the home of the bulk of technical calendrical materials from Qumran. In fact, the only exception is 6Q17, a small parchment fragment published by Baillet in DJD 328 that has often been erroneously described as a papyrus fragment since it was photographed on a plate alongside a number of papyrus fragments. The full publication of the rich calendrical material from Qumran has led to some major re-assessments of the place of the calendar in the movement behind the Scrolls. The full evidence now in front of us paints an intricate and difficult picture many aspects of which are still being debated. Thus one of the editors of the calendrical texts, Jonathan Ben Dov, refers to “multiple 364-day calendar traditions.”29 The complexity of the evidence is apparent already in what I have labelled the “transient nature” of the designations assigned to the calendrical corpus with many a Calendrical Text being re-labelled Mishmarot and vice versa. Other texts previously labelled Mishmarot are now called Historical texts in order to emphasize their interest in particular historical events or figures placed in a framework of priestly courses.30 A number of compositions reach as far back as the creation of the luminaries in Genesis 1:14–16 attesting to the far reach-

•

28  Baillet, in DJD 3:132–33. 29  Jonathan Ben Dov, Head of All Years: Astronomy and Calendars at Qumran in Their Ancient Context (STDJ 78; Leiden: Brill, 2008), 11, cf. also Talmon, in DJD 21:1, 14; James C. VanderKam, Calendars in the Dead Sea Scrolls: Measuring Time (Literature of the Dead Sea Scrolls; London: Routledge, 1998), 69, 86; also Uwe Glessmer, “Calendars in the Qumran Scrolls,” in The Dead Sea Scrolls After Fifty Years: A Comprehensive Assessment (ed. P. W. Flint and J. C. VanderKam; Leiden: Brill, 1999), 2: 213–78 (268). 30  See Stephen J. Pfann, ed., Qumran Cave 4.XXVI: Cryptic Texts; Philip S. Alexander et al., eds.; Miscellanea, Part 1 (DJD 36; Oxford: Clarendon Press, 2000), 275–89, cf. also Talmon, in DJD 21:2.

We noted above the explicit association of the Maskil with the cryptic Qumran corpus in the title of 4Q298 (Words of the Maskil to the Sons of Dawn) and already alluded briefly to an association of the Maskil with calendrical issues. A connection of the Maskil to matters calendrical, especially liturgical aspects of the calendar, is suggested by the Final Hymn attested in several manuscripts of the Community Rule and rightly commonly associated with the Maskil.33 The Hymn includes a liturgical calendar in 1QS 10:1b–8 // 4QSb 19:1–6 // 4QSd 8:11b–9:7a // 4QSf 2:1–5. Moreover both heavenly luminaries and annual festivals feature in the Songs of the Maskil (see 4Q511 2 i). A connection between the cryptic texts and the topic of calendrical lore is evident, finally, by eight currently identified cryptic texts dealing with calendrical matters.34 In sum, we witness a confluence of the three facets of Cave 4Q outlined thus far: the transmission of material in cryptic script, calendrical lore, and the sphere of influence of the Maskil. While the Maskil is clearly not exclusive to Cave 4Q, this confluence strengthens the suggestion that the contents of this cave chime particularly with what we know of the sphere of influence of this figure.35

•

Cave 4Q is further characterised by having contained the largest number of works in multiple copies with 21 compositions attested in 5–9 copies; 5 in 10–14 copies;

31  See Sacha Stern, “Qumran Calendars and Sectarianism,” in The Oxford Handbook of the Dead Sea Scrolls (ed. T. H. Lim and J. J. Collins; Oxford: Oxford University Press, 2010), 232–53 (242). 32  See, e.g., Steven D. Fraade, Legal Fictions: Studies of Law and Narrative in the Discursive Worlds of Ancient Jewish Sectarians and Sages (JSJSup 147; Leiden: Brill, 2011), 255–83. 33  Manfred Weise, Kultzeiten und Kultischer Bunderschluss in der “Ordensregel” vom Toten Meer (SPB 3; Leiden: Brill, 1961), 3. 34  For details see Hempel, The Qumran Rule Texts in Context, 312–17. 35  For an insightful recent study of this figure see Judith H. Newman, “Speech and Spirit: Paul and the Maskil as Inspired Interpretes of Scripture,” in The Holy Spirit, Inspiration, and the Cultures of Antiquity: Multidisciplinary Perspectives (Ekstasis 5; ed. J. Frey and J. R. Levison with the collaboration of Andrew Bowden; Berlin: W. de Gruyter, 2014), 243–66.

84

Hempel

and Isa, Deut, and Psalms represented in 15 or more copies.36 Furthermore, a significant number of texts from Cave 4Q appear less finessed, a characteristic which I have described as the “workaday quality” of a number of Cave 4Q texts.37 Thus, Cave 4Q revealed a considerable number of anthologies of various kinds such 4Q265 Miscellaneous Rules; 4Q159 Ordinancesa; the calendrical anthologies 4QOtot; 4Q320 Calendrical Document and 4Q329 Mishmarot G; 4Q176 Tan�unim, an anthology on the theme of divine comfort;38 4Q174 Florilegium, an exegetical collection39 as well as 4QTestimonia which Steudel suspects to be the personal notes (“Handzettel”) of the compiler.40 Cave 4Q also contained a relatively large proportion of papyri and what I have called “raw data” in the form of calendrical rosters and registers. I contrast the workaday quality of a significant proportion of Cave 4Q with evidence of refinement in Cave 1Q as indicated by the most developed form of documents such as the Community Rule, the War Scroll, and the Hodayot emerging from Cave 1Q.41 I summed up my discussion on this issue as follows,

•

One almost gains the impression that large parts of Cave 4Q constitute the laboratory of a learned group where data, texts, and ideas are collected and experimented with over against the show room quality of Cave 1Q.42 3

The Community Rule as a Test Case for the Character of Cave 4Q

In what follows I will reflect on this picture with a particular focus on the Serekh tradition which is one of the most amply attested compositions that spans across perhaps as many as four caves.43 36  See Hempel, The Qumran Rule Texts in Context, 330. 37  Hempel, The Qumran Rule Texts in Context, 332. 38  See Campbell, The Exegetical Texts, 78–87 and Johann Maier, “Tan�umin and Apocryphal Lamentations,” EDSS 2:915. 39  See George J. Brooke, “Florilegium (4Q174),” EDEJ 646–47 (647). 40  Annette Steudel, Der Midrasch zur Eschatologie aus der Qumrangemeinde (4QMidrEschata.b): Materielle Rekonstruktion, Textbestand, Gattung, traditionsgeschichtliche Einordnung des durch 4Q174 (“Florilegium”) und 4Q177 (“Catena A”) repräsentierten Werkes aus den Qumranfunden (STDJ 13; Leiden: Brill, 1994), 178–81. 41  See Dimant, “The Composite Character of the Qumran Sectarian Literature.” 42  Hempel, The Qumran Rule Texts in Context, 335. 43  Cave 1Q (1QS), cave 4Q (4QSa–j), cave 5Q (5QS[?] or 5Q11), and cave 11Q (11QFragment Related to Serekh ha-Ya�ad [11Q29]).

3.1 Multiple Copies The Community Rule is represented by up to twelve compositions of the Serekh, ten of which hail from Cave 4Q, a tally that may increase as we refine our assessment of very fragmentary compositions.44 3.2 Cryptic Script 4Q259 (Se) contains two phrases written in cryptic letters in successive lines. A rather difficult passage in 4Q259 Se 2:3 (par. 1QS 8:12) was first deciphered by Milik as reading “in Israel” written in cryptic letters, a reading more recently discussed and substantiated at some length by Emile Puech.45 What makes this passage more difficult is that the cryptic writing goes back to a second hand correcting the earlier unencrypted reading “in the community (yaḥad).” A further complication is the fact that the last two cryptic letters (alef and lamed) of beyiśrael are written above the line.46 Puech draws attention to the occurrence of same word “Israel” unencrypted in 4Q259 1 ii 13 (1QS 8:4–5) and partially preserved in 2:18 (1QS 8:9).47 In the immediate context of the cryptic phrases we note a pronounced concern with threatening elements within or on the periphery of the community. Thus 1QS 8:11b–12a // 4QSd 6:5b–6a // 4QSe 3:2 refers to fear of a renegade spirit, and 1QS 8:13 // 4QSd 6:7 // 4QSe 3:3b–4a returns to the issue of distancing community members 44  See Jutta Jokiranta and Hanna Vanonen, “Multiple Copies of Rule Texts or Multiple Rule Texts: Boundaries of the S and M Documents,” in Crossing Imaginary Boundaries: The Dead Sea Scrolls in the Context of Second Temple Judaism (ed. M. S. Pajunen and H. Tervanotko; Helsinki: Finnish Exegetical Society, 2015), 11–60. 45  See Józef T. Milik, “Le travail d’édition des fragments manuscripts de Qumrân,” RB 63 (1956): 60–62 (61) and Emile Puech, “L’alphabet cryptique A en 4QSe (4Q259),” RevQ 71 (1998): 429–35. See also Elisha Qimron, The Dead Sea Scrolls: The Hebrew Writings (Jerusalem: Yad Ben Zvi, 2010), 1:225 and Tov, Scribal Practices, 205–6. For reservations about the identification of cryptic script see Philip S. Alexander and Geza Vermes, eds., Qumran Cave 4.XIX: Serekh Ha-Yaḥad and Two Related Texts (DJD 26; Oxford: Clarendon Press, 1998), 145–46. See also Sarianna Metso, The Textual Development of the Qumran Community Rule (STDJ 21; Leiden: Brill, 1997), 53–54. Recently published multi-spectral images support the readings of Milik and Puech, see The Leon Levy Dead Sea Scrolls Digital Library, images B-295966 and B-314657. 46   Cf. Alexander and Vermes, DJD 26:145–46; Milik, “Le travail d’édition;” Sarianna Metso, “The Primary Results of the Reconstruction of 4QSe,” JJS 44 (1993): 303–8; eadem, The Textual Development of the Qumran Community Rule, 53–54; and Puech, “L’alphabet cryptique A en en 4QSe.” 47  Puech, “L’alphabet cryptique A en 4QSe,” 433; see also Alexander and Vermes, DJD 26:146.

85

The Profile And Character Of Qumran Cave 4q

from the people of injustice already familiar from 1QS 5:10b–18a // 4QSb 9:8b–11 // 4QSd 1:7b–10. The concern with rebutting opponents is a possible motive already identified by Prof. Puech as lying behind the cryptic notes.48 To this we may add the repeated occurrence of the verb “to hide.” The notion of concealed knowledge is referred to in the context of affairs hidden from Israel which, if found by the scholar, the latter is instructed not to hide, cf. 1QS 8:11–12 // 4QSd 6:5–6 // 4QSe 3:2. However, the tenor of what is being said is rather nuanced: while promoting an esoteric framework in referring to matters hidden from Israel, the particular emphasis admonishes the scholar not to hide his insights from the sub-group within Israel, in a sense encouraging more internal openness.49 This again is somewhat counter-balanced by the use of cryptic letters at two key points in the text of 4QSe and—on Prof. Puech’s analysis already its Vorlage—which left our first copyist baffled and necessitated the second more expert hand to step in and first correct and then supplement the space left for the second phrase in cryptic script. Equally disingenuous is the introduction of the cryptic phrases in 4QSe when a much older manuscript such as 1QS already contains the crucial words in standard script, and an earlier occurrence of the phrase “when these exist in Israel” is preserved in standard script in both 1QS 8:4 and 4QSe. As I have argued in detail elsewhere the material on the council of the community in 1QS 8:1–16a // 4QSd 6:1–8a // 4QSe 2:9b–3:6a has clearly evolved over time, in my view successively expanding the earliest core in 1QS 8:1–7a // 4QSd 6:1 // 4QSe 2:9b–16a.50 Several stages in the successive growth of this section are attested materially both in the form of superlinear additions in 1QS 8 as well as the large portion of text corresponding to 1QS 8:15–9:11 not found in 4QSe.51 In other words, a sense of sending mixed messages, scribal supplementation, and repetition is already present in 1QS 8–9. The complex sub-plot that seems to unfold behind the cryptic phrases in 4QSe is thus part of a pattern in a particularly complex tale of literary development. 3.3 The Presence of Technical Calendrical Learning Several copies of the Rule testify to the important link between the creation of the luminaries and the timely performance of festivals and prayers in the final hymn, especially in the opening lines of 1QS 10 // 4QSb // 4QSd. 48  Puech, “L’alphabet cryptique A en 4QSe,” 435. 49  See also Charlotte Hempel, “Interpretative Authority in the Community Rule Tradition,” DSD 10 (2003): 59–80 (68–69). 50  See Hempel, The Qumran Rule Texts in Context, 85–92. 51  See Alexander and Vermes, DJD 26:148 and further literature cited there.

4QSe is the only manuscript of the Serekh that incorporates a technical roster (or collection of rosters) of calendrical data in the form of 4QOtot. 3.4 The Place (Prominence) of the Maskil In 1QS the Maskil features in the heading to the Teaching on the Two Spirits and twice in column 9 where two headings introduce material in 9:12 and 9:21 that deals at some length with this figure. The heading of the Teaching on the Two Spirits is not preserved in any of the Cave 4Q manuscripts. The two headings found in 1QS 9 are attested in 4QSd and 4QSe and in the case of 1QS 9:12 also in CD 12:21 // 4QDa 9 ii 7–8. What sets apart the Cave 4Q copies is the occurrence of Maskil also in the title of 4QSd and at a major juncture in 4QSb where 1QS 5:1 begins with a Serekh heading. Thus, overall we have equivalence of Maskil references in Cave 4Q manuscripts of the Rule that preserved material where he figures in 1QS alongside two important additional references to the Maskil at a key juncture in 4QSb and in the title of 4QSd respectively. The Workaday Quality of Cave 4Q Specimen: Papyrus; Possibility of Early Drafts We certainly have two representatives of the Serekh tradition in Cave 4Q that are early papyrus manuscripts (4QSa and 4QSc). It has even been suggested by several scholars, including the editors of DJD 26, that with 4QSa we may be dealing with a draft of the Serekh copied roughly on the back of another text—4QSa being an opistograph with 4Q433a Hodayot-like text.52 Given both early papyrus manuscripts do not preserve any text corresponding to material after 1QS 4, I have argued elsewhere that, 3.5

It is at least worth considering the possibility that with 4Q255 we are dealing with an early draft of what we now find in the opening columns that was produced with the intention of supplementing existing Serekh material such as is preserved at the start of the short text in 4Q258 (Sd).53

52  See George J. Brooke, “Between Scroll and Codex: Reconsidering the Qumran Opisthographs,” in On Stone and Scrolls: Essays in Honour of Graham Ivor Davies (ed. J. K. Aitken, K. J. Dell, and B. A. Mastin; BZAW 420; Berlin: W. de Gruyter, 2011), 123–38 (123); Milik, “Le travail d’édition,” 61; Eileen Schuller, “4Q433a. 4QpapHodayot-like Text B,” in Qumran Cave 4.XX: Poetical and Liturgical Texts, Part 2 (DJD 29; ed. E. Chazon et al.; Oxford: Clarendon Press, 1999), 237–45 (237 and n. 4); and Tov, Scribal Practices, 68–73. 53  See Charlotte Hempel, “The Long Text of the Serekh as Crisis Literature,” RevQ 105 (2015): 3–24.

86 4 Conclusion To conclude, it seems to me that the complex evidence of the S manuscripts from Cave 4Q is compatible with several features that span across the nature of Cave 4Q more widely. Moreover, of the ten Cave 4Q manuscripts of the Rule 4QSe (4Q259) emerges as the most learned and “avant-garde” exemplar of the Community Rule54 containing an anthology of calendrical data (4QOtot) as well as a number of words written in Cryptic A script. In several respects, 4QSe seems to be particularly representative of the distinctive aspects of the character of Cave 4Q. Finally, the concept of a tiered system of access to knowledge is well attested in ancient Jewish texts55 including the book

54  Hempel, The Qumran Rule Texts in Context, 327. 55  See Samuel I. Thomas, The “Mysteries” of Qumran: Mystery, Secrecy, and Esotericism in the Dead Sea Scrolls (EJL 25; Atlanta, GA: SBL, 2009).

Hempel

of Daniel where access to knowledge privileges those who have it, 4QInstruction, and 4 Ezra 14 to name but a few examples. In the Community Rule in particular the idea is explicit in passages such as those that admonish the Maskil to exercise discretion in his dealings with the people of the pit and the people of injustice, cf. 1QS 9:16b–17a // 4QSd 8:1b–2a // 4QSe 3:13b–15a. Moreover, a fundamental hierarchical principle laid out in the Rule permits access to deliberations only to full members, cf., e.g., 1QS 6:21b–23 // 4QSg 3:1. My suggestion is that in what was clearly a tiered system of access to knowledge Cave 4Q was the place where a large proportion of—if not all— the most restricted material was stored.

CHAPTER 5

Scribal Characteristics of the Qumran Scrolls Emanuel Tov The Qumran scrolls were written at different times and in different places in ancient Israel and it is therefore not expected that they would be of one kind. Indeed, there are many internal differences between the Qumran scrolls, but there are also many similarities between them, as would be expected for scrolls written in roughly the same period between the third century BCE and the first century CE. This study focuses on the special scribal characteristics of the corpus of Qumran scrolls as opposed to those from other sites in the Judean Desert and those dating from other periods. It is not easy to describe the characteristics of the Qumran corpus because it is unclear with what entity it should be compared. At best we can compare it with the corpora found in the other Judean Desert sites even though they are of a different nature. However, this type of comparison may still yield wrong conclusions, for which I want to give one example from a related area. It was long believed, at least in the Jewish tradition, that the section divisions of the MT (open sections, closed sections) were characteristic of that tradition because no comparable material had been found. Furthermore, the Jewish tradition claims that these sections were of divine origin. However, the finds in the Judean Desert and elsewhere have made it clear that this type of paragraphing is not at all characteristic of the Masoretic tradition, but rather reflects a widespread custom used in almost all ancient Hebrew and Aramaic texts. This example illustrates how difficult it is to capture the special characteristics of any one corpus. All these comparisons are correct as long as we do not find a manuscript find elsewhere in ancient Israel. The first step in this journey is addressing the question where the scrolls were produced. 1

Local Production of Written Material in the Judean Desert?

It is hard to know how many of the texts found in the Judean Desert were actually produced locally, that is, both their physical preparation and the copying of the manuscripts. Undoubtedly at least some of the leather scrolls were produced locally (as can be proven with DNA analysis comparing the scrolls with hides of local animals, both

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_007

ancient and present-day), but at present for the majority of the scrolls the place of production cannot be ascertained. It is also impossible to know whether papyri were produced locally in the Judean Desert (Ein Feshkha or elsewhere in Israel) or imported from Egypt. 1.1 Qumran If it could be proven that locus 30 at Qumran served as a room in which documents were written (a scriptorium in medieval terminology),1 the assumption of a Qumran scribal school would receive welcome support, but the reliability of the evidence pointing to the existence of such a scriptorium is questionable. Furthermore, most scholars now believe, on the basis of the content of the scrolls, that some, many, or all of the documents found at Qumran were copied by the Qumran community, which could lend support to the assumption of local production. Stegemann holds a maximalist view on this issue, assuming that most Qumran scrolls were copied on site.2 According to him, one of the main occupations of the Qumran community was the preparation of parchment for writing and the mass-production of written texts. These texts were then sold by the Qumran community to the outside world. Stegemann pinpoints the places in the community buildings in which scrolls were manufactured, stored, and offered for sale.3 Golb, expressing a ­minimalist 1  Thus the majority of scholars ever since the description by Roland de Vaux, L’archéologie et les manuscrits de la Mer Morte (The Schweich Lectures 1959; London: Oxford University Press for the British Academy, 1961), 23–26; idem, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973), 29–33; see also Ronny Reich, “A Note on the Function of Room 30 (the ‘Scriptorium’) at Khirbet Qumran,” JJS 46 (1995): 157–60. 2  Hartmut Stegemann, The Library of Qumran: On the Essenes, Qumran, John the Baptist, and Jesus (Grand Rapids, MI/Leiden: Eerdmans/Brill, 1998), 51–55. 3   This theory was rejected in a detailed analysis by Ferdinand Rohrhirsch, Wissenschaftstheorie und Qumran: Die Geltungsbegründungen von Aussagen in der biblischen Archäologie am Beispiel von Chirbet Qumran und En Feschcha (NTOA 32; Freiburg/Göttingen: Universitätsverlag/Vandenhoeck & Ruprecht, 1996) and idem, “Die Geltungsbegründungen der Industrie-Rollen-Theorie zu Chirbet Qumran und En Feschcha auf dem methodologischen Prüfstand: Relativierung und Widerlegung,” DSD 6 (1999): 267–81.

88

Tov

view, claims that none of the Qumran documents were written locally.4 The contents of the scrolls do not help us in determining where they were copied. The presumably non-Qumranic scrolls could have been copied in Qumran. That position is, however, unlikely, and we have to accept the assumption that some or many scrolls were brought to Qumran later, just like the Greek scrolls were brought to Qumran from elsewhere. 1.2 Masada There is no reason to believe that the Masada fragments were penned at Masada itself, even though the Zealots and presumably also the Essenes remained there long enough to have embarked upon such an endeavor. On the other hand, some evidence apparently points to the tanning of hides at Masada, which could imply some scribal activity.5 Probably only the Hebrew, Aramaic, and Greek ostraca were written at Masada prior to the destruction of the fortress. The Latin ostraca and some Greek papyri were probably inscribed under the Roman occupation, and other papyrus and leather texts may have been imported.6 1.3 Naḥal Ḥever and the Other Judean Desert Sites Except for the literary texts found in the Judean Desert sites, all the documentary texts, mainly papyri, found in these sites relate to local activities and were written on location. However, external information about the Judean Desert texts does not guide us in exploring the special characteristics of the Qumran corpus. Such information should come from the necessarily subjective analysis of the features of that corpus, mainly referring to texts written by the Qumran community, but also to some imported texts. 2

Special Characteristics of the Qumran Corpus

We now focus on the types of texts that are characteristic of the Qumran corpus, mainly referring to the texts written by the Qumran community, but also to some imported texts. I refer to the presence of (a) cryptic and (b) paleo4  Norman Golb, Who Wrote the Dead Sea Scrolls? The Search for the Secret of Qumran (New York: Scribner, 1995). 5  See Ehud Netzer, Masada III: The Yigael Yadin Excavations 1963– 1965; Final Reports; The Buildings, Stratigraphy and Architecture (Jerusalem: Israel Exploration Society/Hebrew University, 1991), 634–35. 6  See my study “A Qumran Origin for the Masada Non-biblical Texts?” DSD 7 (2000): 57–73.

Hebrew texts, (c) the abundance of the texts written on leather, (d) scribal marks, and (e) the special writing of divine names. 2.1 Cryptic Texts from Cave 4 Probably the single most typical group of the Qumran corpus is a group of texts written in the Cryptic A (also named esoteric and Hebrew Hieratic) script probably written by the Qumran community. The writing in this script distinguishes the Qumran corpus not only from the other Judean Desert sites, but as far as we know, also from all of ancient Israel. Until the 2010s, the Cryptic scripts were known only from the Qumran excavations, but an inscribed Second Temple stone cup,7 probably containing an incantation, found at Mount Zion provides the first sample of the Cryptic script outside Qumran,8 possibly also authored by a member of the Essene community. In its published form,9 this group of texts is rather sizeable (according to S. J. Pfann, thirty-six papyri from cave 4Q, two of which are opisthographs containing different texts), but if calculated differently this corpus represents a smaller number of texts. The writing in the Cryptic A script reflects authorship by the Qumran community, but no strong case can be made for each and every composition regarding its sectarian background. This script, described by Pfann as a development from the Late Phoenician scripts,10 was used for several texts of a Qumran sectarian nature as well as for other texts that must have had a special meaning for the Qumran community.11 According to Milik and Pfann, this script was used especially by the Maskil; see especially 4QcryptA Words of the Maskil to All Sons of Dawn (4Q298). If this composition indeed contains the instructions of the Maskil to the Qumran novices, it is understandable that it was written in a special script not to be shared with others. The same case can be made for 4QcryptA Lunisolar Calendar (4Q317) and pos7   Stephen J. Pfann, “The Mount Zion Inscribed Stone Cup: Preliminary Observations,” in New Studies in the Archaeology of Jerusalem and Its Region: Collected Papers (ed. D. Amit et al.; Jerusalem: Israel Antiquities Authority, Jerusalem Region, 2010), 4:32*–41*. 8  See Pfann, “Mount Zion.” 9  Stephen J. Pfann, ed., Qumran Cave 4.XXVI: Cryptic Texts; Philip S. Alexander et al., eds., Miscellanea, Part 1 (DJD 36; Oxford: Clarendon Press, 2000). 10  Stephen J. Pfann, “4Q298: The Maskîl’s Address to All Sons of Dawn,” JQR 85 (1994): 203–35. 11  See also Stephen J. Pfann,“The Writings in Esoteric Script from Qumran,” in The Dead Sea Scrolls Fifty Years After Their Discovery: Proceedings of the Jerusalem Congress, July 20–25, 1997 (ed. L. H. Schiffman, E. Tov, and J. C. VanderKam; Jerusalem: Israel Exploration Society/Shrine of the Book, 2000), 177–90.

Scribal Characteristics Of The Qumran Scrolls

sibly for 4Qpap cryptA Midrash Sefer Moshe (4Q249).12 The Cryptic B and C scripts, in which only a few texts have been inscribed, are less known. Since the Cryptic scripts have been found at Qumran, we wish to examine the scribal background of the documents written in this script. Our conclusion is that writing in this script does not involve the employment of scribal habits different from writing in the square script. The manuscripts written in one of these scripts share with the other Hebrew texts the use of the same scribal features for writing on papyrus (most Cryptic texts are written on papyrus) and leather and for the preparation of the scrolls. For the leather scrolls this involves: horizontal and vertical ruling, stitching of sheets, size and shape of columns, correction systems (superscript letters in 4Q298 and 4Q317), number of columns per sheet, height of columns, and margins. Open and closed sections have been preserved in 4Q249. Exceedingly large spaces (4.0–4.25 cm) are found between the lines of 4QcryptB Unclassified Text (4Q363). For two texts, titles have been preserved, in the square script, in 4QcryptA Words of the Maskil (4Q298) as the first words of the running text, and in 4Qpap cryptA Midrash Sefer Moshe (4Q249) on the verso of frg. 1. The separation of words with dots is reflected also in the only text written in the Cryptic C script, 4QcryptC Unclassified Religious Text (4Q363a). This text is written mainly in paleo-Hebrew letters, intermingled with some cryptic signs, and the use of dots to separate words is typical of the writing in that script. If our analysis is correct, the texts written in the Cryptic scripts could have been written by the same scribes that wrote other Qumran scrolls. 2.2 Paleo-Hebrew Texts The second group of texts that is almost exclusively characteristic of the Qumran site c­ omprises texts that have been written completely in paleo-Hebrew script. At Qumran, fragments of twelve biblical texts written in the paleo-Hebrew script were found as well as three nonbiblical paleo-Hebrew texts, mainly from cave 4, but also from caves 1, 2, 6 and 11. One nonbiblical text was found at Masada,13 but that may have derived from Qumran as well. 12  For the latter text a new name (“Sefer Moshe/Midrash Moshe”) has been suggested by Jonathan Ben Dov and Daniel Stökl Ben Ezra, “4Q249 Midrash Moshe: A New Reading and Some Implications,” DSD 21 (2014): 131–49. 13  Beyond the publications of these texts in DJD 1, 3, 9, 23, see: Mark D. McLean, The Use and Development of Palaeo-Hebrew in the Hellenistic and Roman Periods (unpubl. Ph.D. diss., Harvard University, Cambridge, MA, 1982), 41–47 (University Microfilms); David N. Freedman and Kenneth A. Mathews, The Paleo-Hebrew Leviticus Scroll (11QpaleoLev) (Winona Lake, IN: Eisenbrauns, 1985). According to Raymond L. Edge, “The

89 These texts, rather than predating the use of the square script (with the exception of Mur 17A–B), were written at a relatively late period, possibly but not necessarily as a natural continuation of the earlier tradition of writing in the “early” Hebrew script. They were concurrent with the use of the square script, as can be proven by a paleographical examination of the paleo-Hebrew script.14 Most scholars assume that with the revival of the paleo-Hebrew script in the Hasmonean period, texts were transformed from the square to the paleo-Hebrew script,15 and this is probably correct, although it is not impossible that the practice of writing in the paleo-Hebrew script had never ceased in some circles. The preserved biblical fragments written in the paleoHebrew script contain only texts of the Torah and Job,16 both of which are traditionally ascribed to Moses (cf. manuscripts and editions of S in which Job follows the Torah).17 The longest preserved texts written in the paleoHebrew script are 4QpaleoExodm and 11QpaleoLeva. All texts written in the paleo-Hebrew script reflect a similar scribal approach, but the scribes of these texts often displayed their individuality in specific features. The only external data regarding the background of the writing in the paleo-Hebrew script is of a negative nature. Various statements in rabbinic literature, e.g. m. Yad. 4.5, forbid the use of this script for biblical texts. How then should we describe the background of the writing of complete paleo-Hebrew texts or of individual Use of Palaeo-Hebrew in the Dead Sea Scrolls: Paleography and Historiography” (unpubl. Ph.D. diss., University of Texas at Austin, 1995), 357 (University Microfilms), fifty years separated the writing of entire scrolls in the paleo-Hebrew script in the archaic and Hasmonean periods and the writing of the divine names in paleo-Hebrew characters in the Herodian period. 14  See McLean, Use and Development; Richard S. Hanson apud Freedman and Mathews, Leviticus, 20–23; idem, “Paleo-Hebrew Scripts in the Hasmonean Age,” BASOR 175 (1964): 26–42. For an earlier discussion, see Lajos Blau, “Wie lange stand die althebräische Schrift bei den Juden im Gebrauch?” in Gedenkbuch zur Erinnerung an David Kaufmann (ed. M. Brann and F. Rosenthal; Breslau: Schottländer, 1900), 44–57. 15  Thus Kenneth A. Mathews, “The Background of the PaleoHebrew Texts at Qumran,” in The Word of the Lord Shall Go Forth: Essays in Honor of David Noel Freedman in Celebration of his Sixtieth Birthday (ed. C. L. Meyers and M. O’Connor; Winona Lake, IN: Eisenbrauns, 1983), 549–68. 16  One explanation for the writing of a text of Job in Paleo-Hebrew would be to assume that Job was ascribed to patriarchal times. But it is probably more sound to assume that Mosaic authorship was ascribed to that text, cf. b. B. Bat. 14b–15a. 17  Note, however, 4Qpaleo paraJosh (4Q123). Although this text contains elements from Joshua 21, it is probably not a biblical text in the later sense.

90 words? Since none of the paleo-Hebrew texts shares the characteristics of the Qumran scribal practice, they were probably not written by the Qumran scribes. More generally, there is no major argument in favor of the assumption that the biblical texts written completely in paleo-Hebrew characters were copied by the Qumran scribes.18 It is unlikely that the paleo-Hebrew texts came from the circles of those who produced MT, since the use of the paleoHebrew script was strictly forbidden in the Talmud (see above). We therefore turn to a third possibility based on criteria of script, textual character, and scribal approach: namely that the paleo-Hebrew texts found at Qumran came from the circles of the Sadducees who ascribed great importance to the authenticity of the ancient characters. This explanation should alleviate the difficulty of the apparent contradiction that many of these texts reflect MT, while the rabbis prohibited the use of this script. If this hypothesis holds ground, it is understood that as the rabbis prohibited the use of paleo-Hebrew characters, such texts of proto-Masoretic content were written by others. Likewise, on the basis of Diringer,19 Naveh hesitatingly ascribed the paleo-Hebrew texts from Qumran to the Sadducees, without any arguments.20 This possibility is discussed extensively elsewhere,21 but it should be admitted that the nature and status of the nonbiblical paleo-Hebrew fragments from Qumran and Masada remain unclear. The fact that paleo-Hebrew texts are known almost solely from Qumran lends a certain exclusivity to that 18  Mathews, “The Background,” is not consistent in his approach to this issue. On the one hand this scholar speaks of the Essene origin of the paleo-Hebrew texts (551, 558), but at the same time he also considers these texts as having been brought to Qumran (557). 19  David Diringer, “Early Hebrew Script versus Square Script,” in Essays and Studies Presented to Stanley Arthur Cook (ed. D. W. Thomas; London: Taylor’s Foreign Press, 1950), 35–49. 20  Joseph Naveh, Early History of the Alphabet: An Introduction to West Semitic Epigraphy and Palaeography (2nd ed.; Jerusalem: Magnes Press, 1987), 122. See also Emile Puech, “Notes en marge de 11QpaléoLévitique: Le fragment L, des fragments inédits et une jarre de la grotte 11,” RB 96 (1989): 161–83 (167–68). Little is known about the approach of the Sadducees towards Hebrew Scripture in spite of the analysis of Jean le Moyne, Les Sadducéens (Paris: Librairie Lecoffre, 1972), 357–59. 21  Emanuel Tov, “The Socio-Religious Background of the PaleoHebrew Biblical Texts Found at Qumran,” in Geschichte–Tradition– Reflexion: Festschrift für Martin Hengel zum 70. Geburtstag (ed. P. Schäfer, H. Cancik, and H. Lichtenberger; Tübingen: Mohr Siebeck, 1996), 1:353–74. For an extensive survey of the different explanations of the background of the Qumran Paleo-Hebrew texts, see Edge, “The Use of Palaeo-Hebrew,” especially 334–69.

Tov

site, but this situation may be misleading. If my analysis is correct, these texts were brought to Qumran from elsewhere, so that presumably similar biblical texts were circulating elsewhere in ancient Israel. Furthermore, from a certain point onwards, the Samaritans wrote in a similar script, deriving from the ancient Hebrew script. Finally, in Masada one fragment was discovered inscribed in two different scripts on both sides.22 The content is unclear; the recto has two occurrences of the word renanah (“joy”) as well as the name of the Samaritan holy site hargerizim, written in one word, as in the Samaritan tradition. There is much discussion regarding how this fragment reached Masada, but in view of the discovery at Masada of Qumran type documents, it is quite clear that both Samaritans and Essenes found refuge at Masada hoping to survive the Roman attack.23 2.3 Abundance of Texts Written on Leather A comparison of the types of written texts found at Qumran and the other sites shows that the community that left these texts behind at Qumran differed completely from the other communities. Table 5.1 compares the quantities of the papyrus texts with the leather texts found in the Judean Desert. This Table24 shows that the situation at Qumran differs from that at the other sites in the Judean Desert. At Qumran (and Masada), the papyri form a minority of the texts found there (14%), while at all other sites excluding Sdeir they form a majority. This points to a very basic difference between the Qumran corpus and that of the other sites. The Qumran corpus contains almost exclusively non-documentary texts bearing witness to literary activity, while the other sites evidence the daily life experienced at these places, though with some literary activity recorded as well (only a small number of non-documentary leather texts has been found at these sites). The leather texts from Qumran do not document any daily activity, with the exception of 4QRebukes Reported by the Overseer (4Q477) and the Greek 4QAccount gr (4Q350) written on the back of frg. 9 of the Hebrew 4QNarrative

22  Shemaryahu Talmon, “Hebrew Fragments from Masada,” in Masada VI: The Yigael Yadin Excavations 1963–1965; Final Reports; Hebrew Fragments from Masada (ed. S. Talmon and Y. Yadin; Jerusalem: Israel Exploration Society/Hebrew University, 1999), 138–49. 23  Thus also Talmon, “Hebrew Fragments,” 148–49. 24  Copied from my book, Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (STDJ 54; Leiden: Brill, 2004), 45.

91

Scribal Characteristics Of The Qumran Scrolls Table 5.1

Comparison of Hebrew, Aramaic, Greek, Latin, and Nabatean-Aramaic Papyri and Leather Texts from the Judean Desert (listed from North to South)

Site

Total Papyri and Leather Texts

Jericho Qumran Nar Wadi Ghuweir Murabba‘at Sdeir Ḥever/Seiyal Mishmar Ṣeʾelim Masada

 23+ 930   3+   1 151+   4 179   3   3  45

Papyri No.

Percentage of Total

Leather Texts No.

Perc of Total

 23+ 131 (138)   3+   1 101+   1 166+   3   0  31 (34)

100  14.0 100 100  67  25  93 100   0  70

  0 800   0   0  50   3  13   0   3  14

  0  86   0   0  33  75   7   0 100  30

Work and Prayer (4Q460).25 It should be remarked that the ­percentage of the papyri in Qumran is lower if two caves are disregarded: cave 7 contains only (Greek) papyri (19), and cave 6 contains more papyri (21) than literary texts (12). The Qumran papyri26 are mainly literary and the dichotomy between Qumran and the other sites can thus be expressed in two ways: while the majority of the documents in Qumran were written on leather, in the other sites they were mainly written on papyrus. Furthermore, the 130 Qumran papyri are mainly literary, while all the papyri from the other sites are documentary. A similar situation is reflected at Masada. The Jewish inhabitants left only a few papyri (2 [3]) but fourteen Hebrew leather texts (one of which may contain an Aramaic document). In addition, the Roman army left many documentary papyri (28 [30]).27 It is difficult to ascertain how many valid parallels to the almost exclusively literary corpus of Qumran are known from antiquity. Many collections of Greek papyri from Egypt contain more documentary than non-documentary texts, but the following corpora are valid parallels to the 25  We probably need to exclude also 4Q344–349 and 4Q351–359, but these texts may not have derived from Qumran. 26  See the list in The Texts from the Judaean Desert: Indices and an Introduction to the Discoveries in the Judaean Desert Series (DJD 39; ed. E. Tov; Oxford: Clarendon Press, 2002), 204–10. 27  Including Mas 721 r + v containing one or two lines of Virgil on the recto and one or two lines of an Unidentified Poetical Text on the verso.

Qumran corpus: the philosophical corpus found in the “villa of the papyri” at Herculaneum (terminus ante quem 79 CE), the Oxyrhynchus corpus (if the literary texts from that site came from a specific part of the city), and the Antinoopolis corpus. However, the most valid and obvious parallels are probably the libraries which were lost, that is, the collections stored in Alexandria, Pergamon, and Ephesus from the Hellenistic period, Roman libraries from later periods, and Christian libraries from the fourth century CE in Jerusalem, Alexandria, and Caesarea.28 The Nag Hammadi literary corpus derives from a slightly later period. While it remains unknown what would have been included in the Jerusalem temple library, we can assume that it would have contained at least all the Scripture scrolls on leather.29 28  All these libraries are discussed by Harry Y. Gamble, Books and Readers in the Early Church: A History of Early Christian Texts (New Haven, CT/London: Yale University Press, 1995), 154–70, 176–96 and Alan Millard, Reading and Writing in the Time of Jesus (New York: New York University Press, 2000), 17–22. 29  Note, for example, m. Yom. 1.5 according to which the elders of the court read to the High Priest from Job, Ezra, Chronicles, and Daniel on the day before the Day of Atonement. These books, together with the master copy of the Torah, were probably part of a temple library. The founding of such a library by Nehemiah was mentioned in 2 Macc 2:13–15 (“books concerning kings, prophets, David, and royal letters”). Josephus mentions the temple library on various occasions (e.g. Ant. III 38; IV 303; V 61), once also with regard to the copy of the Jewish Law which was taken by Titus (Bell. Jud. VII 150, 162). For further references to such a library and an analysis, see Albertus F. J. Klijn,

92 I suggest that the collection of Hebrew and Aramaic Qumran papyri mainly reflects personal copies owned by members of the Qumran community, while some may have been imported from other localities. The other Judean Desert sites housed only documentary papyri, and they reflect a completely different life style from that of the Qumran covenanters. I now turn to two peculiarities of the Qumran texts that probably derived from the special nature of many of the texts found at Qumran. They are found in specific texts written in the Qumran scribal practice. Probably these texts were unique in ancient Israel. I do not mention the Qumran scribal practice itself as characterizing the Qumran corpus because it would be difficult to argue from silence that there were no such texts elsewhere in Israel.30 2.4 Scribal Marks We consider any element that is not related to the content of the originally inscribed text, but additional to it, a scribal mark. This definition thus excludes guide dots written at the beginnings and ends of sheets guiding the drawing of lines on the leather, since these were inserted before the writing of the text. The shapes of these signs are mostly distinct from the letters of the script in which the text was written, although some letters in the square, “A Library of Scriptures in Jerusalem?” in Studia Codicologica (TU 124; ed. K. Treu; Berlin: Akademie Verlag, 1977): 265–72. The mentioning of the finding of a Torah scroll in the temple in 2 Kgs 22:8 does not necessarily prove the existence of a library in the seventh century BCE, pace André Lemaire, “Writing and Writing Materials,” ABD 6:999–1008 (1005). See further Shmuel Safrai, “The Synagogue,” in The Jewish People in the First Century: Historical Geography, Political History, Social, Cultural and Religious Life and Institutions (CRINT 1; ed. S. Safrai and M. Stern; Assen/Amsterdam: Van Gorcum, 1976), 2:908–44 (940). 30  The argument runs as follows. The 210–212 Hebrew-Aramaic biblical texts found at Qumran do not share any major textual, linguistic, or scribal characteristics, since they were written in different periods and at different places. However, one group of texts is idiosyncratic and is closely related to the scribes who copied the Qumran sectarian scrolls. Within the Qumran corpus, a group of 167 non-biblical and biblical texts has been isolated as reflecting an idiosyncratic practice, the characteristics of which are visible in peculiarities in orthography, morphology, and scribal features. See Tov, Scribal Practices, 261–73. This group of texts is closely connected with the Qumran community, since it includes virtually all commonly agreed upon sectarian writings. The texts written in the Qumran scribal practice could have been penned anywhere in Palestine, but they were probably written mainly at Qumran. Early scrolls, such as 4QQoha (175–150 BCE), must have been copied elsewhere by similarly oriented scribes as this and a few other texts predate the settlement at Qumran.

Tov

paleo-Hebrew and Cryptic A scripts are also used as signs. Some markings were inserted by the original scribes, but probably a greater number were inserted by later scribes and generations of users, and usually we are not able to distinguish between these three levels. Sometimes the color of the ink or the shape of the sign show that the sign was written after the text was completed. The mere occurrence of scribal marks does not set off the Qumran scrolls from the other corpora. True, many marks are known only from the Qumran corpus, but that situation may be misleading, since the other corpora are not so well documented as that of Qumran. However, in view of the fact that a large number of documents have been preserved in the Judean Desert, we nevertheless note that the non-Qumran documents do not contain as many scribal marks as the Qumran scrolls. The difference must be both in genre (most Qumran scrolls are literary as opposed to the documentary nature of the Judean Desert scrolls) and in material (many signs were used especially on leather). The main characteristic of the scribal signs in the Qumran scrolls is their relative frequency, though not in all the texts. In certain Qumran scrolls (sectarian texts from caves 1 and 4) these signs occur very frequently (mainly texts written in the Qumran scribal practice from caves 1 and 4), and this fact sets off the corpus of the Qumran scrolls from the other corpora. The texts from the Judean Desert, especially those from Qumran, contain various scribal markings, some of which recur often in certain texts. Most signs were intentional even if their meaning is often unclear.31 There are almost no differences between the scribal practices in this respect between biblical and nonbiblical texts. A few signs are known from Aramaic secular sources preceding the time of the earliest texts from Qumran (the paragraphos sign and the paleo-Hebrew waw and aleph). Others are known from Greek sources concurrent with the earliest Qumran texts (the paragraphos sign, cancellation dots, and the sigma and antisigma). Similar signs were probably also used in Hebrew texts preceding the earliest Qumran texts, but our information regarding Hebrew non-lapidary texts from the period preceding the mid-third century BCE is very fragmentary.32 Most of the non-lapidary sources are 31   Similarly, Kathleen McNamee, Sigla and Select Marginalia in Greek Literary Papyri (PapyBrux 26; Brussels: Fondation Égyptologique Reine Élisabeth, 1992), 7 notes that a great number of the signs in Greek literary papyri are obscure. 32  For comparison, note the various signs used in the Samaritan writing tradition, but it is not known in which period they originated. The various grammatical treatises explaining the use and meaning of these signs were collected by Zeev Ben Hayyim, The

93

Scribal Characteristics Of The Qumran Scrolls

ostraca that display different scribal practices from those used on papyrus and leather. The use of signs is almost exclusively limited to the texts written according to the Qumran scribal practice. Similarly, the majority of the texts using paleo-Hebrew letters for the divine names (see below) are written according to the Qumran scribal practice. It should also be noted that the occurrence of these signs is almost exclusively confined to Hebrew, and not Aramaic, texts. In Aramaic texts these signs are very rare. The proportionally largest number of signs is found in 1QIsaa and 1QS-1QSa-1QSb (these three compositions were written by the same scribe who also inserted some corrections in 1QIsaa), 4Q502–511, 4Qpap pIsac (4Q163), and 4QCantb. The scribal marks were inserted into specific copies of a Qumran composition, and are not indicative to the scribal transmission of the composition in all its copies. Thus, the marks in 1QS are not found in the copies of that composition from cave 4Q (e.g. the signs in 1QS V are not found in the parallel 4QSb [4Q256]). Scribal markings may be subdivided into several categories (for a detailed analysis, see my Scribal Practices, 180­–214). 1.

2.

3. 4. 5.

Section markers, almost exclusively in the margin, and other scribal systems pertaining to the division of the text into sections in biblical and non-biblical texts. These signs are found in MasSir (II 8, 24; III 18; IV 16), in Greek literature and in Aramaic texts from the fifth century BCE onwards. The use of the paleoHebrew waw in 11QpaleoLeva and 4QpaleoExodm is paralleled by some Arad ostraca, dating to the end of the First Temple period. Marks pertaining to scribal intervention, mainly for the correction of errors such as cancellation dots/ strokes, crossing out with a line, parenthesis signs, and erasures. Cancellation dots are also known from Greek sources, they are mentioned in the Talmud (Sifre Numbers  § 84), and are known from the Masoretic tradition. Crossing out with a line is also known from Egyptian and Greek sources. Parenthesis signs are also known from Greek sources. Single letters in the Cryptic A script primarily written in the margin. Single paleo-Hebrew letters written in the margin. Marks, including unexplained signs, drawing attention to matters in the text. Literary and Oral Tradition of Hebrew and Aramaic amongst the Samaritans (Jerusalem: Bialik Institute, 1957), Vol. 2 [Hebrew], especially Qanun Dartha fi l-Maqra.

6.

7. 8. 9.

Marks written at the ends of lines as line-fillers in three Qumran texts.33 The X-signs were likewise used for the specific purpose of indicating that a space at the end of a line should not be mistaken for an “open section,” which has a definite contextual meaning.34 Separation dots between words Letters and marks possibly numbering sheets and units Signs for numerals. Paralleled by many sources.

Some of these signs in the Qumran scrolls are paralleled by use in other sources. See the examples mentioned above. 2.5 Special Writing of the Divine Names The writing of the divine names using paleo-Hebrew characters in Hebrew texts otherwise written in the square script is unique to Qumran and most likely needs to be linked with the Qumran scribal practice mainly in texts from caves 1 and 4. Several Qumran texts, mainly of a nonbiblical, sectarian nature, display a special approach toward the writing of divine names, especially the Tetragrammaton. As in rabbinic literature, most sectarian texts avoided representing the Tetragrammaton and elohim as much as possible, finding alternative means of expression. The clearest evidence of this avoidance pertains to the pesharim, which by way of circumlocution often refer to God in the third person.35 Reflecting a similar approach to avoiding the use of divine names, other scribal solutions were invoked for their 33  According to Gregory H. Snyder, “Naughts and Crosses: Pesher Manuscripts and Their Significance for Reading Practices at Qumran,” DSD 7 (2000): 26–48 (especially 42–43) these signs “are best understood as cues for textual performance.” According to him, these signs were meant to guide the reader who read this text orally in disregarding unwanted spaces. Snyder follows our explanation (he quotes my earlier article on this subject), but he connects the writing of the sign with the oral recitation of the text. Gregory L. Doudna, 4Q Pesher Nahum: A Critical Edition (JSPSup 35; Sheffield: Sheffield Academic Press, 2001), 239 follows our explanation. 34  This practice is paralleled by several of the later texts from Naḥal Ḥever included in Hannah M. Cotton and Ada Yardeni, eds., Aramaic, Hebrew and Greek Documentary Texts from Naḥal Ḥever and Other Sites: With an Appendix Containing Alleged Qumran Texts (The Seiyâl Collection II) (DJD 27; Oxford: Clarendon Press, 1997) and in Ada Yardeni, Textbook of Aramaic, Hebrew and Nabataean Documentary Texts from the Judaean Desert and Related Materia. Vols. 1–2 (Jerusalem: Hebrew University, Ben Zion Dinur Center, 2000), e.g. 1:102, 113. 35  For further details, see Tov, Scribal Practices, 239.

94

Tov

safeguarding in the text, especially in biblical quotations. Thus, in addition to the writing of the Tetragrammaton in square characters, which occurs relatively infrequently in the Qumran texts, three scribal s­ ystems were employed for the writing of the divine names, especially the Tetragrammaton. The most frequently used procedure was the writing in paleo-Hebrew characters. The use of this script probably ensured the non-erasure of the divine names, while the two other systems (use of dicolon and Tetrapuncta) indicate a special approach to the Tetragrammaton, possibly alerting against pronouncing it. It is unclear why certain scribes used paleo-Hebrew characters for the Tetragrammaton, while others wrote the Tetragrammaton in square characters. This question is particularly relevant with regard to the texts written according to the Qumran scribal practice, since most texts using the paleo-Hebrew Tetragrammaton are written in this style. Spaces were probably left by the scribe in some manuscripts for the paleo-Hebrew words to be filled in later either by that scribe or by a special scribe who was entitled to write the divine name. 3

Background of the Scribal Peculiarities of the Qumran Corpus

In my view the scribal peculiarities of the Qumran corpus derive from the special nature of the Qumran community. The texts written in cryptic scripts are closely connected with the sectarian nature of the Qumran community. The paleo-Hebrew texts, possibly Sadducean, are probably closely connected with the Qumran community as well. By the same token, the fact that the Qumran corpus includes more literary than documentary texts than the other Judean Desert corpora is connected with the fact that this was a highly intellectual and spiritual community. The other features are intimately connected with the Qumran scribal practice. Ultimately the existence of some special scribal features is closely connected with a subjective interpretation of statistical evidence. The frequency of scribal marks and the appearance of paleo-Hebrew divine names occurs mainly in the texts that are written in a certain orthography and morphology. However, not all details regarding scribal diversity in the Qumran scrolls are clear: A relatively large number of biblical and non-biblical scrolls display guide dots guiding the preparation of straight lines. The distribution of this system of preparing manuscripts is unclear, but one point should be made. While the majority of biblical Qumran texts displaying

guide dots reflect MT, only two of the aforementioned texts in which guide dots are found reflect the protorabbinic MT. As a result, this practice must have been followed in particular by scribes who produced texts other than the proto-rabbinic Scripture texts. 4

Internal Differences between the Qumran Caves?

The contents of the individual Qumran caves are not homogeneous at the scribal level, with the exception of caves 6Q and 7Q. This issue is dealt with more in detail in the papers of F. García Martínez and C. Hempel in this volume.36 As for cave 11Q, using several arguments of types, I suggested that this cave contains more sectarian texts than the other caves.37 Of special relevance is the theory of D. Stökl Ben Ezra.38 His beginning remark is a statement by Webster, “Why do almost all the texts from caves 1 and 7 have BCE dates, while virtually all the cave 2, 3, 6, 8, and 9 texts have CE dates?”39 Ben Ezra wondered what happened to the caves 4, 5 and 11 missing in this preliminary observation.40 Cave 4Q is particularly important having revealed by far the greatest number of manuscripts. In his words, “it is statistically practically impossible that the manuscripts from caves 1 and 4 are random samples coming from the same collection of manuscripts as those from caves 2, 3, 5, 6 and 11.” According to him, “the manuscripts from cave 1 and cave 4 represent the state of the Qumran library at an earlier date than those from caves 2, 3, 5, 6 and 11. The latter caves represent the manuscript collection of the same sect but at a later stage. In a possible scenario, the scrolls from cave 1 were hidden there much before 68 CE, around the turn of the era when Qumran was destroyed by a fire after an 36  Florentino García Martínez, in this volume; Charlotte Hempel, in this volume. 37   “The Special Character of the Texts Found in Qumran Cave 11,” in my volume Hebrew Bible, Greek Bible and Qumran: Collected Essays (TSAJ 121; Tübingen: Mohr Siebeck, 2008), 421–27. Florentino García Martínez, “Cave 11 in Context,” in The Dead Sea Scrolls: Texts and Context (STDJ 90; ed. C. Hempel; Leiden: Brill, 2010), 199–209 advances a different view. 38  Daniel Stökl Ben Ezra, “Old Caves and Young Caves: A Statistical Reevaluation of a Qumran Consensus,” DSD 14 (2007): 313–33. 39  Brian Webster, “Chronological Index of the Texts from the Judean Desert,” in DJD 39:351–446 (377). 40  Stökl Ben Ezra notes that the paleographical dates of the scrolls from caves 4Q, 5Q and 11Q vary highly from 200 BCE to 82.5 CE and therefore, their average age can not be recognized without calculation.

95

Scribal Characteristics Of The Qumran Scrolls

attack. Cave 4 might have been used as an emergency hiding place, library or as a depository already around the same time, though some manuscripts were added later.” Stökl’s thesis is acceptable, and this provides another way of distinguishing between the caves, though a very general one. In my view he puts too much weight on the designation of this collection as a “library,” and he speaks about “librarians,” but this fact does not touch on the general correctness of the theory. This theory does not describe the nature of the contents of each cave and that remains enigmatic, although some replies have been given above.

The paleographical dates applied to the documentary texts range from the fourth century BCE to the first century CE for the Jericho documents, from 250 BCE to 70 CE for the Qumran texts,43 from 150 BCE to 70 CE for the Masada texts, and from 75 BCE to 135 CE for the texts from Wadi Murabbaʿat, Naḥal Ḥever, and Naḥal Ṣeʾelim. However, at least one much older document has been found in the Judean Desert: the two layers of the palimpsest papyrus Mur 17 (A: papLetter, B: papList of Personal Names) were dated by Milik to the eighth century BCE and by Cross to the second half of the seventh century BCE.44 The following chronological assumptions may be made:

5

1.

Chronological Differences between the Corpora?

The corpora found near the Dead Sea are roughly from the same period, the last centuries before the Common Era and the first centuries of the Common Era. With the exception of the dated documents from Murabbaʿat and Naḥal Ḥever, the dates of these documents remain hypothetical, although paleographical analysis and AMS analysis (Accelerated Mass Spectometry; carbon-14) provide an ever-increasing probability regarding their dating.41 The latter procedure, however, has so far only been applied to a very small number of texts.42 41  See a summary analysis of the procedures involved in James C. VanderKam and Peter Flint, The Meaning of the Dead Sea Scrolls: Their Significance for Understanding the Bible, Judaism, Jesus, and Christianity (San Francisco: HarperSanFrancisco, 2002), 20–33. See further my study “The Sciences and the Reconstruction of the Ancient Scrolls: Possibilities and Impossibilities,” in The Dead Sea Scrolls in Context: Integrating the Dead Sea Scrolls in the Study of Ancient Texts, Languages, and Cultures (VTSup 140/1; ed. A. Lange et al.; Leiden: Brill, 2011), 1:3–25. 42  See Georges Bonani et al., “Radiocarbon Dating of the Dead Sea Scrolls,” ʿAtiqot 20 (1991): 27–32 = Radiocarbon 34 (1992): 843–49; for criticisms, see Gregory Doudna, “Dating the Scrolls on the Basis of Radiocarbon Analysis,” in The Dead Sea Scrolls After Fifty Years: A Comprehensive Assessment, ed. P. W. Flint and J. C. VanderKam (Leiden: Brill, 1998), 1:430–65; idem, 4Q Pesher Nahum, 675–82; Barbara Thiering, “The Date and the Order of Scrolls, 40 BCE to 70 CE,” in Schiffman, Tov, and VanderKam, Jerusalem Congress, 191–98.

2.

The oldest Judean Desert document derives from Murabbaʿat. The corpora from Qumran and Masada are roughly from the same period, as both sites were destroyed in 70 of the Common Era, but the documents found at Qumran are older, from 250 BCE to 70 CE for the Qumran texts and from 150 BCE to 70 CE for the Masada texts.

In sum, the focus of this paper is the special scribal characteristics of the Qumran corpus. We described the occurrence in Qumran of the texts written in the Cryptic and paleo-Hebrew scripts, the abundance of the leather texts at Qumran, the frequency of scribal signs, and the special writing of the divine names. I believe that these features derived from the special nature of the Qumran community. Although not all the texts found at Qumran were written by the Qumran community, their nature reflects the features and preferences of that community.

43  Some carbon-14 dates fall outside this range. See the evidence discussed by VanderKam and Flint, The Meaning, 41. 44  Józef T. Milik, “Textes hébreux et araméens,” in Les grottes de Murabba‘ât (DJD 2; ed. P. Benoit, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1961), 93–100, Pl. XXVIII; Frank M. Cross, “Epigraphic Notes on Hebrew Documents of the EighthSixth Centuries B.C. II. The Murabbaʿat Papyrus and the Letter Found near Yabneh-Yam,” BASOR 165 (1962): 34–42.

CHAPTER 6

La paléographie des manuscrits de la mer Morte Emile Puech La paléographie hébraïque en particulier a connu un temps fort avec la découverte des manuscrits de la mer Morte. Si cette nouvelle science avait eu ses premiers débuts confirmés avec la découverte d’ossuaires inscrits et de quelques inscriptions funéraires à Jérusalem et ses environs, elle a fait un grand pas avec la publication des dizaines de milliers de fragments de manuscrits en hébreu et en araméen dans les grottes du Ouadî Murabbaʿât et de Khirbet Qumrân en particulier, ainsi que des documents de quelques autres grottes ou sites du Désert de Judah. 1 Préliminaires Comme aucun manuscrit des grottes de Qumrân ne porte la date de sa copie dans un colophon comme ce sera parfois le cas plus tard, leur datation est réduite pour le ­terminus ante quem à celle de leur dépôt dans les grottes au printemps 68 de notre ère1. On doit aussitôt préciser que la 1  Voir Roland de Vaux, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959 ; rev. ed. ; London : Oxford University Press for the British Academy, 1973), 106-9 “. . . none of the manuscripts . . . placed in the caves after June A.D. 68” (p. 107), et “none of the manuscripts belonging to the community is later than the ruin of Khirbet Qumran in A.D. 68” (p. 109), en accord avec la destruction du site en juin 68 (p. 36-41), et Roland de Vaux, “Esséniens ou Zélotes? A propos d’un livre récent”, RB 73 (1966) : 212-35 “Cette destruction marque la fin de l’occupation juive à Qumrân. L’utilisation des grottes par les juifs est contemporaine de l’occupation des bâtiments, comme le prouve l’identité de la poterie, et cette utilisation a cessé en même temps que l’occupation des bâtiments. Comme les manuscrits proviennent de la communauté juive qui a occupé les bâtiments et qu’on ne peut prouver qu’ils ont été apportés plus tard dans les grottes, cette date marque aussi le dépôt ou l’abandon des manuscrits dans les grottes. Aucun manuscrit des grottes ne peut être postérieur à juin 68 ap. J.-C. . . . toute hypothèse qui situe la composition ou la copie des manuscrits après cette date est fausse” (p. 235). Voilà qui est clairement dit par le fouilleur qui a vu se dérouler sous ses yeux les découvertes des manuscrits dans les grottes. La liste de Brian Webster, “Chronological Index of the Texts from the Judaean Desert”, in The Texts from the Judaean Desert : Indices and an Introduction to the Discoveries in the Judaean Desert Series (DJD 39 ; ed. E. Tov ; Oxford : Clarendon Press, 2002), 351-446, doit être corrigée ; la paléographie ne date aucun manuscrit après 68 AD, malgré ce qui a pu être écrit à propos de 4Q2, 4Q21, 6Q30 et

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_008

datation paléographique d’une copie ne donne pas la date de la composition de l’œuvre littéraire. La date limite d’une copie de manuscrit a trouvé confirmation par la découverte de manuscrits du même type et des ostraca scellés par la destruction de Masada en 732. En outre, certains manuscrits mentionnent des personnages historiques par ailleurs bien connus qui donnent un terminus a quo pour leur composition, et a fortiori pour leur copie, lorsqu’on n’a pas affaire à un original, voir par exemple les mentions du commandant juif Peitholaos à Jérusalem dans le deuxième quart du premier siècle av. J.-C., du gouverneur romain Æmilius (Scaurus) en Syrie en 63-62, de la reine Alexandra/Shlomsion (régnant à Jérusalem de 76 à 67), de ses fils Hyrkan (II) (en 76-67 et 63-40) et Aristobule (II) (en 67-63), de Jonathan (Maccabée) (grand prêtre de 152 à 143), d’Antiochus IV et de Démétrius, ou encore de la prise de Jérusalem et du pillage du temple en 168 av. J.-C. (voir 4Q248). Concernant les rouleaux des livres “bibliques” et des apocryphes retrouvés3, on est certain de n’avoir affaire qu’à des rouleaux ou des copies locales de rouleaux plus 9Q1, l’étude paléographique a fait quelques progrès depuis lors, et ce genre de graphie est connue à l’époque hérodienne. 2  Voir Shemaryahu Talmon, “Hebrew Fragments from Masada”, et Yigael Yadin, “The Ben Sira Scroll from Masada”, in Masada VI : The Yigael Yadin Excavations 1963-1965 ; Final Reports ; Hebrew Fragments from Masada (ed. S. Talmon and Y. Yadin ; Jerusalem : Israel Exploration Society/Hebrew University, 1999). 3  Cette dénomination peut paraître quelque peu anachronique aux yeux de certains, cependant dès le deuxième siècle avant J.-C. était connue une division tripartite de livres normatifs du judaïsme (voir le Prologue de Ben Sira, 1 et 2 M, 4QMMT), même si la liste des “Écrits” n’était pas définitivement fixée, voir Emile Puech, “La constitution de groupements de livres normatifs à Qumrân”, RB 119 (2012) : 45-57. Mais malgré ce qui a pu en être dit ça et là par quelques auteurs, la plupart de ces livres normatifs qui sont cités dans les compositions qumraniennes et autres comme des livres de référence, seront de fait retenus plus tard dans le “canon biblique” ; c’est dire le statut qui était déjà le leur avant d’être définitivement intégrés dans le “canon” ou en être exclus par la suite. En revanche, on maintient la triple répartition entre manuscrits “bibliques”, apocryphes ou péri-testamentaires, et esséno-qumraniens, comme étant de loin la plus à même de regrouper commodément les divers écrits retrouvés dans les grottes, malgré de nouvelles propositions qui n’apportent pas de meilleures solutions, bien loin de là. L’identification de Khirbet Qumrân = Sokoka comme site essénien est assurée.

97

La Paléographie Des Manuscrits De La Mer Morte

anciens, provenant très vraisemblablement de la bibliothèque du temple, de même pour le livre de Ben Sira daté du premier quart du deuxième siècle av. J.-C. Quelques inscriptions de la région donnent d’autres points de comparaison, telle la ketubbah iduméenne de Maréshah datée de 176 av. J.-C.4, ou encore les inscriptions du Garizîm5, et les papyri du Ouadî Daliyeh6, ainsi que des contrats ou des lettres datées d’entre les deux révoltes juives, voir par exemple les manuscrits des grottes de Murabbaʿât avec des dates de la fin du premier et du début du deuxième siècle de notre ère7 et des grottes plus au sud dans le désert de Judah (Ouadîs Khabra et Seiyal), sans compter les nombreuses inscriptions en écriture “lapidaire” sur des objets archéologiques, comme les monnaies, les ossuaires et les sarcophages avec des noms de personnages connus, ainsi que les inscriptions des tombeaux de Jason et des benê-Ḥézîr. Ce sont là déjà des indications fort précieuses pour connaître l’art d’écrire dans la région. Pour davantage de précisions, on en est réduit à l’étude de la main du copiste, sachant que chaque scribe a appris à écrire à l’école de son maître, lui-même un maillon dans la longue chaîne de transmission de l’art d’écrire les lettres de l’alphabet d’une part8 et, d’autre part, qu’au cours de son exercice de copiste, sa propre graphie évolue dans la rapidité d’exécution des lettres suite à une plus ou moins longue pratique.

4  Datée de l’an 136 de l’ère séleucide, voir Esther Eshel et Amos Kloner, “An Aramaic Ostracon of an Edomite Contract from Maresha, Dated 176 B.C.E.”, IEJ 46 (1996) : 1-22. On peut y ajouter les nombreux ostraca de Khirbet el-Qôm = Maqqédah en écriture iduméenne dont plusieurs sont datés par des années de règnes, en particulier dans la deuxième moitié du 3e s. av. J.-C. 5  Voir Jan Dušek, Aramaic and Hebrew Inscriptions from Mt. Gerizim and Samaria between Antiochus III and Antiochus IV Epiphanes (CHANE 54 ; Leiden : Brill, 2012), la plupart datent de la première moitié du deuxième siècle av. J.-C., ou déjà de la fin du troisième siècle (p. 59, 116-18). 6  Voir Frank M. Cross, “The Scripts of the Dâliyeh (Samaria) Papyri”, in Leaves from an Epigrapher’s Notebook : Collected Papers in Hebrew and West Semitic Palaeography and Epigraphy (HSS 51 ; Winona Lake, IN : Eisenbrauns, 2003), 44-46. 7  Voir Józef T. Milik, “Textes hébreux et araméens”, in Les grottes de Murabbaʿât (DJD 2 ; par P. Benoit, O.P., J. T. Milik, et R. de Vaux, O.P. ; Oxford : Clarendon Press, 1961), 67-205. 8  On a de fait retrouvé des exercices de scribes parmi les fragments des grottes, par exemple 4Q234, 4Q341, 4Q360, ainsi que parmi les ostraca de Khirbet Qumrân. Pour d’autres exercices de scribes de la région, voir Emmanuele Testa, Herodion IV : I graffiti e gli ostraka (SBF 20 ; Jerusalem : Franciscan Printing Press, 1972), 78-82, 93-95, et Emile Puech, “Abécédaire et liste alphabétique de noms propres hébreux du début du IIe s. A.D.”, RB 87 (1980) : 118-26.

2

La paléographie

La paléographie des manuscrits en hébreu et en araméen est la science qui s’intéresse au tracé ou ductus des vingtdeux lettres de l’alphabet après sa réduction au douzième siècle av. J.-C., et à son évolution au cours des siècles dans les diverses écoles de scribes, en vue d’établir une chronologie relative. Vers la fin de l’époque perse, apparaissent pour certaines lettres des formes médianes qui se différencient des formes habituelles ou finales qui, elles, conservent leur forme traditionnelle ; et à l’époque hellénistique, restent en usage les formes dites “finales” de cinq lettres, ce qui porte à vingt-sept le nombre des formes des lettres. Aussi, avec ces points de repères internes et externes comme cadre assez bien délimité, il doit être possible de retrouver et de préciser une évolution de l’écriture alphabétique dite araméenne, la plus fréquente sinon habituelle sur ces manuscrits pour écrire l’hébreu (faussement appelé “hébreu carré”) et l’araméen. C’est là un héritage de l’araméen d’empire de l’époque perse, même si l’écriture hébraïque, dite aussi “paléo-hébreu,” n’a pas totalement disparu des cercles de prêtres et de lévites9 ; on la retrouve en effet en particulier dans quelques manuscrits 9  La question est posée de savoir si les manuscrits en paléo-hébreu ont été importés à Qumrân, étant alors considérés comme des produits de scribes sadducéens, ou s’ils ont pu être copiés sur place par les scribes esséniens ? A priori, rien ne s’oppose à ce que des scribes lévites esséniens aient copié ce genre de rouleau d’autant que ne sont guère concernés que des livres du Pentateuque, le Lévitique en particulier, Josué dans 4Q123-paléoJosué (voir Emile Puech, “Les copies du livre de Josué dans les manuscrits de la mer Morte : 4Q47, 4Q48, 4Q123 et XJosué”, RB 122 (2015), 480-506), Job dont le personnage est supposé avoir été plus ou moins un contemporain de Moïse, et quelques fragments non encore identifiés, voir aussi Giancarlo Lacerenza, “Un nouveau fragment en écriture paléo-hébraïque”, RevQ 75 (2000) : 441-47, et Emile Puech, “Note additionnelle sur le fragment en paléo-hébreu”, RevQ 75 (2000) : 449-51. Voir Kenneth A. Mathews, “The Background of the Paleo-Hebrew Texts at Qumran”, in The Word of the Lord Shall Go Forth : Essays in Honor of David Noel Freedman in Celebration of His Sixtieth Birhtday (ed. C. L. Meyers and M. O’Connor ; Winona Lake, IN : Eisenbrauns, 1983), 549-68, et Emile Puech, “Notes en marge de 11QPaléoLévitique : Le fragment L, des fragments inédits et une jarre de la grotte 11”, RB 96 (1989) : 16183 (167-68), où je n’écartais pas les copies esséniennes, compte tenu de l’usage du paléo-hébreu dans plusieurs noms divins, cela jusque dans la période hérodienne. Pour le tétragramme ou des substituts, voir Emile Puech, “Les langues et les écritures dans les manuscrits de la mer Morte”, in Des signes pictographiques à l’alphabet : La communication écrite en Méditerranée ; Actes du colloque, 14-15 mai 1996, Villa grecque Kérylos, Fondation Théodore Reinach (Beaulieu-surMer) (sous la direction de R. Viers : Paris/Nice : Éditions Karthala/ Association Alphabets, 2000), 175-211 (200-4).

98 bibliques, Pentateuque et Lévitique en particulier, et dans l’écriture du tétragramme10 et de quelques autres noms divins11, ainsi que sur des monnaies12. Par ailleurs, les inscriptions et les ostraca provenant de Khirbet Qumrân qui sont comparables aux inscriptions des jarres retrouvées dans les grottes à manuscrits, indiquent eux aussi un ­terminus ante quem13. Quelques considérations générales sont utiles pour l’étude paléographique de la forme des lettres afin de caractériser la main des scribes qui utilisent un même support et un calame de même type. Les différences portent essentiellement sur la dimension des lettres, la décomposition des traits, la direction des tracés et leurs points de rencontre. Et malgré de petites différences entre chaque main, l’identification d’une lettre dans ce qui la différencie d’une autre est généralement possible. En effet, chaque scribe a son propre style ou ductus pour tracer chaque lettre composée d’un nombre limité de traits fondamentaux reproduisant l’image idéale stéréotypée ou “pictogramme” de chacune d’elles à chaque stade de son évolution ; ainsi les lettres composées de deux ou trois traits se distinguent les unes des autres par leurs tracés 10   Le tétragramme est écrit en paléo-hébreu dans des textes bibliques comme 2Q2 (Exode), 3Q3 (Lamentations), 4QIsc, 11QPsa (y compris les Psaumes non canoniques et les Hymnes), dans plusieurs pesharîm, par exemple 1Q14, 1Q15, 4Q161, 4Q171, mais aussi en écriture araméenne habituelle, par exemple dans 4Q162, 4Q163, 4Q174 (un pesher thématique). 11  Voir ’l en 1Q14, 1Q35, 3Q14 frag. 18, 4QIsc, 4Q180, 4Q182, etc., 6Q15 (D), 6Q18 (Hymnes), etc., et le paléo-hébreu est aussi usité dans une forme d’écriture cryptique avec d’autres alphabets, voir 4Q186, qui sont des manuscrits authentiquement esséniens. 12  Utilisée entre autre sur les monnaies de la première révolte, cette écriture “paléo-hébraïque” n’est plus utilisée après la deuxième révolte juive, les rabbins ayant interdit cette écriture, apparemment comme réactions anti-sadducéenne et anti-samaritaine ; mais les Samaritains l’ont toujours gardée même pour leurs inscriptions et les copies des livres bibliques. 13  Voir de Vaux, “Esséniens ou Zélotes ?” 230, et Emile Puech, “L’ostracon de Khirbet Qumrân (KhQ 1996/1) et une vente de terrain à Jéricho, témoin de l’occupation essénienne à Qumrân,” in Flores Florentino : Dead Sea Scrolls and Other Early Jewish Studies in Honour of Florentino García Martínez (JSJSup 122 ; ed. A. Hilhorst, E. Puech, et E. J. C. Tigchelaar ; Leiden : Brill, 2007), 1-29 (20-21 et n. 49), j’ai daté le bol inscrit du locus 89 de la fin de la période 1b en accord avec de Vaux, Archaeology and the Dead Sea Scrolls, 11 et n. 1 malgré Milik, mais avec Frank M. Cross, voir infra n. 15, respectivement 190, n. 9, et 4, n. 9, et Emile Puech, “L’épigraphie de Qumrân : Son apport à l’identification du site”, RevQ 95 (2010) : 433-40, où est soulignée l’identité d’inscriptions sur les jarres du site et celles de la grotte 4. L’étude des lampes des grottes et du site va dans le même sens, voir Jolanta Młynarczyk, “Terracotta Oil Lamps” (dans ce volume).

Puech

et leurs points de rencontre. Sont à distinguer les caractères stylistiques des marques personnelles d’une écriture que traduisent la dimension des lettres, l’angle d’inclinaison des tracés et, jusqu’à un certain point, l’ensemble des traits personnalisés qui reflètent la dextérité ou l’entraînement du scribe jusqu’à son caractère, voire même son âge, des facteurs qui relèvent tout autant de la graphologie. Mais pour déchiffrer et identifier une main, on doit d’abord en rester à la forme conventionnelle des lettres dans une époque donnée. L’évolution des formes au cours des siècles est due à plusieurs facteurs techniques, comme le support de l’écriture et le type et la forme de l’outil, poinçon, stylet ou calame, sa tenue dans la main et la rapidité dans l’exécution des tracés. D’autres facteurs stylistiques peuvent intervenir dans les écoles de scribes qui transmettent leurs manières propres, jusqu’à l’invention d’écritures cryptiques, au nombre de trois parmi les manuscrits de la mer Morte. Afin de déterminer un type d’écriture, le paléographe examine la structure de chaque lettre, ses tracés de base et les fioritures additionnelles s’il y a lieu, l’épaisseur, la direction et les jonctions des pleins et des déliés dans une écriture à l’encre, le module de la lettre donné par sa largeur et sa hauteur, la dimension relative de chaque lettre, leur rapport entre elles, etc. Cet ensemble d’observations permet de comparer une main à une autre, de distinguer des copistes à l’aide de caractéristiques personnelles, mais aussi de relever les traits communs à une école à une époque donnée. Ainsi un examen paléographique détaillé et précis autorise une classification et une datation des écritures selon leur style et parfois même d’identifier un scribe au point de lui attribuer plusieurs ou un ensemble de copies (voir ci-dessous 4). À défaut d’une datation absolue, une analyse paléographique bien menée arrive à déterminer une chronologie relative dans la fourchette d’une génération, fondée qu’elle est sur les changements formels dans le ductus et le module de chaque lettre au cours des siècles, jusqu’à l’époque post-hérodienne tout au moins en ce qui nous concerne ici. Depuis un demi-siècle, l’étude de la paléographie des écritures de chancellerie de l’époque perse à celles de la deuxième révolte juive s’est beaucoup précisée grâce à l’abondance des découvertes d’inscriptions datées comme points de repères. Avec la fin de l’usage de la belle écriture araméenne de la deuxième moitié du quatrième siècle bien connue et datée par les papyri de Samarie retrouvés dans des grottes du Ouadî Daliyeh ou celui des grottes de Jéricho14, l’écriture standard de l’empire perse 14  Voir Hanan Eshel et Haggai Misgav, “A Fourth Century B.C.E. Document from Ketef Yeriho”, IEJ 38 (1988) : 158-76, même si la

99

La Paléographie Des Manuscrits De La Mer Morte

est, avec quelques éléments de la cursive, l’ancêtre direct de la plus ancienne écriture formelle retrouvée à Qumrân. Bien que le processus évolutif des formes des lettres soit graduel et continuel, et variable selon les écoles et les scribes, il a cependant été possible de distinguer quatre grandes phases, quelque peu arbitrairement sans doute mais toutefois en référence à des périodes historiques15 : 1) la phase dite archaïque ou pré-hasmonéenne (circa 250150), 2) la phase hasmonéenne (circa 150 à-31), 3) la phase hérodienne (circa 30 avant à 70 après J.-C.), et 4) la phase post-hérodienne (70-135). En outre, dans ces phases principales, sont à distinguer des sous-types, en particulier : 1) l’écriture formelle ou calligraphique, 2) l’écriture semi-formelle assimilant des influences de la cursive précédente, 3) l’écriture cursive dans le prolongement de la cursive apparue à la fin de l’empire perse, et 4) l’écriture semi-cursive qui est le produit d’un croisement de l’écriture formelle et de l’écriture cursive, sans compter quelques autres sous-catégories. La majorité des manuscrits des grottes de Qumrân datent des phases hasmonéenne et hérodienne. Toutefois, quelques manuscrits semblent appartenir encore à la phase précédente, à la période “archaïque”, ou même la précéder comme étape de transition avec les caractéristiques du quatrième siècle et des amorces de l’écriture proto-cursive, tel 4QEx-Lvf (4Q17)16. La caractéristique générale est dans la recherche d’une certaine standardisation de la dimension d’un module des lettres sous la ligne idéale d’une écriture suspendue.

tendance actuelle serait de le rajeunir quelque peu, voir Cross, “The Scripts of the Dâliyeh (Samaria) Papyri”, 43 et n. 202. 15  Voir principalement Nahman Avigad, “The Palaeography of the Dead Sea Scrolls and Related Documents,” in Aspects of the Dead Sea Scrolls (ScrHier 4 ; ed. C. Rabin et Y. Yadin ; Jerusalem : Magnes Press, 1958) : 56-87, et Frank M. Cross, “The Development of the Jewish Scripts”, in The Bible and the Ancient Near East : Essays in Honor of William Foxwell Albright (ed. G. E. Wright ; Garden City, NY : Doubleday, 1961), 133-202, repris et partiellement mis à jour, idem, Leaves from an Epigrapher’s Notebook, 3-43, avec une bibliographie. Voir aussi Ada Yardeni, Textbook of Aramaic, Hebrew and Nabatean Documentary Texts from the Judaean Desert and Related Material, 2 vols. (Jerusalem : Hebrew University, Ben Zion Dinur Center, 2000), malgré les remarques paléographiques et autres sur des évidences qui lui sont propres de Gregory L. Doudna dans sa communication au colloque (dans ce volume). 16  Cross, “The Scripts of the Dâliyeh (Samaria) Papyri”, 18b–22a et 27b : entre 275 et 200, au mieux vers 250 av. J.-C.

La phase archaïque ou pré-hasmonéenne (circa 250 à 150)17 Le grec remplaçant l’araméen comme lingua franca avec Alexandre le Grand, l’emploi de l’araméen perdit peu à peu de son rayonnement international, et lentement il fut réservé pour les usages locaux d’une population toujours de langue sémitique, donnant graduellement naissance, à travers les écoles scribales d’administrations géographiquement circonscrites, à des écritures nationales, telles le palmyrénien, le nabatéen et l’araméen palestinien, judéen en particulier, ou si l’on veut, à un araméen provincial, ou dialectal en devenir, avec le déclin de l’empire grec dès la mort d’Antiochus IV. En un sens, l’évolution locale de la graphie accompagne la dialectisation de la langue. Ce type d’écriture archaïque formelle avec des formes encore assez étroites et allongées de la tradition de la chancellerie perse, est caractérisé par l’apparition de formes médianes, des kaf, nun, pe en particulier, moins clairement pour mem et ṣadé, des jambages finissent par des lignes de fuite vers la gauche pour attaquer la lettre suivante, sous l’influence de la cursive ou de la rapidité d’exécution du tracé, courbes évidemment absentes dans les lettres à la fin des mots, et où se distinguent aussi assez souvent le lamed et le he avec un allongement du trait horizontal et parfois même le ʾalef. Ce type se retrouve par exemple déjà dans 4QSmb (4Q52) vers 250, dans 4QJra (4Q70) manuscrit un peu plus tardif vers 200, et enfin à un stade un peu plus avancé annonçant la transition en 4QDta (4Q29) de la première moitié du deuxième siècle, qui sont tous des copies de livres “bibliques”. Il en est de même de 4QHena (4Q201) à l’écriture archaïque semi-formelle/semi-cursive avec des éléments proto-cursifs à dater circa 20018 et de 4QQoa (4Q109) à dater peu après dans le deuxième quart du deuxième siècle. 2.1

L’écriture hasmonéenne (circa 150 à 37/31 av. J.-C.) Tout au long de cette phase bien représentée à Qumrân, l’écriture formelle confirme la tendance à une standardisation de la dimension des lettres toujours suspendues, comme si elles étaient délimitées par deux lignes idéales, ce qui sera la règle ensuite de l’écriture hérodienne. Comme exemples de cette écriture formelle hasmonéeenne, on peut citer 4QDta (4Q28) et 4QPsa (4Q83) à la transition ou du début de la phase, circa 150, 4QDtc (4Q30) vers la fin du deuxième siècle, ou circa 100, 4QHene 2.2

17  Cross, “The Development of the Jewish Scripts” ; idem, Leaves from an Epigrapher’s Notebook, la nomme “pre- or proto-Jewish”. 18  Voir Józef T. Milik, The Books of Enoch : Aramaic Fragments of Qumrân Cave 4 (Oxford : Clarendon Press, 1976), 140.

100

Puech

(4Q206), 4QApocalypse messianique (4Q521) peu après 10019, et 4QSma (4Q51) vers le milieu du premier siècle. Parmi les exemples d’écriture semi-formelle hasmonéenne du début de la phase serait à classer 4QParoles des Luminaires – 4QDibrê ha-me’orôt (4Q504)20, et de la fin du deuxième siècle ou circa 100 sont à ranger 1QIsa, et contemporains, copiés par un même scribe, 4QSmc (4Q53), 1QS-Sa-Sb, 4QTestimonia (4Q175). Parmi les écritures hasmonéennes semi-cursives anciennes, j’ai rangé 4QRouleau du Temple (4Q524), et 4QJonathan (4Q523), circa 130 avant J.-C21. L’écriture semi-cursive hasmonéenne médiane et tardive est aussi bien représentée par 4QXII a (4Q76), 4QDnc (4Q114), 4QGéantsb (4Q530) circa 100, puis 6QGéants (6Q8) avec des influences plus marquées de la cursive, 4QHeng (4Q212)22, 4QMMTe (4Q398) et 4Q448 (PsApoc et Prière)23 circa 50 avant J.-C. De cette

phase finale de la semi-cursive hasmonéenne datent les copies de 4Q331-332-333 qui parlent des massacres du gouverneur romain de la province de Syrie, Æmilius Scaurus, de la reine Salomé-Shlomsion, de Hyrkan II et d’Aristobule II, copies plus ou moins contemporaines des événements24. L’écriture cursive type n’est apparemment pas ou peu représentée comme telle par les manuscrits littéraires des grottes, mais par le papyrus Nash en Égypte circa 150, portant les dix commandements et le début du Shema‛, et par des inscriptions, par exemple le texte narratif sur l’ostracon de Murabbaʿât (Mur72) circa la fin du deuxième siècle25. Je propose de classer aussi dans ce type d’écriture cursive tardive, vers le milieu du premier siècle av. J.-C., le manuscrit 4Q468e-Texte historique F, qui mentionne le commandant Peitholaos26. Cet échantillon des types d’écritures hasmonéennes comprend aussi bien

19   Voir Emile Puech, “521. 4QApocalypse messianique”, in idem, Qumrân Grotte 4.XVIII : Textes hébreux (4Q521-4Q528, 4Q576-4Q579) (DJD 25 ; Oxford : Clarendon Press, 1998), 5 : “circa 100-80” (non 125-75, Webster, “Chronological Index”, 365, 372), et les résultats du C14 pour ce manuscrit sont entre 93 av. J.-C. et 80 AD, voir A. J. Timothy Jull et al., “Radiocarbon Dating of Scrolls and Linen Fragments from the Judaean Desert”, Radiocarbon 37 (1995) : 11-19 = ʿAtiqot 28 (1996) : 85-91. 20  Voir Maurice Baillet, “504. Paroles des luminaires (premier exemplaire : DibHama)”, in idem, Qumrân Grotte 4.III (4Q482-4Q520) (DJD 7 ; Oxford : Clarendon Press, 1982), 137 : “L’écriture est une calligraphie asmonéenne, qui peut dater des environs de 150 avant J.-C.”, Cross, “The Development of the Jewish Scripts”, 127 Fig. 1,7 : “An Archaic or early Hasmonaean semiformal script of ca. 175-125 B.C.,” mais idem, “The Scripts of the Dâliyeh (Samaria) Papyri”, 7 Fig. 1.1,7 : “. . . of ca. 175-150 BCE” ! L’usage de ʾdwny au lieu du tétragramme YHWH utilisé encore par Ben Sira par exemple, milite au contraire pour une date circa 150 ou peu après. 21  Voir Emile Puech, “524. 4QRouleau du Temple,” in DJD 25, respectivement 87 et 75. 22  Avec Milik, The Books of Enoch, 246, alors que Cross le date de la seconde moitié du premier siècle av. J.-C., “The Development of the Jewish Scripts”, 149, Fig. 4,5, et “The Scripts of the Dâliyeh (Samaria) Papyri”, 17, Fig. 1.4,5. 23  Ce manuscrit en écriture semi-cursive hasmonéenne tardive peut avoir été copié par deux mains différentes comme le suggèrerait la présentation et la disposition des “deux ou trois (?)” colonnes au-dessous de la copie du Psaume, ou par une même main dans un deuxième temps, voir Esther Eshel, Hanan Eshel, et Ada Yardeni, “448. 4QApocryphal Psalm and Prayer”, in Qumran Cave 4.VI : Poetical and Liturgical Texts, Part 1 (DJD 11 ; ed. E. Eshel et al. ; Oxford : Clarendon Press, 1998), 403-25, mais on ne peut être aussi affirmatif que les éditeurs : “4Q448 includes a prayer for the welfare of King Jonathan, whom we identify as Alexander Jannaeus (103-76 BCE). We believe that 4Q448 was copied during Jannaeus’ lifetime and that it is thus the only [mes italiques] document from Qumran published to date which can be assigned with almost complete certainty to the first half of

the first century BCE. Hence, this document is extremely important for Hebrew palaeography” (404-5), mais on peut en douter ! Comme l’identification du roi est discutable -et elle a été de fait disputée-, et que plusieurs solutions sont envisageables pour identifier “le roi Jonathan”, sa datation paléographique n’est pas aussi ferme que l’affirment les auteurs, loin de là ; la copie assez peu soignée, plus cursive dans les deux colonnes inférieures, peut n’être point contemporaine dudit roi en question, ce que rien ne prouve par ailleurs et une identification avec Jonathan Maccabée est bien plus vraisemblable, comme je l’ai montré. Une date vers le milieu du 2e siècle est tout aussi vraisemblable que dans le premier quart du 1er. 24  Ces copies portent toutes les caractéristiques de l’écriture hasmonéenne semi-cursive tardive, circa 50-35, contrairement à l’analyse erronée de Joseph A. Fitzmyer, “4QpapHistorical Text C”, “4QHistorical Text D”, et “4QHistorical Text E”, in Qumran Cave 4.XXVI : Cryptic Texts, ed. S. J. Pfann ; Miscellanea, Part 1, ed. P. S. Alexander et al. (DJD 36 ; Oxford : Clarendon Press, 2000), 275-89, qui les classe respectivement dans “a typical Hasmonaean script” (276), “early Herodian formal script” (281), “a semi-formal Herodian hand” (287) ! 25  Un autre exemple d’écriture cursive est représenté par l’inscription araméenne du tombeau de Jason à Jérusalem, à dater circa le milieu du premier siècle av. J.-C. ou peu avant, voir voir Emile Puech, “Inscriptions funéraires palestiniennes : tombeau de Jason et ossuaires”, RB 90 (1983) : 481-533. 26  Ada Yardeni, in Magen Broshi, “468e. 4QHistorical Text F”, in DJD 36, 406-22 (406-7), en a fait “a late first century BCE or Early first century CE mixed semi-cursive script”, invoquant l’apparition de l’alef cursif uniquement dans la phase hérodienne, mais ce type d’alef cursif est bien connu par exemple de l’inscription de Jason, déjà auparavant. Ce texte offre un mélange d’écriture semi-cursive et de cursive hasmonéennes tardives, et paraît de peu postérieur aux événements qu’il rapporte ; la lecture de M. Broshi, Potlais = Ptollâs contemporain d’Archélaüs, est clairement fautive mais elle a déterminé cette fois encore l’analyse paléographique de Yardeni ! Voir John Strugnell, “The Historical Background to 4Q468g [= 4Qhistorical B]”, RevQ 73 (1999) : 137-38 : Peitholaos était contemporain de Hyrkan II et d’Aristobule II.

La Paléographie Des Manuscrits De La Mer Morte

des copies de livres “bibliques”, d’apocryphes, et des rouleaux typiquement esséniens27, et il correspond bien à la Période 1 ( = 1a–1b) d’occupation de Khirbet Qumrân, telle que l’a reconnue le fouilleur R. de Vaux28. L’écriture hérodienne (circa 31 av. J.-C. à 70 après J.-C.) L’appellation “hérodienne” englobe de fait tout le dernier siècle de l’occupation essénienne de Khirbet Qumrân, soit la Période II reconnue par le fouilleur R. de Vaux, de circa 31 avant J.-C. à 68 après J.-C. Au cours de cette période, l’écriture formelle va atteindre son plein développement avec l’uniformisation assez rapide du module “carré” des lettres, comme si les scribes avaient maintenant pleinement conscience d’une écriture suspendue mais délimitée par deux lignes idéales, semblant aussi reposer sur une ligne de base bien que celle-ci ne soit jamais tracée, peutêtre sous l’influence de l’écriture onciale grecque connue à Qumrân. Ainsi s’expliquent la longueur égale des jambages et les bases coudées à gauche tendant à un angle droit des lettres médianes, mais sans nulle ligature29, comme il en sera de l’écriture cursive pour quelques lettres tout au moins qui, avec ses nombreuses ligatures, finira plus tard par donner naissance à une écriture posée sur la ligne, et non plus suspendue. Va aussi se développer graduellement la présence de kéraiai ou apices dont l’emploi sera systématique dans une écriture calligraphique à l’époque suivante (de 70 à 135). À côté de l’élégante écriture formelle hérodienne ancienne carrée, bien représentée par 2.3

27  L’ostracon KhQ1996/1 paraît devoir être classé dans l’écriture hasmonéenne vulgaire, semi-formelle tardive avec des influences de la cursive, de préférence à l’écriture hérodienne tardive, comme l’écrivent les éditeurs suite à leur déchiffrement forcé de la date de la deuxième année de la révolte juive, ligne 1, voir Frank M. Cross et Esther Eshel, “1.Khirbet Qumran Ostracon”, in DJD 36 : 497-507, malgré les réactions paléographiques de Cross à la lecture souvent bien préférable de A. Yardeni (505-7). 28  Voir de Vaux, Archaeology and the Dead Sea Scrolls, 5, 48, retenant plus volontiers le début de l’occupation sous Simon Maccabée, suivi par Frank M. Cross, The Ancient Library of Qumran and Modern Biblical Studies (3rd. ed. ; Sheffield : Sheffield Academic Press, 1995), 97, qui retient Simon mais pour des motifs exégétiques, alors qu’il concède les dates archéologiques limites entre 150 et 134 (voir aussi 184), mais de Vaux n’exclut pas un début sous Jonathan Maccabée, comme le proposent Józef T. Milik, Ten Years of Discovery in the Wilderness of Judaea (SBT 26 ; London : SCM Press, 1959), 83-87, 98, (qui a participé à la fouille du site), ainsi que bien d’autres savants avec lui, date qui paraît de loin la meilleure, répondant à la fois aux données des documents et des monuments. 29  Frank M. Cross prend pour ligature le fait que deux lettres se touchent, ce qui est inexact. Une ligature suppose le tracé continu de deux ou plusieurs lettres sans aucun lever de main.

101 1QMilḥama-Rouleau de la Guerre du dernier tiers du premier siècle av. J.-C. ou circa 0 dans la ligne de 4QSma, par 1QHa-Hymnes (prima manu) du tournant de notre ère, par l’écriture formelle hérodienne moyenne (deuxième quart du premier siècle) tel 4QDnb (4Q113), 4QPešerNahum (4Q169) et par l’écriture tardive du troisième tiers du premier siècle comme 4QDtj (4Q47), 4QPsb (4Q84), on a distingué une écriture semi-formelle rustique et une écriture semi-formelle ­vulgaire, qui ont respectivement pris la relève de l’écriture semi-formelle et semi-cursive hasmonéenne. L’écriture semi-formelle rustique hérodienne ancienne (du dernier tiers du premier siècle avant J.-C. au début du premier siècle après (circa de -30 à 20 AD) est très bien documentée parmi les manuscrits de Qumrân et de Masada, voir 4QNbb (4Q27), 4QpešerPsa (4Q171), 1QJérusalem Nouvelle (1Q32), 4QGéants (4Q531), 4QBéatitudes (4Q525) (voir ci-dessous “Un grand copiste du scriptorium de Qumrân”), MasApocryphe de Josué, etc. L’écriture hérodienne semi-formelle vulgaire est représentée par la seconde main de 1QHa-Hymnes, et plus tard en 68 par le Rouleau de cuivre (3Q15). Les écritures hérodiennes semi-cursives et cursives sont représentées par le papyrus Murabbaʿât 18 daté de la deuxième année de Néron en 55/56, par l’inscription incisée sur le vase de ‛Aïn Feshkha 207 datée de l’année 66, et la plupart des inscriptions des grottes et de Khirbet Qumrân30. 2.4 L’écriture post-hérodienne (70-135) La phase de l’écriture post-hérodienne couvre la période d’entre les deux révoltes juives, soit après la destruction de Qumrân, de Jérusalem et de Masada à la fin de la deuxième révolte, de 70 à 135 ap. J.-C. La plupart de la documentation est écrite dans une écriture notariale souvent datée ou cursive, ainsi les actes de vente, contrats divers, actes de répudiation, lettres, les lettres de Bar Koséba, des listes, ou en cursive extrême dans les contrats doubles, avec le premier texte à l’intérieur (scriptura interior), scellé par les sept sceaux. Mais plusieurs manuscrits attestent l’écriture formelle des scribes recopiant des livres “bibliques”, tels les fragments de Genèse-Exode (Mur1) et du Rouleau des XII Petits Prophètes de Murabbaʿât (Mur88), un fragment d’un rouleau de Psaumes du Ouadî Khabra, et des abécédaires de Murabbaʿât (Mur78, Mur80).

30  Voir André Lemaire, “Inscriptions du Khirbeh, des grottes et de ‛Aïn Feshkha”, in Khirbet Qumrân et ‘Aïn Feshkha. Vol. 2 : Études d’anthropologie, de physique et de chimie (NTOA.SA 3 ; présentées par J.-B. Humbert et J. Gunneweg ; Fribourg/Göttingen : Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 341-88, mais bien des lectures sont à revoir, et feront l’objet d’une autre publication.

102 L’analyse paléographique a montré que les manuscrits des grottes appartiennent aux trois premières phases et d’aucuns sans doute à la période précédente, mais aucun manuscrit n’appartient à la période post-hérodienne31. Comme confirmation de la méthode paléographique et de ses résultats, une forme de vérification des datations proposées est maintenant fournie par le carbone 14 (C14) qui, à défaut d’une date précise, donne au moins une fourchette de datation du support matériel à défaut de celle de l’encre. Mais cette analyse donne encore des résultats bien moins précis qu’une datation paléographique bien menée32. Si les manuscrits “bibliques” et Apocryphes ou Péritestamentaires ont été recopiés au moins depuis la phase archaïque et en suivant, il est important de noter que, même si elle n’est pas une autographe, la plus ancienne copie de la Règle de la Communauté (4Q255-papSa) dont le caractère essénien ne peut être mis en doute, date du dernier quart du deuxième siècle av. J.-C. suivie peu après, avec des développements internes de la copie de la Règle de 1QS et 4Q257-papSc, datées circa 100 av. J.-C., les neuf autres copies s’étalant au long des phases hasmonéenne tardive et hérodienne33. C’est là une claire indication de la longue présence essénienne à Qumrân, qui rejoint la 31  Malgré Webster, “Chronological Index”, 351 n. 1, pour 4Q2 et 4Q21, 6Q30 et 9Q1. 6Q23 et 6Q30 sont à ranger au plus tard dans l’écriture semi-cursive ou cursive hérodienne (les restes conservés sont trop réduits pour plus de précisions). 32  Voir George Bonani et al., “Radiocarbon Dating of the Dead Sea Scrolls”, ʿAtiqot 20 (1991) : 27-32, dont les manuscrits 4Q542, 4Q365, 1QIsa, 4QTestLévi, 4QSmc, 11Q19 (RT), 1QApGn, 1QHa, Mas1039/211 (ApocJos) et Mas1747/10 (non identifié), et une autre série d’analyses dans un autre laboratoire par Jull et al., “Radiocarbon Dating of Scrolls”, dont 4Q266 (Da), 1QpHa, 1QS, 4Q268 (Sd), 4Q171 (pPsa), 4Q521 (ApocMessianique), 4Q267 (Db), 4Q249 (MidrašSeferMoše), 4Q317 (Phases de la lune), 4Q208 (HenAstra), 4Q22 (paléo-Ex), 4Q342 (Lettre ar), 4Q344 (Reconnaissance de dette ar), 4Q345 (Vente de terrain), 1QIsa. Cette dernière étude a aussi analysé des tissus en lin. À ce sujet, les conclusions du C14 de Gregory L. Doudna, “Dating the Scrolls on the Basis of Radiocarbon Analysis”, in The Dead Sea Scrolls After Fifty Years : A Comprehensive Assessment (ed. P. W. Flint et J. C. VanderKam ; Leiden : Brill, 1998), 1 :430-71 (460-64), contrediraient son hypothèse, même révisée en Gregory L. Doudna, “Redating the Dead Sea Scrolls found at Qumran : the case for 63 BCE”, QC 8/4 (1999) : 1-96 (62-88), selon laquelle les manuscrits doivent tous appartenir à la période Ib (comprise du règne de Jannée à Pompée), aucun manuscrit ne serait postérieur à cette date. Il est clair que cette hypothèse repose sur une série d’a priori ne tenant pas compte des realia. 33  Il est à noter que la copie 4QSe(- 4Q259) en écriture semi-cursive hasmonéenne tardive use aussi de lettres d’un alphabet cryptique, ce qui accentue le caractère essénien de cette écriture, voir Emile Puech, “L’alphabet cryptique en 4QSe (4Q259)”, RevQ 71 (1998) : 429-35.

Puech

date de composition de 4QMMT au milieu du deuxième siècle34, et qui s’accorde pleinement avec les données archéologiques présentées par le fouilleur, mais qui n’appuient pas la remise en question de leur datation par les réviseurs ultérieurs. En revanche, aucun manuscrit d’une composition essénienne ne date de la phase archaïque ou pré-hasmonéenne, cela est aussi important à souligner. On a là un terminus post quem pour les débuts de l’histoire de la Communauté et de l’occupation essénienne à Qumrân35, tout comme un terminus ante quem pour sa fin, puisqu’il n’y a pas de copies qui soient datées de la phase post-hérodienne, la dernière composition étant le Rouleau de cuivre d’un genre particulier. Les multiples caractéristiques des copies des scribes qumraniens, outre leur invention d’écritures cryptiques propres, font que la plupart des manuscrits retrouvés datant des périodes hasmonéenne et hérodienne sont des copies des scribes du scriptorium qumranien36, non des copies venant du Temple ou d’ailleurs, pour être déposées dans ces grottes cachettes en temps de crise, comme il est parfois suggéré. Enfin, l’écriture formelle ou calligraphique des copies des livres “bibliques” (Mur1 et Mur88) du deuxième siècle AD a été “canonisée” en même temps que le texte pré-massorétique, devenant ainsi la norme paléographique de la copie des livres du “canon” de la Bible. 3

La paléographie comme outil précieux pour le déchiffrement des fragments 

Il est clair que l’analyse des différentes mains des copistes et de leur ductus est une aide précieuse non seulement pour dater une copie, mais aussi pour identifier des lettres très souvent partiellement conservées et pouvoir proposer une restauration d’un texte endommagé, que ce soit des mots et même des lacunes plus importantes. Les exemples sont légions et il est inutile d’insister longuement sur cette aide à la lecture, qui s’impose pour une réédition critique des manuscrits. Deux exemples récents parmi bien d’autres vont illustrer mon propos : 34  Voir Emile Puech, “L’épilogue de 4QMMT revisité”, in A Teacher for All Generations : Essays in Honor of James C. VanderKam (JSJSup 153 ; ed. E. F. Mason et al. ; Leiden : Brill, 2012), 1 :309-39, et Emile Puech, “La Lettre essénienne MMT dans le manuscrit 4Q397 et les parallèles”, RevQ 105 (2015) : 99-135. 35  Les pešarîm qui font allusion à des événements contemporains, sont tous des compositions qumraniennes, grosso modo de la deuxième génération qumranienne, dans le premier tiers ou, au plus tard, le deuxième quart du premier siècle avant J.-C. Cela donne un terminus post quem pour les copies ensuite. 36  Le nombre d’encriers retrouvés à Qumrân témoigne lui aussi d’une importante activité scribale.

103

La Paléographie Des Manuscrits De La Mer Morte

Un premier exemple est donné en 4Q225 2 i 1 : suite au déchiffrement correct des traces de lettres du début du fragment 2, on doit lire à la fin de la ligne 14 de la colonne précédente et à la ligne 1 et au début de la ligne 2, la citation de Gn 17,14a d’après le Texte Hébreu attesté par le Samaritain et par les LXX (comme en Lv 12,3) avec une légère variante dans la forme verbale pour en faire l’obligation morale d’un commandement divin. Cette lecture assurée, parfaite pour la reconstruction du passage et l’espace à la marge droite à 2,5 cm ainsi que pour la provenance de la citation de Genèse, exclut une citation de Nb 15,31 inattendue ici à la suite de l’editio princeps. Abraham a pratiqué la circoncision, une fois entré en Canaan (Gn 17,23-24 ; Jub 15,11-14.23-24), ainsi qu’Isaac, lui-même circoncis “âgé de huit jours comme Dieu avait ordonné”. Cette ligne est la seule attestation de Gn 17,14 dans les manuscrits de Qumrân, qui plus est, dans une leçon en accord avec la Vorlage hébraïque des LXX et avec le Samaritain, soit maintenant trois familles textuelles concordantes contre le Texte Massorétique. C’est en souligner son importance textuelle pour l’étude du verset37. Un autre exemple est donné avec 4Q398 14-17 i 2-4 : un déchiffrement correct de ces lignes a fait difficulté, mais il est capital d’une part pour comprendre son parallèle avec 4Q397 14-21 ii 9-15, et pour retrouver ainsi la séquence originelle des colonnes en 4Q398 ; les éditeurs ont présenté chacun un point de vue différent, mais une seule séquence est possible et elle est confirmée par un bon déchiffrement et une reconstruction du passage parallèle38. D’autre part, le déchiffrement correct est capital aussi pour l’attestation irréfutable de la croyance à la résurrection des justes et au châtiment des méchants lors du jugement eschatologique à la fin des jours, croyance définitivement acceptée dans le courant essénien à la suite de Dn 12,1-3, croyance qu’il partage avec le courant pharisien mais que n’acceptaient pas les Sadducéens, et croyance que la notice de Flavius Josèphe lui refusait et que bien des modernes, à la suite de Flavius Josèphe et sans tenir grand compte des textes, refusent encore d’admettre malgré les évidences39. 37  Voir Emile Puech, “4Q225 revisité : un midrash essénien ?” RevQ 102 (2013) : 169-209. 38  Voir Puech, “L’épilogue de 4QMMT revisité”, 313-15. 39  Le passage va dans le sens de tous les autres manuscrits qumrano-esséniens touchant le sujet, y compris 4Q521, etc., voir par exemple Emile Puech, “Apports des manuscrits de Qumrân à la croyance à la résurrection dans le Judaïsme ancien”, in Qoumrân et le Judaïsme du tournant de notre ère : Actes de la Table Ronde, Collège de France, 16 novembre 2004 (CREJ 40 ; sous la direction de A. Lemaire et S. C. Mimouni ; Leuven : Peeters, 2006), 81-110, où je n’ai toutefois pas fait appel à 4QMMT pour ne pas avoir alors revu de près le déchiffrement de ce manuscrit, voir maintenant Puech, “La Lettre essénienne MMT dans le manuscrit 4Q397 et les

L’affirmation suit la recommandation d’une étude assidue des Livres de Moïse, des Prophètes et de David. Or personne de nos jours ne met en doute la composition manifestement essénienne de 4QMMT. Il est donc clairement prouvé par ces exemples et par bien d’autres que plusieurs conséquences très importantes découlent d’un déchiffrement correct des tracés de quelques lettres, en particulier quand elles sont mal conservées. Une attention très précise à porter à l’écriture et à la direction des traits d’une main de scribe est sans cesse à rappeler à l’intention du paléographe. 4

Un grand copiste du scriptorium de Qumrân40

L’analyse paléographique ci-dessus a révélé une grande diversité de mains de scribes sur plusieurs siècles et permis, par la détermination des traits communs et de leurs variations, de proposer une datation relative dans l’évolution des tracés. Elle a révélé aussi le ductus spécifique de diverses mains dans une longue tradition scribale contribuant à une classification des manuscrits et à leur attribution à tel ou tel scribe plus particulier. En effet, chaque copiste même professionnel a une manière plus personnelle de tenir son calame, d’en affûter son extrémité, d’attaquer ou de finir un trait, bref tout ce qui donne une apparence spécifique à son écriture quant à sa netteté, ses proportions, sa régularité, son élégance même, laissant percevoir sa touche propre comme sa signature personnelle, mais on ne doit pas aussi oublier que l’entraînement de la main finit par induire au fil des années une évolution dans l’enchaînement des tracés et, ça et là, quelques variables inévitables. Il est clair que le tracé d’une lettre par un même scribe ne peut se réduire à la répétition ou décalque d’une matrice, tels les caractères d’imprimerie. Ainsi en est-il du scribe qui a copié un nombre impressionnant de rouleaux de toutes sortes, quant au genre parallèles”, RevQ 105 (2015) 99-135, n. 106. Comme la composition date des tout premiers temps de la fondation de la Communauté, elle revêt de fait une grande importance pour ce sujet, allant dans le sens de la notice d’Hippolyte de Rome et non dans celui, passablement déformé, de celle de Flavius Josèphe Bell. Jud. II 151-58. 40  Même si le terme peut paraître anachronique, il désigne certainement une réalité présente à Qumrân, dans le locus 30 où ont été retrouvées des tables de mortier et des encriers. Il est évident que la copie d’un rouleau à partir d’un autre rouleau exige une surface plane importante autre que les seuls genoux du scribe ou une simple planche plus ou moins stable. Voir Ronny Reich, “A Note on the Function of Room 30 (the ‘Scriptorium’) at Khirbet Qumran”, JJS 46 (1995) : 157-60, qui critique avec raison l’utilisation de ce genre de mobilier comme tables d’un triclinium plus que suspect, à la romaine, de la proposition de P. Donceel-Voute.

104 littéraire, à Qumrân même vers la fin du premier siècle avant J.-C. et peu après, dont on a retrouvé des témoignages jusqu’à Masada, preuve s’il en est, que quelques Esséniens de Qumrân, devant l’attaque romaine du site au printemps 68, trouvèrent refuge en compagnie d’autres résistants dans la forteresse de Masada, en y emportant quelques uns de leurs rouleaux41. Ce scribe se distingue en particulier par une tendance à préférer les courbes gracieuses, à cambrer les jambages des lettres, à tracer la tête du he en “V” replié à gauche parfois même en le prolongeant dans le jambage gauche après un retour à angle aigu, à commencer la hampe du lamed par une boucle plus ou moins marquée et à la poser sur un grand pied replié en forme de trombone, forme qui se retrouve en partie dans la tête du qof, à distinguer waw de yod le plus souvent à tête en triangle, et à préférer souvent un ample kaf médian en position finale, de même pour ṣade, mais le mem final est clairement tracé avec une large tête caractéristique. Certes, de légères différences ou variantes entre ses copies peuvent tenir à la rapidité d’exécution des tracés des lettres, à l’urgence des commandes, à la dimension du rouleau à copier, à des conditions impondérables du scribe affairé à sa table de travail au scriptorium, à des distractions, à une altération de sa vue dans une longue vie de copiste, aux conditions climatiques du moment, etc. Il est à souligner aussi qu’un scribe même entraîné reste un homme avec toutes ses limites du moment. Si ces considérations ont quelque intérêt pour l’identification d’une main et l’attribution des fragments à tel ou tel manuscrit, il n’en reste pas moins, qu’à défaut de joints matériels ou à distance ou de contenu, la répartition de tel ou tel fragment reste encore aléatoire et discutable. Il est tout à fait possible que l’attribution de plusieurs petits fragments à des copies de cette main soit à revoir pour retrouver d’autres joints qui ont pu échapper aux éditeurs. L’attribution essaie de tenir compte aussi du support, de l’épaisseur du cuir, de sa couleur (?), de sa préparation, de réglage, etc. Cela ne diminue pas pour autant la valeur du classement effectué par l’équipe internationale dans les premières années, labeur dont on a du mal à imaginer l’ampleur. Cependant le classement est toujours à poursuivre et à peaufiner.

41  Voir Talmon, “Hebrew Fragments from Masada”, 24-25, 113-16, 119, 148-49, position résumée p. 116, où on a retrouvé, outre des copies de livres bibliques, des restes d’une Admonition du Déluge (ex Apocryphe de la Genèse), d’un Apocryphe de Josué, une copie des Širôt ‛olat ha-šabbat, d’un Pseudo-Jubilés ( ? ), une copie de Ben Sira, et des fragments de compositions non encore identifiées, restes caractéristiques de rouleaux mis au jour à Qumrân, dont des compositions esséniennes.

Puech

Il y a quelques années, A. Yardeni a proposé d’attribuer à cette main très caractéristique quelque cinquante-sept manuscrits et même à lui en rattacher trente-sept autres, ce qui représenterait alors environ un dixième du total des manuscrits retrouvés au vingtième siècle. Ce sont principalement des copies sur cuir, mais aussi sur papyrus. Elles concernent des livres “bibliques” (Pentateuque, Prophètes et Écrits), des compositions apocryphes ou péri-testamentaires, mais aussi d’autres typiquement esséniennes (MMT, Db.d.f.g.h, Hymnes, Prières, Guerre, Calendriers, etc.) sans distinction, mais tous sont des documents r­ eligieux42. Les copies découvertes au vingtième siècle sont réparties dans la plupart des grottes : 1, 2, 3, 4, 6, (8) et 11, ce qui rend totalement futile et hors propos la distinction parfois proposée de plusieurs dépôts successifs dans les cachettes ; les rouleaux déposés en hâte, avant la destruction de Khirbet Qumrân en 6843, appartiennent tous assurément à une même collection et à la même population, qui est un important groupe essénien44. Ils ne sont certainement 42  Voir Ada Yardeni, “A Note on a Qumran Scribe”, in New Seals and Inscriptions, Hebrew, Idumean and Cuneiform (HBM 8 ; ed. M. Lubetski ; Sheffield : Phoenix Press, 2007), 287-98 (289-291 listes), même si on peut disputer telle ou telle analyse paléographique (p. 294-95), la répartition est assurément fort instructive. 43  Voir la conviction arrêtée du fouilleur, de Vaux, “Esséniens ou Zélotes ?” 235. 44  Voir par exemple les statistiques de Daniel Stökl Ben Ezra, “Old Caves and Young Caves : A Statistical Reevaluation of a Qumran Consensus”, DSD 14 (2007) : 313-33 (323 et n. 37) : il n’y a aucune raison de dater 6Q30 après la destruction de 68 AD, tous les tracés sont bien connus au premier siècle, d’une part. D’autre part, c’est faire preuve de beaucoup d’imagination ou de fausse ingéniosité que de prétendre avoir prouvé que les dépôts dans les grottes 1 et 4 proviennent de bibliothèques différentes et d’une population différente à une époque plus ancienne (vers la fin du 1er siècle avant J.-C., entre 8 et 4 av. J.-C. au plus tard) que les dépôts plus récents des bibliothèques dans les autres grottes en 68. Cela ne résiste pas à une analyse paléographique, comme il a été montré ci-dessus. La grotte 1 a été visitée plus tard, puisque le fouilleur y a retrouvé de la céramique postérieure à l’occupation essénienne (contra p. 324). En outre, cette hypothèse assez fantaisiste que certains sont prêts à retenir comme prouvée, contredit toutes les observations du fouilleur lui-même qui s’est prononcé, on ne peut plus clairement à ce sujet (voir note 1). Nos analyses, partagées par bien d’autres paléographes, vont dans le sens des conclusions du fouilleur. Les indications de Webster, “Chronological Index”, doivent être soigneusement vérifiées, car elles ne correspondent pas toujours aux indications des éditeurs. Par ailleurs, Roland de Vaux, “Archéologie”, in Qumrân Grotte 4.II : I. Archéologie, par R. de Vaux ; II. Tefillin, Mezuzot et Targums (4Q128-4Q157) par J.T. Milik (DJD 6 ; Oxford : Clarendon Press, 1977), 9-14, a suffisamment montré que la grotte 4a n’était pas une bibliothèque avec étagères, puisque un très grand nombre de fragments ont été retrouvés au bas des escaliers à

105

La Paléographie Des Manuscrits De La Mer Morte

pas déposés non plus dans des genizôt au cours de l’occupation du site. Leur dépôt répondait à une autre finalité historique, et les nombreux manuscrits de la grotte 4 n’étaient ni enveloppés de lin ni stockés dans des jarres, tout comme les grottes de la terrasse marneuse n’ont pas été creusées pour servir de caches de manuscrits ou des genizôt, comme l’a montré le fouilleur en répondant à ce type d’hypothèse formulée lors des débuts de la fouille des grottes et du site. l’entrée dans la chambre 1 et d’autres dans les plus petits réduits en contre-bas (chambres 3 et 4, avec des fragments de bois dans cette dernière), mais rien dans la grande chambre 2 et sa paroi sud aux trous supposés avoir servi à fixer des étagères, et presque rien dans la grotte 4b avec des lits de palmes. La grotte 4 (en fait deux grottes distinctes, 4a et 4b, avec des entrées indépendantes qui ne communiquaient pas entre elles à l’origine, écrit-il, p. 13) n’a pu servir de bibliothèque aux périodes 1b-II ni à la fin de la période 1b de dépôt non perturbé par des occupants de la période II. En outre, les deux entrées sont celles des deux grottes indépendantes et non deux entrées d’une grotte unique comme issues de secours pour fuir en cas d’attaque, en effet les accès débouchent tous deux à une quinzaine de mètres l’un de l’autre sur une surface parfaitement dénudée et sans échappatoire possible ni à droite ni à gauche devant un ennemi, comme le comprend J.-B. Humbert lors de la discussion au colloque, ce qui va d’ailleurs contre le moindre bon sens. La grotte n’a pas davantage servi de tombe à l’époque du Fer ni ensuite de genizah ou de dépôt de manuscrits en attente de la fin des jours, comme l’ont affirmé plusieurs auteurs, mais, dans sa dernière “utilisation”, uniquement de dépôt-cachette en vrac et en toute hâte et sans grand soin dans les petits réduits et au bas de l’escalier devant le danger imminent qu’était l’arrivée des Romains en 68, sans autre preuve de crises au premier siècle av. J.-C., même si par ailleurs un incendie a détruit le site. Comment expliquer dans le cas de deux dépôts distincts au bas des escaliers la présence de manuscrits plus récents parmi les plus anciens dans cette chambre 1 ? Ils ne peuvent provenir que d’un seul dépôt. Voir aussi Roland de Vaux, “Conclusions et hypothèses”, in Les ‘petites grottes’ de Qumrân : Explorations de la falaise ; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q ; Le rouleau de cuivre (DJD 3 ; par M. Baillet, J. T. Milik, et R. de Vaux, O.P. ; Oxford : Clarendon Press, 1962), 32-36, “Il n’y a aucune grotte dont on puisse affirmer qu’elle n’a été occupée que pendant la période Ib” (p. 32). Ces grottes ne sont pas non plus des grottes refuges pour des fuyards rescapés comme celles des Ouadîs Murabbaʿât et Khabra lors de la deuxième révolte, mais elles sont en relation directe avec le site de Sokoka-Qumrân. Dans ce dernier cas, les occupants ont fui leur lieu d’habitation et ne se sont pas réfugiés dans les grottes où ils auraient laissé une partie de leurs biens, les grottes n’étant point comparables en cela aux dépôts des grottes de Murabbaʿât ou du Ouadî Khabra, comme le suggère Mladen Popović, “When and Why Were the Caves Near Qumran and in the Judaean Desert Used ?” (dans ce volume). L’hypothèse de dépôts anciens et plus récents dans deux types de grottes n’apporte rien à l’histoire de l’occupation du site, bien au contraire.

5 Conclusion En conclusion, la familiarité des diverses mains des scribes dans toutes leurs particularités que donne la fréquentation attentive, régulière et assidue des fragments, s’avère indispensable pour leurs identifications et leurs déchiffrements corrects et, partant, pour leurs datations. L’analyse paléographique est le plus souvent le seul moyen d’obtenir, par comparaison, une datation relative assez exacte malgré tout sur une génération, quand l’étude est bien menée, le terminus ante quem étant la ruine de Khirbet Qumrân en juin 68, terminus confirmé par le siège de Jérusalem en 70 suivi de la ruine de Masada en 73 ; et il est peu probable que des manuscrits aient été copiés à Masada dans les conditions de la première révolte et du siège de la forteresse, ils ont dû y être apportés d’ailleurs quelques années avant le siège. Parfois, on l’a noté, des renvois à des événements ou à des personnages connus fournissent un terminus post quem qui permet de caler, de tester et d’ajuster au besoin la justesse de l’analyse paléographique, puisque les manuscrits dans les jarres ou en vrac étaient sans contexte stratigraphique d’appoint. Mais les manuscrits esséniens s’étalent tout au long des périodes hasmonéenne et hérodienne45. Enfin, l’analyse du C14 du support d’écriture offre un contrôle souvent assez lâche mais précieux tout de même pour ceux que les conclusions paléographiques laisseraient dubitatifs, en attendant la datation par l’analyse de l’encre elle-même, qui serait un argument bien plus probant, car il peut y avoir un décalage entre la datation du support et son écriture (voir par exemple 4Q542-Testament de Qahat)46. Quoi qu’il en soit, l’étude paléographique d’un manuscrit se révèle d’une grande utilité, non seulement pour en proposer une datation mais aussi et d’abord pour son déchiffrement correct et son insertion dans une histoire.

45  L’étude des tissus de lin blancs et de bleu indigo renforce l’identification essénienne des occupants du site et des dépôts des manuscrits dans les grottes, voir Orit Shamir et Naama Sukenik, “Qumran Textiles and the Garments of Qumran’s Inhabitants”, DSD 18 (2011) : 206-25. 46  Voir Emile Puech, Qumrân Grotte 4.XXII : Textes araméens ; Première partie 4Q529-549 (DJD 31 ; Oxford : Clarendon Press, 2001), 257-88, daté circa 125-100, et la datation du C14, voir Bonani et al., “Radiocarbon Dating of the Dead Sea Scrolls”, 30, avec environ deux siècles d’écart ! Mais 4Q542 ne peut aucunement être une écriture du début du quatrième siècle av. J.-C., il y a là une erreur quelque part ou une réutilisation d’un vieux cuir, en effet, la peau n’a pas d’autre point de comparaison parmi les autres manuscrits que j’ai pu étudier ou manipuler.

Part 3 Other Finds

∵

CHAPTER 7

Terracotta Oil Lamps (Roland de Vaux’s Excavations of the Caves) Jolanta Młynarczyk Examples of terracotta oil lamps, whole or fragmentary, dated to the Graeco-Roman period (specifically, to the later Hasmonean and Roman imperial times down to the reign of Hadrian) were found in nine of the Qumran Caves, examined by Father de Vaux’s expedition (1949– 1956). Of these, Cave 1Q yielded three lamps (two of them almost entire, and two fragments apparently pertaining to one lamp),1 Cave 4Q—one lamp,2 Caves 7Q,3 8Q4 and 10Q5—fragments of one lamp each, Cave 8 (the same as Cave 3Q)6 and Cave 97—one lamp each, Cave 17 (just south of 1Q)—fragment of one lamp,8 Cave 29—one

1   Cave 1Q: Gerald L. Harding, “Introductory, the Discovery, the Excavation, Minor Finds” in Qumran Cave 1 (DJD 1; ed. D. Barthélemy and J. T. Milik; Oxford: Clarendon Press, 1955), 6–7; Roland de Vaux, “La poterie,” in DJD 1:11, Fig. 3:1, 4, 5; Roland de Vaux, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973), 49 (four lamps); Jean-Baptiste Humbert and Alain Chambon, eds., The Excavations of Khirbet Qumran and Ain Feshkha: Synthesis of Roland de Vaux’s Field Notes (NTOA.SA 1B; trans. and rev. Stephen J. Pfann; Fribourg/Göttingen: Éditions Universitaires/ Vandenhoeck & Ruprecht, 2003), 66, give the number of three lamps, cf. Yizhar Hirschfeld, Qumran in Context: Reassessing the Archaeological Evidence (Peabody, MA: Hendrickson, 2004), 37 (“three wheelmade lamps”); see also below, nn. 21, 40, 43–44. 2   Cave 4Q: Humbert and Chambon, Excavations of Khirbet Qumran, 66. 3  Cave 7Q: Lamp fragment not mentioned in the description of Cave 7Q, see Roland de Vaux, “Archéologie,” in Les ‘petites grottes’ de Qumrân: Explorations de la falaise; Les grottes 2Q, 3Q, 5Q, QQ, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 27, 30, yet registered as found there on 17.02.1955. 4  Cave 8Q: Roland de Vaux, “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e et 5e campagnes,” RB 63 (1956): 572–73; idem, in DJD 3: Fig. 6:3 and Pl. VIII; cf. Humbert and Chambon, Excavations of Khirbet Qumran, 67. 5  Cave 10Q: de Vaux, in DJD 3: Fig. 6:4 and Pl. VIII; idem, Archaeology and the Dead Sea Scrolls, 53. 6  G Q 8: de Vaux, in DJD 3:8, 15, Fig. 5:6 and Pl. VII; idem, Archaeology and the Dead Sea Scrolls, 51. 7  G Q 9: de Vaux, in DJD 3:8, Fig. 5:5 and Pl. VII. 8  G Q 17: de Vaux, in DJD 3:9, 15, mentioned as a lamp fragment found under collapsed ceiling of the cave; unfortunately, not illustrated and therefore not identifiable.

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_009

lamp.9 In other words, lamp finds from the caves, be they documented or just mentioned in the relevant publications, amount to 11 objects. However, four of them are considered as “lost” (that is, they could not have been identified in the actual material from Qumran), among them, a lamp fragment from Cave 17 that was never (as far as I know) registered or illustrated.10 After the activity of the de Vaux’ team (1949–1956), the search for caves and their examination was continued within the framework of several projects11 that recorded, among others, also the oil lamps. Even if the latter finds are not the main topic of the present paper, two cases are particularly noteworthy. One of them is the discovery of as many as four lamps (representing two different formal types) in a cave situated near Cave 3Q and probably identical with Cave 2 of the 1952 survey; they were picked up on two separate occasions by the Israeli researchers. The cave in question, marked as Cave (P)13 by J. Patrich, initially yielded one lamp,12 and a fragment of another lamp was found during a later research.13 A few years later another survey of the same cave, re-marked then as

9  GQ 29: de Vaux, in DJD 3:11, 15, Fig. 5:2 and Pl. VII. 10  On the other hand, the body of lamp material studied by the present author at the Ecole biblique et archéologique française in Jerusalem, embracing all the lamps discovered during the de Vaux excavations (jointly the Qumran settlement, the caves and Ain Feshkha) include six unregistered lamp fragments without any indication of provenance. It is possible then that one of those “unprovenanced” fragments comes from Cave 17. 11  For the history of the research on the caves, see recent papers by Joan E. Taylor and Jürgen Zangenberg respectively, in this volume. 12   Joseph Patrich and Benny Arubas, “A Juglet Containing Balsam Oil (?) from a Cave near Qumran,” IEJ 39 (1989): 43, Fig. 3:12, Pl. 5C; Joseph Patrich, “Khirbet Qumran in Light of New Archaeological Explorations in the Qumran Caves,” in Methods of Investigation of the Dead Sea Scrolls and the Khirbet Qumran Site: Present Realities and Future Prospects (ANYAS 722; ed. M. O. Wise et al.; New York: New York Academy of Sciences, 1994), Fig. 12; cf. Stephen J. Pfann, “Reassessing the Judean Desert Caves: Libraries, Archives, Genizas, and Hiding Places,” BAIAS 25 (2007): Fig. 14:1. 13  Patrich, “Khirbet Qumran,” 93 (unfortunately, not illustrated).

110

Młynarczyk

Cave X/35, yielded three more lamps.14 The second interesting case is that of Cave (P)24 (just north of Cave 11Q) in which a discus-type lamp was discovered.15 The “cave lamps” pertain to the types well-attested at the habitation phases (“Periods” of R. de Vaux,16 “Levels/Phases” of Jean-Baptiste Humbert17) of the settlement, and recently discussed by the present author.18 One exception is a complete lamp found at the entrance to Cave 4Q: an object whose date clearly corresponds with Humbert’s Level IV at Qumran and which appears to be connected with the Roman (?) presence well after CE 68, perhaps as late as the Second Revolt (ca. CE 132–135); only one small fragment of the same type was discovered in the khirbeh itself.19 The lamp typology in question has been based on the body of 197 registered objects (many of which entered the Qumran register only in the nineties of the 20th century), uncovered by father de Vaux’ team jointly in the Qumran settlement, the caves and Ain Feshkha site.20 It’s logical then to consider the small assemblage of the “cave lamps” within the context of the typological scheme created for this specific region of the north-west part of the Dead Sea, clearly tied up to the area of both Jericho and Jerusalem. The most interesting case, perhaps, is that of three lamps found in Cave 1Q, all of them wheel-made, but representing two quite different types. Two lamps (1Q-43 and

1Q-44, see Fig. 7.1)21 are of Qumran Type 033, pertaining to what we may call the “Qumran lamp family.” The latter is a cluster of five wheel-made types (Qumran 032 through 035-Prime, see Fig. 7.2), sharing a number of visual characteristics and hardly attested outside Qumran.22 The “Qumran family” of lamps has been fairly accurately characterized by Smith: “Although differing widely in form, these lamps have a similarity of ware—gray to buff clay, usually wet-smoothed or having a slip similar to the body color—and a similarity of style and technique of manufacture. There can be no doubt that these lamps are the products of the potter’s shop at Qumran.”23 After him, also R. Donceel, commenting on the lamps of this “Qumran family” found in Loc. 130 in the north-west part of the Qumran settlement (Fig. 7.3), has admitted that they may have been of local manufacture.24 The macroscopic description of the fabric by Smith seems to be very accurate. Additionally, one should point out one interesting detail of the manufacturing technique of the “Qumran family” lamps: the ridged nozzle of those lamps is very obviously a development of (and constitutes a firm link with) the pinched nozzle of the “folded-rim” (or “cornucopia”) lamps (Qumran Type 031) and, contrary to the opinion first expressed by de Vaux and followed by Magness and Bar Nathan, it should not be interpreted (despite some superficial similarity) as imitation of foreign Hellenistic (mostly Ephesian-type) lamps.25 The sug-

14  Michel Itah, Yoni Kam, and Ronny Ben Haim, “Region X: Survey and Excavations of Caves along the Fault Escarpment South of Almog Junction and West of Qalya,” ʿAtiqot 41 (2002): part 1, Fig. 11:2–3 and part 2, 172. 15  Patrich, “Khirbet Qumran,” 90, Fig. 10 (the cave regarded as having been inhabited); cf. Pfann, “Reassessing the Judean Desert Caves,” Fig. 14:5. 16  Cf. Roland de Vaux, “Fouilles au Khirbet Qumrân: Rapport préliminaire sur la deuxième campagne,” RB 61 (1954): 234; idem, “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e, et 5e campagnes,” 569. 17   Jean-Baptiste Humbert, “The Chronology during the First Century B.C., de Vaux and his Method: a Debate,” in Khirbet Qumran et ‘Aïn Feshkha. Vol. 2: Études d’anthropologie, de physique et de chimie (NTOA.SA 3; ed. J.-B. Humbert and J. Gunneweg; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 425–44. 18  Jolanta Młynarczyk, “Terracotta Oil Lamps from Qumran: The Typology,” RB 120 (2013): 99–133. 19  Specifically, KhQ 3136, cf. Młynarczyk, “Terracotta Oil Lamps,” 133, Fig. 5 (Type 047.1–2). 20  Of these, 25 objects have been “lost” and the only information about them is contained in the original register books of the expedition.

21  De Vaux, in DJD 1:Fig. 3:4 and 5 respectively. 22  Młynarczyk, “Terracotta Oil Lamps,” 103–6. Cf. Jodi Magness, “The Community at Qumran in Light of Its Pottery,” in Wise et al., Methods of Investigation of the Dead Sea Scrolls, 41. 23   Robert H. Smith, “The Household Lamps of Palestine in Intertestamental Times,” BA 27 (1964): 124. 24  Robert Donceel, “Poursuite des travaux de publication du matériel archéologique de Khirbet Qumrân: Les lampes en terrecuite,” in Mogilany 1995: Papers on the Dead Sea Scrolls Offered in Memory of Aleksy Klawek (Qumranica Mogilanensia 15; ed. Z. J. Kapera; Crakow: Enigma Press, 1998), 99–104 and n. 35, Figs. 9–12. 25   Magness, “The Community at Qumran,” 41, 43; Rachel Bar Nathan, Hasmonean and Herodian Palaces at Jericho: Final Reports of the 1973–1987 Excavations; Volume III; The Pottery (Jerusalem: Israel Exploration Society, 2002), 110–12. Actually, Smith (“Lamps in Intertestamental Times,” 124) was the first to have spoken about “apparent traces of the influence of the Ephesian lamp” while, on the other hand, he recognized “some connection with wheelmade closed lamps of the 2nd century and the cornucopia lamp” in terms of the manufacturing technique. In the opinion of the present author, Qumran Types 035 and 035-Prime (Młynarczyk, “Terracotta Oil Lamps,” 130, Fig. 2: KhQ 1257 and KhQ 2308 respectively) are the only ones with indisputable influence of Ephesian-type lamps.

Terracotta Oil Lamps

gestion by Humbert that these lamps “could have come from Egypt or were imitations of Egyptian models” has to be rejected as well.26 In terms of the body profile (but not the nozzle type) they may be related to some Hellenistic lamps from Beth Zur and Tell en-Nasbeh.27 Indeed, the sheer statistics are in favour of Qumran as the manufacturing place of these lamps. The excavations by de Vaux’ team brought to light 37 objects of this group coming from the khirbeh and two objects from Cave 1Q; two more lamps were published by Magen and Peleg as coming from “the northwestern refuse dump.”28 This makes 41 objects discovered in the Qumran area (including two lamps from Cave 1Q) versus just four fragmentary objects from Jericho29 and a single lamp fragment (a nozzle) from Masada.30 The fragments from Jericho have been classified by R. Bar Nathan as her type J-LP3, which in fact embraces as many as four Qumran types (032–035). Another example of this “family” (specifically, a lamp of Qumran Type 032) was found at the site of Umm Hadar in al-Kufrayn area, on the way from Jordan River to Philadelphia—Amman.31 Finally, an unprovenanced lamp from a private collection, corresponding to Qumran Type 032, has been published by Smith; to judge by the ware’s description, it comes from a non-Qumran (Idumaean?) source.32

26  Humbert, “The Chronology,” 435. 27  Smith, “Lamps in Intertestamental Times,” 106, Fig. 5. 28  Yizhak Magen and Yuval Peleg, “Back to Qumran: Ten Years of Excavation and Research, 1993–2004,” in Qumran, the Site of the Dead Sea Scrolls: Archaeological Interpretations and Debates; Proceedings of a Conference Held at Brown University, November 17–19, 2002 (STDJ 57; ed. K. Galor, J.-B. Humbert, and J. Zangenberg; Leiden: Brill, 2006), 69, Fig. 3.14. To judge by the pictures, the lamps represent Qumran Types 033.2 and 034.2. 29  Bar Nathan, Hasmonean and Herodian Palaces, 110–12, Ill. 87–8 and Pl. 18. 30  Dan Barag and Malka Hershkovitz, “Lamps from Masada,” in Masada 4: The Yigael Yadin Excavations 1963–1965; Final Reports (ed. J. Aviram, G. Foerster, and E. Netzer; Jerusalem: Israel Exploration Society/Hebrew University, 1994), 71–72, n. 124, Fig. 21. 31  Fawzi Zayadine, “Hellenistic Pottery from the Estate of the Tobiads,” in Lampes antiques du Bilad es Sham: Jordanie, Syrie, Liban, Palestine; Ancient lamps of Bilad es Sham; Actes du Colloque de Pétra-Amman, 6–13 novembre 2005 (ed. D. Frangié and J.-F. Salles; Paris: De Boccard, 2011), 173, 180, Pl. 4, dated to between 50 and 31 BC on the basis of Paul W. Lapp, Palestinian Ceramic Chronology 200 B.C.–A.D. 70 (New Haven: American Schools of Oriental Research, 1961), type 84:B. 32  Smith, “Lamps in Intertestamental Times,” 119, Fig. 18 (buff orange ware, originally covered “with a rich red slip”).

111 In the light of some archaeometric research, however, the Qumran provenance of these lamps is not quite clear. Petrographic study of pottery samples (including lamps) taken from both Qumran and Jericho, carried out by J. Michniewicz, seems to demonstrate that two lamps found in the same Loc. 130 of the khirbeh, classified as one and the same variant of the “Qumran lamp family” (specifically, Qumran Type 034), represent, in spite of the macroscopic similarity of the ware, two separate petrographic groups.33 This apparently means that they are products of two different workshops, which employed different technologies and clays. Also, the sample of lamp Gr 1Q-43 examined by Gunneweg and Balla was described as being similar to the pottery of their Chemical Group III which was “local to Jericho.”34 On the other hand, one should remember that in the case of manufacturing of small ceramic objects such as lamps, which does not require large quantities of raw material, the needed amount of clay might have easily been brought from elsewhere. The present author is of the opinion that, although the so-called “Qumran pottery” does belong to a broader network of regional pottery, the “Qumran lamps family” is specific to Qumran only. And, finally, there comes an important question about the chronology of the “Qumran lamp family.” Bar Nathan supposes that her type J-LP3 (corresponding to Qumran Types 032–035) “may have emerged already in the midfirst century BCE,” but at the same time she questions the attribution of comparable Qumran lamps to the Hasmonean period only, since in Jericho they occur in the later Herodian-period contexts.35 Yet is seems very plausible that the few Jericho fragments were simply residual, the more so that actually all the types pertaining to the “Qumran family” (Qumran Types 032, 033, 034, 035-Prime) make their appearance in the same context of khirbeh. Specifically, this is Loc. 130 in the north-western part of the Qumran settlement (see Fig. 7.3), which yielded nine items of the group (seven of them registered by de Vaux’ team and two more added later).36 According to R. Donceel, the pottery of Loc. 130 does not post-date Period Ib of de Vaux, who attributed the assemblage to the reign of Alexander 33  Jacek Michniewicz, Qumran and Jericho Pottery: A Petrographic and Chemical Provenance Study (Seria Geologia 20; Poznan: Wydawnictwo Naukowe UAM, 2009), 37 (Petrographic Groups 1 and 3). 34  Jan Gunneweg and Marta Balla, “The Provenance of the Pottery,” in Humbert and Gunneweg, Khirbet Qumran et ‘Aïn Feshkha, 23 and table 9, sample Qum 286. 35  Bar Nathan, Hasmonean and Herodian Palaces, 110–12. 36  Four of them illustrated in de Vaux, “Fouilles de Khirbet Qumrân,” 553, Fig. 1:1–4; see also Donceel, “Poursuite des travaux,” 99–104, Fig. 9–12.

112

Młynarczyk

Iannai (103–76 BCE) as based on the numismatic evidence.37 Humbert, while stating that “the ­pottery of Loc. 130 ( . . .) constitute the oldest ceramic deposits found at the site,” suggests its dating to ca. 50 BCE.38 Such dating of the “Qumran lamp family”, associated with Loc. 130, seems to be safer than the later one (“the post-31 BCE phase of Period Ib [that is, during Herod’s reign]”), as proposed by Magness who, on the other hand, is perfectly right in describing them as “a regional type” with “a very limited distribution.”39 1

Description of the Lamps of the “Qumran Family” Found in Cave 1Q Gr 1Q-43 (Qumran Type 033.1, see Figs. 7.1–7.2).40 Lamp lacking loop handle, once attached to carination. Dense burning traces at wick hole. Turning marks on shoulder interrupted by the nozzle ridge. Pseudo ring base is created by a double groove, deep and irregular. Root of the nozzle ridge at the wick hole is flanked by two angular projections, a feature doubtlessly borrowed from bronze lamps;41 it occurs also in a fragmentary lamp from Masada.42 Fabric: dense, fired dark grey, with tiny white grits; surface: beige-grey to very pale brown. Gr 1Q-44 (Qumran Type 033.2, see Fig. 7.1).43 Lamp restored in plaster, including lug handle. Rounded

37  Donceel, “Poursuite des travaux,” 99. 38  Humbert, “The Chronology,” 435–36. 39  Jodi Magness, The Archaeology of Qumran and the Dead See Scrolls (Grand Rapids, MI: Eerdmans, 2002), 79. 40  Cf. de Vaux, in DJD 1:11, 16, Fig. 3:5 and Pl. III:b; Smith, “Lamps in Intertestamental Times,” 123, Fig. 23 (“ca. second and third quarters of 1st century BC”); Donceel, “Poursuite des travaux,” 103, n. 36, Fig. 11; Magness, Archaeology of Qumran, Fig. 21:3; Humbert and Chambon, Excavations of Khirbet Qumran, 66; Rachel Bar Nathan, “Qumran and the Hasmonean and Herodian Winter Palaces of Jericho: The Implication of the Pottery Finds on the Interpretation of the Settlement at Qumran,” in Galor, Humbert, and Zangenberg, Qumran, the Site of the Dead Sea Scrolls, Fig. 15.3:6. 41  See lamp from Tell Sandahannah (Maresha), in Smith, “Lamps in Intertestamental Times,” 117, Fig. 15, “possibly of 2nd century BC” and p. 124, where he suggests that the Qumran lamps may have had “metallic prototypes.” 42  See n. 30 above. 43  Cf. de Vaux, in DJD 1:11, 16, Fig. 3:4 and Pl. III:1 (before its reconstruction in plaster); Donceel, “Poursuite des travaux,” 103, n. 36, Fig. 12; Magness, Archaeology of Qumran, Fig. 21:4; Humbert and Chambon, Excavations of Khirbet Qumran, 66; Bar Nathan, “Qumran,” Fig. 15.3:5.

base, thick out-folded rim to filling hole; wheel-marks on shoulder, soot on rim of wick hole. Fabric: break not available (terre rose, according to Barthélemy and Milik); surface: beige with pink spots with some white grits’ eruptions, semi-rough in feel. In contrast to the mid-1st century BCE dating of the two above-discussed lamps of Qumran Type 033 (Gr 1Q-43 and 44), the third, fragmentary lamp found in Cave 1Q in two non-joining parts (Gr 1Q-56, see Fig. 7.4),44 today unfortunately lost, could be as late as de Vaux’ Period II (ca. 4 BCE/CE 6–68), corresponding to Humbert Level III B–C at the Qumran settlement. Unlike the two representatives of the “Qumran lamp family” with its very limited distribution, this third lamp pertains to a large, diversified family of the wheel-made “Herodian” lamps, also known as “­ knife-pared” or “spatulate,” enormously common throughout Judea. The “Herodian” lamps and their fragments account for 64,5% of all the Qumran lamps (the Caves and the Ain Feshkha site included).45 They constitute Qumran Type 036 which can be subdivided into as many as seven formal groups (036.1 through 036.7).46 Six “Herodian” lamps found in the caves by the de Vaux’ team represent five different groups of the “Herodian” family. The lost “Herodian” lamp from Cave 1Q (Gr 1Q-56, see Fig. 7.4) alongside another lost lamp, GQ 29–1 (Fig. 7.5), seem to pertain to Qumran Type 036.3. Other cave lamps of the same “Herodian family” are lamp nozzle Gr 7Q-10 (Qumran Type 036.1), lamp GQ 8–12 (Type 036.2, see Fig. 7.6), lamp GQ 9–1 (Type 036.4, see Fig. 7.7) and fragmentary lamp Gr 10Q-3 (Type 036.7, see Figs. 7.8–7.9). To these, one should add two lamps from Cave (P)13 (= Cave X/35), pertaining to Qumran Types 036.2 and 036.4/5 respectively, as well as a small fragment 44  De Vaux, in DJD 1: Fig. 3:1 (“deux fragments d’une lampe”); de Vaux, Archaeology and the Dead Sea Scrolls, 49, 102, counts these fragments as two separate lamps (“two lamps of the Roman period”); Jean-Baptiste Humbert and Alain Chambon, eds., Fouilles de Khirbet Qumrân et de Aïn Feshkha. Vol. 1: Album de photographies; Répertoire du fonds photographiques; Synthèse des notes de chantier du Père Roland de Vaux (NTOA.SA 1; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 1994), 343: Q 56 et 57: “Deux fragments de lampes ‘hérodiennes’,” Humbert and Chambon, Excavations of Khirbet Qumran, 66: “Two fragments of a Herodian lamp.” See also Magness, Archaeology of Qumran, Fig. 21:1. 45  Donceel, “Poursuite des travaux,” 93, has evaluated their proportion as amounting to only 50%. 46  Młynarczyk, “Terracotta Oil Lamps,” 108–16, versus Donceel, “Poursuite des travaux,” 93–94, who speaks about “10 groupes et sous-groupes des lampes ‘hérodiennes’ nettement distincts.”

Terracotta Oil Lamps

of lamp top of undetermined “Herodian” (?) type from Cave C (north of the khirbeh).47 In R. Donceel’s opinion, the “Herodian family” lamps found in the Qumran area come, in fact from a number of workshops the majority of which were located outside Qumran.48 Indeed, there can be no doubt that such lamps were being made in several Judaean ateliers. Actually, R. H. Smith has been the first to suggest that the “Herodian” lamps “may have evolved in Jerusalem or nearby.”49 Analytical research on such lamps found at some sites in the north of Israel has proved that most of them did come from the Jerusalem area.50 It seems, however, very logical that at least some of the lamps of Qumran Type 036 were made in Qumran itself, be it from imported or from locally acquired raw material.51 As to the important question of dating, while there is a general consensus that the production of the “Herodian” lamps covered the period between the end of the 1st century BCE and the first half of the 2nd century CE, some details of their chronology are still far from being clear. The present author finds it difficult to accept the view of R. Donceel that les répresentants de différentes familles de lampes hérodiennes étaient en usage à Khirbet Qumran de manière strictement contemporaine,52 in other words, that there was no chronological development that would be reflected by nuances of the shape. Already Smith, who was the first to provide an in-depth study of the “Herodian” lamps, divided them into two basic types, one succeeding the other: Type 1 in exclusive use between ca. 37 BCE and CE 35; a period of transition during which Type 1 developed into Type 2 ca. CE 35–50, and Type 2 in exclusive 47  See n. 12 and 14 above and, for Cave C, see Magen Broshi and Hanan Eshel, “Residential Caves at Qumran,” DSD 6 (1999): 330, Fig. 5:4 and Table 1. 48  Donceel, “Poursuite des travaux”. 49  Smith, “Lamps in Intertestamental Times,” 4. 50  David Adan-Bayewitz et al., “Preferential Distribution of Lamps from the Jerusalem Area in the Late Second Temple Period (Late First Century B.C.E.–70 C.E.),” BASOR 350 (2008): 41–42 (Figs. 2–3) and 77. 51   Gunneweg and Balla (“The Provenance,” 16), mention NAA analysis of three “Herodian” lamps of which one “was locally made in Jerusalem” (sample Qum 294); Michniewicz (Qumran and Jericho Pottery, 37–60 and table 4) having analyzed four “Herodian family” samples from the khirbeh, recognized two different petrographic groups of the material. Unfortunately, the information about results of some other petrographic readings, mentioned by Donceel (“Poursuite des travaux,” 94 and nn. 18–20), does not refer to the actual register numbers of the lamps so it can not be confirmed. 52  Donceel, “Poursuite des travaux,” 93–94 and n. 17.

113 use between ca. CE 50 and 135.53 This sequence of formal types has been confirmed by the archaeological contexts both at Machaerus (destruction layer of CE 72) and at the Lower Herodion (upper floor of the service building dated to CE 48–70) which contained lamps of Smith Type 2 without any examples of Smith Type 1.54 In terms of the terminal date of the “Herodian” lamps, there is no doubt that they are still present in the contexts of the Bar Kokhba Revolt.55 However, the initial date of their production remains disputable. Smith himself, in his slightly later overview of the Palestinian lamps, shifted the beginning of his Type 1 of the “Herodian lamps” into the early decades of the 1st century CE.56 Indeed, the evidence from the Jewish Quarter, Jerusalem, and from Jericho “demonstrates that it first appeared at the very end of the 1st century BCE or at the very beginning of the 1st century CE.”57 It is to be noted, however, that in his publication of the material from Machaerus, Loffreda 53  Robert H. Smith, “The ‘Herodian’ Lamp of Palestine: Types and Dates,” Berytus 14 (1961): 65; see also idem, “The Household Lamps of Palestine in New Testament Times,” BA 29 (1966): 5. For the discussion of the “Herodian” (“knife-pared”) lamps, see also (in chronological order): Barag and Hershkovitz, “Lamps from Masada” (their types C I–IX); Bar Nathan, Hasmonean and Herodian Palaces, 112–13 and 188–89; Adan-Bayewitz et al., “Preferential Distribution,” 38–40; Varda Sussman, “The Oil Lamps,” in Archaeological Excavations at Caesarea Maritima: Areas CC, KK and NN; Final Report; Volume I; The Objects (ed. J. Patrich; Jerusalem: Eisenbrauns, 2008), 219–20; Jodi Magness, “The Pottery from the 1995 Excavations in Camp F at Masada,” BASOR 353 (2009): 80–81. 54  Stanislao Loffreda, La ceramica di Macheronte e dell’Herodion (90 a.C.–135 d.C.) (SBF.CMa 39; Jerusalem: Franciscan Printing Press, 1996), 111–12; Rachel Bar Nathan, “Pottery and Stone Vessels of the Herodian Period (1st century B.C.–1st century A.D.),” in Greater Herodium (Qedem 13; ed. E. Netzer; Jerusalem: Hebrew University, 1981), 65 (type 2b) and 138, n. 88. 55  E.g. Murabbaʿat Cave 1: Roland de Vaux, “Archéologie,” in Les grottes de Murabba‘ât (DJD 2; ed. P. Benoit, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1961), 31–33, Fig. 8:12, lamp standing on disc base, untypical for the “Herodian” lamps; also Yigael Yadin, The Finds from the Bar Kokhba Period in the Cave of Letters (Jerusalem: Israel Exploration Society, 1963), 115, Fig. 42, Pl. 35 (lamp RC.4). 56  Smith, “Lamps in Intertestamental Times,” 4–5. 57   Hillel Geva and Malka Hershkovitz, “Local Pottery of the Hellenistic and Early Roman Periods,” in Jewish Quarter Excavations in the Old City of Jerusalem conducted by Nahman Avigad, 1969–1982. Vol. 3: Area E and Other Studies, Final Report (ed. H. Geva; Jerusalem: Israel Exploration Society, 2006), 115. For the recent résumé of the discussion on the chronology, see Magness, “Camp F at Masada,” 80–81 (late 1st century BC to the Bar Kokhba revolt).

114

Młynarczyk

a­ pparently accepts Smith’s initial dating of the introduction of Type 1 to the early years of Herod the Great.58 Unfortunately, the context datings of de Vaux’ Period II at Khirbet Qumran are not precise enough to shed light on either contemporaneity or any sequence of the “Herodian family” lamps, which means that chronological research should largely be based on comparative studies. Accordingly, we may perhaps forward the hypothesis that the “Herodian” lamps with ring handles, represented in the caves by Qumran Types 036.3 (Gr 1Q-56, GQ 29–1) and 036.7 (Gr 10Q-3), might belonged to an early phase of their production (1st century BCE?). Their ribbed strap handles added to the wheel-thrown bodies strongly resemble the handles of some Late Hellenistic lamps in the Eastern Mediterranean (Cnidus, Ephesus, even Egypt), while they differ from those of the Italian-style lamps, made in bipartite moulds together with the lamp body. In terms of the lamp finds from Cave 1Q, the presence of two distinct types of lamps of rather different dating (Qumran Type 033 of ca. 50 BCE and Type 036.3 of the 1st century CE, unless the latter indeed dates from the second half of the 1st century BCE) seems to reflect either a lengthy period of human activity in the cave, or exceptionally extended usage (up to several decades?) of the two lamps of Qumran Type 033. Either way, it is certain that Hirschfeld could not have been right in dating the whole pottery assemblage of Cave 1Q to the 1st century CE only.59 2

Description of the “Herodian Family” Lamps from the Caves Gr 1Q-56 (lost; apparently pertaining to Qumran Type 036.3, see Fig. 7.4); terre grise, fine et tres cuite (according to DJD 1), estimated L. ca. 11 cm. The ribbed strap handle resembles that of Qumran Type 036.7 (see 10Q-3 below); the nozzle has a double groove across the top.60 GQ 29–1 (lost; Qumran Type 036.3, see Fig. 7.5). Lamp almost complete; just tip of nozzle missing; nozzle top plain. Fabric: terre rouge. Extant L. 11.3, W. 7.3; H. body 2.8.61

58  Loffreda, La ceramica di Macheronte, 111–12. 59  Hirschfeld, Qumran in Context, 37. In the words of de Vaux (Archaeology and the Dead Sea Scrolls, 50): “examples from two distinct periods of the Khirbeh, Ib and II, are represented here.” 60  De Vaux, in DJD 1: Fig. 3:1. 61  Roland de Vaux, “Exploration de la région de Qumrân: Rapport préliminaire,” RB 60 (1953), Fig. 3:4 and Pl. XXII b, 1; de Vaux,

Wide flat shoulder and rather narrow inner flange around the filling hole. Ring handle, applied onto the shoulder, is adorned with two ribs divided by a deep groove. Gr 10Q-3 (Qumran Type 036.7, see Figs. 7.8–7.9). Several fragments of an upper part of a large-size, very thin-walled lamp, including parts of shoulder with handle, parts of central depression and of slightly splayed nozzle with traces of burning.62 L. 17.2, Diam. 12.2, extant H. (with handle) 5.5. Shoulder decorated with three to two (partly overlapping) rows of grooved zig-zags or chevrons made à la roulette. The strap handle is adorned with alternating ribs and grooves. Fabric: hard and dense, light red with some small white grits; surface of “metallic” feel, fired from glossy red to pale beige with many white grits’ eruptions. This lamp, apparently related to the “Herodian family” Qumran Type 036.1, may at the same time have been inspired by the Cnidian lamps of the 1st century BCE.63 A somehow comparable lamp from Beth Shean resembles our lamp by its size, form of the handle and its zig-zag decoration on the shoulder incised by hand; unlike our lamp, however, it has two nozzles adorned with punctured lines and stamped dotted circlets.64 Gr 7Q-10 (Qumran Type 036.1). Short nozzle with flat top; dense burning. Extant L. 3.2; W 3,0. Fabric: light red, with small to medium-size white grits; surface smooth, pale orange with some large white eruptions; dense traces of burning on nozzle. This small fragment seems to have pertained to Smith Type 1, Masada group C:1, which embraces in DJD 3:11, 15, n. 3, Fig. 5:2, Pls. V and VII (photos); Florentino Diez Fernandez, Cerámica común romana de la Galilea: Aproximaciones y diferencias con la cerámica del resto de Palestina y regiones circundantes (Jerusalem/Madrid: Escuela Biblica, 1983), no. 77, his type 3.4c; Humbert and Chambon, Fouilles de Khirbet Qumrân, 345; Humbert and Chambon, Excavations of Khirbet Qumran, 71; Pfann, “Reassessing the Judean Desert Caves,” Fig. 14:4; Młynarczyk, “Terracotta Oil Lamps,” Fig. 3. 62  Cf. de Vaux, in DJD 3: Fig. 6:4 and Pl. VIII:4; Humbert and Chambon, Fouilles de Khirbet Qumrân, 344; Humbert and Chambon, Excavations of Khirbet Qumran, 67; Pfann, “Reassessing the Judean Desert Caves,” Fig. 14:8; Młynarczyk, “Terracotta Oil Lamps,” Fig. 4. 63  Such as a large-size lamp in the British Museum, see Donald M. Bailey, Catalogue of the Lamps in the British Museum: 1. Greek, Hellenistic and Early Roman Pottery Lamps (London: British Museum Publications, 1975), 151, Q 325. 64  Smith, “Lamps in New Testament Times,” 14–15, Fig. 4.

Terracotta Oil Lamps

lamps with a rounded wall profile on slightly convex base, with a long inner flange around filling hole; the splayed nozzle is short, with pronouncedly concave sides and plain top. This shape variant has been considered by Smith as the early one, originally dated by him to the period of 37 BCE–35 CE, and then to (mainly) first third of the 1st ­century CE.65 GQ 8–12 (Qumran Type 036.2, see Fig. 7.6). Nozzle with a part of wall and lamp top; traces of burning at wick hole. Single incision across the nozzle top above the wick hole.66 Extant L. 4.9, extant W. 4.6; W. nozzle 2,9. Fabric: gritty but dense, dark reddish brown with partial dark grey core and many minute white grits; surface: beige to light grey (interpreted as couverte blanche by Baillet, Milik and de Vaux), rather smooth. GQ 9–1 (Qumran Type 036.4, see Fig. 7.7). Several fragments of thin-walled lamp preserving nearly full profile; nozzle top apparently plain(?). Surface severely mutilated; dense burning at nozzle.67 L. 7.9, Diam. (est.) 5.2, H. (est.) 2.6. Fabric: hard, light red with light grey core at the bottom and with abundant tiny white grits; surface: light reddish brown, with occasional white grits’ eruptions, smooth. As to the mould-made lamps, the caves under discussion yielded just two examples, representing two distinctly different types: fragmentary lamp 8Q-12 of Qumran Type 045 and lamp 4Q-22 of Qumran Type 047, the latter unfortunately lost. Gr 8Q-12 (Qumran Type 045.1, see Fig. 7.10). Fragments of lamp, missing large part of body and handle. L. (est.) 12.3, Diam.body 9.2, H. 3.3.68 Shallow 65  Smith, “The “Herodian” Lamp,” 60–61, 65; Smith, “Lamps in New Testament Times,” 4–5, Fig. 2. 66  Published as a complete lamp and described as characteristic of Qumran Period II: de Vaux, in DJD 3:15, Fig. 5:6 and Pl. VII; Humbert and Chambon, Excavations of Khirbet Qumran, 66; Pfann, “Reassessing the Judean Desert Caves,” Fig. 14:3. To exactly the same formal type (illustrated by Młynarczyk, “Terracotta Oil Lamps,” Fig. 3: KhQ 2168) there pertains a lamp from Cave (P)13, the latter regarded as a hideout of CE 68/70, cf. n. 12 above. 67  Cf. de Vaux, “Exploration,” Fig. 4:5; de Vaux, in DJD 3:8 (Cave GQ 1 identified as crevasse du rocher, not inhabited), Fig. 5:5 and Pl. VII; Humbert and Chambon, Fouilles de Khirbet Qumrân, 344; Humbert and Chambon, Excavations of Khirbet Qumran, 69; Pfann, “Reassessing the Judean Desert Caves,” Fig. 14:2. 68  De Vaux, “Khirbet Qumran,” 572–73; de Vaux, in DJD 3: Fig. 6:3 and Pl. VIII:12; Diez Fernandez, Ceramica comun romana, no. 85 (his type L6); Humbert and Chambon, Excavations of Khirbet

115 body, mildly curved in profile, standing on ring base; filling hole within a “discus;” voluted nozzle with a dotted circlet on either fluke; shoulder decorated with vegetal scrolls. Thin walled, from fresh mould. Fabric: hard, very dark grey/black, both at break and surface (the latter with semi-glossy self-slip). This lamp has been classified with Qumran Type 045 which embraces lamps with circular body on base ring, “spatulate” nozzle and ring handle. In terms of the shape, such lamps are clearly mould-made counterpart/development of the wheel-made “Herodian” lamps. Their shoulder is either adorned with a floral wreath or plain; the top of the nozzle has volutes or fins projecting at the joint with the body and repeated also on the underside. Indeed, P. Lapp described the type as the decorated “Herodian” lamp and dated it (probably based on the stylistic similarities to some Hellenistic lamps) to 50–31 BCE.69 A much later date of the introduction of the type has been set down by Rosenthal and Sivan, specifically, CE 70–75.70 This type is distinct from the typical “Darom” lamps which are so common in the contexts dating from the Bar Kokhba Revolt.71 Qumran Type 045 corresponds to some lamps of type D I–III from Masada (“Moulded Lamps with Floral Decoration”), which are considered by the authors as the objects manufactured ca. 50–68 CE in Jerusalem, where they were found in a burnt destruction layer (CE 70) of the Citadel.72 Indeed, the very fact that Type 045 is the best represented among the mould-made lamps of the entire Qumran assemblage, amounting to ten items, is in favour of dating this type’s floruit to before CE 68/70. Lamp Gr 8Q-12, which shares several features with a few lamps from Masada,73 falls into Qumran subtype 045.1.2, in which the lamp top has a small flat “discus” surrounded

Qumran, 67; Pfann, “Reassessing the Judean Desert Caves,” Fig. 14:7. 69  Lapp, Palestinian Ceramic, type 82.2, dating accepted also by Smith, “Lamps in New Testament Times,” 3. 70  Renee Sivan, Ancient Lamps in the Schloessinger Collection (Qedem 8; Jerusalem: Hebrew University, 1978), 82–85. 71  E.g., Yohanan Aharoni, “Expedition B—the Cave of Horror,” IEJ 12 (1962): 194, Fig. 4:1–4. 72  Barag and Hershkovitz, “Lamps from Masada,” 64–71, nos. 102–18. 73  Barag and Hershkovitz, “Lamps from Masada,” no. 111 (the lamp shape and the decoration of its nozzle), shoulder fragment no. 106 and two-nozzled no. 120 (both with the scroll decoration on the shoulder closely similar to that of Gr 8Q-12). For the same shape and similar decoration as in Gr 8Q-12, see also Rosenthal and Sivan, Ancient Lamps, 84, no. 345.

116

Młynarczyk

by a ridge.74 To exactly the same variant there pertain two mould-made lamps discovered in cave X/35. The fact that they were stratigraphically associated with a “Herodian” lamp of Qumran Type 036.4/5 and a pruta of the First Revolt, is strongly in favour of placing their deposition at ca. 70 CE.75 The same cave, published as Cave (P)13, had already yielded, alongside a “Herodian” lamp of Qumran Type 036.2, a fragment of lamp “with a vertical handle, made of black ware;” although not illustrated, this may have been a missing part of one of two aforementioned lamps of Qumran Type 045.76 The discovery of four lamps in a single cave eloquently attests the contemporaneity of two different lamp types and strongly suggests that the place was used as a hide-out or refuge during the years of the First Revolt. Gr 4Q-22 (lost; Qumran Type 047.2, see Fig. 7.11) Complete lamp, found at the entrance to the cave. L. 8.5, W. 7.0, H. 2.4.77 Double-convex circular body on ring base, with small rounded nozzle; discus (with filling hole off the centre) adorned with multipetalled rosette; shoulder with symmetrical “double axes;” a pair of double volutes flanking the root of the nozzle. Fabric: pâte rosé en dessous avec des traces plus rouges (coulées de peinture du dessus?). This lamp, classified as Qumran Type 047.2 and fairly closely paralleled by a lamp fragment from Jerusalem,78 is a Levantine version of Loeschcke Type VIII which started in Italy around the mid-1st century CE.79 It was very common in Cyprus/Cilicia and Syro-Palestine (from the Phoenicia/Galilee in the north to Idumaea in the south) between the end of the 1st century CE and the 3rd/4rd century. Dated by Smith to between the late 1st and the 74  Młynarczyk, “Terracotta Oil Lamps,” 122. 75  Itah, Kam and Ben Haim, “Survey and Excavations,” part 1, Fig. 11:2–3 and part 2, 172, obviously have been mistaken in dating the three lamps to the 2nd century CE. 76  See n. 13 above. 77  De Vaux, Archaeology and the Dead Sea Scrolls, 52; Humbert and Chambon, Excavations of Khirbet Qumran, 66; Pfann, “Reassessing the Judean Desert Caves,” Fig. 14:6; Młynarczyk, “Terracotta Oil Lamps,” 123–24, Fig. 5. 78  Renate Rosenthal-Heginbottom, Römische Bildlampen aus Östlichen Werkstätten (Wiesbaden: Harrasowitz, 1981), 24, Abb. X:4; other related Palestinian finds are John W. Hayes, Ancient Lamps in the Royal Ontario Museum (Toronto: Royal Ontario Museum, 1980), nos. 355–56, Pl. 42. 79  Donald M. Bailey, Catalogue of the Lamps in the British Museum: 2. Roman Lamps Made in Italy (London: British Museum Publications Ltd, 1980), 294 and Pl. 56 (Type O, group I).

2nd century CE,80 it corresponds to Kennedy type 5 dated to the 2nd and 3rd century,81 Dor type 26 dated to between the late 1st century CE and CE 150 (and later),82 Bet Shean type 7/ Variant 1 dated to the late 1st and the 2nd century CE.83 Many objects of this type (“more than one hundred lamps”) were found in the necropolis of Maresha and attributed to the 1st/2nd century CE “with a possible extension into the third century.”84 R. RosenthalHeginbottom, who noted that the earliest occurrence of this Levantine version of Loeschcke Type VIII had coincided with the First Jewish Revolt, drew logical conclusion that the type was introduced to Judea by the Roman soldiers.85 Moreover, originally it was used by the pagan rather than the Jewish population; the latter, before using a lamp of this type, would break its discus, probably out of need to observe the rules of ritual purity. Given the chronology of the type, it is highly improbable that lamp Gr 4Q-22 could make its way to the cave earlier that during de Vaux’ Period III (Roman outpost at Qumran) or Level 4 of khirbeh according to Humbert. Even more plausibly, the widespread use of this lamp type in Judea should be placed not earlier than the first quarter of the 2nd century CE. Actually, J. W. Hayes attributes the “model” South Syrian/North Palestinian series (to which Gr 4Q-22 doubtlessly belongs) to the first half of the 2nd century CE.86 Indeed, three fragmentary lamps of this very type were found in a perfectly narrowly dated context of the Cave of the Letters,87 and a fragment of a similar lamp comes from Cave 1 at Murabba⁠ʿat.88 Another lamp 80  Smith, “Lamps in New Testament Times,” 25 and Fig. 16. 81   Charles A. Kennedy, “The Development of the Lamp in Palestine,” Berytus 14 (1963): 73–75. 82   Renate Rosenthal-Heginbottom, “Imported Hellenistic and Roman Pottery,” in Excavations at Dor: Final Report; Volume IB; Areas A and C; The Finds (QR 2; ed. E. Stern; Jerusalem: Institute of Archaeology, Hebrew University, 1995), 244–45, Fig. 5.22. 83  Shulamit Hadad, The Oil Lamps from the Hebrew University Excavations at Bet Shean (QR 4; Jerusalem: Institute of Archaeology, Hebrew University/Israel Exploration Society, 2002), 1:16–22. 84  Eliezer Oren and Uriel Rappaport, “The Necropolis of Maresha— Beth Govrin,” IEJ 34 (1984): 123, Figs. 14:5–6, 17:3–4, Pl. 14:A. 85  Cf. Rosenthal-Heginbottom, Römische Bildlampen, 8, 26–27, 131. 86  Hayes, Ancient Lamps, 86–87 and nos. 355–56. Also Donald M. Bailey, Catalogue of the Lamps in the British Museum: 3. Roman Provincial Lamps (London: British Museum Publications Ltd, 1988), 280, Pl. 58, dates the beginning of the type in the Levant to not earlier than the end of the 1st century CE. 87  Yadin, The Finds from the Bar Kokhba Period, 114, Fig. 42 (lamps: 3.1, B.1, II.2). 88  De Vaux, in DJD 3:31–34, Fig. 8:13, dated (apparently too early) to the end of the 1st century CE.

117

Terracotta Oil Lamps

of exactly the same type and clearly the same dating (its discus broken, no doubt by a Jewish user) was discovered in Cave (P)24, at short distance to the north of Cave 11Q; it has been attributed by Patrich to an “even a post-70 CE date.”89 What conclusions can we draw from the study of the “cave lamps” (see map Fig. 7.12)? What questions should we pose and what are the odds we will get the right answers? The number of lamps is rather insignificant in comparison to other artifacts recovered from the “Qumran caves.” And, unfortunately, their presence, or absence, in each individual cave cannot be considered a sound evidence in determining the nature of the cave’s use (be it habitation, temporary or “permanent”? or a storage space?). On the other hand, unlike the Qumran settlement, where some of the lamp fragments were found in secondary contexts, all the “cave lamps” should be regarded as a reliable witness to the period of actual use of the respective grottos. In other words, the lamp finds provide safe chronological markers for visiting of the caves which they were to light up (regardless of the purpose of those visits). Two chronologically different lamp types from Cave 1Q strongly

­suggest that the place was visited more than once, ­ robably at an interval of several decades. Four lamps p found in cave P 13=X 35 characterize that cave as a habitation, most probably a Jewish hideout in the times of the First Revolt. Of the two lamps representing the Levantine version of Loeschcke Type VIII (Qumran Type 047), the one found in Cave (P) 24 testifies, in all probability, to the Jewish habitation during the Bar Kokhba Revolt, while the lamp found at the entrance to Cave 4Q must have been left there by a Roman soldier searching the caves, apparently during the very same period. In terms of their direct provenance, there can be little doubt that, regardless of their actual manufacturing place, nearly all the lamps (perhaps with the exception of the lamp from Cave 4Q as well as that from Cave (P) 24) must have come there via the settlement of Qumran. This is proved by the fact that they are most closely paralleled by the objects discovered in the immediately neighboring khirbeh. In other words, the “cave lamps” must have originated in exactly the same manufacturing sources as the “settlement lamps,” that is, in a local (Qumran) w ­ orkshop, a Jericho area workshop and a Jerusalem workshop as well.

Figure 7.1 Lamps Gr 1Q-43 (left) and Gr 1Q-44 (right). Photo by Jean-Baptiste Humbert and Jolanta Młynarczyk.

89  See n. 15 above.

118

Figure 7.2 “Qumran family” of lamps. Qumran Types: 032 (KhQ 171), 033.1 (G 1Q-43), 033.2 (KhQ 2210, KhQ 1015), 034.1 (KhQ 2206), 034.2 (KhQ 5087), 035 (KhQ 1257) and 035-Prime (KhQ 2308). Drawings by Kiyoshi Inoue and Mariusz Burdajewicz.

Młynarczyk

119

Terracotta Oil Lamps

Figure 7.3 Lamps of the “Qumran family” from Loc. 130. Photo by Jolanta Młynarczyk.

Figure 7.4 Fragmentary lamp Gr 1Q-56. Drawing by Mariusz Burdajewicz after Barthélemy and Milik, Qumran Cave 1, Fig. 3: 1.

Figure 7.5 Lamp GQ 29-1. Drawing by Kiyoshi Inoue after a sketch in the original register book.

120

Młynarczyk

Figure 7.6 Lamp GQ 8-12. Drawing by Mariusz Burdajewicz after Baillet, Milik and de Vaux, Les “petites grottes”, Fig. 5: 6.

Figure 7.8 Lamp Gr 10Q-3 before and after restoration. Photo by Jean-Baptiste Humbert and Jolanta Młynarczyk.

Figure 7.7 Lamp GQ 9-1. Drawing by Mariusz Burdajewicz after an original photograph.

Figure 7.9 Fragmentary lamp Gr 10Q-3. Drawing by Kiyoshi Inoue.

121

Terracotta Oil Lamps

Figure 7.10 Fragments of lamp Gr 8Q-12. Photo by Jolanta Mlynarczyk.

Figure 7.11

Lamp Gr 4Q-22. Drawing by Mariusz Burdajewicz after an original photograph.

122

Figure 7.12

Młynarczyk

Distribution of lamp finds in the caves in the Qumran area according to their types. Drawing by Mariusz Burdajewicz.

CHAPTER 8

The Unpublished Textiles from the Qumran Caves Mireille Bélis The so called “Qumran textiles”1 are far from being limited in number to the seventy-five cloths and fragments that were presented by Grace Crowfoot in DJD 1.2 At the time, Crowfoot was very ill and she was also in the process of completing another study on the textiles, found at Murabbaʿat. As a result of her illness, only the preliminary report on the Qumran findings was published. Crowfoot had also intended to undertake a comparison between the two sets of textiles, but unfortunately she died in 1957, leaving this work unfinished: her final report was then completed by her daughter, Elisabeth. In 1995, we resumed work on the textile collection that was stored at EBAF, and we added several hundred more fragments to the corpus that has already been registered. My primary objective in this article is to provide a synopsis of the “Qumran textiles” that have so far been identified. This is no easy task, as the original collection has been dispersed, to the benefit of various museums and private collections. I will also attempt to answer the following questions that are closely linked with the “history of the caves” and which provide the topic of this conference: 1) What is so particular about the Qumran textiles? Why was flax (linen) the only material to have been used; and what was the purpose of these textiles if they are too small to have been used as garments?3 Some textiles were clearly used as jar covers, while others were used as manuscript wrappers, hence the conveniently-shaped “cigarlike” scrolls. Other fragments appear to be too large to have been used as jar covers, although it also unlikely that they were used as “scroll wrappers.” 2) Could these textiles that have been studied recently and stored in Amman, enhance our knowledge concerning this deposit? Can they help us to determine whether this vast concealment of manuscripts, jars and artefacts was the result of a ­project to store over the long-term and preserve them—so 1  Only one textile fragment survived in the khirbeh itself, in locus 96, and another was found in a tomb; the others have been found in the caves in the vicinity of the settlement. 2  Grace M. Crowfoot, “The Linen Textiles,” in Qumran Cave 1 (DJD 1; ed. D. Barthélemy and J. T. Milik; Oxford: Clarendon Press, 1955), 18–38. 3  They are too small to be used either for the front or the back of a tunic. © koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_010

thoroughly prepared and completed as they appear to be? Or do these textiles give us an indication, as Jean-Baptiste Humbert has suggested, that a hasty, emergency scenario took place?4 1

The Unpublished Collection from the École biblique et archéologique française de Jérusalem (EBAF) / Rockefeller Museum

1.1 The Rockefeller Vault In the summer of 1995, when I asked Joseph Zias, the curator of the Rockefeller Collection, to examine the Qumran textiles, he mentioned that the contents of drawer had not been checked for twenty-seven years. Moreover, the drawer had been incorrectly labelled “Aïn Feshkha,” despite the fact that Roland de Vaux had not found any organic remains in Feshkha.5 Before de Vaux had made the connection between the khirbeh and the scrolls, the first cave he discovered was given several names, such as “the Cave of the Hebrew Manuscripts,” the “Cave near the Jordan Valley” and the “Cave near Feshkha.” In other words, the ruins nearby were considered less important, at least at first, although it became quite clear early in de Vaux’s work that these could not be ignored. It was also understood that an initial survey of the minor linen fragments found by de Vaux (these were never studied by Crowfoot) needed to be published. The site from which they had been removed also needed to be established. I filled any gaps in information wherever possible and was thus able to record the fragments. I will not go into further detail here on this, as I have already given a description of these fragments in one of the chapters contained in the Volume II of the final publication on Qumran.6 4  See Joan E. Taylor, The Essenes, the Scrolls and the Dead Sea (Oxford: Oxford University Press, 2012), 272 on the so-called “quick hiding scenario,” as favoured among scholars. 5  Except for a hoard of coins covered by the fossilized remains of linen. See Bélis, “Des textiles: Catalogue et commentaires.” 6  Mireille Bélis, “Des textiles: Catalogue et commentaires,” in Khirbet Qumrân et ‘Aïn Feshkha. Vol. 2: Études d’anthropologie, de physique et de chimie (NTOA.SA 3; ed. J.-B. Humbert and J. Gunneweg; Fribourg/ Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 207–76.

124

Bélis

Figure 8.1 The original boxes and drawer as found in the Rockefeller Museum containing the Ebaf textile collection. © Jean-Michel de Tarragon EBAF

Figure 8.2 A parasite in a 11Q textile under DinoLite.

Several fragments of these textiles were found still covered in dried mud and debris. They also contained visible traces of the context in which they were discovered. I also found remains of fauna, flora and several parasites in the folds or on the surface of the fragments. These were set aside and examined systematically, in order to progress with the investigation about the context. Inside the folds of one particular cloth was a parasite, which thrives only

on lice that live on human beings. This provided evidence that the fragment had originally been worn as a garment. The tow labelled 3Q indicates that raw flax was a valuable material, as demonstrated by the fact that they were found hidden among other precious goods. It is possible that the material was intended to be spun after being recovered from the cave, which would also indicate that this may have been a refugee deposit.

The Unpublished Textiles From The Qumran Caves

1.1.1 Bitumen and Stains This “cigar-like” object could be the remains of the only scroll to have been found in 1Q, still stuck to the neck of a jar. Alternatively but less likely, it could be from a stack of “blackened linen,” bound with a string and found in a recess of cave GQ12 in March 1952. Two photographs of these are available: the first in DJD 1, plate 2.3;7 the second is kept in the BASOR archives. Several scholars have suggested that the bitumen on the jar covers was used as a form of glue, to waterproof the jar and therefore protect its contents. The problem is that no scientific reports provide any basis for this suggestion: the blackish substance looks like bitumen, but the deposit cannot be desiccated asphalt, even if ed-Dheeb described his first findings as “bundles wrapped inside smelly cloths ‘like mummies’.” In fact, ed-Dheeb was often pressed (and was paid to do so) into giving further details about the findings of the cave, to the point that there are as many stories as they are people who questioned him. He may have lied, changed, added and forgot details but he never talked about bitumen. In my opinion it would be highly improbable that a Ta‘amireh would have been unaware of the natural asphalt that materializes sporadically on the surface of the Dead Sea. It is also inconceivable that scholars would not have recognized this as bitumen as soon as the word “mummies” was mentioned, as they would have been fully aware of the techniques for preparing taricheutai linen bandages and that both natrum and bitumen are required for mummification. It is also highly unlikely that a young Ta‘amireh shepherd in the 1950s knew anything about Egyptian funerary practices. In his very first statement on the “sealed” jars, ed-Dheeb only mentioned red clay: not a blackish substance, but a red one. Moreover, Grace Crowfoot’s testimony seems to corroborate the Bedouin’s story: “The first box was unpacked on 3 July 1949 (. . .). When it was opened, the odour it gave off was like that of an Ancient Egyptian tomb and Mr. Hamilton, after taking one whiff of it, suggested that further examination should take place in the garden!”8 7  Dominique Barthélemy and Jóseph T. Milik, eds., Qumran Cave 1 (DJD 1; Oxford: Clarendon Press, 1955). 8  Crowfoot, in DJD 1:19. The Bedouin told his story in an often forgotten paper by William Hugh. It is helpful to read it again, in William H. Brownlee, “Muhammad ed-Dheeb’s Own Story of His Scroll Discovery,” JNES 16, (1957): 236–39, with notes. Brownlee published a second paper, “Edh-Dheeb’s Story of His Scroll Discovery,” RevQ 12 (1962): 483–94.  “In 1945, I was tending a flock of fifty-five head in the Wilderness, along with two other herdsmen who were also with their flock. The three of us were sleeping out in the Wilderness. As is the custom, each one of us would count his flock in the evening. But it so hap-

125 Bitumen remains prompted mere guesswork without any archaeological evidence found on the linen (or on pottery) to be able to verify the substance. Even among the numerous textiles in the Amman Collection, no trace of asphalt has been found on the fragments. As is the usual practice, it is important to re-read the exact words of the scientist who handled the boxes that were sent to him. He described their contents in great detail as follows: 3 boxes Box 1: (. . .) a series of small fragments which appeared to be blank . . . a series of small fragments showing Hebrew script Box 2: (. . .) half a dozen fragments contaminated with black adhesive and textile.

pened that I had not counted my flock for two days. So at about 11 a.m. on the third day, I counted the flock and found that one goat was missing. I went to my companions and told them that I wanted to leave my flock with them because I wanted to go out and search for the lost *goat*. I left them and went in search of the goat. I had to climb hills and go down into valleys. I went very far away from the [other two] herdsmen. As I was wandering around, I came upon a cave with its entrance open at the top, like a cistern. Supposing that the goat had fallen into the cave, I started throwing in stones. Every time I threw a stone into the cave, I could hear a sound, like pottery breaking. I was puzzled as to what the sound was and wanted to know what was in the cave. So I went down into the cave and found pottery jars. I began to break the jars with my staff, thinking I would find some treasure. However, inside the first nine jars that I had broken, I found small seeds, reddish in colour. There was nothing else inside them. When I broke the tenth jar, which was the smallest of the jars, I found some rolled leather with some ‘scrawling’ on it. The nine jars which I had broken first had lids covering on, but these were unsealed; but the smallest tenth jar had a lid which was sealed with something like red clay. I was puzzled as to what to do; whether to take the rolled leather or to leave it where it was. I finally said I would take it, as I remembered that my companions and I needed straps for our sandals. I wrapped the leather up in my cloak, put it on my back and left. When I reached my companions, I showed them what I had found and gave each of them a piece of leather so that they might use it for their sandal straps. Through my lack of good fortune, I did not find the goat. Afterwards, when it was late at night, we returned to the place where we were staying in the Wilderness. I kept the leather with me until I returned to our house, where I put it in a skin bag and hung it up in a corner. The skin bag stayed hanging [there] for more than two years. Afterwards, an uncle of mine came to our house and asked that he might [take the leather and] show it to dealer in antiquities at Bethlehem, to see if he might be of any value. Bethlehem, October 23, 1956. Muhammad ed Deeb, One of Bedouins of the Ta‘amireh.”

126

Bélis

Box 3: a sample of textile and black adhesive (. . .). The black adhesive was tested with solvents (. . .). Similar tests, applied to fragments of parchment, showed that some pieces reacted towards solvents in the same way as the black material itself. When a fragment of parchment was examined, and found to have decomposed at one edge of this pitch-like material, it became evident that the black substance was in fact, the end product after the decomposition of animal membrane: in effect, this was a form of glue.9 The adhesive is not bitumen but the substance that remains after the skin decays. Conversely, the stains and decay on the sides of cloth (in this instance, not “borders” or “selvedges”) remain unexplained. The reddish stains could be wax and the curves along the sides could be the result of burning as linen disappears entirely when put in direct contact with a flame. An analysis of the substance will be undertaken with Christophe Moulhérat (Conservateur au Département des textiles anciens, Musée Jacques Chirac, Paris). 2

described by Crowfoot. The question was then to determine to which cave they belonged: either to the first Cave (1Q) or to other sites; or more importantly, to the “minor caves” of the cliffs that were excavated in 1952. 2.1.1 The Jar When we returned to the list of caves in which textiles had been found, the problem still remained that de Vaux’s elliptical description of the finds scarcely mentioned these materials. Even when he noted that some textile fragments had been preserved inside a cave, he did not consider them worthy of publication: “. . . mais les seuls objets dignes d’être présentés en dehors de la poterie sont les étuis à phylactères, qui seront décrits à propos des textes qu’ils renfermaient.”10 The provenance of the textiles inside the jar cannot be firmly determined: all we can say for certain is that they belong to the Qumran caves.

The Jordanian Collections

In 2004 I spent three weeks, and again a further three weeks in 2007, studying the linens that might have been stored by Gerald Lankester Harding in the Citadel Museum in Amman. The jar and its linen contents were exhibited in the Museum. The linen needed to be checked, sorted and documented. The goal was to establish in which sites, these somewhat mishandled objects, had been discovered. Without doubt, this constitutes the largest unpublished lot of materials ever unearthed in the natural caves in the Qumran cliffs. The Citadel owned a very fine collection, enriched by several of beautiful scroll wrappers and jar-stoppers, published by Crowfoot, and then placed under Perspex frames and then put on permanent display in the Citadel Museum. 2.1 The Amman Collection Apart from the cloths, which were brought to the Citadel Museum by Harding and put on display, there was also a broken jar, with gaps through which scraps of linen could be seen. Once opened, in 2004, we found a large collection of linen fragments similar to the Qumran textiles as 9  Harold J. Plenderleith, “Technical Note on unwrapping of Dead Sea Scroll Fragments,” DJD 1:39–40.

Figure 8.3 Jar GQ39-2 in the Amman Museum, with linen through its lacunae.

10   Roland de Vaux, “Archéologie,” in Qumrân Grotte 4.II: I. Archéologie; II. Tefillin, Mezuzot et Targums (4Q128–4Q157) (DJD 6; ed. R. de Vaux and J. T. Milik; Oxford: Clarendon Press, 1977), 15.

127

The Unpublished Textiles From The Qumran Caves

The jar is the typical cylindrical type from Qumran and appears to be at least three quarters full of flax remains. Pieces of white- and blue-dyed linen, similar to those found at Qumran, were found protruding through a hole in the body of the jar. Does the jar itself provide any useful information? According to the number ascribed to it, the jar comes from Cave GQ 39, south of Qumran near Ain Feshkha, whereas the lid labelled 14–25 comes from Cave 14, i.e. the Manuscript Cave 1Q. Originally, these had not been matched together. Fortunately, the respective dates on which they were discovered are known. Both the lid and the jar were discovered during the expedition carried out on the Qumran cliffs in 1952. The lid was discovered on 18th March and the jar on 16th March 1952.11 The lid has been placed on the top of the jar in its display, so as to demonstrate to visitors how it was used. The lid could have come from the remains left over after the first excavation of 1Q.12 Nevertheless, linen has not only been found in Caves 14 and 39, although this is an hypothesis that can be discarded in the light of the report published after the 1952 Archaeological Campaign, which makes no mention of any important textile findings in either of these two caves.13 As for the museum archives, I checked and found a registration number for the pottery but found no record of the linen.

Figure 8.4 Just empting of the jar.

Elements Relating to the Archaeological Context Together with linen, which was sometimes found entangled among “floating” or erratic threads, we found organic or inorganic material that reportedly belonged to the cave deposits. These we set aside. They included one wooden

stick, some palm fibres, a potsherd (about 1 cm²), elements of a piece of string, leather sewn onto a piece of linen, several blue threads, a seed (perhaps a piece of corn?), two pieces of debris that could be papyrus (?) [G2062] and several fragments with two blue threads. The colour of one of the blue threads was still beautifully preserved; a twisted corner, knotted with a narrow linen strip, was also well preserved.14 Little can be said about any damage done to the lot, whether deliberate or not, although rats seem to have piled up some of the fragments to make their nests. We also found one corner of a linen fragment, cut neatly and diagonally across (about 16 cm in length) and measuring at most 12 cm × 10.5 cm. All the fragments were covered with dust and sometimes with dried mud. This was probably due to the seasonal infiltrations of the caves (both natural and artificial) along the cliff face. Technically, these linens do not differ from the 1Q lot: The flax is S-spun; the weave is always a plain-weave pattern, sometimes with an even-weave; the average count is between 11 and 15 threads per cm. All the linens have the same direction of stitch on the hem (\\\), with the exception of one fragment (///). On one piece, there is a fine indigo hem, sewn in regular stitches with

11  Roland de Vaux, “Archéologie,” in Les ‘petites grottes’ de Qumrân: Explorations de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 9. 12  Although de Vaux specified: “Quelques tessons on été recueillis mais n’ont pas été gardés, car ils n’enseignaient rien de plus.” De Vaux, DJD 3:9. 13  See de Vaux, DJD 3:9, 12.

14  Cf. Yigael Yadin et al., The Documents from the Bar Kokhba Period in the Cave of Letters: Hebrew, Aramaic, and Nabatean-Aramaic Papyri (JDS 3; Jerusalem: Israel Exploration Society, 2002), Pl. 2, 3, 5, 7, 8, 10 and 12, which shows the bundle of the Babatha Archives before they were opened. Narrow bands of linen and strings were rolled around the documents and then secured by tying a knot, to keep the bundle tightly closed.

2.1.2 Sorting The total weight of the material extracted from the jar was more than 566 grams. We had to divide the original lot into seven parts to conserve them while we studied the individual pieces and fragments. 114 pieces of cloth or fragments were registered in total. These were intended to be kept in the storage rooms of a new museum. 2.1.3

128 two threads and preserved along 15 cm on one edge of the corner.15 2.1.4 Outlines We cannot present the entire collection that has been registered here, as further work needs to be resumed. This would enable us to make a more accurate assessment and determine the extent to which the main characteristics of the spinning and weaving of the flax, and the type and measures of the cloths resemble those from Qumran. As a short preliminary report, we must emphasise that some of the characteristics belonging to several fragments, out of the hundreds that have already been studied or are still remain to be studied, are closely parallel to the 1Q Collection. The quality of the materials varies from a coarse weave to a very fine weave, of which the best specimen is woven in 18 × 18 threads per cm. The lowest weave count is 10/11 × 9/10 threads per cm. G2074 is the largest piece of linen, although it is incomplete. It has three fragments and measures at least 50 cm × 30 cm. As only one corner has been preserved, its original dimensions cannot be determined. Its weave count per cm is: warp: 12, weft, 13/14, which is rare among the Qumran textiles and indicates that it might have been used as a “scroll wrapper.” I cannot go further into a description of starting-borders, selvedges, edges etc. as they are too technical to be discussed here. Bands: among the Amman textiles, we found the remains of several narrow strips of linen that had been folded and sewn. G2048: is 12 cm long × 4 cm wide but torn at the sides; G2106 is the longest band, 22 cm in length; G0270 is weft-faced and is interesting because of the narrowness of the band: 25 cm in length × 0.6 cm in width. These bands were clearly intended to keep their shape. If we need to hypothesize about their use, they were perhaps belts or were used, as with those found in the Bar Kokhba caves, as bands in which papyrus documents were found rolled and kept tightly closed.16 Some tiny scraps of papyrus were also found in the same lot. 2.1.5 Twisted Corners As a rule, Mrs. Crowfoot and other specialists have always asserted that the twisted corners were typical of jar cov15  Inventory G2001B. G2001A belongs to the same piece of cloth, which was at least 30 cm × 20 cm in size. 16  A parallel to a similar piece of cloth from IQ will be described in a paper (to be published later).

Bélis

ers. Five of them indicate that the Amman Collection comes from the Manuscript Cave(s), or from so-called “small caves”, in which jars or jar fragments, but no scrolls, were found. G2006 is a corner piece tied with a string around it. The corner is drawn and certainly must have been attached to the ear-lug handle of a jar. The fragment measures 25 × 23 cm but is incomplete. Five other corners are twisted in the same manner, but they are very small, measuring less than 5 cm in width and length. Inside one fragment, G2085, a tiny fragment of leather was found. 2.1.6 Indigo Pattern All the fragments follow exactly the same pattern as in the 1Q system, as described in DJD 1. The stripes are narrow and never have more than two threads on the weft, then, a series of natural coloured (écru) wefts, and rarely a new double-threaded blue line. There is, however, one exception to this: a single blue line in the warp, which could be the remains of a rectangular pattern, as in DJD 1 No.1 and No.46. As the sorting still has to be finished, it is possible that more blue-lined fragments may be found among them. These will require further study, in addition to the other heaps of flax that are still heavily covered in dust and debris. 2.1.7 Remains of Fringes We have also found a few fringed cloths (only on fragments), as has been the case for six fragments. One of these has a fringe of 8 cm long that hangs below the edge, and which is in close parallel with DJD 1, No.17. 2.1.8 Fibres Inside and among the flax fragments, we found several fibres and two “half-processed” kenaf remains that we had already seen in the EBAF/Rockefeller Museum collection. Several stems of this exogenous plant from which the fibres had come, had been dug out from under a thick layer of dust and rumble in the main chamber of cave GQ29. The Kenaf—as far as I know—was unparalleled in ancient Palestine at the time that Qumran was flourishing. Despite its many properties, its only use inside a dark cave would have been to help to light the wick of the lamp found alongside it. These two useless objects were left behind after this ultimate and respectful visit. One beautiful but puzzling fragment, G 2020, might provide us with positive proof that the jar textiles definitely belonged to the Qumran caves. This particular piece is the exact parallel of an up to the minute “hapax”

129

The Unpublished Textiles From The Qumran Caves

Figure 8.5 Cloth 1Q n°19 having a parallel, G2020 in the Amman collection.

textile from 1Q. Clearly inserted in between the two blue lines, a larger blue weft overlaps, and is woven over and under the warps alternately. Seven stitches have been preserved, the others are lost due to tearing. There is a possibility, which cannot be disregarded or overlooked, that the fragment was recovered in 1Q after the archaeologists returned to 1Q, which was re-numbered GQ14 in 1952 (see discussion). In her notice Crowfoot writes: At 4.3 cm below the edge there are two throws of blue weft, followed by four natural and then two more blue wefts; a line of blue is darned ornamentally along the middle of the natural band, over five and under five warps, with an occasional four, where the threads happen to be thicker. There is no doubt that this line is embroidered.17 According to Orit Shamir, however, Crowfoot’s description is not quite accurate: the blue threads were inserted during the weaving and not added afterwards.

17  Crowfoot, in DJD 1:32 No 19.

No. Weave: Description Inventory Threads/cm

G 2020

20 × 20 21 × 21

Remarks

Two series of indigo HAPAX and // 1Q, DJD 1, No. 19 blue lines. Between the two blue stripes, 5 cm below the rolled hem, over 6, under 4 and over 6 threads, is inserted a line of two blue threads. A second row of 2 blue lines occur at 3.3 cm below the first and a third ones. The edge is torn.

This fragment cannot be joined to No.19 as blue rows of threads usually occur in wefts, whereas the Amman fragment clearly shows some blue threads in the vertical warps. No.19 contains a line that regularly crosses over 5 and under 5 warps between the two lines of blue. The

130 Amman linen fragment is very fine, with a high weave count per centimetre, whereas the Crowfoot fragment is woolly and has a lower weave count. The Amman G2020 is a “hapax,” although it does not show evidence of having belonged to 1Q. The only conclusion we can draw is that No.19 and G2020 were dyed and woven in the same workshop and/or according to a pre-determined pattern. The colour of the linen ranges from off-white (or bleached), cream, brown and dark-brown to dark-blackish. 2.2 The “Job 1357” Box In August 2007, Mrs. Arwa Massabeh, Jordan Archaeological Museum, brought me a cardboard box that had hitherto been stored in the Museum basement. In her estimation, the box had been kept for a long time inside a drawer that had been jammed. This new collection was similar to other Qumran findings. It seemed that these fragments could have been taken apart for further study, as they had already been cleaned. An ancient textile would never have been as flat and neat as this if it had been freshly unearthed.

Figure 8.6 Fragments of a cloth G516, in the Amman collection.

The lid of the box was labelled as follows: “Different types of weaving and mending” written in Mrs. Crowfoot’s handwriting. The Arabic word “unpublished” was also written on the box, as Mrs. Massabeh has pointed out. It seems likely therefore, that these fragments belonged to sites that had been excavated several years prior to Mrs. Crowfoot’s death in 1957. 2.2.1 Contents of the Job Box Five layers of fragments were carefully piled inside the box and flattened between sheets of paper to separate them. All the pieces are damaged but are clearly recognizable as typical of the Qumran plain-weave linen. However, there

Bélis

has been no mention of them, even in Mrs. Crowfoot’s publications. It is very possible then that she died before she had begun to write a report on these. The box was subsequently returned to the museum and put into storage. Some fragments have been mended with sewn patches, a technique used at the time in Palestine but later abandoned, when darning was adopted as a better method of repairing fabric. The exclusive use of plain-weave does not provide sufficient evidence, however, to be able to label these fragments as “Qumran.” Fortunately, a hand-written remark was made on a sheet placed between the first and the second layers of the pieces of linen: “these fragments may belong to No. 42 with blue stripes.” “No. 42” obviously stands for Qumran Inventory No. 42, which is described as follows: Fragment with blue lines. Length 18.5 cm, breadth 12 cm at its largest portion. Three sides torn away, the edge preserved above for 6 cm, rolled and whipped with a single thread, with the warp ends hanging out as in No.1. There are three blue lines of two wefts each, the first 4.6 cm. below the rolled edge, with eighteen natural wafts between the first and second and seventeen between the second and third. The spinning of this piece is extremely good, and the flax has a smooth and silky look and slightly greenish colour. Count 16 × 12 per cm.18 This note is proof that Mrs. Crowfoot’s studied these Qumran textiles. As the DJD 1 Qumran Cave 1 series amounts to 77 (75 ancient) pieces and fragments of cloths, the “Amman Box” collection adds a significant number of 22 more fragments to the textile corpus from Qumran. 2.2.2

Characteristics of the Jar and Job Box 1357 Collection As far as it is possible to draw any conclusions after the survey of this large amount of textiles, we can say that they tally with previously known linen cloths and fragments found in the vicinity of Qumran. They consist of flax with an average S-spun, in a plain weave without/ or with patterns made of indigo threads typical of the Qumran 1Q textiles. The state of conservation of the fragments is due to the poor conditions in which they were stored. The jar protected those fragments that were inside the jar but neon light has badly altered and faded the indigo colours of 18  Crowfoot, in DJD 1:35 No. 42.

131

The Unpublished Textiles From The Qumran Caves

those textile fragments that were found protruding from the lacunae in the jar. After sorting, the collection yielded pieces of linen typical of those found in the Qumran caves. All these textiles had obviously been placed together with manuscripts. We even found one potsherd among the textiles.19 The decorations are the same as those as those found in 1Q: the weave patterns (plain weave and occasionally even-weave), fringes, scarcely bare warps, and occasional blue lines, (2 or 3, repeated twice or three times in the weaving) are comparable or similar to those observed on the linens of 1Q. However, no new patterns were found (such as blue warps and blue wefts crossing each other or at right angles, evoking rectangles), although there was one exception: a tiny fragment showing a single line of two blue threads, on the weft, which could be the vertical line of a blue rectangle as in No.1 DJD 1. The fragment could be joined to No. 42 DJD 1 to form a rectangular pattern measuring 52.5 cm × 41 cm plus Frg.No.42. It is one of the largest cloths preserved in the Amman Collection.

deposits and various organic materials which were saved for further analysis), which 103 pieces of linen, that have now been registered. The study is far from complete but hopefully will be resumed during a third mission to Amman, with the help of Christophe Moulhérat.

2.2.3 The Crowfoot Box The Crowfoot Box contains the unbleached fabrics chosen by Mrs. Crowfoot for their unbalanced number of threads warp and weft i.e. with more warps (G0523) than wefts. In one example there are 20 × 15 threads. These fabrics show an irregular weave, with a Z-twist thread of S + S coming out the frayed edge.20 One of the fragments, No. G0521, measuring 30 cm × 18 cm shows an exceptional density of threads: 23 × 15 threads per cm. This density could be the result of “crowding” the threads along the selvedge. No. G0520 shows an outstanding quality of weaving, approaching the “professional” level of even-weave: 13 × 13/cm and occasionally 14 warps × 13 wefts. Mud debris is embedded amongst the fibres. The repairs are made of patches, of the type known in the region at that time, and are a typical feature of the secondary use of worn clothes i.e. divided or cut up into smaller parts, as suggested by Orit Shamir and Naama Sukenik. Sometimes the patches sewn over the holes are in better state than the clothes they repair. The Jordanian inventory currently includes up to 112 new fragments. Besides the 24 fragments sorted by Crowfoot before 1957, there are 136 new items (including

In May 2014, Christophe Moulhérat and I studied fragments belonging to 8Q, 11Q and the enigmatic “SPI” lot. Some of these fragments differ from the textiles from 1Q and the “minor caves.” At least one of them is Z-spun and, according to C. Moulhérat, seems to have been imported.

19  The sherd could be an element of another potsherd that came from a broken jar, which was found stuck to the wrapper enclosing a text. 20  This characteristic is present on a piece of linen cloth in the Louvre (paper to be published).

2.2.4 Initial Synthesis of the Amman Collection Of the 112 remains that have been sorted, 103 consist entirely of textiles or textiles associated with leather. On the 103 pieces of linen, 5 show blue lines, in stripes or in weft yarn. In other words, less than 15% of the remains are dyed with blue. However, as we are referring to fragments and not complete pieces of fabric, even this figure of 15% could be misleading. After sorting through the whole collection, I chose first to examine the fragments containing blue. 3

The Manuscript Caves other than 1Q21

3.1 SPI The name “SPI” refers to a group of letters or numbers that J.-B. Humbert and I tried to decipher from a note written on the outside of a cardboard box that was filled with textiles. The note had been carelessly written with a pencil and had almost faded.22 There was no other label or description of its contents to be found. These textiles were certainly found in a cave, but we cannot ascertain in which one. It is important to note that these textiles do not fit Qumran chronology, as the first set of C14 dates point to the medieval period.23 Moreover, one of the textiles does not respect the Jewish law of sha⁠ʾatnez with regard to mixing wool and linen. On that basis, we could discard the entire lot as not being Jewish and therefore not ­belonging to the period of the deposit of the manuscripts. As for 21  Once again, we invite all scholars to adhere to the full method of naming each cave, adding the appropriate GQ prefix for the minor caves or Q for the manuscripts caves. 22  Jean-Baptiste Humbert has suggested that this could be “Sans Provenance Identifiée” (“unidentified provenance”)—personal communication. 23  Humbert and Gunneweg, Khirbet Qumrân et ‘Aïn Feshkha.

132

Figure 8.7 Violet from 8Q.

Figure 8.8 Blue fibers from 11Q under DinoLite.

Bélis

The Unpublished Textiles From The Qumran Caves

identifying their provenance, they could have come from one of the southern caves, in the neighborhood, west of Ain Feshkha. 3.2 Cave 3Q 3.2.1 The Cave It is important to remember that 3Q is one of the caves that were found by archaeologists. It was excavated for twelve days (14th–25th March 1952) and given the number GQ8. To add more confusion to the numberings, it was the G team who excavated this cave, which was the eighth cave into which they entered. This cave was called G8 or GQ8. Because it had not been discovered by Beduins, archaeologists therefore had the rare opportunity to excavate the cave according to the standard excavation practices of the time. They were thus able to collect a number of untouched artefacts. Among them were the following: Dans la chambre arrière de la grotte, niveaux stratifiés de cailloux avec quelques fragments de tissus, de cuir noirci et des fragments écrits. Très peu de tessons dans cette région. Cette chambre se prolonge par un passage étroit et montant vers une cavité presque complètement remplie par des nids de rats contenant des morceaux de tissus, quelques bouts de cuir et un fragment inscrit. Juste à l’angle nord de la chambre, deux rouleaux de cuivre . . .24 No doubt these textiles have been carefully preserved and safeguarded in the Rockefeller Museum. Of the missing textiles, none remains except for those removed from inside the rats’ nests. More textiles were gathered by two Israeli expeditions and published separately. We wish we would reevaluate the IAA collection, together with EBAF/Rockefeller corpus of 3Q as a single corpus, which they originally formed. Even though the latter seems to be of “Qumran type” and the former belongs to the late Roman or Byzantine periods, both come from 3Q. 3.2.2 The Exceptional Case of Cave 8Q The fragments of textiles discovered in 8Q deserve a special mention. In DJD 3, de Vaux records—and this is rare—that they found “restes d’étoffe et de ficelle” (“fabric and string remains”). The fragments are made only of flax (linen) and two of them are dyed, although not with indigo (See below). The analysis of these is still ongoing and we will await the results in due course. We had hoped that these might 24  De Vaux, in DJD 3:7.

133 have been available by mid-February (date of the conference), but unfortunately, I have not received the final report in time to add these to this paper, nor were the results available at the time of writing. It is however fairly plausible, according to Moulhérat, that this same piece of textile, found stuck in some marl mud, is, if not dyed purple, then at least dyed in a violet-purple. Purple is unusual when compared with the Qumran textile corpus and, more importantly, does not fit in with the Essene way of life. Here we are confronted with another aspect regarding the nature and function of the artificial caves. Humbert has raised the question as to whether these caves were used as a complete system of caves used by refugees. As such, the purple could be the remains of a cloth left in situ by an outcast, an outlaw or a man who has escaped from the town. Any cloth dyed in purple indicates a particular social status and is reserved for high-ranking notability. In accordance with Roman decorum, purple was forbidden to ordinary people. The owner of the garment was therefore of high rank, maybe from Alexandria, a traveller who would have been aware that this was a cave that was used as a refuge. However, he would not have been a pious ascetic. Access to the cave was risky but it is close to or from the khirbeh. It is unsuitable for long stays and a fortiori for the practice of any craft activity, due to its lack of natural light. Only very small amounts of textile and a few fragments of ­manuscripts were found here, along with phylacteries, mezouzah, leather (straps, a sandal), date pits and olives. Do these data contribute to the already known pieces of evidence in favor of a short-term stay in Cave 8Q of a lone visitor? It does. Did the visitor have any connection with those who came here to hide the manuscripts belonging either to themselves or their family, and did he return later to recover the manuscripts? When would this have happened? We shall have to wait the result of the (ongoing) C 14 dating process to be able to form a clearer idea about the date, unless the marl has contaminated the sample and blurred the results (See Conclusion below). 3.2.3 Cave 11Q The cave has yielded fabrics of all kinds. Of special note is a linen fragment (D034 = 11Q -32 EBAF), which is very fine and rare example of its kind. It is torn on three sides and has two strings of indigo and several white threads, knotted around a twisted corner of the cloth. The knot was found loose and could be removed. 3.2.4 General Features The best preserved portion of the fragment measures 6.2 cm × 3 cm, although it is impossible to determine its

134

Bélis

original dimensions. The first piece of string measures 5.5 cm long; the second is 1 cm long. The warps are thick and are woolly in texture, although the wefts are thin: 16 wefts × 12 warps per cm. This is not common among the Qumran textiles. Its edges are torn and frayed. In the middle of the cloth a round hole has destroyed the intersection of two sets of warps and 3 sets of thicker wefts, which creates a visual effect of self-bands. This is the only decoration in the weaving. Roland de Vaux insisted that he never found coins that could likely date the concealment of the manuscripts. Libby was given 4 ounces (114 grams) of flax found in 1Q. The textiles were burnt to provide a date however the dates 16 BCE–233 CE that were given were disappointing. There have been other attempts to narrow the gap between the dates.25 When de Vaux excavated the first cave near Qumran, he did not imagine how rich the desert could be elsewhere, because the cliff appeared to be so bare and austere. However, as a result of his assumption, we are now able to compare and connect the larger corpus of textiles that have been discovered. This also suggests that treating the settlement at Qumran as a secluded group is no longer tenable. The settlement of Qumran and a majority of its caves must be studied together, as underlined by the fact that the entire complex is visible to anyone who views it from the top of the tower. Linen has been found in three minor caves GQ12, GQ20, and GQ29 and in eight out of the eleven Manuscript caves, the provenance of which is certain. A spindle-whorl has also been found in 3Q. 3.3 A Homogenous Corpus Up to now, all the known Qumran textiles follow a basic and unique pattern, which can be described as “sober.” In this respect the same decorative pattern is never repeated twice on another piece of cloth. 1.

The use of the flax and the final shape of the cloths are the best way to preserve the scrolls inside a cave.

25  Willard F. Libby, “Radio Carbon Dates, II.” Science 114 (1951): 291–296; Kaare L. Rasmussen, et al., “Cleaning and Radiocarbon Dating of Material from Khirbet Qumran,” in Bio- and Material Cultures at Qumrân from a COST Action G8 working group meeting held in Jerusalem, Israel on 22–23 May 2005 (ed. J. Gunneweg, C. Greenblat, and A. Adriaens; Stuttgart: Fraunhofer IRB, 2006), 139–163; Joan E. Taylor, et al., “Qumran Textiles in the Palestine Exploration Fund, London: Radiocarbon Dating Results,” PEQ 137 (2005): 159–167.

2.

3.

The blue pattern is a “signature” pattern (apart from Egyptian weavings). The dyeing of threads takes place after the spinning and before the weaving, which implies that the workshops were interconnected and all the weavers followed the same distinctive pattern, blue on white linen. One exception is the cloth found in 8Q: the stripes here are repeated but with varying spaces between them. Once they were cut but sometimes left unfinished (at the seams), all these textiles, served as wrappers for scrolls, whereas the smaller, rougher and plainer pieces were used as jar covers.

The Links between the Textiles, Manuscripts, Jars and the Essenes Of course, the broad repertoire of unpublished textiles requires us to reassess our approach to placing these finds contextually. Together with the IAA collection from 11Q, the number of textiles within this corpus is now significantly closer to the number of manuscripts. Indeed, we can say that there are over 300 or as many as 400 fragments and “complete” cloths, whereas there is still much more that needs to be studied in Amman. In other words, there are more linen, jar covers and wrappers to protect the scrolls than scholars had previously supposed. The cloths found in the caves often belonged to fine garments. The socio-economic class of those who wore these garments must have been rather high. The pattern and means of decoration using indigo dyes are the result of a deliberate choice and can be compared with the rich Hellenistic or Roman use of poikilon colors that were used to demonstrate the personal status of high-ranking families. Indigo is an expensive pigment, and flax requires a process of dyeing by which the fabric is dyed twice for it to obtain a permanent deep blue (cf. Pliny the Elder). Orit Shamir and Naama Sukenik have noted that the original garments were of fine quality and were sometimes bleached.26 The exclusive use of white linen echoes historical sources which describe the “Essene dressed in white.” But are these sources reliable, and must we have confidence in writers who have never mentioned a principal Essene settlement, except the famous infra hos? That is to say, do the textiles from all the caves except 8Q establish a connection with an Essene settlement in the khirbeh? It is true that no weaving device or loom-weight has been found in the ruins. The only evidence of textile production is the spindle-whorl found in 3Q and at least another one in Ain Feshkha. It would have taken a consid3.4

26  Personal communication from Naama Sukenik and Orit Shamir.

135

The Unpublished Textiles From The Qumran Caves

erable amount of time to select the right type of linen with the dimensions required to compile the corpus of textiles suitable for a long-lasting deposit inside such “safe-caves.” If the scrolls appear to come from different places or social class, it is not the case with the textiles. The characteristics of the textiles are homogenous, but why is this so? Is this because the custom at the time was to use linen wrappers to store and protect the manuscripts? Or was it common practice in the first century CE to store and hide away books in caves, in the hope of a better future, waiting for the right moment to recover them when the danger has passed? From this perspective, every refugee fleeing the Roman army could have deposited his or her belongings in a cave. There is nothing specifically Essene or even religious about these materials either. Even if the Essenes happened to be in the surrounding area, it is likely that others too made their contribution to saving books. Nobody watching the Qumran area from the top of the watchtower could ever ignore what was happening below in the valley or in the surrounding slopes. Moreover, the inhabitants of the buildings must also have been aware of the arduous climb that was required to carry these heavy jars and manuscripts up to the caves. It is hard to imagine that such a risky climb could be undertaken at night-time, although Peleg and Magen consider that this was possible: It (the Qumran cliffs) is the last spot where they [refugees] could hide their scrolls before descending to the shore. The complete lack of order in the way the scrolls were hidden in the various caves (. . .) ­indicates that concealing the scrolls was not an orderly project undertaken by members of the sect, but rather a random, hasty act, probably performed at night.27 We disagree with Magen and Peleg in this respect. Neither have they checked what was written on the Manuscript caves nor have they paid enough attention to the materials found in the “small caves.” Not only the textiles and cave GQ29, but also the “orderly” caves, provide us with sufficient evidence to be able to undermine their arguments. Just before and during the revolts, the manuscripts were placed in safe keeping. Some of the Bar Kokhba followers could have come back and recovered their scrolls,

27   Yitzhak Magen and Yuval Peleg, The Qumran Excavations 1993–2004: Preliminary Report (JSP 6; Jerusalem: Staff Officer of Archaeology, Civil Administration of Judea and Samaria/Israel Antiquities Authority, 2007), 64–65.

or at least some of them, when they hoped they could succeed in their fight against Romans.28 4 Conclusion 4.1 Caves—Empty or Emptied? The hapax found in 1Q and its parallel found in the Amman jar, and the pieces repaired and kept aside in the Crowfoot box, all provide a strong link between the Amman Collection and 1Q. If these textile fragments were not found in 1Q, we have to reconsider the possibility that the “empty” or “minor” caves were also originally Manuscript caves containing the same “Qumran type” linen cloths but which had been “emptied” and labelled “empty,” simply because de Vaux had not found any scrolls or scroll fragments in them. This is the case for GQ29. Located on a very difficult and dangerous slope, the cave has a very small entrance. The jars were found in a part of a cave that was accessed only after “wriggling” and crawling along a channel nearly 2 metres long. If they had still been intact, the jars would have nearly reached the roof of the cave. They were found placed together with a lamp and a linen cloth which was still folded, but contained no text. Why undertake such a difficult climb only to hide empty jars? The same question can be asked for all the caves where pottery and linen were found. It would have been unlikely for the jars to have been empty, when cave GQ29 was sealed with a stone slab that Milik found against the wall of the tunnel. Half a dozen lids were found in a pile. Milik described “the cave floor . . . as covered with a thick layer of dust that had accumulated over the years, but here and there he could see ‘slight protuberances.’ As he began to brush away some of the sand with his hand he realized that he had seen a very large number of unbroken scroll jars that were neatly arranged but were, much to his disappointment, empty. As he cleaned away more dirt he also saw a neat stack of jar covers placed to one side. The whole scene impressed him: someone had respectfully emptied the jars and preserved the lids, and this had been done a very long time ago indeed, as shown by the depth of the dust that had accumulated . . . The care with which the jars had been treated showed as much respect for the contents as for the receptacles (as indicated in the Talmud).”29 28  Coins from the second Jewish revolt, found in hoards in the khirbeh, substantiate this hypothesis. 29  Weston W. Fields, The Dead Sea Scrolls: A Full History. Vol. 1: 1947– 1960 (Leiden: Brill, 2009), 136–37.

136 He was the first to imagine—and which concurs with the evidence—that the “minor” caves had indeed been the Manuscript caves. Our study of the linens that have been recovered strengthens his educated guess. Grace Crowfoot reports that she received “packs of folded cloths.” Among them there was one piece of linen, No. 30, which was photographed still unfolded. Once unfolded and flattened out, it appeared that No. 30 was devoid of any text. One side had been neatly cut and a small triangular piece of linen had been removed from the cloth. The reason cannot be anything else than to allow the content (a scroll) to be “freed” from its wrapper, without damaging the leather throng sewn to the hem of the linen. In at least two of the caves, one natural, the other artificial, a fascinating fragment was discovered, showing leather sewn onto a linen wrapper as a means of giving better protection to a scroll.30 The leather fastening on the 4Q fragment is preserved only along 4 cm. I had the idea of checking the other textiles to see whether it was possible to recognize the cut after a small part of the edge had been removed. A neat slit was made and the consequent damage limited to a minimum. In 1Q, cloth No. 30 shows a similar cut, but is triangular in shape. At least we can estimate that someone knew where they were, and came back to empty the jars, without breaking them. This could have been the case for Timothy’s Cave, GQ29, and others like it. De Vaux imagined “une violation antique.” A burglar or looter does not leave a place he plunders in good order. These caves were visited at least twice by the same group or individual, who returned after whatever danger had passed. They could have been religious enough to have treated with great respect the wrapper of the texts which they wanted to take with them. 4.2 A Quick or Slow Hiding Scenario? The mere presence of textiles, together with jars and manuscripts, bears de facto witness to a “slow-hiding scenario” of the scrolls, a process that took place over an extended period of time and was completed in different phases. The caves in which the “complete storage plan” is documented are rare: IQ and 4Q, and the caves(s) from which the Amman Collection originates. It is axiomatic that if linen was found in a cave, then this cave must also have 30  (Calibrated age: 1σ = 160–41 BCE 2σ = 193 BCE–11 CE—A. Baginski).

Bélis

contained scrolls. The blue-striped cloths were deliberately woven for a specific size of scroll and suitable for its height. Some cloths were attached to a manuscript matching in size, with a leather thong that was sewn onto it. Worn-out garments were cut to the required shape, their edges hemmed and their corner(s) sewn around with thin string so that they could be then attached to the lugs of the handled jars; some were made into thick and waterproof stoppers and placed under the lid of the “Qumrantype” jars. Some of the “small natural caves” were reshaped (carved) inside or a low wall built, behind which a hole was dug and then filled up with objects. This would have taken a tremendous amount of time, unless we can imagine that there was a workshop that was specialized in the production of such fabrics. This, however, is pure speculation because the “Qumran linen with blue” was never unearthed anywhere else in the area around the Dead Sea except in the Qumran caves. However, if a potential threat of danger increased, or if a number of scrolls that were to be hidden exceeded the number of available textiles and pottery, the last remaining cachettes (hiding places) might have been filled haphazardly, emergency. Nevertheless, despite the hurry, no other textile materials were used for this except for linen. Not a single piece of woolen cloth has been found. This was not because of religious proscriptions, but for pragmatic reasons: wool would have been quickly eaten by vermin, thus endangering the contents instead of protecting them. It is likely then, that both a “slow and rapid deposit scenario” took place, one after the other. On a final note: At the end of our presentation in Lugano, a priest came up to me and thanked me for telling his family’s personal history. He went on to explain that as a child, he and his family had had to flee the war that was raging at that time, in a country somewhere in Asia, and search for a safe place to hide their belongings. They chose a small cave in a cliff side to hide their goods, in hope of coming back later when times were better times. They did so years later but could not recognize their cave among so many crevices and rock holes. However, another family, who had hidden their goods in the same way, found their cave. They read the name of the priest’s family, and brought the family here to reclaim all their belongings that they had left years before. This is the same story that the “minor caves” also tell us.

CHAPTER 9

Miscellaneous Artefacts from the Qumran Caves: An Exploration of their Significance* Dennis Mizzi The Qumran caves have practically become synonymous with the Dead Sea Scrolls. Most scholarship on the caves focuses on the history of deposition of the Scrolls and their distribution pattern within the different caves.1 The pottery—especially the cylindrical jars—and the textiles have also received some attention.2 This paper deals with * This is an abridged version of the paper presented at the conference entitled The History of the Caves of Qumran, which was held in Lugano in February 2014; originally, the paper was entitled “NonCeramic/Textual Artefacts from the Qumran Caves.” I would like to thank Marcello Fidanzio for his invitation to this conference. I also express my gratitude to Daniel Stökl Ben Ezra for his comments and suggestions. 1  The literature on this topic is, of course, substantial. For some recent works, see, for example, Daniel Stökl Ben Ezra, “Old Caves and Young Caves: A Statistical Reevaluation of a Qumran Consensus,” DSD 14 (2007): 313–33; Stephen J. Pfann, “Reassessing the Judean Desert Caves: Libraries, Archives, Genizas and Hiding Places,” BAIAS 25 (2007): 147–70; Florentino García Martínez, “Cave 11 in Context,” in The Dead Sea Scrolls: Texts and Context (STDJ 90; ed. C. Hempel; Leiden: Brill, 2010), 199–209; idem, “Reconsidering the Cave 1 Texts Sixty Years After Their Discovery: An Overview,” in Qumran Cave 1 Revisited: Texts from Cave 1 Sixty Years after their Discovery: Proceedings of the Sixth Meeting of the IOQS in Ljubljana (STDJ 91; ed. D. W. Parry and D. K. Falk; Leiden: Brill, 2010), 1–13; Joan E. Taylor, “Buried Manuscripts and Empty Tombs: The Qumran Genizah Theory Revisited,” in ‘Go Out and Study the Land’ ( Judges 18:2): Archaeological, Historical and Textual Studies in Honor of Hanan Eshel (JSJSup 148; ed. A. M. Maeir, J. Magness, and L. H. Schiffman; Leiden: Brill, 2012), 269–315; Mladen Popović, “Qumran as Scroll Storehouse in Times of Crisis? A Comparative Perspective on Judaean Desert Manuscripts Collections,” JSJ 43 (2012): 551–94. 2  The pottery from the Qumran caves is frequently discussed in general studies on the archaeology of Qumran. For studies on the textiles, see Mireille Bélis, “Des textiles: Catalogue et commentaires,” in Khirbet Qumrân et ‘Aïn Feshkha. Vol. 2: Études d’anthropologie, de physique et de chimie (NTOA.SA 3; ed. J.-B. Humbert and J. Gunneweg; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 207–76; Orit Shamir, “Textiles and Garments from Qumran: Chalcolithic and Roman Period,” in Bio- and Material Cultures at Qumran: Papers from a COST Action G8 Working Group Meeting held in Jerusalem, Israel on 22–23 May 2005 (ed. J. Gunneweg, C. Greenblatt, and A. Adriaens; Stuttgart: Fraunhofer IRB Verlag, 2006), 285–96; Orit Shamir and Naama Sukenik, “Qumran Textiles and the Garments of Qumran’s Inhabitants,” DSD 18 (2011): 206–25.

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_011

the rest of the material artefacts, which have largely been neglected in studies of the Qumran caves. Admittedly, these artefacts are quite meagre, which partly explains why they rarely attract any scholarly attention. Nonetheless, as this paper will illustrate, this body of material is of some significance to the study of the Qumran caves and, accordingly, it should not be ignored. For the purposes of this paper, these “other” artefacts are collectively and conveniently referred to (for lack of a better term) as miscellaneous artefacts or finds.3 This paper offers a presentation of the miscellaneous artefacts from the Qumran caves with the aim to make these data more accessible and more widely known. Moreover, because a comparative approach helps the data to be more clearly defined, these artefacts are collectively analyzed within the context of other caves in the Judaean region. Finally, a discussion is undertaken which outlines the contribution of these objects to the understanding of the history of the Qumran caves, with a focus on the late Second Temple Period. 1

Caves in the Region of Qumran

The term Qumran Caves requires proper definition as it could have both geographic and conceptual connotations and, sometimes, these two notions clash. The Qumran caves can be defined as caves situated in the geographic region of Qumran—how ever this is demarcated (see further below)—or as caves that might have had a special connection with the settlement at Qumran, a connection which other caves in the same region might not have had. 3  In this work, this term incorporates glass, stone, wooden, and metal vessels, tools and weapons, jewellery, clothing articles, small implements of various kinds, and coins. Therefore, it only incorporates human-made objects (i.e. artefacts), and excludes biofacts/ecofacts (e.g. food remains, hearths, ash layers, etc.), even when the presence of these in the caves was the result of human activity. As a term, miscellaneous artefacts (or finds) is preferable than small finds since the latter gives the false impression that such finds are of little significance; moreover, it would be hard to justify the labelling of some objects (e.g. glass, wooden, stone, and metal vessels) as small finds.

138 If there are indeed caves, in the area of Qumran, which had no connection whatsoever with the settlement, we would have a situation where the same term—Qumran Caves— is used with reference to different conceptual entities, which might be confusing. Consequently, an important endeavour in the quest to understand the history of the Qumran caves is to identify and distinguish between caves that have different histories and, to this end, the miscellaneous artefacts can contribute. However, this can only be done after the caves are analyzed as one entity, that is, as a series of caves within the same geographic region. For the purposes of this paper, the Qumran region may be broadly defined as the region between latitudes 124° and 131°, which covers the region north and south of the Qumran settlement at a maximum distance of 3–4 km from the settlement. The geographical scope of the Qumran region, therefore, consists of the immediate area of Wadi Qumran (at circa 127–128° latitude), the area to its north and south of Almog (up to circa 131° latitude), and the area to its south and north of Wadi el-Samara (up to circa 124° latitude), which includes the oasis of ‘Ein Feshkha. The eastern border of this region is largely defined by the shore of the Dead Sea, whereas the western border is demarcated by the limestone cliffs in which the majority of the caves have been found. This broad geographical outline is quite maximalist in its definition of the Qumran region, but it has the advantage of incorporating the area that has witnessed various surveys and investigations in connection with the study of Khirbet Qumran. Indeed, many studies (for which, see the following paragraph) have conceptualized caves in this broad geographic region as “Qumran caves.” Therefore, in order to present a line of continuity with previous studies, this paper will adopt this maximalist definition of the Qumran region. The first systematic investigation of this area was conducted by Roland de Vaux in the 1950s, in an expedition which he claimed not to have been exhaustive. During this expedition, around 270 natural caves and crevices were surveyed and investigated in the limestone cliffs to the north, west, and south of the built settlement at Qumran, although only forty of these caves contained traces of human activity.4 This same area has been partly 4  See Gerald L. Harding, “Introductory, the Discovery, the Excavation, Minor Finds,” in Qumran Cave 1 (DJD 1; ed. D. Barthélemy and J. T. Milik; Oxford: Clarendon Press, 1955), 3–7; Roland de Vaux, “La poterie,” in DJD 1:8–17; Grace M. Crowfoot, “The Linen Textiles,” in DJD 1:18–38; Roland de Vaux, “Archéologie,” in Les ‘petites grottes’ de Qumran: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de

Mizzi

r­ e-investigated by Joseph Patrich and his team5 and, more recently, by various other teams during an archaeological survey conducted in the northern part of the Judaean Desert.6 These different explorations have produced further new data concerning limestone caves in the region of Qumran. Besides the aforementioned natural caves, de Vaux and his team also investigated a number of artificially hewn marl caves in the very immediate vicinity of the buildings,7 and further marl caves have been discovered in recent archaeological explorations conducted by Hanan Eshel and Magen Broshi.8

Vaux; Oxford: Clarendon Press, 1962), 3–36; idem, “Exploration de la region de Qumrân: Rapport préliminaire,” RB 60 (1953): 540–61; idem, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973), 51. 5  See Joseph Patrich, “Khirbet Qumran in Light of New Archaeological Explorations in the Qumran Caves,” in Methods of Investigation of the Dead Sea Scrolls and the Khirbet Qumran Site: Present Realities and Future Prospects (ANYAS 722; ed. M. O. Wise et al.; New York: New York Academy of Sciences, 1994), 73–95. 6  For the results from the survey as a whole, see ʿAtiqot 41 (2002). For the investigations in the region of Qumran, see Michel Itah, Yoni Kam, and Ronny Ben Haim, “Region X: Survey and Excavations of Caves along the Fault Escarpment South of Almog Junction and West of Qalya,” ʿAtiqot 41/1 (2002): 175–86 [Hebrew], ʿAtiqot 41/2 (2002): 169–76 [English]; Yuval Baruch, Gabriel Mazor, and Debora Sandhaus, “Region XI: Survey and Excavations of Caves along the Fault Escarpment above Ḥorbat Qumran,” ʿAtiqot 41/1 (2002): 189– 98 [Hebrew], ʿAtiqot 41/2 (2002): 177–83 [English]; Fawzi Ibrahim, “Region XII: The Excavation of Cave XII/49,” ʿAtiqot 41/1 (2002): 202–5 [Hebrew], ʿAtiqot 41/2 (2002): 185 [English]; idem, “The Excavation of Cave XII/56,” ʿAtiqot 41/1 (2002): 215–16 [Hebrew], ʿAtiqot 41/2 (2002): 186 [English]; Rudolf Cohen and Yigal Yisraeli, “The Excavations of Rock Shelter XII/50 and in Caves XII/52–53,” ʿAtiqot 41/1 (2002): 207–13 [Hebrew], ʿAtiqot 41/2 (2002): 185–86 [English]; Fawzi Ibrahim and Nadav Hameiri, “The Excavation of Cave XII/61,” ʿAtiqot 41/1 (2002): 217–18 [Hebrew], ʿAtiqot 41/2 (2002): 186 [English]; Nir Tal and Gershon Oron, “Region XIII: Survey and Excavations of Caves along the Fault Escarpment South of Qumran,” ʿAtiqot 41/1 (2002): 221–27 [Hebrew], 187–93 [English]. 7  De Vaux, in DJD 3:26–31; idem, “Archéologie,” in Qumrân Grotte 4.II: I. Archéologie; II. Tefillin, Mezuzot et Targums (4Q128–4Q157) (DJD 6; ed. R. de Vaux and J. T. Milik; Oxford: Clarendon Press, 1977), 3–22; idem, Archaeology and the Dead Sea Scrolls, 52–53. 8  Magen Broshi and Hanan Eshel, “Residential Caves at Qumran,” DSD 6 (1999): 328–48; Magen Broshi and Hanan Eshel, “How and Where did the Qumranites Live?” in The Provo International Conference on the Dead Sea Scrolls: Technological Innovations, New Texts, and Reformulated Issues (STDJ 30; ed. D. W. Parry and E. C. Ulrich; Leiden: Brill, 1999), 266–73; Hanan Eshel and Magen Broshi, “Excavations at Qumran, Summer of 2001,” IEJ 53 (2003): 61–73.

Miscellaneous Artefacts From The Qumran Caves

2

Miscellaneous Artefacts from the Qumran Caves

Miscellaneous artefacts retrieved during the aforementioned expeditions are listed in Table 9.1, according to the cave in which they were found.9 It should be noted that this paper is intended to present an analysis of these objects and not a final report on them. Therefore, Table 9.1 provides only the bare essentials in terms of data. The study of this material is based on Roland de Vaux’s official inventory (published for the first time as an appendix to this volume), which I accessed at the École Biblique et Archéologique Française de Jérusalem,10 and on the various published reports on the investigations of the Qumran caves (for which see the references cited in Section 1). The data at hand are very limited—for example, illustrations are not available for a large number of the finds— and, hence, the study of this material is restricted on various fronts. An attempt has been made to establish a date for these miscellaneous artefacts, although this is highly problematic owing to the fact that the finds generally lack any stratigraphic context within the caves and many of them cannot be dated through typology. The problem is exacerbated by evidence for the multi-period use of many of the caves. As a result, a dating methodology has been established in order to categorize these finds chronologically. The principles behind this methodology cannot be explained in detail here, but the reasoning behind the dating of specific artefacts is elaborated on in Table 9.1. A cursory look at the data from the Qumran caves reveals a number of significant features. One of the most conspicuous is the fact that, from all the caves that preserved traces of human activity, only a small percentage

9  The designation of the caves is based on the systems implemented by de Vaux, in DJD 3; by Patrich, “Khirbet Qumran;” and on the system implemented in the survey of the northern part of the Judaean Desert, published in ʿAtiqot 41 (2002). In the case of de Vaux’s system, a distinction must be made between the designation that incorporates a “Q” and which is reserved for the manuscript caves, and the numerical designation implemented for both the manuscript and the non-manuscript caves; therefore, for example, Cave GQ1 is not equivalent to Cave 1Q (which is equivalent to Cave GQ14), and so on and so forth. The designation of the caves excavated by Patrich and his team is here preceded by a “P” so as to distinguish Patrich’s labels from those of de Vaux, due to the fact that Patrich applied a similar numerical system for designating the caves. 10  I would like to thank Jean-Baptiste Humbert for kindly giving me permission to access this material.

139 contained miscellaneous artefacts.11 The percentage of caves containing miscellaneous artefacts dating to the late Second Temple Period is even smaller.12 Moreover, when present, the number of miscellaneous artefacts in individual caves is extremely limited if not negligible, except for some cases, such as Caves GQ37 = P37 and 11Q. Glass, stone, and wooden vessels are exceptionally rare, and few personal items and small implements (e.g. jewellery pieces, clothing articles, spindle whorls, cosmetic and medicinal utensils, combs, etc.) and weapons and tools have been found. Similarly, basketry, textiles, pieces of leather, and biofacts/ecofacts (e.g. date and olive pits as well as other food remains, bones, hearths, ashy deposits, etc.) occur in a relatively small number of caves.13 Indeed, pottery is 11  Evidence for human activity has been detected in circa seventytwo caves in the region of Qumran (as defined above). Unfortunately, owing to space constraints, the data from these caves could not be included in Table 9.1, which lists only those caves containing objects here classified as miscellaneous artefacts. From seventy-two caves, only seventeen contained such artefacts (= 24%). 12  From seventeen caves containing miscellaneous artefacts, six have yielded material that probably dates to the late Second Temple Period (i.e. Caves GQ3, 11Q, GQ14 = 1Q, GQ17, A, and XI/16); four caves contained material that dates to the late 1st century or the early 2nd century CE, but which might date to around 70 CE (i.e. GQ37 = P37, P13 = X/35, P24, and X/51); and one cave contained objects that might date to the late Second Temple Period (i.e. Cave GQ4). See Table 9.1 for more details. 13  From seventy-two caves, twenty contained basketry, pieces of leather, or textiles (i.e. Caves GQ8 = 3Q, 4Q, 8Q, 9Q, 11Q, GQ12, GQ14 = 1Q, A, P13 = X/35, P24, X/42, X/51, X/57, X/60, XI/16, XI/18, XII/49, XII/61, XIII/2, and a cave excavated by Broshi and Eshel close to Caves 7Q–9Q). See de Vaux, in DJD 3; idem, in DJD 6; idem, “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e, et 5e campagnes,” RB 63 (1956): 533–77; Harding, in DJD 1; Crowfoot, in DJD 1; Patrich, “Khirbet Qumran;” Itah, Kam, and Ben Haim, “Region X;” Baruch, Mazor, and Sandhaus, “Region XI;” Ibrahim, “Region XII;” Ibrahim and Hameiri, “The Excavation of Cave XII/61;” Tal and Oron, “Region XIII;” Eshel and Broshi, “Excavations at Qumran.” The only basketry, pieces of leather, and/or textiles that could be dated with some certainty to the late Second Temple Period come from Caves GQ8 = 3Q, 8Q, 9Q, 10Q, 11Q, GQ12, GQ14 = 1Q, A, XI/16, a cave near Caves 7Q–9Q excavated by Broshi and Eshel, and (probably) P24.  From seventy-two caves, twelve had a layer of ash or remains of charcoal (i.e. Caves GQ1, GQ6, 11Q, GQ23, GQ24, GQ31, GQ37 = P37, P24, X/60, XI/6, XI/16, and XI/18); seven caves yielded animal bones (i.e. Caves 5Q, GQ20 = XI/20, P13 = X/35, XI/7, XI/16, XI/18, and XII/61); seven caves contained food remains (i.e. Caves GQ14 = 1Q, 8Q, 9Q, P13 = X/35, P24, XII/61, and a cave excavated by Broshi and Eshel close to Caves 7Q–9Q); and four caves had traces of hearths (i.e. Caves GQ20 = XI/20, P24, XI/14, and XIII/2). See de Vaux, in DJD 3; idem, “Fouilles de Khirbet Qumrân;”

140

Mizzi

the only type of artefact that is virtually present in every cave that contained traces of habitation. Nevertheless, the number of pottery in individual caves is typically very small, mostly consisting of a few jar fragments and, to a lesser extent, bowls, cooking pots, and jugs; in a number of instances, human activity is merely attested by the presence of undiagnostic sherds.14 In fact, although few caves contained manuscripts, quantitatively, the latter dominate the total material assemblage retrieved from all the different caves at Qumran. 3

The Miscellaneous Artefacts from the Qumran Caves in Context

Before the significance of the miscellaneous artefacts for the history of the Qumran caves can be discussed, it is important that the state of the archaeological record within the caves is assessed. This ensures the methodological integrity of the discussion in Section 4. To this end, a comparative approach is called for, since this highlights the similarities and differences between the situation at Harding, in DJD 1; Patrich, “Khirbet Qumran;” Itah, Kam, and Ben Haim, “Region X;” Baruch, Mazor, and Sandhaus, “Region XI;” Ibrahim and Hameiri, “The Excavation of Cave XII/61;” Tal and Oron, “Region XIII;” Eshel and Broshi, “Excavations at Qumran.” In many instances, it is difficult to attribute these biofacts/ecofacts to any particular period owing to the lack of stratigraphic evidence and/or to the evidence for the caves’ use in different periods. Exceptions are Caves 11Q (ashy deposit dating to the Chalcolithic Period), Cave GQ14 = 1Q (food remains associated with late Second Temple pottery), 8Q–10Q and a cave near Caves 7Q–9Q excavated by Broshi and Eshel (food remains associated with late Second Temple pottery), GQ20 = XI/20 (a hearth, animal bones, and organic material dated to the Islamic Period), P13 = X/35 (food remains and animal bones, probably dating to around 70 CE), P24 (hearths, layers of ash, and food remains, probably dating to around 70 CE), X/60 (layer of ash from the Late Roman/Byzantine Period), XI/6 (2 hearths, charcoal, and organic material associated with late Second Temple pottery), XI/16 (animal bones, charcoal, and organic material associated with Early Roman pottery), and XIII/2 (hearths dating to the Iron Age). 14  The only notable exceptions are Caves GQ8 = 3Q (fragments from 35 cylindrical jars and 26 bowl-lids as well as a few other vessels), GQ14 = 1Q (fragments from 50 cylindrical jars and 35 bowl-lids as well as a few other vessels), GQ19 = 2Q (fragments from 9 cylindrical jars and other pottery items), GQ29 (fragments from 7 cylindrical jars, 6 unidentified jars, 19 bowl-lids and other pottery items), GQ37 = P37, P13 = X/35, and P24 (various pottery vessels), XII/49 and XII/53 (various pottery vessels from different periods). See de Vaux, in DJD 3; idem, in DJD 1; Patrich, “Khirbet Qumran;” Itah, Kam, and Ben Haim, “Region X;” Ibrahim, “Region XII;” Cohen and Yisraeli, “The Excavations.”

Qumran and that at other caves in Judaea and the Judaean Desert. In this way, the nature of the corpus from Qumran can be properly defined. As noted, one of the most striking features is the overall dearth of miscellaneous artefacts. However, the significance of this feature can only be truly gauged if this corpus is collectively set within the context of similar corpora from other caves in the region. The result is that the virtual absence of miscellaneous finds from the majority of the Qumran caves becomes all the more conspicuous. Indeed, a comparison between the quantity and type of objects found in the caves at Qumran and in some sub­ terranean complexes and caves in Judaea and the Judaean Desert reveals various non-correlations. A number of the latter caves have yielded both a large number and a large variety of objects, comprising numerous and varied pottery vessels—storage jars, cooking pots, bowls, kraters, jugs, juglets, and lamps—as well as glass, stone, wooden, and metal vessels, various metal implements, weapons, tools, jewellery, clothing articles, cosmetic and medicinal utensils, spindle whorls, various textiles and basketry, coins, food remains and hearths, as well as written documents. The most pertinent examples—containing material dating to the 1st centuries BCE and CE and to the early 2nd century CE—include Cave 2 in Wadi ed-Daliyeh,15 Cave IV/12 along Jebel Abu Saraj Cliff,16 Cave VIII/28 (or 15  See Nancy L. Lapp and George W. E. Nickelsburg Jr., “The Roman Occupation and Pottery of ‘Arâq en-Na‘saneh (Cave II),” in Discoveries in the Wâdī ed-Dâliyeh (AASOR 41; ed. P. W. Lapp and N. L. Lapp; Cambridge, MA: American Schools of Oriental Research, 1974), 49–54, Pls. 24–30; Gladys D. Weinberg and Dan Barag, “Other Finds: Glass Vessels,” in Lapp and Lapp, Discoveries in the Wâdī ed-Dâliyeh, 103–5; Pls. 38–39; George W. E. Nickelsburg Jr., “Other Finds: Miscellaneous Small Finds,” in Lapp and Lapp, Discoveries in the Wâdī ed-Dâliyeh, 101–2, Pls. 37, 97–100b; Ben B. Cox, “Other Finds: Beads,” in Lapp and Lapp, Discoveries in the Wâdī ed-Dâliyeh, 105–6, Pl. 101b; Elisabeth Crowfoot, “Other Finds: Textiles,” in Lapp and Lapp, Discoveries in the Wâdī ed-Dâliyeh, 60–78. In this cave, the following artefacts were found: at least 65 bag-shaped storage jars, 16 cooking pots, 10 jugs, 6 bowls, and 2 lamps, as well as 19 glass vessels, 3 wooden bowls, 1 bronze jug handle, 2 wooden combs, 1 kohl stick, 1 pin, 90 stone beads, 1 key, 1 blade, 1 iron pike, 1 bronze coin, and pieces of linen. These finds have been dated to the early 2nd century CE, and the cave interpreted as a refuge cave which probably housed a large number of people. 16  Samuel Wolff, “The Excavation of Cave IV/12,” ʿAtiqot 41/2 (2002): 73–83; Yael Gorin-Rosen, “Glass,” in Wolff, “The Excavation of Cave IV/12,” 81; Donald T. Ariel, “The Numismatic Finds,” in Wolff, “The Excavation of Cave IV/12,” 80. From this cave, the followed artefacts, dating to the Early Roman Period, were retrieved: 9 bag-shaped jars, 3 cooking pots, 2–4 jugs, 2 flasks, 3 juglets, 1

Miscellaneous Artefacts From The Qumran Caves

141

the so-called “Cave of the Sandal”) in Ketef Jericho,17 the so-called “Abi’or Cave” in Ketef Jericho,18 the so-called “Magharat el-Jai Cave” in Na�al Michmash,19 Cave 1 in Wadi Murabbaʿat,20 Cave 2 in Wadi Murabbaʿat,21 and

various caves in Wadi Murabbaʿat,22 the so-called “Cave of the Pool” in Na�al David,23 the so-called “Har Yishai Cave” in Na�al David,24 the so-called “Cave of the Treasure” in Na�al Mishmar,25 the so-called “Cave of Letters” in Na�al

krater, 9 bowls, and 2 lamps, as well as various pieces of glass vessels, 1 kohl stick, 1 bone object, 1 iron implement, 1 piece of lead, and 2 coins. 17   Hanan Eshel and Boaz Zissu, “Excavations and Surveys: Ketef Jericho, 1993,” IEJ 45 (1995): 292–98; idem, “Jericho: Archaeological Introduction,” in Miscellaneous Texts from the Judaean Desert (DJD 38; ed. J. H. Charlesworth et al.; Oxford: Clarendon Press, 2000), 3–20 (14–18), table 5, Figs. 13, 14, 15:5–7; Yael Gorin-Rosen, “The Glass Vessels from Cave VIII/28,” ʿAtiqot 41/2 (2002): 143–45; Victoria Ladizhinskaya, “Jewelry and Metal Artefacts of Middle Bronze Age II and the Early Roman Period from Cave VIII/28,” ʿAtiqot 41/2 (2002): 147–55. The material from this cave, which was dated to the early 2nd century CE, comprises: jars, cooking pots, 1 jug, 1 bowl, and lamps, as well as 4 glass vessels, 1 needle, 2 gold finger rings, 1 gold earring, 1 silver spoon, various arrowheads, 26 coins, pieces of leather sandals, fragments of leather, ropes, strings, mats, basketry, and 2 papyri. 18  Eshel and Zissu, “Jericho,” 5, 12, 13–14, 18, 19, table 4, Figs. 12:3–21, 15:2, 3–4, 10–13. From this cave, the following items were discovered: 4 jars, cooking pots, jugs, 2 bowls, and 1 lamp, as well as glass vessels, 1 wooden comb, 1 needle, nails, 1 functional ring, pieces of linen, ropes, leather sandals, food remains, and papyri. The cave also contained the skeletal remains of 38 individuals. The excavators have dated this material to the early 2nd century CE. 19  Hanan Eshel and Boaz Zissu, “Na�al Mikhmash, Magharat el-Jāi,” ESI 110 (1999): 75–76 [Hebrew], 56*–57* [English]; Hanan Eshel, Boaz Zissu, and Amos Frumkin, “Two Refuge Caves in Wadi Suweinit,” in Refuge Caves of the Bar Kokhba Revolt (ed. H. Eshel and D. Amit; Tel Aviv: Israel Exploration Society, 1998), 1:93–109 [Hebrew]. The objects from this cave comprise: 8 jars, 4 cooking pots, 1 lid, and 2 jugs, as well as 3 glass vessels, 2 needles, 1 buckle, 2 functional rings, 2 nails, 4 silver coins and 11 bronze coins. These finds have been dated to the early 2nd century CE. 20  Roland de Vaux, “Archéologie,” in Les Grottes de Murabbaʿât (DJD 2; ed. P. Benoit, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1961), 3–50 (29–47), Figs. 7–12, Pls. VIII–XVIII. The following artefacts were excavated from Cave 1 in Wadi Murabbaʿat: 4 bag-shaped storage jars, 2 jar stoppers, 2 juglets, 1 flask, 3 bowls, and 3 lamps, as well as 4 stone vessels, 1 spatula, 1 needle, 1 spoon, 1 bone die, 1 decorated bone button, 2 finger rings, 1 javelin, 2 arrowheads, 1 key, 2 wooden tablets, 2 bronze coins, many sandals, 1 phylactery case, and manuscript fragments. The finds have been dated to the early 2nd century CE. As far as the manuscripts are concerned, these were originally ascribed to the First Jewish Revolt by Milik, but re-dated to the Bar Kokhba Period by Eshel. See Józef T. Milik, “Textes hébreux et araméens,” in DJD 2:67–205; Hanan Eshel, “Murabbaʿat, Wadi: Written Material,” EDSS 1:583–86. 21  De Vaux, in DJD 2:29–47, Figs. 7–12, Pls. VIII–XVIII. The finds from Cave 2 in Wadi Murabbaʿat include: 1 bag-shaped storage

jar, 1 jar stopper, 1 cooking pot, and 1 casserole, as well as 1 weight, 2 javelins, 1 blade, 1 bone blade, 1 hook, 2 nails, many sandals, and manuscript fragments, all of which have been dated to the early 2nd century CE. 22  De Vaux, in DJD 2:29, 39, 41, 43, 44–45, Figs. 8–12, Pls. XII–XIV; Ephraim Stern, “Murabbaʿat, Wadi: Archaeology,” EDSS 1:581–83; Grace M. Crowfoot and Elisabeth Crowfoot, “The Textiles and Basketry,” in DJD 2:51–63. There are a number of objects the provenance of which is unknown, either because they were acquired from the Bedouin or because their provenance is not specified in the report. Therefore, these finds could come from any of the caves in Wadi Murabbaʿat. All these finds are here grouped together, and they include: 1 juglet and 4 bowls, as well as 2 needles, 1 lancet, 1 medicinal or cosmetic stick, 2 finger rings, 1 decorated bone spoon, 1 decorated bone button or spindle whorl, 1 wooden seal, 11 spindle whorls, 1 wooden spindle, 4 spindles, 1 wooden comb, various wooden implements, 2 javelins, 2 arrowheads, 1 chisel, 1 blade, 1 knife, over 200 silver coins, pieces of linen and woollen textiles, and manuscript fragments. In addition, 2 medicinal or cosmetic sticks and 2 blades are specifically identified as coming from Cave 3, and further manuscript fragments are reported to come from Cave 5. All the aforementioned objects have been dated to the early 2nd century CE. 23   Nahman Avigad, “Expedition A,” IEJ 11 (1961): 6–10; idem, “Expedition A—Na�al David,” IEJ 12 (1962): 169–83. The ­following artefacts were excavated from the “Cave of the Pool”: 3 jars, 2 pithoi, 4 cooking pots, 1 juglet, 1 Nabataean sherd, and lamps, as well as 14 glass-vessel fragments, 2 wooden combs, 1 needle, 2 wooden spindle whorls, 1 bead, arrowheads, 2 bronze coins, food remains, and woven fabrics consisting of flax and wool. The material from this cave has been dated to the early 2nd century CE. 24  Roi Porat, Hanan Eshel, and Amos Frumkin, “Finds from the Bar Kokhba Revolt from Two Caves at En Gedi,” PEQ 139 (2007): 35–53. This cave contained the following artefacts: 6 jars, 3 cooking pots, 1 casserole, 1 jug, and 1 lamp, as well as 2 glass vessels, 1 stone vessel, 12 arrowheads, arrow shafts, 11 bronze coins, pieces of woven textiles and leather, and 2 papyri. These are all dated to the early 2nd century CE. 25  Pesach Bar Adon, “Expedition C,” IEJ 11 (1961): 25–35; idem, The Cave of the Treasure: The Finds from the Caves in Naḥal Mishmar (Jerusalem: Israel Exploration Society, 1980), 2–3, 205–11. The items from this cave include: jars, numerous cooking pots, many jug fragments, many juglet fragments, bowls, and many lamp fragments, as well as many glass-vessel fragments, stone vessels, 1 buckle, 1 hook, 1 strip, 1 lock, grinders and mortars, food remains, animal bones, hearths, basketry, clothing, and papyri. This rich assemblage has been dated to the early 2nd century CE.

142 Ḥever,26 the so-called “Cave of Horror” in Na�al Ḥever,27 the so-called “Cave of the Tetradrachm” in Na�al Ḥever,28 the subterranean complexes at Tel Goded,29 the hiding complex at Modi‘in,30 and the hiding complex at ʿAin Arrub.31 26  Yohanan Aharoni, “The Caves of Na�al Ḥever,” ʿAtiqot 3 (1961): 148–62; Yigael Yadin, “Expedition D,” IEJ 11 (1961): 36–52; idem, “Expedition D—The Cave of the Letters,” IEJ 12 (1962): 227–57. The objects from this large cave consist of the following: jars, pithoi, cooking pots, and jugs, as well as glass vessels, wooden vessels, 22 metal vessels, 1 wooden pyxis, 1 comb, 2 mirrors, 1 spindle, 2 spindle whorls, 1 seal, several beads, 1 key, pieces of furniture, 1 awl, 1 handle, 3 large knives, 3 knives, 2 sickles, arrowheads, 1 coin, food remains, pieces of leather, shoes, sandals, pieces of coloured textiles, baskets, ropes, unspun wool, coloured wool skeins, 1 large net, pieces of cloth with geometric designs, and various manuscript fragments. This rich assemblage has been dated to the early 2nd century CE. 27  Aharoni, “The Caves of Na�al Ḥever”; idem, “Expedition B— The Cave of Horror,” IEJ 12 (1962): 186–99; Levi Y. Rahmani, “The Coins from the Cave of Horror,” IEJ 12 (1962): 200; Hanan Eshel, “Ḥever, Na�al: Archaeology,” EDSS 1:357–59. The finds from this cave include: jars, cooking pots, casseroles, jugs, and lamps, as well as wooden objects, spindle whorls, wooden sticks, gaming counters, 7 wooden combs, needles, awls, 1 knife, 1 axe, nails, arrowheads, 4 bronze coins, pieces of leather, textiles, basketry, ropes, sandals, shoes, hairnets, numerous fabrics, manuscript fragments, and the skeletal remains of at least 40 individuals. The occupation of this cave has been dated to the early 2nd century CE. 28  David Amit and Hanan Eshel, “Bar Kokhba Period Finds from the Tetradrachm Cave,” in idem, Refuge Caves of the Bar Kokhba Revolt, 189–204 [Hebrew]; idem, “A Tetradrachm of Bar Kokhba from a Cave in Na�al Ḥever,” INJ 11 (1993): 33–35; Hanan Eshel, personal communication (January–February 2008). This cave yielded the following objects: 70 jar rims, 20 cooking pots, jugs, bowls, and lamps (with a total pottery count of circa 120 fragments), as well as many glass-vessel fragments, arrowheads, various metal objects, and 1 tetradrachm. All these have been dated to the early 2nd century CE. 29  Nahum Sagiv, “Tel Goded,” ESI 14 (1995): 112–14. This subterranean complex contained: jars, cooking pots, jugs, juglets, 1 bottle, kraters, and eastern terra sigillata vessels, as well as glass vessels, stone vessels, 1 fibula, and nails. These finds have been dated to the 1st–2nd centuries CE. 30  Keren Nahmias and Tzahi Gal, “Modi‘in,” ESI 111 (2000): 63–64 [Hebrew], 49*–50* [English]. This hiding complex yielded the following objects: 3 jars, 4 cooking pots, 2 jugs, and bowls, as well as fragments of glass vessels, grinding tools, and coins, all of which have been dated to the Late Hellenistic and Early Roman Period. 31  Yoram Tsafrir and Boaz Zissu, “A Hiding Complex of the Second Temple Period and the Time of the Bar Kokhba Revolt at ʿAinʿArrub in the Hebron Hills,” in The Roman and Byzantine Near East: Volume 3 (JRASS 49; ed. J. H. Humphrey; Ann Arbor, MI:

Mizzi

In this light, the material remains recovered from the large majority of the Qumran caves look meagre indeed, and the dearth of miscellaneous finds is especially striking. Nevertheless, this dearth of artefacts may possibly be the result of natural and cultural formation processes that have affected the archaeological record in the Qumran region; accordingly, it is highly probable that the evidence from the Qumran caves is incomplete and distorted, and this might explain its limited nature. Thus, for example, the western part of Cave 10Q as well as the southern part of Cave 7Q and part of its floor collapsed into Wadi Qumran; almost two thirds of Cave 9Q have disappeared, and only a part of Cave 8Q survives today.32 This means that material from these caves has spilled out into the wadi below, and that it is lost to us. Parts of Caves 4Qa and 4Qb have also subsided, with material spilling over into the ravine.33 Furthermore, many of the Qumran caves might have also been looted, either in antiquity or in more recent times. For example, it is almost certain that the Period III occupants of Qumran would have visited and disturbed the contents of some of the nearby caves.34 Explicit evidence for post-68 CE activity can be found in Cave 4Q, in which a late 1st century CE round lamp has been found35 and evidence for the re-use of some of the cave’s manuscripts in the early 2nd century CE has been identified.36 In addition, the northern marl caves could have been disturbed by Roman soldiers, whose activity in this area is indicated by the presence of numerous sandal nails,37 and one must not forget that some of the limeJournal of Roman Archaeology, 1995), 7–36. The objects from this hiding complex comprise: jars and cooking pots, as well as glass vessels, stone vessels, spindle whorls, jewellery, and coins, all of which have been dated to the 1st century CE. 32  See de Vaux, in DJD 3:29–31. 33  See de Vaux, in DJD 6:9, 13. 34  See also Joan E. Taylor, The Essenes, the Scrolls, and the Dead Sea (Oxford: Oxford University Press, 2012), 293, 298–300; eadem, “Buried Manuscripts” 292, 298–303. 35  See de Vaux, in DJD 6, Fig. 6:10. 36  See Hannah M. Cotton and Erik W. Larson, “4Q460/4Q350 and Tampering with Qumran Texts in Antiquity?” in Emanuel: Studies in Hebrew Bible, Septuagint, and Dead Sea Scrolls in Honor of Emanuel Tov (VTSup 94; ed. S. M. Paul et al.; Leiden: Brill, 2003), 113–25. But see Mladen Popović, “Roman Book Destruction in Qumran Cave 4 and the Roman Destruction of Khirbet Qumran Revisited,” in Qumran und die Archäologie: Texte und Kontexte (WUNT 278; ed. J. Frey, C. Claussen, and N. Kessler; Tübingen: Mohr Siebeck, 2011), 239–91 (241–49). 37  Joan E. Taylor, personal communication (August 2010). Broshi and Eshel, “Residential Caves,” 340 suggest that these nails are remnants of the Qumranites’ sandals. However, Joan Taylor has shown that the typical civilian sandal would have been

Miscellaneous Artefacts From The Qumran Caves

stone caves contained Late Roman/Byzantine and Islamic pottery, indicating that these caves (and perhaps others in the region) continued to experience disturbance and looting even in later periods. This is further corroborated by accounts of ancient manuscript discoveries near Jericho (e.g. Eusebius, Ecclesiastical History 6:16:3),38 some of which may quite possibly relate to Qumran.39 The prolific monastic and anchorite activity in the Judaean Desert, during the Byzantine Period, could have been the source of further disturbance.40 Furthermore, the Qumran caves have also experienced significant disruption in modern times; indeed, all the teams that have investigated the caves in this region have detected and noted clear evidence of modern looting.41 Therefore, we should bear in mind that, as far as the Qumran caves are concerned, we are dealing with a corpus of material that has been considerably affected by various natural and cultural formation processes. On the other hand, it should be pointed out that some of the above mentioned caves from the Judaean Desert have also been looted and/or re-used in later periods and, yet, a large quantity and a wide range of objects were still found within them. Examples include Cave IV/12 along the Jebel Abu Saraj Cliff,42 “Abi’or Cave” in Ketef Jericho,43 without nails and that the type of sandals with nails were distinctive of the Romans or other military personnel; see Joan E. Taylor, “Khirbet Qumran in Period III,” in Qumran, the Site of the Dead Sea Scrolls: Archaeological Interpretations and Debates; Proceedings of a Conference held at Brown University, November 17–19, 2002 (STDJ 57; ed. K. Galor, J.-B. Humbert, and J. Zangenberg; Leiden: Brill, 2006), 133–46 (140–41). 38  See further Godfrey R. Driver, The Judaean Scrolls: The Problem and a Solution (New York: Schocken Books, 1965), 7–15. 39  See, for example, Hartmut Stegemann, The Library of Qumran: On the Essenes, Qumran, John the Baptist, and Jesus (Grand Rapids, MI: Eerdmans, 1998), 68–71, 76–77; Taylor, The Essenes, 279–80, 298–300; eadem, “Buried Manuscripts,” 278–79, 298–303. 40  For a detailed discussion, see Taylor, The Essenes, 298–300; eadem, “Buried Manuscripts,” 298–303. 41  See de Vaux, in DJD 3, passim; Harding, in DJD 1:6; Broshi and Eshel, “Residential Caves,” 340–41; Itah, Kam, and Ben Haim, “Region X,” passim; Baruch, Mazor, and Sandhaus, “Region XI,” passim; Cohen and Yisraeli, “The Excavations of Rock Shelter XII/50,” 185. Also see Taylor, The Essenes, 277–78. 42  In addition to the Early Roman material, Cave IV/12 also contained evidence for its use in the Middle Bronze Age II and the Mamluk-Ottoman Period. See Wolff, “The Excavation of Cave IV/12.” 43  The excavators remark that the deposit within this cave was ­disturbed, and the cave seems to have experienced activity in the Mamluk Period as well as in the mid-20th century. See Eshel and Zissu, “Jericho;” also see Popović, “Qumran as Scroll Storehouse,” 557.

143 “Magharat el-Jai Cave” in Na�al Michmash,44 and the caves in Wadi Murabbaʿat.45 In the so-called “Large Cave Complex” (Cave VIII/9), in Ketef Jericho, pottery vessels, sixteen fragments of glass vessels, a bronze ring, three coins, and papyri fragments, all dating to the period between the 1st century and the early 2nd century CE were found,46 despite the fact that this cave complex was reused in the Byzantine Period (as attested by the later pottery) and that it was accessible.47 The excavators suggest that the reason Cave VIII/9 is devoid of any other finds is looting, facilitated by the cave’s accessibility; however, the example of the “Large Cave Complex” also illustrates that later occupations within a cave or looting do not obliterate all traces of the objects that were once hidden or deposited there. There are other similar examples from the Judaean Desert, such as Cave III/8 in Lower Wadi elMakkuk48 and Cave VI/5249 along the Jebel Abu Saraj Cliff. 44  The excavators state that the cave was looted, and that most of the finds were found near the entrance of the cave. Moreover, the cave yielded artefacts from various periods, indicating its multi-period use. See Eshel and Zissu, “Na�al Mikhmash, Magharat el-Jāi.” 45  The caves in Wadi Murabbaʿat appear to have also been visited during the Islamic Period, as attested by pottery and coins from that period. Moreover, the caves were looted by Bedouin when they were discovered in the 1950s. See de Vaux, in DJD 2; idem, “Les grottes de Murabba⁠ʾat et leurs documents,” RB 60 (1953): 245–67 (260). Also see Popović, “Qumran as Scroll Storehouse,” 557–59. 46  Hanan Eshel and Boaz Zissu, “The Excavation of Cave VIII/9 (‘The Large Caves Complex’),” ʿAtiqot 41/1 (2002): 151–68 [Hebrew], ʿAtiqot 41/2 (2002): 125 [English]; idem, “Jericho,” 4, 12; Ruth E. Jackson-Tal, “The Glass Finds from Cave VIII/9 (‘The Large Caves Complex’),” ʿAtiqot 41/1 (2002): 167–68 [Hebrew], ʿAtiqot 41/2 (2002): 127 [English]. 47  Eshel and Zissu, “Jericho,” 4, 12. 48   Dror Barshad and Idan Shaked, “Region III: Survey and Excavations of Caves in Lower Wadi el-Makkuk,” ʿAtiqot 41/1 (2002): 33–41 [Hebrew], ʿAtiqot 41/2 (2002): 35–41 [English] (36–37 [English]); Orit Shamir and Alisa Baginski, “The Later Textiles, Basketry and Cordage from Caves in the Northern Judean Desert (‘Operation Scroll’),” ʿAtiqot 41/2 (2002): 241–55 (241, 243). No traces of looting were discovered in this cave, but evidence that it has been visited in the past was detected. Nonetheless, numerous objects were still unearthed, including pottery from the Chalcolthic, Hellenistic, Early and Late Roman Periods, 1 glass vessel, 1 possible weight, 2 hooks, 1 metal pin, 1 stone bead, 2 glass beads, 11 bronze coins, food remains, pieces of wood, 12 undyed woollen textiles, 10 undyed linen textiles, 2 undyed woollen threads, and basketry. 49   Ofer Sion, “Regions IV and VI: Survey and Excavations of Caves along the Jebel Abu Saraj Cliff,” ʿAtiqot 41/1 (2002): 45–89 [Hebrew], ʿAtiqot 41/2 (2002): 43–70 [English] (62–63); Ruth E. Jackson-Tal, “The Glass Vessels,” in Sion, “Regions IV and VI,”

144 Therefore, natural and cultural formation processes might not be adequate to explain the relatively meagre archaeological record and the overall dearth of miscellaneous artefacts within the Qumran caves; the virtual absence of miscellaneous finds is unlikely to be simply the result of disturbance and/or looting.50 Accordingly, it seems that the archaeological record of the caves in the Qumran region, despite its fragmentary and haphazard nature, still reflects a somewhat reliable picture as to what type of artefacts were originally deposited in the caves. This is further corroborated by the overwhelmingly consistent pattern that emerges from the finds in all of them.51 Therefore, we are now in a better ʿAtiqot 41/1 (2002): 78 [Hebrew], ʿAtiqot 41/2 (2002): 63 [English]. In this cave, pottery sherds (especially jar fragments), 13 fragments of glass vessels, 1 faience button, metal objects, animal bones, and hearths were found, dating to the 1st century CE and the early 2nd century CE, and Islamic textiles. Despite the fact that the cave was re-visited in the Islamic Period and despite evidence for modern looting, relatively substantial traces of miscellaneous artefacts and other finds were still detected. 50  This scarcity of miscellaneous artefacts is also unlikely to be the result of shortcomings in the methodology used in the 1950s. Recent surveys of the caves in the Qumran region have similarly produced very few objects classifiable as miscellaneous finds. 51  A limited number of caves—most notably Caves P13 = X/35, P24, GQ37 = P37, and 11Q—do not fit neatly into this pattern owing to the variety of artefacts and/or miscellaneous finds within them. Interestingly, Cave GQ37 is quite accessible, and although “the cave is located high in the slope, and the ascent is quite difficult and steep . . . no ladder or ropes are required” (Patrich, “Khirbet Qumran,” 92). Likewise, the caves in Region X, of which Cave P13 = X/35 forms a part, are easily accessible (Itah, Kam, and Beh Haim, “Region X,” 169). No information is available as to the accessibility of Cave P24; in view of its distance from Qumran, Cave P24 might fall within Region X as well—in fact, the recent investigations in Region X have surveyed caves with traces of previous excavations, and the surveyors have linked these with Patrich’s investigation of the caves (Itah, Kam, and Ben Haim, “Region X,” 169, 172)—which means that probably it is easily accessible as well. In addition, Cave P13 = X/35 also contained traces of later occupation, in the form of textiles dating to the Early Islamic Period and the Middle Ages (Itah, Kam, and Ben Haim, “Region X,” 171–72). Cave P24 was located “at the southern end of a 70 m long terrace that served until recently as a sheepfold” (Patrich, “Khirbet Qumran,” 90), which means that Cave P24 was in an area that must have experienced considerable disturbance as far as the archaeological record is concerned. As for Cave 11Q, this was certainly looted in the 1950s and now we also have evidence (in the form of textiles) of visits made to the cave in the Middle Ages and in Early Modern times (Marcello Fidanzio and Joan Taylor, personal communication). All this continues to underline the relative reliability of the archaeological record of the Qumran caves. Caves P13, P24, GQ37, and 11Q

Mizzi

position to analyze the significance of the miscellaneous artefacts for the history of the Qumran caves. 4

The Function/s of the Qumran Caves: The Contribution of the Miscellaneous Artefacts52

The question of function remains one of the central questions with regard to the caves at Qumran. On the basis of the overall material evidence, Roland de Vaux concluded that both the limestone caves and the artificially hewn marl caves, together with tents set up in the surrounding area, were used by the inhabitants of Qumran (henceforth the Qumranites) for permanent habitation and that the settlement at Qumran was largely used as a community centre.53 Hanan Eshel and Magen Broshi have advanced de Vaux’s hypothesis further by their discovery of more marl caves in the vicinity of the settlement, but they propose that only these marl caves were used for habitation in antiquity,54 a position which has largely been accepted by

do not disrupt the archaeological pattern; rather, they prove that the noted pattern is actually reliable since, despite their accessibility and their later re-use, these four caves still yielded a relatively rich and varied assemblage compared to the rest of the Qumran caves. 52  This section focuses on the use of the caves in the late Second Temple Period. Accordingly, general statements about the material remains from the Qumran caves pertain only to late Second Temple material. 53  De Vaux, in DJD 3:32–36; idem, Archaeology, 50–57. Also see Jodi Magness, The Archaeology of Qumran and the Dead Sea Scrolls (Grand Rapids, MI: Eerdmans, 2002), 69–71. 54  Broshi and Eshel, “Residential Caves;” Broshi and Eshel, “How and Where did the Qumranites Live?” Broshi and Eshel point out that most of the limestone caves in the Qumran area are either inaccessible or too small to have served as dwelling quarters. Moreover, the unbearable warmth and humidity (particularly during the summer season), the lack of proper ventilation, the unlevelled state of the caves’ floors, and their darkness make even the largest and most accessible of these caves uninhabitable in the long term (Hanan Eshel, personal communication [January 2008]; also see Broshi and Eshel, “Residential Caves,” 332). The marl caves, on the other hand, are usually better lit and ventilated, and the water-retaining capability of the marl deposits also means that there are cooler temperatures within them; in addition, unlike many of the limestone caves, the marl caves have levelled floors, they are artificially hewn, and they were easily accessible via hewn stairs within the marl terrace, rope ladders, or via pathways (Broshi and Eshel, “Residential Caves,” 332–33).

Miscellaneous Artefacts From The Qumran Caves

the scholarly community.55 Eshel and Broshi also suggest that the discovery of some collapsed marl caves in the area may be indicative of the existence of more such dwelling caves in the marl deposits around the settlement,56 caves which might have collapsed as a result of the frequent earthquakes in this region and because of rains and flash floods.57 Accordingly, they estimate that between twenty and forty artificially hewn marl caves might have existed in the environs of Qumran and that between 150 and 200 individuals lived in the Qumran area, the majority of whom lived in these caves.58 However, there are a few scholars who have questioned the habitability of the caves (the marl caves included). Joseph Patrich, for example, claims that the material remains emanating from the marl caves are too meagre to constitute the remains of permanent or long-term domestic occupation and, therefore, he concludes that most of the caves were used as temporary or short-term shelters.59 55   See, for example, Sidnie White Crawford, “Qumran: Caves, Scrolls, and Buildings,” in A Teacher for all Generations: Essays in Honor of James C. VanderKam (JSJSup 153; ed. E. F. Mason et al.; Leiden: Brill, 2012), 1:253–73 (258–59). But see García Martínez, “Cave 11 in Context,” 206–7, who supports de Vaux’s original hypothesis that some of the natural limestone caves were used for permanent habitation. García Martínez’s view has been subsequently criticized by Daniel Stökl Ben Ezra, “Further Reflections on Cave 1 and 11: A Response to Florentino García Martínez,” in Hempel, The Dead Sea Scrolls, 211–23 (220) and by Taylor, The Essenes, 279, n. 30. 56  Eshel and Broshi, “Excavations at Qumran,” 68. 57  Broshi and Eshel, “Residential Caves,” 334–35. 58   Broshi and Eshel, “Residential Caves,” 334–35. In the past, various other scholars had already suggested that the number of inhabitants in the Qumran area was high, although the numbers posited had been different. See, for example, Józef T. Milik, Ten Years of Discovery in the Wilderness of Judaea (SBT 26; trans. J. Strugnell; London: SCM Press, 1959), 97; Ernest-Marie Laperrousaz, Qoumrân: L’établissement essénien des bords de la Mer Morte; Histoire et archéologie du site (Paris: A. & J. Picard, 1976), 99–107; Bryant G. Wood, “To Dip or Sprinkle? The Qumran Cisterns in Perspective,” BASOR 256 (1984): 45–60; Magen Broshi, “The Archaeology of Qumran—A Reconsideration,” in The Dead Sea Scrolls: Forty Years of Research (STDJ 10; ed. D. Dimant and U. Rappaport; Leiden: Brill, 1992), 103–15 (114). 59  Joseph Patrich, “Did Extra-Mural Dwelling Quarters exist at Qumran?” in The Dead Sea Scrolls Fifty Years after their Discovery: Proceedings of the Jerusalem Congress, July 20–25, 1997 (ed. L. H. Schiffman, E. Tov, and J. C. VanderKam; Jerusalem: Israel Exploration Society/Shrine of the Book, 2000), 720–27. Patrich also notes the impracticality of these dwellings for community members who had to return to their dwelling cave in the dark, and he questions why a community that was skilled enough to construct such an extensive and complex built-up area at

145 Yizhar Hirschfeld and Lena Cansdale likewise reject the above interpretation of the caves and conclude that they merely served as temporary living quarters for passersby, shepherds, or hermits.60 Other scholars, such as Joan Taylor, Daniel Stökl Ben Ezra, and Stephen Pfann, have emphasized the function of the caves as depositories of manuscripts and/or as workshops.61 It is not the purpose of this paper to discuss the function of the Qumran caves; rather, the intention is to highlight the contribution of the miscellaneous artefacts to this debate. These finds are particularly important for the role they play in putting the debate about the residential function of the caves into perspective and, therefore, this section consists of an evaluation of this hypothesis in light of this repertoire of objects; nonetheless, it should be noted that these finds do not (and perhaps they cannot) offer absolute conclusions. The significance of the miscellaneous finds lies in the fact that they give an unambiguous domestic dimension to a cultural assemblage. Personal items (e.g. pieces of jewellery, articles of clothing, needles, mirrors, combs, and cosmetic utensils), small implements (e.g. spindle whorls, loom weights, and grinding equipment), and tools are the vivid remnants and reflectors of daily life. Artefacts of the sort have been excavated from various late Second Temple sites, either in clear domestic contexts or in burial contexts.62 Domestic assemblages are also typically characterized by the presence of glass and stone vessels. By the late Second Temple Period, glass vessels became commonly widespread owing to the invention of

Qumran (or, at least, was skilled enough to modify and expand pre-existing structures to the level of Period II Qumran) would not build enough rooms to accommodate the entire community, and why the Qumranites would live in caves when they could easily have expanded the settlement, either to the south or to the north of the site, where ample space was available. 60  Yizhar Hirschfeld, Qumran in Context: Reassessing the Archaeological Evidence (Peabody, MA: Hendrickson, 2004), 43, 45; Lena Cansdale, Qumran and the Essenes: A Re-evaluation of the Evidence (TSAJ 60; Tübingen: Mohr Siebeck, 1997), 94–97. 61  Taylor, The Essenes, 272–303; eadem, “Buried Manuscripts,” 269– 315; Stökl Ben Ezra, “Old Caves and Young Caves;” idem, “Further Reflections,” 211–23; Pfann, “Reassessing the Judean Desert Caves.” 62  It is well known that graves say as much about the dead as they do about the living. In this way, they can be important mirror images of social reality, especially as far as personal items are concerned. Consequently, items found in burial contexts reveal important information on what type of objects would have been present in a domestic setting.

146 free-blowing;63 likewise, stone vessels—particularly stone mugs—appear to have enjoyed a very wide distribution.64 Consequently, the presence of the aforementioned finds provides a very strong indication that we are dealing with a domestic setting, in which all manner of daily activities were carried out. As far as caves are concerned, the Judaean subterranean complexes and the Judaean Desert caves mentioned in Section 3 above serve as excellent examples. It is widely acknowledged that these were refuge caves—used either during the First or the Second Jewish Revolt—which were inhabited intensively and on a permanent basis, albeit in the short-term. The ubiquitous presence of various miscellaneous artefacts65 solidly sustains their interpretation as refuge caves, which were used for dwelling purposes for a few years and in which various daily activities were performed. In many ways, these caves are like time capsules of daily life in the 1st and early 2nd centuries CE. In this context, it is important to refer to many other Judaean Desert caves which contained only a handful of pottery vessels dating either to the late Second Temple Period or the late 1st–early 2nd century CE; in addition to the limited pottery, some caves also had a very limited number of miscellaneous finds; there are also caves which yielded only coins or just a few miscellaneous artefacts, but no pottery whatsoever.66 Some of these caves have

63  For a detailed discussion of the distribution of glass vessels in late Second Temple Palestine, see Dennis Mizzi, “The Glass from Khirbet Qumran: What Does it Tell us About the Qumran Community?” in Hempel, The Dead Sea Scrolls, 99–198. 64  See Yitzhak Magen, The Stone Vessel Industry in the Second Temple Period: Excavations at Hizma and the Jerusalem Temple Mount (JSP 1; ed. T. Levana; Jerusalem: Israel Exploration Society/ Israel Antiquities Authority/Staff Officer of Archaeology—Civil Administration of Judea and Samaria, 2002). 65  These caves and complexes yielded large quantities of objects and/or a large range of finds. Besides numerous pottery vessels, a wide range of miscellaneous finds was discovered, including glass, stone, wooden, and metal vessels, tools and weapons, articles of clothing, jewellery pieces, combs and mirrors, cosmetic and medicinal utensils, needles, spindle whorls, coins, textiles, and food remains. See Section 3 above for more details. 66   See, for example, various caves in Upper Wadi el-Makkuk (Yizhar Hirschfeld and Shimon Riklin, “Region II: Survey and Excavations in the Upper Wadi el-Makkuk Caves,” ʿAtiqot 41/1 [2002]: 9–30 [Hebrew], ʿAtiqot 41/2 [2002]: 5–19 [English]), caves along the Jebel Abu Saraj Cliff (Sion, “Regions IV and VI”), and caves in Ketef Jericho (Hanan Eshel and Boaz Zissu, “Region VIII: Survey and Excavations of Caves along the Cliff Slopes of Triangulation Point 86, on the Fringes of Jebel Ma‘ar el-Bas,” ʿAtiqot 41/1 [2002]: 141–50 [Hebrew], ʿAtiqot 41/2 [2002]: 117–23 [English]).

Mizzi

also been interpreted as refuge caves67—and they may well have been used as such—but the dearth of archaeological material hinders one from reaching any confident conclusions, and other interpretations are certainly possible. Problems of interpretation especially arise in the case of caves containing pottery alone. Certainly, such caves might have been used as refuge caves; perhaps the people who sought refuge there could not take too many items with them, or perhaps later looting and/or disturbance erased most traces that would have indicated a more permanent (albeit still short-term) occupation. Nonetheless, the lack of any other finds might also be indicative of the use of these caves as mere shelters by shepherds and travellers, or as temporary places of refuge the occupation of which was quite ephemeral (as opposed to the relatively permanent occupation, lasting a number of years, in some of the above-mentioned refuge caves). The interpretative process is clearly complex and riddled with many difficulties. The problem is that, on the basis of a few pottery vessels, it is practically impossible to identify the type of occupation carried out within a cave: was it a domestic occupation or was the cave used for something else (e.g. for storage, for the deposition of manuscripts, etc.)? In the case of a domestic occupation, was it a long-term or a short-term one? And was it a permanent/continuous longterm occupation or was it an intermittent type of occupation, but one which stretched out over a long period of time? Was it a permanent/intensive short-term occupation that lasted for a few years or was it a transient occupation that lasted only a few days or weeks? The presence of miscellaneous finds, in addition to pottery, might alleviate these problems because this would strengthen the possibility that a cave could have been used for domestic purposes, although one should acknowledge that such finds might get lost even during transient or ephemeral visits or that some finds could have been merely hidden away in such caves for safe keeping; consequently, the larger the number and the wider the range of miscellaneous artefacts, the firmer such conclusions. With this in mind, the dearth of miscellaneous finds in the majority of the caves of Qumran—which is especially striking when compared to the “rich” Judaean Desert caves cited in Section 3—needs to be properly assessed. Outside the framework of any hypothesis, the archaeology of the caves casts serious doubts on the notion that these caves were used for long-term, permanent or continuous domestic habitation. This is because the large majority of these caves contained only pottery, but no miscellaneous artefacts, and this applies to both the limestone and the 67  See, for example, Hirschfeld and Riklin, “Region II,” 16; Sion, “Regions IV and VI,” 65–66.

Miscellaneous Artefacts From The Qumran Caves

marl caves. Therefore, in many ways, the Qumran caves recall the “poor” Judaean Desert caves just cited, which lack evidence that could unequivocally determine their function/s.68 The contrast with the “rich” Judaean Desert 68  Although scholars consider pottery—especially the presence of storage jars, cooking pots, and table vessels—as the indicator par excellence for determining the domestic function of the Qumran caves (see, for example, de Vaux, in DJD 3:32–36; Broshi and Eshel, “Residential Caves;” Magness, Archaeology of Qumran, 70; eadem, Stone and Dung, Oil and Spit: Jewish Daily Life in the Time of Jesus [Grand Rapids, MI: Eerdmans, 2011], 101; Stökl Ben Ezra, “Further Reflections,” 220), this can be a very ambiguous indicator. Pottery is the most common archaeological artefact and it is found in virtually every site—whatever its function—that is excavated. This is also true of caves in Judaea and the Judaean Desert, where pottery is essentially the common denominator: whether the caves are large or small, and whether or not they contained numerous quantities and types of other artefacts, pottery is virtually always present. Pottery is multi-functional and, therefore, pottery alone—even if it includes jars, cooking pots, and table vessels—indicates human activity but not necessarily a domestic occupation.  The physical features of caves and their accessibility are not definite indicators for the caves’ residential function either (pace Broshi and Eshel, “Residential Caves”). Therefore, the fact that the marl caves have many characteristics that make them suitable for habitation does not automatically make them residential caves. Many of these characteristics, such as the presence of levelled floors and ease of accessibility, would also have been important for caves that would have been used for storage or as workshops, for instance.  To this, one could add other ambiguous indicators that are sometimes cited in favour of the caves’ interpretation as permanent residential quarters, such as the presence of hearths and food remains, and of mats and mezuzot. Food remains in perfectly habitable caves—such as Caves 8Q, 9Q, and a collapsed cave near them (for which see Eshel and Broshi, “Excavations at Qumran”)—might be indicative not of residential caves but of ones used as storage facilities; in addition, one should note that few caves—including marl caves—actually contained traces of food (see n. 13). As for hearths, while these are clearly indicative of human occupation, they do not reveal whether this was a long-term or a short-term one and whether this was permanent or transient; more importantly, only two caves in the Qumran region contained hearths associated with late Second Temple material (see n. 13). Likewise, mats are rather uncommon at Qumran, and they do not say anything regarding the length or type of occupation. One should also note that a mat was discovered in Cave GQ12, which is deemed to be uninhabitable; indeed, de Vaux tentatively suggested that this mat might have been used as a cover for the cylindrical jars deposited there (de Vaux, in DJD 3:8). The presence of a mezuzah (8Q4) in Cave 8Q is also problematic as a possible pointer for troglodytic habitation (pace Broshi and Eshel, “Residential Caves,” 334) as it is equally possible (and perhaps more likely) that the mezuzah was brought from the Qumran settlement and deposited in Cave 8Q

147 caves and the Judaean subterranean complexes, which were certainly used for permanent domestic occupation and which were occupied for just a few years, cannot be greater; the Qumran caves simply lack clear evidence reflecting the carrying out of daily activities within them or evidence for their intensive use over a prolonged period of time.69 The possibility that this picture is solely the result of natural and cultural formation processes that have affected the archaeological record has already been ruled out above. One could also argue that most of the Qumran caves are relatively small and that, therefore, they could not have housed entire families, unlike some of the “rich” refuge caves. This is certainly correct, but if it might explain the small quantity of finds in the Qumran caves, it would still not explain the virtual absence of miscellaneous finds, especially if we are talking about a longterm, permanent domestic occupation. after the buildings of Qumran were abandoned. The fact that there are many mezuzot in Cave 4Q (4Q149–155), but none in the rest of the caves is, in fact, very telling: 4Q149–155 might well represent mezuzot that were once affixed to the various doorposts within the Qumran settlement.  Therefore, all the evidence that has been used to support the hypothesis that the Qumran caves (especially the marl caves) might have been used for long-term, permanent or continuous domestic occupation is ambiguous, and the aforementioned indicators never occur all together in the same cave. The interpretation of the caves as residential quarters can only be strengthened if there is a significant convergence of evidence in such a way that ambiguous or circumstantial evidence becomes much less problematic. The presence of finds here classified as miscellaneous artefacts would help to strengthen the case— because they are important components of a domestic assemblage and, collectively, they constitute the remnants of daily domestic life—but such objects are virtually absent in the Qumran caves and (quite significantly) none have been found in the marl caves. 69  Overall, the archaeological evidence indicates that the use of the caves extended from the 1st century BCE to the 1st century CE (see the pottery published by de Vaux, in DJD 1; idem, in DJD 3; idem, in DJD 6; Broshi and Eshel, “Residential Caves,” 340–42, Figs. 4–5), although the strongest evidence comes from the marl caves. The problem with the limestone caves is that the pottery from the majority of these natural caves is too meagre to suggest long-term use; otherwise, the typology of the pottery within such caves is not adequate to reach any conclusions regarding the matter. Therefore, unequivocal evidence for the long-term use of the limestone caves is available for only a few caves.  However, the attestation of pottery from both the 1st centuries BCE and CE is not necessarily indicative of a continuous or permanent occupation spanning these two centuries, especially not when the number of pottery in the majority of the caves is so meagre. Rather, the data could also be suggestive of various bouts of occupation or use at different points in the late Second Temple Period.

148 This observation is corroborated by the remains of a tent encampment excavated circa 100m north of the Qumran settlement. Remains of stones placed in a circular pattern—probably the traces of tents in the area70—were detected at point 2X and to the south-west of Cave D,71 and in this general area a number of finds were unearthed, including two complete bowls, three complete goblets, twenty-five rim and base fragments of bowls, as well as fragments from at least five jars, dating to the 1st century CE. At point 2X, three 1st century CE coins (including one from the First Jewish Revolt) were also found, together with sandal nails, an iron peg, and a fragment of a stone mug.72 This tent encampment provides us with a snapshot of an extra-mural domestic context in the Qumran region and it presents some minor differentials in comparison with the Qumran caves, namely the presence not only of pottery but also of stone vessels and coins. This becomes especially significant when one realizes that the level of natural and cultural disturbance that this tent encampment must have experienced is much higher than that experienced by the caves. These encampments have been exposed to the elements for centuries,73 and they are situated in an open, accessible area, in a place which must have seen countless foot traffic. Therefore, it is not only significant that some material from these tents survived, but it is also very likely that the associated material culture was more substantial. In addition, according to the excavators, these tent encampments were most probably of a seasonal nature,74 which further underscores the significance of the excavated artefacts, however limited in number. The scarcity of miscellaneous finds in the majority of the Qumran caves is further accentuated when it is seen within another context, namely that of the settlement at Qumran, situated less than 3 km away. Although architecture and pottery have been the main focus of most studies on the archaeology of Qumran, the settlement has yielded a substantial amount of other material, including glass, stone, and metal vessels, small metal implements, clothing articles, some jewellery, tools and weapons, as well as a large number of coins.75 The differential between the set70  Broshi and Eshel, “Residential Caves,” 336–39. 71  See Broshi and Eshel, “Residential Caves,” map 1. 72   Broshi and Eshel, “Residential Caves,” 336, 338–39, 343–44, Figs. 2–3, Pls. 1–3, tables 3–4. 73  Also see Broshi and Eshel, “Residential Caves,” 338, n. 10, who remark that “as the marl here is very soft, erosion has obliterated much of the evidence.” 74  Broshi and Eshel, “Residential Caves,” 338–39, n. 10. 75  See the lists in Jean-Baptiste Humbert and Alain Chambon, eds., Fouilles de Khirbet Qumrân et de Aïn Feshkha. Vol. 1: Album de photographies; Répertoire du fonds photographiques;

Mizzi

tlement at Qumran and the majority of the Qumran caves is quite significant, which is somewhat of a paradox considering the close and clear connections between the caves and the settlement. The marl caves are physically linked, through pathways and stairs, with the settlement; some of them, like Caves 7Q–9Q, were only accessible through the buildings on the marl plateau. Therefore, the marl caves must have certainly been used by the inhabitants of the settlement at Qumran.76 Likewise, many of the limestone caves appear to have been used by the Qumranites, as attested by the presence of cylindrical jars and bowllids—paralleling those from the Qumran ­settlement—in these caves (although ovoid and/or cylindrical jars have been excavated from other sites around the Dead Sea, Synthèse des notes de chantier du Père Roland de Vaux (NTOA. SA 1; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 1994); Roland de Vaux and Ferdinand Rohrhirsch, Die Ausgrabungen von Qumran und En Feschcha: IA: Die Grabungstagebücher (NTOA.SA 1A, trans. and notes F. Rohrhirsch and B. Hofmeir; Fribourg/Göttingen: Éditions Universitaires/ Vandenhoeck & Ruprecht, 1996); Jean-Baptiste Humbert and Alain Chambon, eds., The Excavations of Khirbet Qumran and Ain Feshkha: Synthesis of Roland de Vaux’s Field Notes (NTOA.SA 1B; trans. and rev. Stephen J. Pfann; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003). Also see Robert Donceel and Pauline Donceel-Voûte, “The Archaeology of Khirbet Qumran,” in Wise et al., Methods of Investigation, 1–38; Yitzhak Magen and Yuval Peleg, “Back to Qumran: Ten Years of Excavation and Research, 1993–2004,” in Galor, Humbert, and Zangenberg, Qumran, the Site of the Dead Sea Scrolls, 55–113 (59–64, 68–72); Yitzhak Magen and Yuval Peleg, The Qumran Excavations 1993–2004: Preliminary Report (JSP 6; Jerusalem: Staff Officer of Archaeology, Civil Administration of Judea and Samaria/ Israel Antiquities Authority, 2007), 5–11, 15–23; Helena Wouters et al., “Antique Glass from Khirbet Qumrân: Archaeological Context and Chemical Determination,” BIRPA 28 (1999/2000): 9–40; Helena Wouters, “Archaeological Glass from Khirbet Qumran: An Analytical Approach,” in Gunneweg, Greenblatt, and Adriaens, Bio- and Material Cultures at Qumran, 171–90; Mizzi, “The Glass from Khirbet Qumran;” Jean-Baptiste Humbert, “L’agglomérat métallique KhQ 960 et son context,” in Humbert and Gunneweg, Khirbet Qumrân et ‘Aïn Feshkha, 405; Noël Lacoudre et al., “L’amas métallique KhQ 960 du site de Khirbet Qumrân,” in Humbert and Gunneweg, Khirbet Qumrân et ‘Aïn Feshkha, 397–404. Also see Dennis Mizzi, The Archaeology of Khirbet Qumran: A Comparative Approach (Ph.D. diss., University of Oxford, 2009). Several of the data presented and analyzed in this dissertation have been obtained from the École Biblique, and they comprise de Vaux’s official inventory, illustrations of the pottery, the metal artefacts, and other small finds, and preliminary notes on these data made by Jean-Baptiste Humbert and Alain Chambon. I would like to offer my heartfelt thanks to Jean-Baptiste Humbert for generously giving me access to all these data. 76  See also Taylor, The Essenes, 276–77.

Miscellaneous Artefacts From The Qumran Caves

it is very significant that only the caves in the region of Qumran have yielded such jars).77 This means that there 77  In the Dead Sea region, besides ‘Ein Feshkha, variations of the ovoid jar have also been found at ‘Ein Boqeq, ‘Ein Gedi, Machaerus, and possibly Herodium. See Roland de Vaux, “Fouilles de Feshkha,” RB 66 (1959): 225–55 (Fig. 1:4); Moshe Fischer and Oren Tal, “Pottery,” in ‘En Boqeq: Excavations in an Oasis on the Dead Sea; Volume 2; The Officina; An Early Roman Building on the Dead Sea Shore, ed. M. Fischer, M. Gichon, and O. Tal (Mainz: Verlag Philipp Von Zabern, 2000), 29–67 (37, Fig. 2.7:1–4); Stanislao Loffreda, La Ceramica di Macheronte e dell’Herodion (90 a.C.–135 d.C.) (SBF.CMa 39; Jerusalem: Franciscan Printing Press, 1996), 53, Figs. 19:1, 55:79; images of ovoid jars from ‘Ein Gedi were available online, but the webpage no longer exists. Both ovoid jars and cylindrical jars have been found at Jericho and Masada. See James L. Kelso and Dimitri C. Baramki, Excavations at New Testament Jericho and Khirbet en-Nitla (AASOR 29–30; New Haven: American Schools of Oriental Research, 1955), 29–30, Pl. 23:A115; Rachel Bar Nathan, Hasmonean and Herodian Palaces at Jericho: Final Reports of the 1973–1987 Excavations; Volume III; The Pottery (Jerusalem: Israel Exploration Society, 2002), 24–26, ills. 12–13, Pls. 1–2:6–11; Rachel Bar Nathan and Ronit Gitler-Kamil, “Typology of the Herodian 3 Pottery,” in Bar Nathan, Hasmonean and Herodian Palaces at Jericho, 145–92 (150); Rachel Bar Nathan, Masada VII: The Yigael Yadin Excavations 1963–1965; Final Reports; The Pottery of Masada (Jerusalem: Israel Exploration Society/Hebrew University, 2006), 67–71, Fig. 19, Pls.15:78–87. Beyond the Dead Sea, cylindrical jars have been found at Qalandiya; see Yitzhak Magen et al., “The Finds,” in The Land of Benjamin (JSP 3; ed. Y. Magen et al.; Jerusalem: Staff Officer of Archaeology, Civil Administration of Judea and Samaria/ Israel Antiquities Authority, 2004), 29–144 (85, Fig. 102, Pls. 3:7, 4:19, 7:16–17, 10:22). There are also a few jar rims from the Jewish Quarter, in Jerusalem, which have very similar morphological features to cylindrical and ovoid jars, although these vessels have gone unnoticed; see, for example, Hillel Geva, “Hellenistic Pottery from Areas W and X-2,” in Jewish Quarter Excavations in the Old City of Jerusalem: Volume 2; The Finds from Areas A, W and X-2 (ed. H. Geva; Jerusalem: Israel Exploration Society, 2003), 113–75 (Pl. 5.4:13).  Although cylindrical and/or ovoid jars have also been found at the aforementioned sites, these jars remain characteristic of and synonymous with Qumran. This is indicated by the large quantity of such jars and by the wide variety of types attested at Qumran (this observation is based on the data and illustrations I have accessed at the École Biblique); in contrast, ovoid and/or cylindrical jars occur in very limited quantities at other sites. This suggests that this type of storage jar was an innovation of Qumran and that it was largely produced there. Indeed, it appears that Magen and Peleg have excavated wasters of such jars at Qumran (Rachel Bar Nathan, “Qumran and the Hasmonean and Herodian Winter Palaces of Jericho: The Implication of the Pottery Finds on the Interpretation of the Settlement at Qumran,” in Galor, Humbert, and Zangenberg, Qumran, the Site of the Dead Sea Scrolls, 263–77 [275]). Accordingly, the presence of such jars in the Qumran caves, together with the accompanying bowl-

149 is a stark contrast between the caves and the settlement: the clear, unambiguous evidence of daily life attested in the latter is not mirrored in the former. This continues to suggest that, quite possibly, the Qumran caves were used for purposes other than dwelling or that, at most, they only experienced temporary, intermittent occupations.78 If they were used intensively as permanent residential quarters for circa 150 years, one would expect more material and many more miscellaneous finds to have been retrieved from within these caves. It should be pointed lids, creates a very strong connection between the caves and the settlement. See also Taylor, The Essenes, 275–76.  The results from INAA, which determined that the clay used for the production of some of the Qumran cylindrical jars comes from the Motsa Clay Formation and from the Jericho area are in no way indicative of the existence of workshops (producing these jars) in the Hebron and the Jericho regions, contra Jan Gunneweg and Marta Balla, “Neutron Activation Analysis: Scroll Jars and Common Ware,” in Humbert and Gunneweg, Khirbet Qumrân et ʿAïn Feshkha, 3–53 (13–14, 18, 26–27); Bar Nathan, Masada VII, 71; idem, “Qumran and the Hasmonean,” 275. If this were the case, then one would expect large numbers of such jars to surface in excavations in the regions of Hebron, Jerusalem, and Jericho, which is not the case. Therefore, the archaeological evidence suggests that it was the clay (and not the jars) that was imported to Qumran. The discovery of a few ovoid and cylindrical jars in late 1st century and early 2nd century CE contexts in Jericho does not diminish the strength of this conclusion. For the publication of these jars, see Rachel Bar Nathan and Irinia Eisenstadt, “The Ceramic Corpus from the Roman Estate at Jericho: Late 1st–Early 2nd Centuries C.E.,” in Hasmonean and Herodian Palaces at Jericho: Final Reports of the 1973–1987 Excavations; Volume V; The Finds from Jericho and Cypros (ed. R. Bar Nathan and J. Gärtner; Jerusalem: Israel Exploration Society/Hebrew University, 2013), 3–84 (11–12, Pls. 1.1:553–555; 1.9:677–678; 1.10:689–692). The presence of these jars in post-70 CE contexts might indicate that they are remnants from the earlier 1st century CE. Nonetheless, the ware of some of these jars as well as their shape is not paralleled by the pre-70 CE. examples, which suggests the existence of a post-70 CE workshop, in the area of Jericho, continuing this jar tradition (Bar Nathan and Eisenstadt, “The Ceramic Corpus,” 11–12), but this does not necessarily mean that such workshops also existed before 70 CE.  In the end, as far as caves are concerned, the fact remains that cylindrical jars have only been found in caves in the area of Qumran, and this underlines the special connection that these caves must have had with the settlement in the same region, where these jars are most commonly found. 78  Within a framework provided by the Scrolls, one could argue that some caves might have been used by novices, who were still undergoing their admission process, or by ritually impure persons, the impurity of whom necessitated their seclusion, or by persons who were temporarily expelled from communal life. Such intermittent use is not likely to leave much trace in the archaeological record (especially in the form of miscellaneous artefacts).

150 out that, despite seemingly archaeological evidence to the contrary, this conclusion is not at all implausible.79 This is one possible approach to the data. However, there is another approach, framed within the most popular of the Essene/sectarian hypotheses, namely the one mentioned above, which sees the settlement of Qumran as a community centre and the Qumranites living in caves and tents in the surrounding region.80 The Yaḥad of the Community Rule (1QS) has frequently been identified exclusively with Qumran, a supposition that continues to underline the notion of a large community inhabiting the region.81 Within the framework of this hypothesis, the general dearth of artefacts (especially miscellaneous finds) and biofacts/ecofacts in the Qumran caves could 79  There are three main objections that could be raised against this scenario, namely: 1) the cemetery; 2) L.77 and the thousands of pottery vessels excavated throughout the site of Qumran; and 3) the large number of ritual baths and the large volume of water stored in these and other cisterns at the site. It is impossible to discuss these objections in detail and, therefore, the following brief remarks do not do justice to this question, which deserves a detailed treatment of its own.  As far as the cemetery is concerned, the fact that we are dealing with an unrepresentative sample, the fact that some tombs were apparently empty, and the possibility that some of the interred could have come from outside means that the data from the cemetery cannot be used to determine the population size of the Qumranites. Regarding the second and the third objections, the evidence could be interpreted from a different perspective; the large number of pottery vessels and ritual baths could reflect a small group deeply concerned with ritual purity, to the point that it consumed an exorbitant amount of pottery (which once impure had to be broken or thrown away) and that it ensured continued yearlong access to ritually pure water. 80  For a synthesis of this view, see de Vaux, Archaeology; Magness, Archaeology of Qumran. 81  See, for example, Sarianna Metso, “In Search of the Sitz im Leben of the Community Rule,” in Parry and Ulrich, The Provo International Conference on the Dead Sea Scrolls, 306–15; eadem, “Whom does the Term Ya�ad identify?” in Biblical Traditions in Transmission: Essays in Honour of Michael A. Knibb (JSJSup 111; ed. C. Hempel and J. M. Lieu; Leiden: Brill, 2006), 213–35; Charlotte Hempel, “Emerging Communal Life and Ideology in the S Tradition,” in Defining Identities: ‘We’, ‘You’ and ‘the Others’ in the Dead Sea Scrolls (STDJ 70; ed. F. García Martínez and M. Popović; Leiden: Brill, 2008), 43–61. For an opposing view, see John J. Collins, Beyond the Qumran Community: The Sectarian Movement of the Dead Sea Scrolls (Grand Rapids, MI: Eerdmans, 2010); Alison Schofield, From Qumran to the Yaḥad: A New Paradigm of Textual Development for the Community Rule (STDJ 77; Leiden: Brill, 2009); eadem, “Between Center and Periphery: The Yaḥad in Context,” DSD 16 (2009): 330–50. Also see Michael A. Knibb, “The Community of the Dead Sea Scrolls: An Introduction,” DSD 16 (2009): 297–308.

Mizzi

be explained by the possibility that most activities involving their use were conducted within the main building at Qumran and not within the caves. For example, the scarcity of cooking pots, tablewares, food remains, and hearths may be explained by the possibility that cooking and ­eating activities took place in the main building, in a communal context, and not individually in caves.82 The lack of any miscellaneous finds could be explained in similar fashion. For example, the virtual absence of coins within the caves, in contrast to the hundreds of coins found within the confines of the buildings, could be accounted for by the possibility that the Qumranites pooled their resources into one collective fund, kept within the confines of the main building.83 The lack of stone and glass vessels, of tools and other implements, and of small personal items could be explained along similar lines: their use could have been limited to activities performed within the buildings of Qumran, or they could have simply been part of the communal resources. Moreover, small personal items and other precious objects could have been carried away by their owners when they fled the area during the First Jewish revolt, although long-term, permanent occupation would still leave some traces in the form of earlier (lost or broken) items. Therefore, because of the various links between the caves and the buildings at Qumran, the Qumran caves cannot be simplistically compared to other caves in Judaea and the Judaean Desert; unlike the latter, the Qumran caves are located in the immediate vicinity of a settlement with which they were intricately connected. As a result, if Qumran is approached within the framework of the classic Essene/sectarian hypothesis, the virtual absence of miscellaneous artefacts from the caves would not necessarily be incongruent with their interpretation as long-term dwelling quarters.84 One may also add that, 82  This would adhere with the widely accepted view that the Qumranites ate communal meals in L.77. See, for example, de Vaux, Archaeology, 11–12; Magness, Archaeology of Qumran, 116– 26; Bart Wagemakers and Joan E. Taylor, “New Photographs of the Qumran Excavations from 1954 and Interpretations of L.77 and L.86,” PEQ 143 (2011): 134–56. 83  See, for example, de Vaux, in DJD 3:17, 35–36; Magness, Stone and Dung, 99–101. 84  In fact, an interesting comparison may be sought, albeit with caution (due to the chronological and cultural gaps), from the monastic cells discovered in the Judaean Desert and in the Sinai, dating to the Byzantine Period, which appear to have been occupied for several decades by ascetics and monks associated with nearby monasteries. Excluding a few exceptions, generally, the finds discovered from these caves are very limited, typically consisting of a few pottery vessels, if anything at all; there are

Miscellaneous Artefacts From The Qumran Caves

within the framework of this hypothesis, one could argue that these caves functioned as mere sleeping quarters, which would further explain the sparse archaeological record in general and the scarcity of miscellaneous finds in particular. It is evident, that the evidence is such that it can be interpreted in very different ways. One of the principal problems is the lack of miscellaneous finds: if, for example, such artefacts were found—together with pottery, food remains, and hearths—in the marl caves (which are ideal for habitation, are easily accessible, and are directly linked to the Qumran settlement), it would have been very natural to interpret the caves’ function as residential quarters; the evidence would have been quite unequivocal. But without the miscellaneous finds, and without any significant intersection of the other indicators for a possible domestic function, there remains an element of ambiguity, even though the absence of miscellaneous artefacts does not necessarily invalidate such a conclusion. Therefore, the questions raised by the absence of miscellaneous finds in the Qumran caves need to be taken into consideration in an assessment of the caves’ function/s. So far, the Qumran caves have been treated as a collective entity. And as far as the miscellaneous artefacts are concerned, with the exception of a few examples (see below), the caves of Qumran do indeed share (more or less) the same archaeological pattern. This notwithstanding, it is most likely that different types of caves had different functions, and the distinction between marl and limestone caves remains fundamental.85 Moreover, partly on the basis of the miscellaneous finds, it is also possible to differentiate between limestone caves themselves and, usually no hearths, very few cooking pots, and a very limited number of other material remains, even if the community to which these monks belonged did possess such material possessions, as attested by such finds in the central monastic buildings and their associated churches and chapels. I would like to thank Jodi Magness (personal communication [August 2010]) for bringing this point to my attention. Also see, for example, Yizhar Hirschfeld, The Judean Desert Monasteries in the Byzantine Period (New Haven: Yale University Press, 1992); Joseph Patrich, Benny Arubas, and Benny Agur, “Monastic Cells in the Desert of Gerasimus near the Jordan,” in Early Christianity in Context: Monuments and Documents (SBFCoMa 38; ed. F. Manns and E. Alliata; Jerusalem: Franciscan Printing Press, 1993), 277–96; Uzi Dahari, Monastic Settlements in South Sinai in the Byzantine Period: The Archaeological Remains (IAA Reports 9; Jerusalem: Israel Antiquities Authority, 2000). Despite the chronological and cultural discrepancies, the comparison with the Byzantine monastic cells is quite apt since, like the majority of the caves at Qumran, these cells were closely tied to a main settlement. 85  See, especially, Taylor, The Essenes, 281.

151 thus, these artefacts can make our typology of the Qumran caves more nuanced. It is possible to isolate a small number of limestone caves—namely, Caves P13 = X/35, P24, GQ37 = P37, 1Q, and 11Q—from the rest of the natural caves. These are the caves which break the “Qumran pattern” as far as the archaeological record of the caves (that pertaining to the Early Roman Period) is concerned, owing to the fact that they contained either a relatively large number and/or variety of miscellaneous artefacts, and/or certain distinctive objects indicative of a domestic setting, and/or an overall material culture that is quite dissimilar from the majority of the other caves. Most probably, the history of these five caves, and by extension their function, was different from that of the marl and the other limestone caves, regardless of the respective functions of both the latter. Cave GQ37 = P37 is the richest cave in the Qumran region, and it breaks the “Qumran pattern” by the presence (among its material assemblage) of glass, stone, and wooden vessels, and of a late 1st century CE coin. In addition, an iron nail, two arrowheads, a substantial number of pottery—including three cooking pots—and a layer of charred debris were also discovered.86 Cave GQ37 also contained fragments from two Jerusalem Painted bowls, and it yielded no fragments of cylindrical jars. The illustrated bowl and goblet87 from de Vaux’s exploration of Cave GQ37 fit a 1st century CE date,88 whereas the rest of the pottery sherds are simply described as belonging to the Qumran type.89 In the case of the items unearthed during Patrich’s investigation, it is stated that “the corpus as a whole belongs to the last years of the Second Temple Period, but some types started earlier and some can be post-70 CE.”90 The description of two glass vessels as having folded decorated rims suggests that these should be dated to the late 1st-early 2nd century CE,91 86  See Patrich, “Khirbet Qumran,” 91–93. 87  De Vaux, in DJD 3: Figs. 5:16–17. 88  The lack of any rim fragments makes it difficult to determine their typology with any precision, however. 89  De Vaux, in DJD 3:12. This description implies that the pottery dates to the 1st centuries BCE and/or CE. 90  Patrich, “Khirbet Qumran,” 93. 91  From the description, it seems that these glass vessels belong to a series of glass bowls that have a folded rim or bowls that are characterized by a hollow/tubular or a double-hollow/tubular fold below the rim; some of these vessels were also decorated with what is described as a crimped trail (see, for example, Ruth E. Jackson-Tal, “Glass Vessels from En-Gedi,” in En-Gedi Excavations II: Final Report (1996–2002) (ed. Y. Hirschfeld; Jerusalem: Israel Exploration Society/Hebrew University, 2007), 474–506 [474–75, 479, Pls. 1:3–6, 4:1–3]; eadem, “The Glass Finds

152 while the bronze coin belongs to a group of coins minted in Antioch, typically dated to the 1st and 2nd centuries CE.92 It is quite possible that the corpus from Cave GQ37 dates to around 70 CE but an early 2nd century CE date should not be excluded.93 Considering this, and in view of the various miscellaneous finds as well as the pottery, it is most probable that this was a refuge cave, probably used by a few individuals. The absence of cylindrical jars and the presence of glass vessels unparalleled at the Qumran settlement—not to mention the possibility that the archaeological remains might date to the 2nd century CE—also indicate that Cave GQ37 should probably not be considered as a “Qumranite” cave.94 Cave P24 differs from the other Qumran caves by the presence of a bronze nose/ear ring and a bronze needle. It is also one of only two caves to yield fragments of Jerusalem Painted Ware, and the only one to yield fragments of Nabataean Cream Ware. The presence of six cooking pots and of various hearths provides further contrast with the rest of the caves of Qumran. Other pottery vessels (including storage jars), food remains, basketry, pieces of ropes, and textiles complete the corpus,95 which is highly indicative of Cave P24 having been used as a refuge cave,96 either around 70 CE or in the early 2nd century CE (see the discussion in Table 9.1). The discovery of personal items, together with semi-fine wares and a number of hearths, clearly demonstrates that a small group of people used this cave for dwelling purposes. It is interesting that cylindrical jars were found in Cave P24, which begets two possible conclusions: either that Qumranites sought refuge here or that non-Qumranites occupied a cave that had been previously used by Qumranites. Cave P13 = X/35 breaks the “Qumran pattern” only because of the presence of a coin (First Jewish Revolt)— and this is highly significant given that coins are lacking from the Hasmonean and Herodian Palaces at Jericho,” in Bar Nathan and Gärtner, Hasmonean and Herodian Palaces at Jericho: 100–29 [107–8, 112–13, Pls. 3.5:37–39, 42–45; 3.8:2–6]. These types of bowls typically occur in late 1st century CE and early 2nd century CE contexts—they are particularly common in Bar Kokhba caves in the Judaean Desert—as well as in later 3rd century CE contexts (Jackson-Tal, “Glass Vessels from En-Gedi,” 474–75, 479; eadem, “The Glass Finds,” 107–8, 112). 92  Patrich, “Khirbet Qumran,” 92. 93  Sherds of the “Qumran type” could, in theory, be post-70 CE in date, since some pottery types remained in circulation throughout the duration of the 1st century CE and the early 2nd century CE. 94   Pace Patrich, “Khirbet Qumran,” 92–93. 95  See Patrich, “Khirbet Qumran,” 90. 96  See also Patrich, “Khirbet Qumran,” 90.

Mizzi

in the majority of the Qumran caves—and a bone plaque (the purpose of which is uncertain). In addition, this cave yielded pieces of wood, various food remains (including dates stored in jars), pieces of cloth and ropes, as well as pottery, some of which is not paralleled in the other caves or in the Qumran settlement.97 Most probably, this is another refuge cave situated in the Qumran region, dating to around 70 CE98 or later (see the discussion in Table 9.1), but the lack of any cylindrical jars and the presence of pottery atypical of Qumran suggests that Cave P13 may not have been used by Qumranites.99 Therefore, this represents another possible instance of a Qumran cave not being a “Qumranite” one. Cave 1Q may also be set apart owing to the discovery of a personal item, a wooden comb, which, in addition to food remains, palm fibre, pieces of wood, and pottery (including a cooking pot),100 may be indicative of a very brief occupation that resulted in the loss or intentional deposition of this personal object. It is highly possible that Cave 1Q was originally used for the deposition of manuscripts,101 and that it was subsequently used for refuge (by Qumranites?), around 70 CE. This also helps explain the presence of a cooking pot in an otherwise uninhabitable cave, unless the pot was used for something other than it was originally intended. Finally, Cave 11Q is unique because of the tools and other metal implements discovered within it, including an iron blade, an iron chisel, and an iron pick.102 Cave 11Q 97  See Joseph Patrich and Benny Arubas, “A Juglet containing Balsam Oil (?) from a Cave near Qumran,” IEJ 39 (1989): 43–59 (48–49, Fig. 3); Patrich, “Khirbet Qumran,” 91. 98  See also Patrich, “Khirbet Qumran,” 91. 99  Patrich, “Khirbet Qumran,” 91 reaches a similar conclusion but on the basis of the discovery of a juglet containing what may be balsam oil. Patrich states that “it is doubtful if they were members of the sect, since according to the sect’s laws oil defiles the body.” However, it should be pointed out that balsam could have been used for medicinal purposes, and that juglets (some of which must have contained oil) are quite common in the ceramic corpus of the Qumran settlement. For the use of balsam in medicine, see Taylor, The Essenes, 311–13; eadem, “ ‘Roots, Remedies and Properties of Stones’: The Essenes and Dead Sea Pharmacology,” JJS 60 (2009): 226–44. 100  See Harding, in DJD 1; de Vaux, in DJD 1. 101  For the meticulous way in which the scrolls were stored in Cave 1Q, which may be indicative that they were not quickly hidden around 68 CE, see Taylor, The Essenes, 272–303; eadem, “Buried Manuscripts.” 102  De Vaux, “Fouilles de Khirbet Qumrân,” 574. Contrary to popular understanding, there is no arrowhead from Cave 11Q. The presumed existence of this object emerges from a misreading of the documentation from the original excavations. This issue

153

Miscellaneous Artefacts From The Qumran Caves

yielded no traces of food, hearths, or cooking pots; neither did it contain any fine or semi-fine wares, coins, or traces of any personal items (a bead was found, but this is of uncertain date). The discovery of a collection of tools in the same cave is very interesting, however. Among the metal artefacts an iron key was also found, which suggests that whoever used Cave 11Q had links to a particular building (perhaps the Qumran settlement?) and that there was “the expectation of a return to one’s home.”103 It is possible that a small group of people (probably Qumranites) sought refuge in this cave, but only for a very short period of time. The presence of manuscripts in Cave 11Q could be related to this same transient occupation; the refugees did not just seek temporary shelter in this cave, but they also brought precious documents with them for safe keeping.104 Otherwise, it is possible that Cave 11Q was used as a depository for manuscripts before 68 CE, and briefly re-used for refuge during the First Jewish Revolt. In the aforementioned examples, the physical properties of the caves, and/or the small number (but wide range) of discovered artefacts, and/or the First/Second Jewish Revolt dates gauged from some of the objects hint at a domestic occupation, but one that was of a limited and short-term nature, hence the conclusion that these caves might have been briefly used for refuge. Here, the miscellaneous finds are not necessarily the exclusive indicators for this type of domestic habitation—pottery, food remains and hearths, as well as textiles and basketry are other key pointers, all of which intersect in the same caves—but they still play an important and somewhat of a determining role. Moreover, the presence of miscellaneous artefacts (that are distinctively suggestive of a domestic setting) in caves the overall evidence from which is indicative of a short-term domestic occupation creates an important corrective for how to interpret the other Qumran caves. In this light, it is quite unlikely that any of the other caves in the region of Qumran were used as places of refuge, for example. Rather, the majority of the caves at Qumran were probably used either as depositories for manuscripts, or as storage facilities, or as workshops, or as periodic dwelling quarters, or a combination is discussed in more detail in a forthcoming report by Dennis Mizzi, Annalisa Faggi, and Vincenzo Palleschi. 103   For the latter observation, see Popović, “Qumran as Scroll Storehouse,” 562. 104  See, for example, García Martínez, “Cave 11 in Context;” Stökl Ben Ezra, “Further Reflections;” idem, “Old Caves and Young Caves.” The stylus in the Schøyen Collection, if it does come from Cave 11Q, as it is claimed, would further corroborate this possibility. For more details on this stylus, see http://www.schoyencollec tion.com/dsscrolls.html#5095-3 [accessed January 2014].

of these; the caves’ use as permanent residential quarters cannot be ruled out, although this can only be sustained within the framework of the version of the Essene/sectarian hypothesis outlined above. It is also highly significant that the only two 1st century CE coins happen to emanate from caves that were probably used as refuge places and from what may be described as “non-Qumranite” caves, whereas a third coin dates to the 5th century CE and two others are of unknown date; accordingly, this continues to underscore the interpretative significance of the absence of any late Second Temple coins in the rest of the caves.105 5

Excursus: Women in the Qumran Caves?

If some of the Qumran caves were used as places of refuge, or as permanent or periodic residential quarters by the Qumranites, do any of the finds reveal anything about the gender of their occupants? Most artefacts are gender neutral except for some of the miscellaneous finds and, thus, the latter are important because they can shed light on the question of whether or not there could have been women at Qumran. The literature on the archaeology of gender is vast,106 and it is not possible to delve into this theoretical issue here. However, within the context of ancient Judaism, scholars have generally associated jewellery, mirrors, cosmetic articles, spindle whorls, needles, and combs with women,107 although the issue is not as straightforward.108 105  This contrasts the large number of coins found in various caves throughout the Judaean Desert. See references in Section 3. Also see Donald T. Ariel, “The Coins from the Surveys and Excavations of Caves in the Northern Judean Desert,” ʿAtiqot 41/2 (2002): 281– 304. The significance of the virtual absence of coins from the Qumran caves has already been noted by de Vaux, in DJD 3:17, 35–36, and Magness, Stone and Dung, 99–100. 106  For a detailed discussion of this issue in relation to the archaeology of Qumran, and for further bibliography on studies dealing with archaeology and gender, see Joan E. Taylor, “The Cemeteries of Khirbet Qumran and Women’s Presence at the Site,” DSD 6 (1999): 285–323 (317, n. 112); Jodi Magness, “Women at Qumran?” in What Athens Has to Do with Jerusalem: Essays on Classical, Jewish, and Early Christian Art and Archaeology in Honor of Gideon Foerster (ISACR 1; ed. L. Rutgers; Leuven: Peeters, 2002), 89–123; repr. in Debating Qumran: Collected Essays on its Archaeology (Leuven: Peeters, 2004). 107  See Taylor, “The Cemeteries of Khirbet Qumran;” Magness, “Women at Qumran?” Ann E. Killebrew, “Cosmetics,” EDSS 1:148–49. 108  For example, while loom weights and needles may largely be associated with women (see Hirschfeld, Qumran in Context, 149), explicit evidence of male weavers and tailors does exist.

154 The only possible gendered objects from the Qumran caves are the spindle whorl in Cave GQ3, the nose/ear ring and (perhaps) the needle from Cave P24, the bead from Cave 11Q and the two bone beads from Cave B, and (perhaps) the wooden comb from Cave 1Q. With the exception of the nose/ear ring from Cave P24, all these small finds present interpretative problems. The problem with the spindle whorl and the beads is that they cannot be dated with any certainty (see Table 9.1), whereas the challenge See Magness, “Women at Qumran?” 143–44, 146, nn. 151, 166; Miriam B. Peskowitz, Spinning Fantasies: Rabbis, Gender and History (Berkeley: University of California Press, 1997), 61, 86–87.  Combs present a similar ambiguous case. Some scholars have argued that combs are indeed gendered objects, reflecting the presence of women and children at a site (see, for example, Taylor, “The Cemeteries of Khirbet Qumran,” 318; Magness, “Women at Qumran?” 122–23; Kosta Y. Mumcuoglu, “Human Parasites from Qumran and the Surrounding Regions in Israel,” in Gunneweg, Greenblatt, and Adriaens, Bio- and Material Cultures at Qumran, 57–61 [57, 59]). Mumcuoglu, in particular, notes that the “examination and grooming to remove lice from the head of an infested individual was practically always a social interaction between mother and child.” This conclusion is based on a recent study in modern-day Israel, where it was shown that this activity was performed by the mother in 95% of the households surveyed; however, this ethnographic case study is chronologically too far removed, and neither archaeological evidence nor any literary sources are cited to further support this claim. Indeed, other scholars have argued that, during the Roman Period, hair care was given importance by both men and women (see, for example, Hirschfeld, Qumran in Context, 149; Killebrew, “Cosmetics,” 148).  The reliability of spindle whorls as gendered objects has similarly been questioned (see Peskowitz, Spinning Fantasies), although scholars, in general, still consider spindle whorls as some of the most unambiguous objects which reflect female presence at a site as it is apparent that spinning was an activity exclusive to women in antiquity (see Taylor, “The Cemeteries of Khirbet Qumran,” 318–21 [n. 127]; Alexandra T. Croom, Roman Clothing and Fashion [Charleston, SC: Tempus, 2000], 18; Ronny Reich, “Women and Men at Masada: Some Anthropological Observations based on the Small Finds [Coins, Spindles],” ZDPV 117 [2001]: 149–63 [150]).

Mizzi

with the wooden comb and the needle lies in the fact that whether these are gendered objects or not is unclear. The only unambiguous gendered artefact that might also date to the 1st century CE is the jewellery piece from Cave P24, which may have been used for refuge by Qumranites. Therefore, the evidence for women from the Qumran caves is not clear-cut and, even if Cave P24 was used by Qumranites, the evidence in general pales in comparison with the clear evidence for women in other Judaean Desert caves. Does this mean that female presence was minimal at Qumran? Or are the data from the caves irrelevant to this question (perhaps this further confirms that the majority of the Qumran caves were not used for any sort of intensive domestic occupation)? 6

Concluding Remarks

In this paper, I have tried to highlight the significance of the miscellaneous artefacts for the history of the Qumran caves. It has been shown that this seemingly unpromising corpus plays an important role in an evaluation of the residential function of the caves, even though it does not necessarily provide us with conclusive answers. Indeed, the archaeology of the Qumran caves is very complex, and the sobering fact is that the evidence is perhaps insufficient for us to make any confident judgements as to the caves’ function/s. To end on a positive note, however, it should be noted that, on the basis of the miscellaneous artefacts as well as other finds, it is possible to single out some caves as “nonQumranite” (and there may be more such caves besides the ones discussed above), which continues to underline the fact that there may be many specific histories for the different caves in the region of Qumran and that these different histories may be discernible not only along a diachronic continuum (cf. the finds from different periods in different caves) but also along a synchronic one. This is, in itself, an important conclusion, which emphasizes the necessity of defining properly the term Qumran Caves.

155

Miscellaneous Artefacts From The Qumran Caves Table 9.1 Caves in the region of Qumran containing miscellaneous artefacts. The following conventions are used: ‘normal text’ represents late Second Temple material (1st century BCE to ca. 70 CE); ‘italicized text’ represents prehistoric material (Neolithic, Chalcolithic, Bronze Age); ‘bold text’ represents Iron Age II material; ‘italicized and bold text’ represents Hellenistic material; ‘text in small capitals’ represents Early Roman material dating to the late 1st–early 2nd centuries CE, which probably dates to around 70 CE or to the Second Jewish Revolt; ‘text in Cave Designation

Pottery Vesselsa

GQ3d

4 jars (c) 1 jar (b) 4 bowl-lids 1 goblet 1 cooking pot

GQ4e 11Qf

GQ14 = 1Qg

GQ15h

GQ17i

xn sherds x n sherds 1 jar (c) 2 bowl-lids 1 juglet xn sherds 1 small jar xn sherds 1 jar 1 juglet 2 lamps xn sherds

50 jars (c) 35 bowl-lids 3 bowls 1 plate 1 cooking pot 1 juglet 3 lamps 1 jar (c) 1 jar 1 bowl 2 jars (c) 2 jars (b) 1 bowl-lid xn bowls 1 cooking pot 1 juglet 1 lamp

Other Vesselsb Small Objectsc

italicized small capitals’ represents Late Roman/Byzantine material; and ‘text in bold small capitals’ represents Islamic material. Any later material, including modern one, is not listed. Unless otherwise stated in the footnotes, the dating of the material is based on the excavators’ original assessment. Material the date of which is uncertain is represented by a ‘text in italicized bold small capitals’ and followed by a question mark (?). The question mark (?) is also used when the dating of the material to a particular period is very tentative. Tools/ Weapons

Coins

Biofacts/ Textiles/ Ecofacts Basketry

Manuscripts

1 functional bronze 2 flint blades ring

1 wooden vessel (?)

1 perforated stone disc = spindle whorl (?)

1 iron key 1 nail 1 copper buckle = handle loop? 1 bead (?)

1 iron blade 1 iron chisel 1 iron pick 1 iron rod = awl?

an ashy deposit

fragments found of leather many pieces of cloth fragments of linen

1 wooden comb

small & large pieces of wood

1 flint blade 1 wooden point (?) 5 wooden posts

date & olive stones

fragments of basketry & rope palm fibre found 6 phylactery cases pieces of linen

156 Table 9.1

Mizzi Caves in the region of Qumran containing miscellaneous artefacts. (cont.)

Cave Designation

Pottery Vesselsa

Other Vesselsb Small Objectsc

GQ37 = P37j

6 jars 3 cooking pots 2 bowls 2 bowls ( JPW) 1 goblet 12 juglets 1 lamp x n sherds xn sherds 1 juglet 1 lamp xn sherds xn sherds

4 glass plates

Ak

Bl

P13 = X/35m

P24n

xn sherds xn sherds xn sherds

2 jars 1 cooking pot 1 frying pan 1 bowl 1 plate/lid 2 juglets 5 lamps 1 churn 2 cooking pots 2 bowls

1 jar (c) 1 jar (b) 8 jars (c/b) 6 cooking pots 1 bowl (JPW) 1 jug (NCW) 5 juglets 1 lamp

1 iron nail

1 wooden plate 2 stone vessels

1 wooden bowl

Tools/ Weapons

Coins

Biofacts/ Textiles/ Ecofacts Basketry

2 iron arrow­ heads 1 arrow xn wooden beams

1 bronze a layer coin of charred debris (?)

1 bronze rod

2 bone beads (?)

fragments of cloth & leather

1 stone pestle 2 flint blades

1 bone plaque

1 bronze needle 1 bronze ear/ nose ring

1 flint arrowhead pieces of wood

1 bronze animal pieces of bones cloth coin (67 CE) pieces of date, olive, leather & & fruit cloth (?) stones pieces of textiles & dried rope (?) dates walnut shells xn pieces of hearths rope thick layers of ash

pieces of basketry

date stones

pieces of cloth

Manuscripts

157

Miscellaneous Artefacts From The Qumran Caves

Cave Designation

Pottery Vesselsa

X/51o

1 bowl x n sherds (?)

XII/53r

XII/56s

XIII/2t

1 cooking jug xn sherds

Tools/ Weapons

Coins

Biofacts/ Textiles/ Ecofacts Basketry

1 arrow­ head

1 pendant (?)

X/57p

XI/16q

Other Vesselsb Small Objectsc

1 lance head (?)

large quantity of leather (?) 1 coin (402–408 CE)

worked bone

1 jar (c) 1 jar (b) 1 cooking pot 1 jug 2 jars 1 jar (?) 4 bowls 3 handles 1 bowl 2 jars 1 juglet xn sherds xn sherds xn sherds x n sherds x n sherds

a Note, that some of the listed vessels were found as fragments. However, the pottery count presented in this table does not indicate the number of fragments found but the minimum number of vessels that these fragments represent. The same logic is applied to the other types of finds. The following abbreviations are used throughout the table: (c) refers to cylindrical/ovoid jars; (b) refers to bag-shaped jars (the

Manuscripts

2 flint arrowheads

large quantity of textiles, string, and ropes (?) organic pieces of material cloth & string charcoal animal bones

1 flint lunate

2 coins 2 flint spearheads (?) 1 fan scraper 1 knife (?)

xn hearths 1 rope (?) 1 braid (?)

absence of either of these designations means that the jar fragments found are of an indeterminable type); (JPW) refers to Jerusalem Painted Ware; (NCW) refers to Nabataean Cream Ware (the absence of any of these designations means that the pottery in question is a common ware). b This category includes glass vessels, stone vessels, metal vessels, and wooden vessels.

158 c This category includes jewellery, cosmetic and medical utensils, spindle-whorls, weights, nails, etc. d De Vaux, in DJD 3:6–7, 16, Figs. 5.12, 19, Pls. V, VII. The bronze ring is plain, with an inner diameter of 3.25cm. The inner diameter is too large for a jewellery piece and, therefore, this plain bronze ring is probably a functional ring. For similar functional rings in the region, see Elena Chernov, “Metal Objects and Small Finds from En-Gedi,” in Hirschfeld, En-Gedi Excavations II, 507–543 (here 509); Ravit Nenner-Soriano, “The Metal Artifacts from the Hasmonean and Herodian Palaces at Jericho and Cypros,” in Bar-Nathan and Gärtner, Hasmonean and Herodian Palaces at Jericho, 270–84 (271, Pl. 11.1:5). The diameter of these rings ranges from 2.1cm to 10cm. As a functional ring, the object from Cave GQ3 could have been used for any number of practical purposes; functional rings were used as fixtures within furniture (Chernov, “Metal Objects and Small Finds,” 509; Nenner-Soriano, “The Metal Artifacts,” 271) or attached to pieces of equipment (de Vaux, in DJD 2:35), for example. Nonetheless, the actual function of the ring in Cave GQ3 is difficult to discern. The perforated stone disc is most probably a spindle whorl (for a similar assessment, see Bélis, “Des textiles,” 219, 222). The diameter of the pierced hole is ca 0.75cm, whereas the diameter of the disc itself is ca. 4cm. These dimensions fit those of items identified as spindle whorls at other sites. The outer diameter of the spindle whorls from Masada, for example, ranges from 2cm to 3cm, whereas the diameter of their perforated hole is typically 0.5cm (see Ronny Reich, “Spindle Whorls and Spinning at Masada,” in Masada VIII: The Yigael Yadin Excavations 1963–1965: Final Reports [ed. J. Aviram et al.; Jerusalem: Israel Exploration Society; The Hebrew University of Jerusalem, 2007], 171–94 [172, table 1]). Several whorls from the Burnt House in Jerusalem have an outer diameter which ranges from 1.75cm to 4.6cm, and a perforation diameter ranging from 0.2cm to 1cm (see Ravit Nenner-Soriano, “Spindle Whorls,” in Jewish Quarter Excavations in the Old City of Jerusalem conducted by Nahman Avigad, 1969–1982: Volume IV: The Burnt House of Area B and Other Studies: Final Report [ed. H. Geva; Jerusalem: Israel Exploration Society; Institute of Archaeology, Hebrew University of Jerusalem, 2010], 276–82). Three spindle whorls from Jericho have an outer diameter of 2.1cm, 2.4cm, and 1.8cm, and an inner diameter of ca. 0.5cm, 0.4cm, and 0.2cm (not drilled through), respectively (see Ravit Nenner-Soriano, “The Miscellaneous Finds from the Hasmonean and Herodian Palaces at Jericho and Cypros,” in BarNathan and Gärtner, Hasmonean and Herodian Palaces at Jericho, 285–89 [286, Pls. 12.1:9–11]). Two spindle whorls from ‘Ein Gedi share these same approximate dimensions (see Chernov, “Metal Objects and Small Finds,” 525–26, Fig. 65, Pl. 11:1–2). It should be pointed out that while spindle whorls are typically conical, flat examples are also attested (see, for example, Nenner-Soriano, “Spindle Whorls,” Pls. 9.1–9.2). Therefore, the flat shape of the perforated disc from Cave GQ3 does not rule out its interpretation as a spindle whorl. The fact that the pottery from Cave GQ3 is exclusively late Second Temple in date—and the repertoire includes Qumran-type cylindrical jars—suggests that Cave GQ3 was occupied most intensively during the 1st centuries BCE and CE. Therefore, it is possible that the bronze ring and the spindle whorl date to this same period. However, despite the lack of any prehistoric pottery, Cave GQ3 still contained two prehistoric flint blades. This puts the Second Temple

Mizzi dating of the spindle whorl into question, as spinning activity goes back to prehistoric times, and stone spindle whorls have been found in such contexts in Israel (see Philip J. King and Lawrence E. Stager, Life in Biblical Israel [Library of Ancient Israel, Louisville, Kentucky: Westminster John Knox Press, 2001], 152–53; also see Bar Adon, The Cave of the Treasure, 190 and Tamar Schick, “The Early Basketry and Textiles from Caves in the Northern Judean Desert,” ʿAtiqot 41 [2] [2002]: 223–39 for various textiles from the Neolithic and Chalcolithic Periods and the Early Bronze Age, which attest to prehistoric textile technologies in the Judaean Desert). As a result, the dating of the spindle whorl from Cave GQ3 remains uncertain. As far as the bronze ring is concerned, a late Second Temple date is quite possible. This ring is paralleled by several similar rings excavated from Period I and II contexts at Khirbet Qumran (see the list of finds in Humbert and Chambon, Fouilles de Khirbet Qumrân et de Aïn Feshkha). Moreover, the probable dating of the flint blades to the Chalcolithic Period—for which there is plenty of other evidence in the Qumran caves—naturally rules out the attribution of the bronze ring to this period. Although, a post-Second Temple date is also possible, there is no other evidence that unequivocally indicates that Cave GQ3 was used after 70 CE. e De Vaux, in DJD 3:7. De Vaux dates the wooden vessel to the Islamic Period, although he expresses uncertainty on this issue. Indeed, in light of several wooden vessels found in Second Temple contexts at ‘Ein Gedi and other sites in the Judaean Desert, it is most probable that this wooden vessel belongs to the same tradition. Therefore, it could be tentatively dated to the late Second Temple Period. For the wooden vessels from ‘Ein Gedi, see Gideon Hadas, “Nine Tombs of the Second Temple Period at ‘En Gedi,” ʿAtiqot 24 (1994): 3–65 [Hebrew], 1*–8* [English Summary] (5*, Figs. 14:7, 9–10, 15, 18–19; 50:19, 61:13, 15, 18–19, 21–23). Nili Liphschitz, “Timber Analysis of Household Objects in Israel: A Comparative Study,” IEJ 48 (1998): 77–90 lists a number of wooden artefacts found in sites situated in arid regions, such as Masada, Mo’a, ‘En Rahel, ‘Ein Gedi, the Judaean Desert Caves, Oboda, and Nessana, coming from Hellenistic and Roman contexts. f Roland de Vaux, “Fouilles de Khirbet Qumrân: rapport préliminaire sur les 3e, 4e, et 5e campagnes,” RB 63 (1956): 533–77 (573–74); Humbert and Chambon, Excavations of Khirbet Qumran, 67–68. In his notes, de Vaux does not mention the discovery of any cylindrical jars in Cave 11Q; however, there must have been at least one such jar, which is the one in which the Temple Scroll was found. This cylindrical jar was looted by the Bedouin and was later placed in Kando’s shop in Jerusalem. Hanan Eshel and Joan Taylor, personal communication (February 2008 and June 2009, respectively). See also Taylor, The Essenes, 278, n. 29. The identification of the iron rod as a possible awl and the re-classification of the copper buckle as a possible handle loop are based on a study of the metal objects carried out by Dennis Mizzi, Annalisa Faggi, and Vincenzo Palleschi. The study will appear in the forthcoming final report on Cave 11Q. In the process, we also discovered that, contrary to what appears in the list of objects published in Humbert and Chambon, Excavations of Khirbet Qumran, 67–68, there is neither a Roman arrowhead nor an axe in the assemblage. These two objects were inserted into the list as a result of a misreading of de Vaux's documentation. De Vaux states that, within Cave 11Q, three layers of occupation could be distinguished: a layer containing Chalcolithic material,

Miscellaneous Artefacts From The Qumran Caves found on top of bedrock; a layer containing Iron Age II material, found on top of a natural yellow deposit, which lay on top of the Chalcolithic level; and a layer containing late Second Temple material, found on top of the Iron Age II level. According to de Vaux’s preliminary report, all metal objects and all the textiles and basketry come from the upper layer, that is, the one dating to the late Second Temple Period (de Vaux, “Fouilles de Khirbet Qumrân,” 574). In some instances, this conclusion is supported by the details in de Vaux’s excavation notes. For example, de Vaux states that underneath the Bedouin’s dump in front of the cave, Qumran-type pottery, many pieces of cloth, fragments of leather and basketry, a nail, a blade, and clumps of manuscript fragments were found between various blocks of stone. According to de Vaux’s official (unpublished) inventory, the copper handle loop and the chisel likewise come from the area of the Bedouin’s dump, near the cave’s entrance; since these carry the same contextual label—“entreé: déblais”—as the finds which de Vaux explicitly locates underneath the Bedouin’s dump, it is highly probable that the handle loop and the chisel were found together with the rest of the aforementioned items. The close association of all these small finds with Qumrantype pottery strengthens their attribution to the late Second Temple Period. Likewise, the iron key was found together with sherds typical of Qumran. See Humbert and Chambon, The Excavations of Khirbet Qumran, 67–68. The only item that remains unaccounted for, with regard to dating, is the bead; in view of the fact that Cave 11Q was also used in the Chalcolithic Period and in the Iron Age II, the date of this bead remains uncertain. A forthcoming analysis of de Vaux’s documentation concerning Cave 11Q reveals a more complicated picture than the one portrayed by de Vaux in his RB preliminary report. Nonetheless, the dating of the metal artefacts remains unaffected: with a high degree of probability, these can be attributed to the late Second Temple Period. g Harding, in DJD 1:3–7, Pls. I.4–7; de Vaux, in DJD 1:8–17, Figs. 2–3; Pls. II–III; Crowfoot, in DJD 1:18–38. The number of registered cylindrical jars is twelve, as seen in Humbert and Chambon, The Excavations of Khirbet Qumran, 66 as well as in de Vaux’s official (unpublished) inventory. However, in his final publication of the Cave 1Q pottery, de Vaux states that “les éléments d’au moins cinquante jarres ont été conservés” (de Vaux, “La poterie,” 8). In an early report, de Vaux had claimed that 2nd and 3rd century CE pottery was found in Cave 1Q (quoted in Bélis, “Des textiles,” 239), a claim which he later refuted in his final publication of the material (de Vaux, in DJD 1:11). Indeed, the published pottery clearly dates to the late Second Temple Period. The large number of pottery from this period suggests that the rest of the finds from Cave 1Q probably date to the 1st centuries BCE and CE. This is corroborated by the absence of any pottery from other periods, which probably indicates that Cave 1Q was not re-used or disturbed at a later time; the fact that Cave 1Q is in an inaccessible location (see Broshi and Eshel, “Residential Caves,” 332) continues to strengthen the latter probability. In the case of the linen fragments, a Second Temple date is also confirmed by the fact that some linen pieces were found adhered to manuscript fragments which date to this period. h De Vaux, in DJD 3:9, 16, Fig. 5.11, Pls. V, VII. In theory, the wooden point could belong to any period whatsoever, and the presence of a prehistoric flint blade underscores the fact that Cave GQ15 experienced more than one occupation. The fact that the material from

159 Cave GQ15 is predominantly 1st-century BCE–CE in date might favour a Second Temple date, but this remains uncertain. The function of this wooden point is unclear. i De Vaux, in DJD 3:9, 16–17, Pl. VII. The five wooden posts were found together with the pottery, which dates to the late Second Temple Period, underneath a collapse of rocks. Therefore, these wooden posts are probably contemporaneous with this pottery deposit (de Vaux, in DJD 3:16–17). Two of the posts are forked at one end. According to de Vaux, these posts could have been components from tent equipment, which were deposited or stored in Cave GQ17 (de Vaux, in DJD 3:16–17). Bélis disagrees with de Vaux’s interpretation, pointing out that if these wooden posts are indeed the components of a tent, then one would also expect to find remains of the tent’s canvas, ropes, and other components (Bélis, “Des textiles,” 267). Accordingly, Bélis offers an alternative interpretation and suggests that the five wooden posts comprise the components of a vertical loom. Bélis illustrates a reconstruction of a Gallo-Roman vertical loom as supporting evidence for her interpretation (Bélis, “Des textiles,” 267–68, Fig. 6). Moreover, Bélis makes a compelling case in arguing that a loom would have been more precious than tent equipment, which explains the need for hiding the wooden posts. Nevertheless, Bélis’ interpretation suffers from the same critique that she makes regarding de Vaux’s conclusion, namely that if these wooden posts are components of a vertical loom, then one would also expect to find traces of the loom weights and perhaps even pieces of textiles in the process of production (cf. such textiles found in Cave X/51). Of course, one has to take into account the random and fragmentary nature of the archaeological record, but this applies equally to both proposed interpretations. j De Vaux, in DJD 3:12, Figs. 5.16–17; Patrich, “Khirbet Qumran,” 91–93; Joseph Patrich, “Hideouts in the Judean Wilderness,” BAR 15/5 (1989): 32–42 (34). Cave GQ37 seems to be identical to Patrich’s Cave FQ37 (here listed as P37), investigated by Patrich and his team. Cave GQ37 is the largest of all the surveyed caves, and it really comprises a cave complex, with various chambers opening from the main large chamber. In one of these chambers, a layer of charred debris was found. Near the entrance, there is also a large platform or terrace, parts of which preserve traces of a mud floor mixed with palm fronds. All the finds come from this platform or from its immediate vicinity. The four glass plates are not illustrated, but they are described as having folded decorated rims (Patrich, “Khirbet Qumran,” 91–93). From the description, it seems that these glass vessels belong to a series of glass bowls that have a folded rim or bowls that are characterized by a hollow/tubular or a double-hollow/tubular fold below the rim; some of these vessels were also decorated with what is described as a crimped trail (see, for example, Jackson-Tal, “Glass Vessels from En-Gedi,” 474–75, 479, Pls. 1:3–6, 4:1–3; eadem, “The Glass Finds,” 107, 108, 112–13, Pls. 3.5:37–39, 42–45; 3.8:2–6). These types of bowls typically occur in late 1st century CE and early 2nd century CE contexts—they are particularly common in Bar Kokhba caves in the Judaean Desert—as well as in later 3rd century CE contexts (Jackson-Tal, “Glass Vessels from En-Gedi,” 474–75, 479; eadem, “The Glass Finds,” 107, 108, 112). According to de Vaux, the pottery is of the Qumran type. In the case of the items uncovered by Patrich’s team, it is stated that “the corpus as a whole belongs to the last years of the Second Temple Period, but some types started earlier and some can be post-70 CE”

160 (Patrich, “Khirbet Qumran,” 93). The description of two glass vessels as having folded decorated rims suggests that these should be dated to the late 1st-early 2nd century CE. The bronze coin belongs to a group of coins minted in Antioch, typically dated to the 1st and 2nd centuries CE. Therefore, the Early Roman material from Cave GQ37 belongs to the late 1st-early 2nd century CE, and it could date either to around 70 CE or to the Second Jewish Revolt. k De Vaux, in DJD 3:13, 16. According to de Vaux, the majority of the pottery sherds from this cave date to the late Second Temple Period, whereas some sherds date to the Iron Age II. The date of the miscellaneous finds and the textiles is uncertain, but they can tentatively be attributed to the late Second Temple Period on the basis of a number of factors. First, there is a well known tradition of wooden vessels dating to the Second Temple Period; secondly, bronze rods are commonly attested in 1st century BCE–CE contexts at Khirbet Qumran; thirdly, fragments of cloth and leather in Judaean Desert caves typically date to the Chalcolithic, the Early Roman, the Late Roman/Byzantine, and the Islamic Periods, and from these phases, only the Early Roman Period is represented in Cave A; finally, the largest amount of pottery from Cave A dates to the late Second Temple Period. For all these reasons, therefore, these artefacts can be tentatively attributed to the 1st centuries BCE–CE. l De Vaux, in DJD 3:13, 16; idem, “Fouilles de Khirbet Qumrân,” 574. m Patrich and Arubas, “A Juglet containing Balsam Oil(?)”; Patrich, “Khirbet Qumran,” 91; Itah, Kam, and Ben Haim, “Region X,” 171–72, Fig. 11; Donald T. Ariel, “The Numismatic Find,” in Itah, Kam, and Ben Haim, “Region X,” 180, 182 [Hebrew], 172 [English]. This cave is also known as the “Cave of the Balsam Oil Juglet.” It is suggested that this cave might correspond to de Vaux’s Cave GQ2 (see Patrich and Arubas, “A Juglet containing Balsam,” 43–44, n. 3; Patrich, “Khirbet Qumran,” 91). However, owing to the implicit reservations of the investigators, Cave P13 and Cave GQ2 are here distinguished from one another. Bélis suggests that Cave P13 should be identified with Cave GQ8 = 3Q (see Bélis, this volume), but this seems unlikely. Cave P13 was re-investigated in the recent survey of the northern Judaean Desert, where it was designated as Cave X/35 (see Itah, Kam, and Ben Haim, “Region X,” 171–72). As far as dating is concerned, Patrich attributes the majority of the pottery to the late Second Temple Period, noting that “most of it is early, representing period Ib in Qumran, although some types continued in the later period as well. As for the juglet with oil and the jar with the dates, they may date as late as 70 CE (or 68 CE in this region)” (Patrich, “Khirbet Qumran,” 91). Indeed, many of the pottery items published in Patrich and Arubas, “A Juglet containing Balsam,” 48–49, Fig. 3, seem to come from the 1st century CE. One should also note that three of the lamps have a date ranging from the late 1st century CE to the early 2nd century CE (see Itah, Kam, and Ben Haim, “Region X,” 172, Fig. 11). Likewise, the cooking pot is of a type that continues to appear in early 2nd century CE contexts (see Patrich and Arubas, “A Juglet containing Balsam,” 48, Fig. 3:9).

Mizzi Therefore, the material from this cave belongs to the 1st and early 2nd centuries CE, and it is possible that the cave experienced multiple visits or occupations; however, the probability that all the Early Roman material belongs to one occupation—dating to around 70 CE or to the Second Jewish Revolt—should not be excluded. As far as the textiles are concerned, they are dated by the excavators to the Roman and Early Islamic Periods and to the Middle Ages, but no exact details are provided (see Itah, Kam, and Ben Haim, “Region X,” 171–72). n Patrich, “Khirbet Qumran,” 90. Finds of the Pre-Pottery Neolithic B and the Chalcolithic Period are also recorded, but no details are given in Patrich’s report. According to Patrich, all the finds he lists in his brief note on Cave P24 belong to the Roman Period, and he states that “the entire ceramic corpus from this cave . . . suggest the last phase of the Herodian period or even a post 70 CE date.” Therefore, the miscellaneous objects, the biofacts/ecofacts, and the textiles may be dated to around 70 CE or later. o Itah, Kam, and Ben Haim, “Region X,” 172–73 [English], 182–83 [Hebrew], Fig. 15. This cave is also known as the “Cave of Leather.” The arrowhead is paralleled by similar arrowheads found in early 2nd century CE contexts in the “Cave of the Letters” (Itah, Kam, and Ben Haim, “Region X,” 173); however, it should be noted that such arrowheads were also used during the 1st century CE (see, for example, Guy D. Stiebel and Jodi Magness, “The Military Equipment from Masada,” in Masada VIII: The Yigael Yadin Excavations 1963– 1965: Final Reports [ed. J. Aviram et al.; Jerusalem: Israel Exploration Society; The Hebrew University of Jerusalem, 2007], 1–94 [22–26]). The bowl could date to the late Second Temple Period or to the late 1st century CE and the early 2nd century CE. The tentative dating of the leather fragments is based on its correlation to late 1st–early 2nd century CE material, and on the fact that the other sherds in Cave X/51 are non-diagnostic (therefore, there is no certain evidence for another occupation). p Itah, Kam, and Ben Haim, “Region X,” 173 [English], 183–84 [Hebrew], Fig. 18; Donald T. Ariel, “The Numismatic Find,” in Itah, Kam, and Ben Haim, “Region X,” 184 [Hebrew], 173 [English]. This cave is also known as the “Cave of the String.” The finds in this cave are tentatively dated to the Late Roman and Byzantine Periods on the basis of the 5th century CE coin found in the same cave. q Baruch, Mazor, and Sandhaus, “Region XI,” 180 [English], 194 [Hebrew], Fig. 10. The worked bone and the biofacts/ecofacts were found together with Early Roman pottery, within a fill in a hewn cell in the western wall of Cave XI/16; therefore, they are probably all contemporaneous. r Cohen and Yisraeli, “The Excavations of Rock Shelter XII/50 and in Caves XII/52–53,” 207–13 [Hebrew], 185–86 [English]. s Ibrahim, “The Excavation of Cave XII/56,” 215–16 [Hebrew], 186 [English]. t Tal and Oron, “Region XIII,” 188–89 [English], 222–23 [Hebrew], Fig. 2. This cave is fronted by a terrace.

CHAPTER 10

The Distribution of Tefillin Finds among the Judean Desert Caves Yonatan Adler 1 Preface Since the publication of the final remaining texts from Qumran Cave 4Q, scholarship has debated the relationship between the manuscript corpora from the eleven caves near Qumran in which scrolls were found. Did all of these assemblages derive from a single library, or should the corpus from each cave be viewed as reflecting a distinct collection? Can the corpora from the various caves be distinguished on the basis of chronology, subject matter or perhaps religio-ideological orientation? Closely related to these problems is the debate regarding the nature of these collections; are some or all of these scroll assemblages the result of emergency concealment? Should any of these collections be viewed as “genizah” repositories? Did some of these caves serve as active, functioning libraries? Do any of these corpora represent the personal collections left behind by individuals who may have utilized these caves for human habitation? When dealing with these questions, scholars have usually focused on the texts themselves, and more infrequently on archaeological artifacts, such as pottery, found together with the texts. The present article seeks to contribute to the discourse regarding the enigmatic nature of the various Qumran manuscript corpora and their depositional backgrounds by focusing on one specific type of artifact found in some of the manuscript caves but which has often been overlooked: tefillin (phylacteries).1 As

Research for this study was conducted while serving as a Lady Davis postdoctoral fellow at the Institute of Archaeology of the Hebrew University, under the kind and able aegis of Prof. Zeev Weiss.  I would like to extend my appreciation to all of the authorities who have graciously provided me complete access to the artifacts whose study forms the basis of this article: Dr. Zvi Greenhut, head of the Artifacts Treatment and Conservation Department at the Israel Antiquities Authority (IAA); Dr. Orit Shamir, curator and head of the Organic Materials Division at the IAA, together with her assistant Dr. Naama Sukenik; Mrs. Pnina Shor, curator and director of the Dead Sea Scrolls Projects Unit of the IAA; and Dr. Adolfo Roitman, curator and director of the Shrine of the Book. 1  As the English word “phylacteries” (from the Greek: φυλακτήρια) is a loaded term, implying an apotropaic function, I prefer here to use the more neutral Hebrew/Aramaic “tefillin.” For a detailed investigation into a possible magical function of tefillin in ancient times,

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_012

t­ efillin are manifestly ritual objects which fulfilled a specific function, their very presence in any particular cave may potentially reflect both on the nature of the textual deposits found in the same cave together with the tefillin, as well as on the motivations of the people who made these deposits in antiquity. Recognizing the existence of typological differences between the tefillin exemplars allows us to go one step further, and to examine how the distribution of the various types of tefillin among the caves may shed light on differences in the nature of the associated textual assemblages found in these caves in terms of chronology and possibly religio-ideological orientation as well. Our study opens with a survey of all of the tefillin findspots among the caves, both at Qumran and elsewhere in the Judean Desert. The investigation will proceed by identifying two typological classifications within the corpus of tefillin found in these caves—one relating to the morphology of the leather tefillin cases and the other involving the choice of texts included on the tefillin slips and their arrangement—with the aim of mapping out the distribution of distinct tefillin types according to find-spot among the various caves. Our study concludes with an analysis of how these data may contribute to our understanding of the functions each of various caves may have served at the time when the scrolls were placed in them, as well as the nature of the diverse scroll assemblages themselves, especially with regard to chronology and the possibly diverse religio-ideological backgrounds of the individuals who deposited the scrolls in the each of the different caves. A cautionary note regarding the provenance of the material to be discussed in the present study is in order here. While a large number of the finds which are the focus of our investigation were discovered by archaeologists, many were purchased from Bedouin clansmen or their middlemen after having been clandestinely looted from Judean Desert caves. While the scholars who purchased these materials certainly tried to the best of their see: Yehudah B. Cohn, Tangled Up in Text: Tefillin and the Ancient World (BJS 351; Providence, RI: Brown University, 2008); see also: Yonatan Adler, Review of Tangled Up in Text: Tefillin and the Ancient World, by Yehudah B. Cohn, Zion 66 (2001): 383–87 [Hebrew].

162 abilities to determine the true provenance of the materials through questioning the Bedouin finders or their middlemen, caution must always be exercised whenever dealing with finds which were not uncovered within the framework of scientific archaeological excavations.2 Experience has proven that not all of the provenance claims of the Bedouin finders or their intermediary dealers are reliable.3 The only way to be sure that a purchased find was discovered where it was alleged to have been found is if fragments of the same document were subsequently uncovered at the reported site within the framework of controlled scientific excavations. To the best of my knowledge, such was never the case with purchased tefillin finds. With this in mind, I have been careful here to note clearly which tefillin finds were uncovered through proper archaeological excavation (I refer to these simply as having been “found” at a certain location) and which

2  Stephen Reed has recently provided an excellent discussion of this problem together with an historical survey of the first six decades of scholarship’s handling of this troublesome issue; see: Stephan A. Reed, “Find-Sites of the Dead Sea Scrolls,” DSD 14 (2007): 199–221. 3  For example, the dealer who sold the first documents later determined to have been found in the Wadi Murabbaʿat caves (Khalil Iskandar Shahin, known as “Kando”) at first claimed that they were found in Qumran Cave 1Q, and only after having been bombarded by de Vaux with questions recanted, alleging instead that in truth they came “from elsewhere, but . . . very nearby” (Roland de Vaux, “Historique des Découvertes,” in Les grottes de Murabbaʿât [DJD 2; ed. P. Benoit, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1961], 4). For an example of a document purchased from the Bedouin finders who claimed to have discovered it in Qumran “Cave 4Q,” but which in reality proved to derive from Na�al Ṣeʾelim (Wadi Seiyal), see: Ada Yardeni, “Introduction to the ‘Qumran Cave 4’ Documentary Texts,” in Aramaic, Hebrew and Greek Documentary Texts from Naḥal Ḥever and Other Sites: With an Appendix Containing Alleged Qumran Texts (The Seiyâl Collection II) (DJD 27; ed. H. M. Cotton and A. Yardeni; Oxford: Clarendon Press, 1997), 283. For purchased texts purported to have been found at Na�al Ṣeʾelim but which were later shown to have come from Na�al Ḥever, see: Jonas C. Greenfeld, “The Texts from Na�al Ṣeʾelim (Wadi Seiyal),” in The Madrid Qumran Congress: Proceedings of the International Congress on the Dead Sea Scrolls, Madrid 18–21 March, 1991 (STDJ 11; ed. J. C. Trebolle Barrera and L. Vegas Montaner; Leiden: Brill, 1992), 2:661–65; Hannah M. Cotton and Ada Yardeni, “General Introduction,” in DJD 27:1–6; James Charlesworth et al., “Biblical Scrolls from Na�al Ḥever and ‘Wadi Seiyal’: Introduction,” in Miscellaneous Texts from the Judean Desert (DJD 38; Oxford: Clarendon Press, 2000), 133–34. For a tefillin case into which the Bedouin finders (or the antiquities dealer who made the final sale) inserted a tefillin slip from elsewhere in order to be able to sell a “complete set,” see: Yigael Yadin, Tefillin from Qumran (XQ Phyl 1–4) (Jerusalem: Israel Exploration Society/ Shrine of the Book, 1969), 11–13.

Adler

of these finds were purchased from the Bedouin or their intermediaries. 2

Overview of the Tefillin Find-Spots

In discussing “tefillin,” we are actually referring to artifacts composed of two distinct components: (1) leather cases, intended to house (2) rolled-up slips made of thin skins inscribed with biblical texts. In only a limited number of instances were inscribed slips found in-situ inside of a leather case; usually tefillin slips were found unassociated with any case, while cases were typically found empty of any contents. Twenty-three leather tefillin cases have been reported upon from the caves near Qumran: four were found in Qumran Cave 1Q, three were found in Cave 4Qa and another eleven cases which were purchased from Bedouin are attributed to “Cave 4Q,” one was found in Cave 5Q, three in Cave 8Q, and one purchased from Bedouin was said to derive from an unspecified cave at Qumran.4 Elsewhere in 4  Qumran Cave 1Q: Gerald L. Harding, “Introductory, the Discovery, the Excavation, Minor Finds,” in Qumran Cave 1 (DJD 1; ed. D. Barthélemy and J. T. Milik; Oxford: Clarendon Press, 1955), 7, Pl. I:5–7. Aside from the four cases which appear in the report, Harding noted that remains of at least two other cases similar to those shown in Pl. I:6–7 were found in Cave 1Q; as no other description is provided and no record of these cases could be located, they are not counted here. Qumran Cave 4Q: Thirteen cases (only the first three of which were discovered in situ in Cave 4Qa, the remainder having been purchased from Bedouin who claimed to have found them in “Cave 4Q”) were reported upon in: Józef T. Milik, “Tefillin, Mezuzot et Targums,” in Qumrân Grotte 4.II: I. Archéologie; II. Tefillin, Mezuzot et Targums (4Q128–4Q157) (DJD 6; ed. R. de Vaux and J. T. Milik; Oxford: Clarendon Press, 1977), 33–35, Pl. VI:1–13. An additional case attributed to “Cave 4Q,” (it is unclear if it was purchased or if it was discovered during excavations) was brought to the University of Leeds for analysis in 1958 along with other uninscribed skins from Qumran, and was assigned the internal serial designation “4Q45” (John B. Poole, The Nature, Origins and Techniques of Manufacture of those of the Dead Sea Scrolls which are Made from Animal Skins [Ph.D. diss., University of Leeds, 1959], 100–2, 115–16, Pl. XL; Ronald Reed and John B. Poole, “A Study of Some Dead Sea Scroll and Leather Fragments from Cave 4 at Qumran: Part I— Physical Examination,” Proceedings of the Leeds Philosophical and Literary Society [Scientific Section] 9/1 [1962]: 1–13, [7, 9, Fig. 1, 13]; Ronald Reed, Ancient Skins, Parchments and Leathers [London: Seminar Press, 1972], 299–300, Fig. 11; this exemplar has gone completely unnoticed in all previously published treatments of the Judean Desert tefillin, probably due to the specialized nature of the publications where it was reported upon. Another purchased fragment allegedly from “Cave 4Q” which might derive from another tefillin case appears in the upper right-hand side of an archival

Distribution Of Tefillin Finds Among The Judean Desert Caves

the Judean Desert, one tefillin case was found in Cave 1 at Wadi Murabbaʿat and another—only recently identified as such—was found in Cave 34 at Na�al Ṣeʾelim (Wadi Seiyal).5 As noted, most of these cases were found empty; rolledup slips were found in situ inside only five of the cases: four slips in a case found in Qumran Cave 4Qa, three slips in an additional fragmentary case found in this same cave, four slips in a case purchased from Bedouin and alleged to derive from “Cave 4Q,” three slips in the fragmentary case found in Cave 5Q, and four slips in a case alleged to derive from an unspecified cave near Qumran (of which one slip is apparently not original to the case).6 To date, slips from only two of these cases have been unrolled and successfully deciphered: three slips from the abovementioned fragmentary case found in Cave 4Qa, and another three slips from the case reported to derive from an unspecified cave at Qumran.7 Twenty-eight additional slips were found without any clear association with any leather case: one slip was found in Qumran Cave 1Q, eighteen slips are alleged to derive from “Cave 4Q” (aside from the aforementioned slips found photograph (PAM 43.319, to the left of a fragment which probably derives from Milik’s case 9). It is impossible to know which of the purported “Cave 4Q” materials purchased from the Bedouin may be attributed to Cave 4Qa and which to Cave 4Qb. Qumran Cave 5Q: Józef T. Milik, “Textes de la grotte 5Q: Phylactère,” in Les ‘petites grottes’ de Qumrân: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 178, Pl. XXXVIII:8. Qumran Cave 8Q: Roland de Vaux, “Archéologie,” in DJD 3:31, Pl. VIII:5–6. Unspecified Qumran Cave: Yadin, Tefillin. 5  Wadi Murabbaʿat: Roland de Vaux, “Archéologie: La période Romaine,” in DJD 2:44, Pl. XIV:4. Naḥal Ṣeʾelim: The case was discovered in Cave 34 at Na�al Ṣeʾelim during Yohanan Aharoni’s excavations at the site on January 30–31, 1960. The find is mentioned nowhere in any of Aharoni’s publications on the excavations; it was only recently recognized as a tefillin case by Dr. Orit Shamir, curator and head of the Organic Materials Division at the Israel Antiquities Authority, under whose care the organic materials from this excavation are entrusted. I thank Dr. Shamir for bringing this artifact to my attention in February 2014, and for providing me access to study it. Full scientific publication of this tefillin case will appear in: Yonatan Adler and Ada Yardeni, “Remains of Tefillin from Naḥal Ṣeʾelim (Wadi Seiyal): A Leather Case and Two Inscribed Fragments (34Se 1 A–B).” Forthcoming. 6  Qumran Cave 4Q: Milik, in DJD 6:34, 55–57; Qumran Cave 5Q: idem, in DJD 3:178; Unspecified Qumran Cave: Yadin, Tefillin. Yadin believed that one of the four slips found in this case had been placed there by the Bedouin who discovered the case in order to produce a “complete” set which might be better sold at the antiquities market (Yadin, Tefillin, 12–13). 7  Milik, in DJD 6:55–57; Yadin, Tefillin.

163

in situ inside cases discovered in Cave 4Qa), four slips were found in Cave 8Q, two slips purchased from Bedouin are purported to derive from a cave at Wadi Murabbaʿat, two slips were found in Cave 34 at Na�al Ṣeʾelim, and an additional slip was discovered by Bedouin and purported to have been found at Na�al Ṣeʾelim (bisected at the time by the 1949 Israel-Jordan armistice line), although there is reason to believe that it may actually have been discovered at Na�al Ḥever (in Israel).8 In Qumran Cave 11Q, a number of small wads and fragments with undeciphered texts were found.9 Although the publishers noted the similarity of these wads to tefillin slips from the other caves in terms of the extremely thin skin upon which they are written (ca. 0.05–0.1 mm), the manner in which the skins were folded, and the diminutive size of the letters (ca. 1 mm) with hardly any space 8  Qumran Cave 1Q: Dominique Barthélemy, “Textes Bibliques,” in DJD 1:72–76, Pl. XIV. Qumran Cave 4Q: Milik, in DJD 6:48–79, Pl. VII–XXV; For a preliminary publication of fragments from four of these slips, see Karl G. Kuhn, Phylakterien aus Höhle 4 von Qumran, Abhandlungen der Heidelberger Akademie der Wissenschaften, Philosophisch-Historische Klasse 1 (Heidelberg: C. Winter, 1957). For a recent treatment of these fragments, see: Anna Busa, Die Phylakterien von Qumran (4Q128.129.135.137) aus der Heidelberger Papyrussammlung (PHeid Neue Folge 15; Heidelberg: C. Winter, 2015). All of these were purchased from the Bedouin (Kuhn, Phylakterien, 33). Qumran Cave 8Q: Maurice Baillet, “Textes des grottes 2Q, 3Q, 6Q, 7Q, à 10Q,” in DJD 3:149–57, Pls. XXXII–XXXIII. Although Baillet presumed that the four groups of slips derived from a single tefillin set, Yadin (Tefillin, 15 [n. 28], 34) appears to have been correct in viewing these slips as independent of one another; cf. Maurice Baillet, “Nouveaux phylactères de Qumrân (XQ Phyl 1–4): à propos d’une édition récente,” RevQ 7 (1970): 414. Wadi Murabbaʿat: Józef T. Milik, “Textes hébreux et araméens,” in DJD 2:80–85, Pl. XXII–XXIV. Milik believed that these two slips originated from a single tefillin set. Both fragments were purchased from Bedouin; see: de Vaux, in DJD 3:6. Naḥal Ṣeʾelim: Yohanan Aharoni, “Expedition B,” IEJ 11 (1961): 22–23, Pl. 11:B–C (note mistaken caption); Adler and Yardeni, “Remains.” Aharoni believed that both slips belonged to a single tefillin set. Naḥal Ḥever/Naḥal Ṣeʾelim: Matthew Morgenstern and Michael Segal, “6. XḤev/Se: Phylactery,” in DJD 38:183–91, Pl. XXX. The slip consists of two fragments which, according to the publishers, were probably joined at the time of writing. For problems with establishing the provenance of the texts from the so-called “Seiyal collection,” see: Cotton and Yardeni, in DJD 27:1–6; Charlesworth et al., in DJD 38:133–34. 9  Florentino García Martínez, Eibert J. C. Tigchelaar, and Adam S. Van der Woude, “11QUnidentified Wads,” in Qumran Cave 11.II: 11Q2–18, 11Q20–31 (DJD 23; ed. idem; Oxford: Clarendon Press, 1998), 445–46, Pl. LII. It is unclear if some or all of this material was purchased from the Bedouin, along with most of the material purported to derive from Cave 11Q, or if it was found along with the scanty textual remains uncovered in de Vaux’s excavations at the site in February 1956 after the cave had already been looted.

164 between the lines, they objected to the identification of the wads as tefillin for three reasons: (1) the wads are larger than the tefillin slips found in other caves; (2) the wads were not found in a tefillin case and no such cases were in fact found in Cave 11Q; (3) there are too many wads to have belonged to one tefillin set. Yehudah Cohn has already noted that none of these arguments is compelling, as a similar wad found in Cave 1Q was deciphered and identified as a tefillin slip.10 It should be noted that the editors of the Cave 11Q material mistakenly note that elsewhere at Qumran “all the phylacteries were found in tefillin boxes;” as we have seen, the majority of tefillin slips from Qumran and elsewhere in the Judean Desert were actually not found inside tefillin cases. It may be suggested that these wads are actually formed of numerous slips which may have originally been placed in multiple tefillin cases (now lost), but later removed and folded together prior to their final deposit in the cave. Certainly any conclusive verdict on the matter must await the opening and deciphering of these very brittle and damaged wads in order to determine if their texts are otherwise compatible with tefillin found in the other caves.11 To summarize, tefillin remains (cases and slips) have been found at Qumran Caves 1Q, 4Q, 5Q, 8Q and possibly 11Q, as well as at caves in Wadi Murabbaʿat, Na�al Ṣeʾelim and possibly also Na�al Ḥever. Each of the aforementioned Qumran caves (with the exception of Cave 11Q) produced tefillin finds discovered within the framework of controlled archaeological excavations, as did Cave 1 at Wadi Murabbaʿat and Cave 34 at Na�al Ṣeʾelim. Here it is important to be reminded that absence of evidence is not evidence of absence; that is to say that lack of tefillin finds from any of the other Judean Desert caves should never be taken as evidence that tefillin were not deposited there in antiquity.12 This is particularly true of tefillin slips, which 10  Cohn, Tangled Up, 58. 11  For the possibility that other fragments from Qumran and Wadi Murabbaʿat may have originally derived from tefillin, particularly those identified by the original publishers as “mezuzot,” see: Hartmut Stegemann and Jürgen Becker, “Zum Text von Fragment 5 aus Wadi Murabbaʿat,” RevQ 11 (1962): 443–48; Cohn, Tangled Up, 60–62, 98, 103. See also Cohn, Tangled Up, 67, 75–77 for other Biblical excerpts from Qumran which resemble tefillin in terms of content. 12   Contra Stephen J. Pfann (“Reassessing the Judean Desert Caves: Libraries, Archives, Genizas and Hiding Places,” BAIAS 25 [2007]: 158, 164, 168), who has asserted that the absence of tefillin finds from Caves 2Q, 3Q and 11Q is particularly significant for his proposed classification of the various caves as “priestly Essene,” “lay Essene,” “priestly rebel” and “lay rebel.” Aside from the significant methodological problem of argumentum e silentio noted

Adler

when rolled up are miniscule in size (measuring no more than a few millimeters in length, width and girth) and are a beige-grey color which is easily camouflaged in the types of soil prevalent in both the limestone and the marl caves found in the area. It is quite likely that numerous tefillin remains deposited in caves have simply gone unnoticed by excavators and hence are now lost to scholarship. 3

The Distribution of Two Types of Tefillin Cases

The Judean Desert tefillin cases are constructed of a small rectangular piece of leather, folded over to create a pouch into which inscribed slips might be placed, and sewn shut on all four sides. In some of these cases, the pouch is composed of a single bulge in the leather into which one or more slips might have been placed.13 In the remaining cases the pouch is divided into multiple bulges, or compartments, into which slips were to be placed.14 All of the complete cases of this type have four compartments of more or less equal size, however in two poorly preserved cases only three compartments have survived.15 here, it is quite unclear what leads Pfann to believe that presence or absence of tefillin might be associated with one of these supposed groups more than with another. 13  Two such cases were found in Qumran Cave 1Q (Harding, in DJD 1: Pl. I:6–7), at least two others are purported to derive from “Cave 4Q” (Milik, in DJD 6: Pl. VI:11–2, and perhaps fragments 3 and 13 there as well), and one more was found in Cave 8Q (de Vaux, in DJD 3: Pl. VIII:6). No Slips were found in situ in any of these cases. 14  Two such cases were found in Qumran Cave 1Q (Harding, in DJD 1: Pl. I:5), two were found in Cave 4Qa and another eight are alleged to have been found in “Cave 4Q” (Milik, in DJD 6: Pls. VI:1–2, 4–10; Reed and Poole, “A Study,” 7), one was found in Cave 5Q (Milik, in DJD 3: Pl. XXXVIII:8), another two were found in Cave 8Q (de Vaux, in DJD 3: Pl. VIII:5), one was alleged to have been found in an unspecified Cave near Qumran (Yadin, Tefillin), one was found in Cave 1 at Wadi Murabbaʿat (de Vaux, in DJD 3: Pl. XIV:4), and one was found in Cave 34 at Na�al Ṣeʾelim (Adler and Yardeni, “Remains”). 15  Milik, in DJD 3: Pl. XXXVIII: 8; idem, in DJD 6:35, Pl. VI:1. The fact that both of these cases are poorly preserved and not complete is apparent upon visual inspection of the actual artifacts. Cohn (Tangled Up, 57–58) has suggested that the two cases with only three surviving compartments belong to a distinct “threecell type;” in personal communication, however, he has indicated to me that this assessment relied only on the description provided in Milik’s report rather than on any personal inspection of the actual artifacts. David Nakman (“The Contents and Order of the Biblical Sections in the Tefillin from Qumran and Rabbinic Halakhah: Similarity, Difference, and Some Historical Conclusions,” Cathedra 112 [2004]: 30 [Hebrew]) has claimed

Distribution Of Tefillin Finds Among The Judean Desert Caves

Members of the original scroll team differentiated between these two types of cases, referring to those with a single compartment as “arm-tefillin” and those with multiple compartments as “head-tefillin.”16 While the rationale for this classification is never stated explicitly, it seems likely that these scholars interpreted these early finds on the basis of their familiarity with significantly later practices, and perhaps also in light of somewhat later rabbinic literature which differentiates between tefillin worn on the head and tefillin worn on the arm.17 Recently, Cohn has contested this classification, suggesting that both the single-compartment type and the multiple-compartment type may reflect alternative understandings of how the practice was to be observed, with individual practitioners wearing only one kind of case.18 While possible, it does that “the photograph [sic]” of the tefillin case from Cave 4Qa with three surviving compartments shows the remains of a fourth compartment. It appears that the “photograph” he refers to is the poor quality reproduction of two photographs, PAM 43.446 and PAM 43.447, which appears in Milik’s report (DJD 6: Pl. VI:1). An inspection of the actual PAM images reveals that what evidently appeared to Nakman like a protrusion of leather in the published photographs is actually the dangling end of the flax thread used to sew the case closed. 16  Harding, in DJD 1:7; Milik, in DJD 6:34–35; cf. Yadin, Tefillin. Note that the original title of the monograph, which first appeared as an article in Hebrew, is actually: “‫ראש מקומראן‬-‫של‬-‫”תפילין‬. 17  No single rabbinic source presents an unambiguous description of exactly how the construction of a tefillin case worn on the arm differs from that of one worn on the head. It is implied in the Tosefta (t. Kelim BB 4:1 [ed. Zuckermandel 593]) that head-tefillin are distinct from arm-tefillin in that the former are comprised of separate units, each called a ‫( קציעה‬or perhaps ‫ ;קציצה‬see: Saul Lieberman, Tosefeth Rishonim: A Commentary [Jerusalem: Bamberger and Wahrmann, 1939], 3:81 [Hebrew]) which are individually stitched. According to Mekhilta de-Rabbi Ishmael (Tractate Pis�a 17 [ed. Lauterbach2, 102]; 18 [ed. Lauterbach2, 114]), tefillin worn on the arm are to be written on one “roll” (‫)כרך אחד‬, whereas tefillin worn on the head are to be written as four “ṭôṭāpōt” which are to be made as “one pouch of four [scriptural] portions” (‫ ;כיס אחד של ארבע פרשיות‬cf. Sifre Deut 35 [ed. Finkelstein 63–64]; see also m. Sanh. 11:3 where the practice of using four “ṭôṭāpōt” is alluded to). According to a beraita recorded in b. Menaḥ. 34b, tefillin are to be written on four separate slips and placed within “four housings in one leather” (‫ ;)ד' בתים בעור אחד‬since the prooftext here is the word ‫“( ולזכרון‬and as a reminder [between your eyes]”; Exod 13:9), this pericope could be understood as referring specifically to tefillin worn on the head, although this is by no means the certain intention. The Talmud (loc. cit.) then quotes another beraita which indicates that arm-tefillin are to be written on one slip which is placed in a single housing. 18  Cohn, Tangled Up, 58–59. One of the reasons Cohn cites for contesting the possibility that the Qumran tefillin cases reflect

165

not seem unlikely that a differentiation between multicompartment head-tefillin and single-compartment armtefillin may have been practiced for some centuries prior to the redaction of the rabbinic sources where such a practice is already assumed.19 If in fact these two types of cases reflect nothing more than two elements of a single practice rather than two competing practices, it should come as no surprise that in every Judean Desert cave where more than one tefillin case was found (Qumran Caves 1Q, 4Q and 8Q), both kinds of cases were brought to light.20 As I have argued elsewhere, an additional taxonomic distinction should be made among the Judean Desert tefillin cases between two discrete subtypes of multi-compartment cases which quite likely reflect two competing tefillin practices.21 In the first subtype, which we might call the “simple-type,” a single line of stitching separates the compartments from one another. In the second subtype, which I call the “split-type,” the compartments are separated by incisions in the leather, and each compartment is stitched closed individually.22 The difference between these two morphological subtypes is easily apparent upon the kind of differentiation alluded to in rabbinic sources is that some of the Qumran cases clearly do not coincide with rabbinic prescriptions, namely—the supposed “three-cell type.” However, see above, n. 15. 19  See above, n. 17, especially t. Kelim BB 4:1 (ed. Zuckermandel 593), where the morphological distinction between multi-­ compartment head-tefillin cases and single-compartment armtefillin cases is assumed as opposed to prescribed, which may indicate that the practice was not a new one. 20  It should be noted that all of the single-compartment cases from “Cave 4Q” were purchased; the three cases found in the excavations at Cave 4Qa are all multi-compartment exemplars (Milik, in DJD 6:33–35). 21  Yonatan Adler, “Two Types of Multi-Compartment Tefillin Cases from the Judean Desert Caves” (forthcoming). 22  Although Milik noted the incisions on the three cases of this type which he published (Milik, in DJD 6:35), he made no mention of the marked difference in stitching found on these two types. Based on Milik’s description, David Rothstein has commented: “It is not certain whether the incisions are part of the capsule’s original construction or simply the result of age” (David Rothstein, From Bible to Murabbaʿat: Studies in the Literary, Textual and Scribal Features of Phylacteries and Mezuzot in Ancient Israel and Early Judaism [Ph.D. diss., University of California, Los Angeles, 1992], 365). Although indiscernible on the published photographs, sewing-holes indicating the individual stitching surrounding each of the compartments on the split-type cases are readily apparent upon visual inspection of the actual objects. Close examination of the incisions reveals that the leather was cut with a sharp object, and the intentional nature of the incisions is confirmed again by the stitching design.

166

Adler

causal observation: whereas the simple-type case retains its rectangular appearance, the split-type case displays a splayed, fan-like appearance. There is no readily apparent functional explanation for the differences between these two designs; whereas the method used to partition the compartments in the simple-type case is straightforward, the technique ­whereby the individual compartments in the split-type case are separated appears significantly more labor intensive, while not providing any obvious functional advantage. In consideration of the incontrovertibly ritual nature of the objects under consideration, it seems likely that some kind of ritual issue is at stake. In my typological study of these two kinds of tefillin cases, I further suggested the possibility that an allusion to these two types as competing halakhic practices, with the split-type rejected by the rabbis, may be seen in two parallel tannaitic sources.23 The majority of multiple-compartment tefillin cases found in the Judean Desert belong to the simple-type variety; two were found in Qumran Cave 1Q, two were found in Cave 4Qa and another four purchased cases are attributed to “Cave 4Q,” one was found in Cave 5Q, two in Cave 8Q, one was alleged to have been found in an unspecified Qumran cave, one was found in Cave 1 at Wadi Murabbaʿat, and one in Cave 34 at Na�al Ṣeʾelim.24 The only known examples of split-type tefillin cases are three cases among the purchased artifacts attributed to “Cave 4Q.”25 The significance of this distribution will be explored below. 4

The Distribution of Two Types of Tefillin Slips

While all of the tefillin slips from the Judean Desert contain scriptural pericopes excerpted exclusively from Exodus, Deuteronomy or both, the texts displayed by the various inscribed slips are incredibly diverse in terms of the exact verses chosen to be included, their order, textual variants, and the orthography and morphology of individual words. As differences in orthography, morphology and textual character between the various tefillin slips simply parallel such diversity among other literary texts 23   Mekhilta de-Rabbi Ishmael (Tractate Pis�a 17 [ed. Lauterbach2, 102]): ‫] או יעשה ארבעה כיסין של ארבע פרשיות? תלמוד‬. . .[ ‫ 'ולזכרון בין עיניך'—כיס אחד של ארבע פרשיות‬:‫“( ”לומר‬Perhaps, then, one should make separate containers for the four scriptural sections? Against this Scripture says: ‘And for a memorial between your eyes’ [Ex. 13:9], meaning one container for all the four scriptural sections”). In the parallel source found in Sifre Deut (35 [ed. Finkelstein 63–64]), the enigmatic term “ṭôṭāpōt” is found in place of “scriptural sections.” 24  See above, nn. 4–5. 25  Milik, in DJD 6: Pls. 6:5, 6, 9.

found in the Judean Desert, I have chosen to concentrate here on features unique to the tefillin ritual, namely—the choice and exact scope of scriptural pericopes included, as well as the order whereby the selected passages are arranged.26 After analyzing how the various tefillin slips may be profitably classified into distinct types with regard to these features, I shall map out the distribution of findspots among the Judean Desert caves of each type of ­tefillin slip. It seems beyond doubt that early tefillin practitioners based their observance of this ritual on a practical understanding of the four very similar verses in the Pentateuch where we find an imperative that something should “be as,” or should be “tied as,” a “sign” upon the arm, and concurrently that this same thing should “be as a reminder,” or “be as ṭôṭāpōt” (however this enigmatic word may have been understood), between the eyes: (1) Exod 13:9; (2) Exod 13:16; (3) Deut 6:8; (4) Deut 11:18. While in the Exodus verses it is not completely clear what exactly it is that aught “be as”/“tied as” “a sign”/“a reminder”/“ṭôṭāpōt,” the Deuteronomy verses evidently refer back to “these words” (Deut 6:6), or “these words of Mine (i.e. God’s)” (Deut 11:18), which are to be both tied to the arm and also “as ṭôṭāpōt” between the eyes. All of the tefillin texts discovered in the Judean Desert (with the special exception of 4Q141) contain exclusively one or more of these four verses, and/ or a large selection of verses found preceding these four verses such as were probably understood as the referent of “these words.”27 To this extent the choice of texts included 26  For the possible significance of differences between the various tefillin slips in terms of orthography, morphology and textual character, see: Emanuel Tov, “Tefillin of Different Origin from Qumran?” in A Light for Jacob: Studies in the Bible and the Dead Sea Scrolls in Memory of Jacob Shalom Licht, ed. Y. Hoffman and F. S. Pollak (Jerusalem: Bialik Institute, 1997), 44*–54*; idem, “Further Evidence for the Existence of a Qumran Scribal School,” in The Dead Sea Scrolls Fifty Years after their Discovery: Proceedings of the Jerusalem Congress, July 20–25, 1997 (ed. L. H. Schiffman, E. Tov, and J. C. VanderKam; Jerusalem: Israel Exploration Society/Shrine of the Book, 2000), 213–16; idem, Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (STDJ 54; Leiden: Brill, 2004), 270–71; Yonatan Adler, “Identifying Sectarian Characteristics in the Phylacteries from Qumran,” RevQ 89 (2007): 79–92. Essential to both authors’ arguments is Tov’s theory regarding a specific “Qumran scribal practice” with regards to orthography and morphology. For an important recent critical assessment of this theory, see: Eibert J. C. Tigchelaar, “Assessing Emanuel Tov’s ‘Qumran Scribal Practice,’ ” in The Dead Sea Scrolls: Transmission of Traditions and Production of Texts (STDJ 92; ed. S. Metso, H. Najman, and E. Schuller Leiden: Brill, 2010), 173–207. See also below, n. 34. 27  Although some of the slips which feature the Deuteronomic Decalogue also include verses from Exod 20, these should be

Distribution Of Tefillin Finds Among The Judean Desert Caves

in tefillin was quite uniform. Beyond this, however, there seems to have been a substantial range of options as to the exact scope of the scriptural selections preceding the four constitutive verses and the order according to which these pericopes were inscribed. While the diversity of texts found in the Judean Desert tefillin exemplars confounds any simple effort to classify all of the slips comprehensively into a limited number of clearly defined taxonomic groups, previous studies have recognized that the corpus can be viewed roughly in terms of two distinct types of tefillin slips: (1) Slips which contain exclusively the four entire scriptural sections (as delineated in the medieval Masoretic division of the Pentateuch) in which the verses understood as mandating the practice of tefillin are found, namely: (1) Exod 13:1–10; (2) Exod 13:11– 16; (3) Deut. 6:4–9; (4) Deut. 11:13–21. This type includes slips that contain either one or more (usually all four) of these sections. (2) Slips which contain verses from the scriptural sections which immediately precede the scriptural sections delineated above, namely: (1) Exod 12:43–51; (2) Deut. 5:1–6:3; (3) Deut. 10:12–11:12. Tefillin slips of this type often (although not always) also include verses from the scriptural sections delineated for the first type.28

167

known examples of this type, verses found on a single are written consecutively, with excerpts from Exodus always preceding excerpts from Deuteronomy.30 In contrast, the second type is characterized by its heterogeneity, remarkable for the diversity it displays in terms of the range of contained texts, the ways these texts are split up and ordered, and the choice of various words and verses omitted—often seemingly deliberately.31 The choice and order of the texts found in the first type appear to coincide precisely with the rabbinic prescription found in Mekhilta de-Rabbi Ishmael for both arm-­ tefillin and head-tefillin: ‫ "והיה‬,"‫ "קדש לי‬:‫בארבעה מקומות מזכיר פרשת תפלין‬ ‫ מצות‬:‫ "והיה אם שמוע"; מכאן אמרו‬,"‫ "שמע‬,"‫כי יביאך‬ ‫ ארבעה פרשיות‬.‫ הן כרך אחד‬,‫תפילין—ארבע פרשיות של יד‬ ‫ "והיה כי‬,"‫ "קדש לי‬:‫ ואלו הן‬.‫ הן ארבע טוטפות‬,‫של ראש‬ ‫ כותבן כסדרן; ואם כותבן‬."‫ "והיה אם שמוע‬,"‫ "שמע‬,"‫יביאך‬ .‫ הרי אלו יגנזו‬,‫שלא כסדרן‬

The first type is extremely homogeneous, always including one or more of the constitutive scriptural pericopes in their entirety, with no omissions, and to the exclusion of any verses from any adjacent scriptural portions.29 In all

In four places (the Torah) records the scriptural portion of tefillin: “Consecrate to Me,” (Exod 13:2), “And when the Lord has brought you” (Exod 13:11), “Hear, (O Israel)” (Deut 6:4), “If, then, you obey” (Deut 11:13). On this basis they have said: The commandment of tefillin (consists of) four scriptural portions of arm(tefillin) which are (written on) one roll (of skin). The four scriptural portions of head-(tefillin) are (made as) four ṭôṭāpōt. And these are the (four scriptural portions): “Consecrate to Me;” “And when the Lord has brought you;” “Hear, (O Israel);” “If, then, you obey. One must write them in their order; and if

viewed simply as harmonizations with Deut 5; see: Esther Eshel, “4QDeutn—A Text that has Undergone Harmonistic Editing,” HUCA 62 (1991): 122–23; George J. Brooke, “Deuteronomy 5–6 in the Phylacteries from Qumran Cave 4,” in Emanuel: Studies in Hebrew Bible, Septuagint and Dead Sea Scrolls in Honor of Emanuel Tov (VTSup 94; ed. S. M. Paul et al.; Leiden: Brill, 2003), 57–70. 4Q141 is singular in that it displays Deut 32:14–20, 32–33 (Milik, in DJD 6:72–74). For an argument against identifying this text as a tefillin slip, see: Nakman, “Contents,” 35–37. For a critique of Nakman’s position as “reductionist and highly problematic,” see: Cohn, Tangled Up, 75–77. 28  Nakman argued that this type be referred to as “expanded,” while the first type he claimed should best be called “abridged” (Nakman, “Contents”). This terminology is imprecise, though, as not all of the Type 2 tefillin slips include, and therefore cannot be said to “expand” upon, verses from the scriptural sections found in Type 1 tefillin slips (e.g. 4Q137 and 4Q138). 29  While according to Yohanan Aharoni’s reconstruction of the missing portion from 34Se 1A, this slip begins with Exod 13:2 (Aharoni, “Expedition B,” 22), it seems likely that the original

slip did in fact contain Exod 13:1 (Adler and Yardeni, “Remains”; contra Cohn, Tangled Up, 125). 30  Despite the unusual arrangement of 8Q3 (Group I), it seems likely that it too was written in the regular order of the verses as found in the Pentateuch; see Yonatan Adler, “The Content and Order of the Scriptural Passages in Tefillin: A Reexamination of the Early Rabbinic Sources in Light of the Evidence from the Judean Desert,” in Halakhah in Light of Epigraphy (JAJSup 3; ed. A. I. Baumgarten et al.; Göttingen: Vandenhoeck and Ruprecht, 2011), 205–29 (222–23). The order that verses were inscribed on a single slip is different from the question of the order in which individual slips were placed inside of a multiple-compartment tefillin case; for the latter, see: Adler, “Content and Order,” 212–20. 31  For a suggestion that some of the omissions found in this type of tefillin slip are the result of intentional scribal abbreviations in order to avoid repetitive content, see: Brooke, “Deuteronomy 5–6,” 64–65. For a theory which views deliberate apotropaic intentions behind many of these omissions, see: Cohn, Tangled Up, 75–77, 94–95.

168

Adler

he has written them not in their order, they must be hidden away.”32 The basic taxonomic distinction between these two types of inscribed tefillin slips was first made by Józef Milik, who presumably had the rabbinic literature in mind when he consistently referred to the first type as “Pharisaic” and the second type as “Essene”—apparently assuming a Pharisaic origin for the rabbinic prescriptions and an Essene identity for the Qumran sectarians.33 Tov further developed Milik’s classification by showing how, at least with regard to the tefillin finds attributed to “Cave 4Q,” those belonging to the second type in terms of content all follow Qumran practice with regard to orthography, morphology and scribal habits, whereas those belonging to the first type do not follow practices of the Qumran scribal school.34 While subsequent studies have pointed out significant problems with Milik’s choice of “Pharisee” and “Essene” labels, his basic taxonomy of tefillin slips into two groups remains essentially valid.35 There seems little doubt that the two types of tefillin slips that we have just defined reflect two distinct practical approaches toward the observance of the tefillin ritual. Type 1 slips represent a practice of incorporating a very carefully circumscribed set of verses, all of these verses and only these verses, and always in their proper order. The only substantial variety found among slips of this type relates to the number of pericopes found on a single slip, although this is almost certainly related to the type of case in which the slip was intended to be housed, with singlepericope slips meant for multi-compartment cases and multi-pericope slips intended for single-compartment cases.36 Type 1 slips exactly prefigure the kind of tefillin called for by the tannaitic rabbis, who explicitly rejected 32   Mekhilta de-Rabbi Ishmael, Tractate Pis�a 18 (ed. Lauterbach2, 114). Cf. b. Menaḥ. 34b. 33  Milik, in DJD 6:36–37, 47–48, 51, 53, 56, 58. It remains unclear how seriously Milik actually took his “Pharisee” and “Essene” labels, as at one point he claimed to use these simply “for convenience” (Milik, in DJD 6:47). 34  Tov, “Tefillin,” esp. n. 22. For a critique of Tov’s exclusive focus on the finds from “Cave 4Q,” see: Adler, “Identifying,” 91, n. 44. 35  For critique of Milik’s classification of tefillin along sectarian lines, see: Nakman, “Contents;” Adler, “Identifying;” Cohn, Tangled Up, 73–75. 36  Thus, e.g., the three Type 1 slips 4Q131–133, each of which include only a single scriptural pericope, were found in situ in a multi-compartment tefillin case (Milik, in DJD 6:55–57). Milik (DJD 3:80–85) was likely correct in assuming that Mur4 frag. 1, a Type 1 slip featuring three pericopes, was meant to be housed in a single-compartment tefillin case together with Mur4 frag. 2, a Type 1 slip which includes the fourth pericope missing from the first slip.

the inclusion of any scriptural verses falling outside of the precise boundaries found in this type, and invalidated any slip whose texts are written out of their scriptural order.37 The second type reflects a far more liberal approach, calling for the inclusion of verses culled from a much larger scope of texts. So diverse is the choice of texts found on these slips that no two slips from this group can be shown definitively to include exactly the same scriptural sections.38 Coupled with the haphazard nature of the textual content, it is impossible to discern on Type 2 tefillin slips any consistent concern that scriptural sections should be inscribed in their Pentateuchal order. It is hardly intuitive to imagine that the differences between these two types of tefillin slips would have been unimportant to contemporary tefillin practitioners. The texts inscribed on the slips were presumably the most important component of the tefillin ritual, and accordingly the exact choice of these texts (and likely their order as well) would have been of critical significance. Whether the two types of tefillin slips we have classified may reflect competing halakhic approaches of different contemporary groups or perhaps a diachronic evolution of the practice unconnected to religio-ideological differences is a question to be explored below; first we must attend to the distribution of these two types of tefillin among the Judean Desert caves. Type 1 tefillin slips were found at the following locations: three slips were found in situ in a single tefillin case discovered in Qumran Cave 4Qa; one slip belongs to the group of purchased finds ascribed to “Cave 4Q;” one slip was found in Cave 8Q; two slips (which may have belonged to a single tefillin set) were found in Cave 34 at Na�al Ṣeʾelim; two slips (which likely belonged to a single tefillin set) were among the purchased texts attributed to the Wadi Murabbaʿat Caves; one slip was discovered by Bedouin who alleged to have found it at Na�al Ṣeʾelim, but there is reason to believe that it was actually discovered in Na�al Ḥever.39 37   Sifre Deut (35 [ed. Finkelstein 63]) specifically singled out the Decalogue as a pericope not to be included in tefillin; see: Adler, “Content and Order,” 224–28. For prescriptions regarding the order of the pericopi, see above note 32. 38  Nakman conjectured that all of the tefillin slips classified here as Type 2 were originally housed in tefillin cases together with other slips such that each “set” always contained a precisely-­ defined range of text (Nakman, “Contents,” 29–35). For a serious critique of this hypothesis, see: Cohn, Tangled Up, 72–73, n. 51. 39   Cave 4Qa: Milik, in DJD 6:55–57, Pl. XII–XIV (4Q131–133). Attributed to Cave 4Q: Milik, in DJD 6:53–55, Pl. X–XI (4Q130); Cave 8Q: Baillet, in DJD 3:149–52, Pls. XXXII:1–11 (8Q3 [Group I]). Naḥal Ṣeʾelim Cave 34: Aharoni, “Expedition B”: 22–23, Pl. 11:B–C; Adler and Yardeni, “Remains” (34Se 1A–B). Attributed to Wadi

Distribution Of Tefillin Finds Among The Judean Desert Caves

Type 2 tefillin slips were discovered at the following sites: one slip was found in Qumran Cave 1Q; thirteen slips are among finds attributed to “Cave 4Q;” three slips were found in Cave 8Q; three slips were found inside a single purchased case which was alleged to have come from an unspecified cave near Qumran.40 5

Analysis of the Data

How Did the Finds Come to be Deposited in the Caves? While the deposits discovered in the so-called “Bar Kokhba caves” are easily explained as the physical remains of human occupation left over by refugees seeking shelter from the ravages of war, little consensus has been achieved in explaining how scrolls came to be deposited in the caves near Qumran.41 Some have suggested that the Qumran caves, or at least some of them, served as some sort of a “genizah,” a depository for sacred texts which had become worn-out or otherwise unusable.42 Alternatively, some of 5.1

Murabbaʿat: Milik, in DJD 3:80–85, Pl. XXII–XXIV (Mur 4); Attributed to Naḥal Ṣeʾelim /Naḥal Ḥever: Morgenstern and Segal, “6. XḤev/Se: Phylactery,” 183–91, Pl. XXX (XHev/Se 5). 40   Qumran Cave 1Q: Barthélemy, in DJD 1:72–76, Pl. XIV (1Q13). Attributed to Cave 4Q: Milik, in DJD 6:48–53, 58–76, Pl. VII–IX, XV–XXIII (4Q128–129, 4Q134–144). Another four slips attributed to “Cave 4Q” (4Q145–148) are too fragmentary to assign to either type. Cave 8Q: Baillet, in DJD 3:152–57, Pls. XXXII:12–29, XXXIII (8Q3 [Group II–IV]). Unspecified Qumran Cave: Yadin, Tefillin (XQ1–3). 41  The principle phase of occupation at both the Wadi Murabbaʿat and Na�al Ṣeʾelim caves is conventionally identified as the period of the Bar Kokhba revolt (132–135 CE). For a recent discussion of First Revolt finds deriving from the Wadi Murabbaʿat caves, see: Michael O. Wise, “Murabba‛āt and the First Jewish Revolt,” in Pesher Naḥum: Texts and Studies in Jewish History and Literature from Antiquity through the Middle Ages Presented to Norman (Naḥum) Golb (SAOC 66; ed. J. L. Kraemer and M. G. Wechsler; Chicago: Oriental Institute, University of Chicago, 2012), 293–315. For a discussion on the dating of the finds from Naḥal Ṣeʾelim Cave 34, see: Adler and Yardeni, “Remains”. 42  Eliezer L. Sukenik appears to have been the first to publicize this suggestion shortly after the initial discovery of the scrolls in Cave 1Q: Eliezer L. Sukenik, Megilot genuzot mitokh genizah keduma shenimtse’ah bemidbar Yehudah [Hidden Scrolls from the Genizah Found in the Judaean Desert] (Jerusalem: Bialik Institute, 1948–49), 10 [Hebrew]. For a recent and thorough review of this theory, including a discussion of what might be meant by the term “genizah,” see: Joan E. Taylor, “Buried Manuscripts and Empty Tombs: The Qumran Genizah Theory Revisited,” in ‘Go Out and Study the Land’ ( Judges 18:2): Archaeological, Historical and Textual Studies in Honor of Hanan Eshel (JSJSup 148, ed.

169

the caves (especially Cave 4Qa) have been interpreted as serving the function of an active library for the residents of nearby Khirbet Qumran, or perhaps of library “stacks” for the remote storage of books not in immediate use by residents of the settlement.43 Another suggestion is that at least some of the caves may have served as emergency hiding places for the scrolls, deposited at a time of crises to protect them from harm.44 Finally, some of the caves (especially Caves 5Q, 7Q–9Q) have been interpreted as dwelling spaces meant for temporary, semi-temporary or even long-term human habitation, with the scrolls found in these caves interpreted accordingly as the personal

A. M. Maeir, J. Magness, and L. H. Schiffman; Leiden: Brill, 2012), 269–315. 43  It should be noted that not all scholars who speak of the corpus of Qumran texts as a “library” have made the additional claim that any one of the caves themselves actually served as the permanent location where this library was regularly housed and accessed. Devorah Dimant suggested that “Cave 4Q” (it is unclear if she had in mind Cave 4Qa, Cave 4Qb, or perhaps both) served as a library of the community living at Qumran, while the minor caves should be viewed as: “adjacent smaller, ‘micro-libraries,’ perhaps for private use” (Devorah Dimant, “The Qumran Manuscripts: Contents and Significance,” in Time to Prepare the Way in the Wilderness: Papers on the Qumran Scrolls by Fellows of the Institute for Advanced Studies of the Hebrew University, Jerusalem, 1989–1990 [STDJ 16; ed. D. Dimant and L. H. Schiffman; Leiden: Brill, 1995], 36. Lawrence Schiffman has argued that only “Cave 4Q” (he seems to mean Cave 4Qa, although perhaps Cave 4Qb is intended as well) served as an active library, while the other caves served as hiding places for scrolls endangered by the invading Romans (Lawrence H. Schiffman, Reclaiming the Dead Sea Scrolls: The History of Judaism, the Background of Christianity, and the Lost Library of Qumran [Philadelphia, PA: Jewish Publication Society, 1994], 56–57). For a recent suggestion that “Cave 4Q” (both Cave 4Qa and 4Qb seem to be intended) served as “library stacks,” see: Daniel Stökl Ben Ezra, “Old Caves and Young Caves: A Statistical Reevaluation of a Qumran Consensus,” DSD 14 (2007): 329–31. 44  Roland de Vaux made this suggestion quite early on for the finds from Cave 1Q: “[. . .] ce sont des archives ou une bibliothèque, cachées dans moment critique” (Roland de Vaux, “La cachette des manuscrits hébreux,” RB 56 [1949]: 236). Later, he suggested that the finds from “Cave 4Q” represent the library normally located within the settlement at Khirbet Qumran but which was hastily hidden immediately before the destruction of the settlement in 68 CE (Roland de Vaux, “Archéologie,” in DJD 3:34). Most recently, see: Sidnie White Crawford, “Qumran: Caves, Scrolls, and Buildings,” in A Teacher for All Generations: Essays in Honor of James C. VanderKam (JSJSup 153; ed. E. F. Mason et al.; Leiden: Brill, 2012), 1:272–73; Mladen Popović, “Qumran as Scroll Storehouse in Times of Crisis? A Comparative Perspective on Judaean Desert Manuscript Collections,” JSJ 43 (2012): 551–94 (593–94).

170 collections of the caves’ occupants.45 A combination of some or all of these suggestions is of course possible as well, with any one cave interpreted as serving different functions at different times, or possibly even multiple functions at one and the same moment.46 How can our survey of the tefillin find-spots contribute towards elucidating this problem? 45  This suggestion was made by de Vaux for the textual finds from the marl caves (excluding “Cave 4Q”) and from Cave 11Q as well (de Vaux, in DJD 3:34; idem, Archaeology and the Dead Sea Scrolls [The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973], 105). Magen Broshi and Hanan Eshel have essentially agreed with de Vaux’s assessment that the marl caves were intended for permanent human habitation, arguing that the majority of the members of the Qumran community lived in these and other caves carved into the marl plateau surrounding the site (Magen Broshi and Hanan Eshel, “How and Where did the Qumranites Live?” in The Provo International Conference on the Dead Sea Scrolls: Technological Innovations, New Texts, and Reformulated Issues (STDJ 30; ed. D. W. Parry and E. C. Ulrich; Leiden: Brill, 1999), 267–73; Magen Broshi and Hanan Eshel, “Residential Caves at Qumran,” DSD 6 [1999]: 328–48, (330–35). Joseph Patrich believes that none of the limestone caves served as dwelling quarters for the permanent members of the Qumran community, although he admits that it is possible that some of the caves may have served as temporary living quarters for postulants requesting to join the community, for those expelled from the community for violating its laws, or for itinerant shepherds (Joseph Patrich, “Khirbet Qumran in Light of New Archaeological Explorations in the Qumran Caves,” in Methods of Investigation of the Dead Sea Scrolls and the Khirbet Qumran Site: Present Realities and Future Prospects (ANYAS 722; ed. M. O. Wise et al.; New York: New York Academy of Sciences, 1994), 73–95, [93]; idem., “Did Extra-Mural Dwelling Quarters Exist at Qumran?” in The Dead Sea Scrolls Fifty Years after their Discovery: Proceedings of the Jerusalem Congress, July 20–25, 1997 (ed. L. H. Schiffman, E. Tov, and J. C. VanderKam; Jerusalem: Israel Exploration Society/Shrine of the Book, 2000), 720–27 (724). Patrich seems to agree with de Vaux and others that the marl caves south of Qumran were used for human habitation (“Did Extra-Mural Dwelling Quarters,” 726). See also: Jodi Magness, The Archaeology of Qumran and the Dead Sea Scrolls (Grand Rapids, MI: Eerdmans, 2002), 69–71. 46  For example, both de Vaux and Milik believed that “Cave 4Q” was originally used for human habitation, but that the scrolls found in the cave were hidden there hastily slightly before the destruction of the settlement in 68 CE (de Vaux, Archaeology and the Dead Sea Scrolls, 56, 105; Józef T. Milik, Ten Years of Discovery in the Wilderness of Judaea [SBT 26; London; SCM Press, 1959], 20). Florentino García Martínez has recently argued that a similar scenario best explains the finds from Cave 11Q (Florentino García Martínez, “Cave 11 in Context,” in The Dead Sea Scrolls: Texts and Context [STDJ 90; ed. C. Hempel; Leiden: Brill, 2010], 199–209).

Adler

Our initial observation relates to the simple presence of tefillin cases and slips in the caves together with the scrolls. Considering the nature of these ritual objects, which were meant to be worn on the body—and not opened up and read—their presence in the caves does not accord readily with the identification of some of these caves as functioning libraries.47 Admittedly, the discovery of tefillin cases and slips in the caves does not in and of itself discredit the theory that some of these caves may have served as libraries, however proponents of such theories are certainly duty-bound to account for the apparently anomalous presence of these objects in the context of a library. Daniel Stökl Ben Ezra has recently addressed this problem by suggesting that while “Cave 4Q” could have served as an active library, some of the manuscripts in the cave— including tefillin and mezuzot—may have been “thrown there in an emergency.”48 The view that some or all of the manuscript corpora in the Qumran caves are themselves emergency caches appears better suited to explain the presence of tefillin among the scroll assemblages. Like the scrolls, it is easy to imagine that tefillin might have been regarded as cherished objects whose owners would have wanted to protect them from harm at a time of crisis. Unlike the scrolls, however, tefillin are small objects which could have been worn on the body or otherwise carried and protected on the person of their owners, and as such seem somewhat less likely to have necessitated any sort of emergency storage in caves. The suggestion that at least some of the Qumran caves served as “genizah” repositories appears somewhat better equipped to explain the presence of tefillin amongst the scrolls. As ritual objects to which a certain amount of sanctity was likely attributed, it is conceivable that some Second Temple period Jews may have already followed a 47  This point has already been raised by Emanuel Tov, who sees the presence of tefillin (and certain other texts) as an indication that not only did the caves not serve as active libraries, but also that the corpora of texts found in these caves are unlikely to have derived from a library located elsewhere, such as within the settlement at Khirbet Qumran (Tov, Scribal Practices, 4–5). For an opposing response to Tov’s argument (but without attending specifically to the question of tefillin finds within the corpora), see: Armin Lange, “The Qumran Dead Sea Scrolls—Library or Manuscript Corpus?” in From 4QMMT to Resurrection: Mélanges qumraniens en hommage à Émile Puech (STDJ 61; ed. F. García Martínez, A. Steudel, and E. J. C. Tigchelaar; Leiden: Brill, 2006), 177–93. 48  Stökl Ben Ezra, “Old Caves and Young Caves,” 329, n. 60. He also suggests that the tefillin and mezuzot may have been placed in “Cave 4Q” “as in a depository,” which I take to mean a “genizah.”

Distribution Of Tefillin Finds Among The Judean Desert Caves

practice similar to that found in later rabbinic writings and subjected tefillin cases and slips to being “hidden away” if they were determined to be unfit for ritual use.49 Like scrolls, tefillin which had become physically worn or damaged might have been regarded as ritually unfit and hence subject to being “hidden away.” It is also possible that tefillin might have been put away if their owners came to believe that they displayed characteristics which did not fully conform to the “correct” halakhic regulations (see further discussion of this possibility below).50 Of course, whether or not this is the best hypothesis for explaining the rest of the textual finds in the caves near Qumran is an entirely different question which cannot be addressed within the limited framework of the present study. If, as a number of scholars have argued, at least some of the caves near Qumran were used for human dwelling (whether temporary or longer-term), we must consider the final possibility that the tefillin found in these caves might belong to the material remains left over from such habitation. Like the pottery found in these caves, the tefillin might simply represent everyday objects left over by the Jewish troglodytes who dwelled in these caves. This certainly seems to be the explanation for the tefillin deposits in the Wadi Murabbaʿat and Na�al Ṣeʾelim caves, and accordingly such a scenario is hypothetically possible for at least some of the tefillin remains found in the caves near Qumran as well. This explanation is most relevant for Caves 5Q and 8Q in the marl plateau, but may also be appropriate for the finds from the limestone Caves 1Q and 11Q as well.51 With regard to “Cave 4Q,” while the hundreds of scrolls found in the cave (many represented by multiple 49   B. Meg. 26b; Tractate Soferim 5:16 (ed. Higger 160). Mladen Popović (“Qumran as Scroll Storehouse,” 584) is doubtlessly correct in cautioning against simple interpretations of the Qumran finds on the basis of significantly later and contextually incongruent conceptualizations of “genizah.” 50  As we have seen above, the tannaitic rabbis mandated that tefillin written in the wrong order must be “hidden away” (Mekhilta de-Rabbi Ishmael, Tractate Pis�a 18 [ed. Lauterbach2, 114]). 51  For de Vaux’s assessment that Cave 1Q was only questionably habitable, see: Roland de Vaux, “La grotte des manuscrits hébreux,” RB 56 (1949): 586–609; idem, “La poterie,” in DJD 1:13; idem, Archaeology and the Dead Sea Scrolls, 56–57. In contrast, Cave 11Q was believed by de Vaux to have been definitely inhabited; see: idem, “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e et 5e campagnes,” RB 63 (1956): 574; idem, Archaeology and the Dead Sea Scrolls, 56–57. Recently, Stökl Ben Ezra has argued that Cave 11Q could not have been inhabited for any prolonged period, while García Martínez has defended de Vaux’s assessment; see: Stökl Ben Ezra, “Old Caves and Young Caves,” 322, n. 31; Daniel Stökl Ben Ezra, “Further Reflections on Caves 1 and 11: A Response to Florentino García Martínez,”

171

copies of the same work) could hardly be viewed as the private reading material of individuals living in the cave, de Vaux and Milik have suggested that this cave originally served as living quarters and only later was filled with scrolls hidden immediately prior to the Roman attack in 68 CE.52 Possibly, the tefillin found in the cave are remains left in the cave by its inhabitants prior to its having been turned into an emergency repository. This scenario is somewhat problematic, however, since the large number of tefillin cases and slips allegedly found in “Cave 4Q” would imply the presence of a rather larger number of cave-dwellers than the space would seem to comfortably allow; this problem is alleviated only slightly by conjecturing that the tefillin found here by Bedouin may have actually been divided between two caverns, Cave 4Qa and Cave 4Qb.53 It should also be stressed that such a suggestion would see the tefillin finds, deriving from an earlier stage of occupation in the cave, contextually divorced from the corpus of hundreds of scrolls conjectured to have been deposited in the cave only later. Finally, we should note the special circumstances of Qumran Cave 8Q, where a disproportionately large number of reinforcing tabs and thongs were found together with fragments of only three literary scrolls, one slip identified as a mezuzah, three tefillin cases and four tefillin slips. Milik reportedly suggested that this cave may have served as the working space for a specialized leather worker who manufactured the tabs, straps and tefillin cases found there, or alternatively that material from such a worker may have been stored in this cave prior to the Roman attack on Qumran, similar to the way that the scrolls of the Qumran library were placed in emergency storage in other caves.54 While both suggestions are certainly possible, and in fact do well in explaining the presence of the tabs, thongs and tefillin cases, neither hypothesis adequately explains the presence of the tefillin slips. One possible solution is to imagine that the tefillin belonged to occupants of the cave (leather workers?), who for whatever reason left them behind in the cave, never returning

in Hempel, The Dead Sea Scrolls: Texts and Context, 220; García Martínez, “Cave 11 in Context,” 206–7. 52  See above, n. 46. 53  For a physical description of Caves 4Qa–b, see: Roland de Vaux, “Archéologie,” in DJD 6:9–13, Pl. I–II. Milik seemed to believe that “Cave 4Q” (Cave 4Qa?) was inhabited by only one individual, calling it: “a cell for a hermit” (Milik, Ten Years, 20). 54  Milik’s suggestion is quoted in: John Carswell, “Fastenings on the Qumrân Manuscripts,” in DJD 6:24, n. 1. See also: Broshi and Eshel, “How and Where,” 271.

172 afterwards to retrieve them.55 In any event, the paucity of other textual finds in the cave seems to preclude viewing the tefillin deposits in Cave 8Q as belonging to an organized “genizah” collection, and certainly not a library. The Distribution of Divergent Types of Tefillin among the Caves However we may wish to explain the presence of tefillin in the caves, it remains to be determined why tefillin of different types were found in the various caves. Is it possible to discern any depositional patterns which might allow us to identify specific types of tefillin cases or slips with particular caves? As we have seen, one or more simple-type tefillin cases were found in every cave with tefillin finds (Caves 1Q, 4Qa, 5Q and 8Q at Qumran, Cave 1 at Wadi Murabbaʿat, and Cave 34 at Na�al Ṣeʾelim), whereas split-type cases were alleged to have been found only in Qumran “Cave 4Q.” Type 1 tefillin slips were found in Qumran Caves 4Qa and 8Q, at Na�al Ṣeʾelim Cave 34, allegedly at Wadi Murabbaʿat and perhaps also at Na�al Ḥever, while Type 2 slips were found in Qumran Caves 1Q and 8Q, are attributed to “Cave 4Q,” and are alleged to have derived from an unspecified cave near Qumran. Our first observation is that in the so-called “Bar Kokhba caves,” only simple-type cases and Type 1 slips were found, whereas both types of cases and slips were found in the caves near Qumran. Among the Qumran caves themselves, “Cave 4Q” displays the widest variety in terms of both tefillin cases and slips, although Cave 8Q also produced both kinds of slips. How are we to account for such a plurality of practices at Qumran, especially where evidence for divergent practices is found within one and the same cave? One possibility is that the material evidence represents a diachronic development of halakhic practices. 5.2

55  Pottery finds in Cave 8Q also attest to some sort of human occupation in the cave; see: de Vaux, “Archéologie,” in DJD 3:28–31. For a recent suggestion that the caves in this area (Caves 7Q–9Q) served as a “processing area” for the preservation of scrolls prior to burial, see: Taylor, “Buried Manuscripts,” 294–95. For an argument that the finds in Cave 8Q do not reflect a workshop of any sort, see: Popović, “Qumran as Scroll Storehouse,” 572. Recent assertions (e.g. Broshi and Eshel, “Residential Caves,” 334; Magness, Archaeology of Qumran, 70; White Crawford, “Qumran: Caves, Scrolls, and Buildings,” 259) that the presence of a mezuzah among the finds indicates that Cave 8Q was a dwelling place are unfounded as there is no way to know if the slip identified as a mezuzah was affixed to the cave’s “doorpost” (or anywhere else in the cave for that matter), nor if mezuzah practitioners at the time would have used mezuzot strictly for dwelling places and not, e.g., a workshop.

Adler

The admittedly limited evidence is such that while both simple-type cases and Type 1 slips are known from as late as second century CE archaeological contexts and third century rabbinic sources, use of split-type cases and Type 2 slips are known only from pre-70 CE archaeological contexts.56 If we conjecture a linear process of halakhic evolution, the spilt-type cases and Type 2 slips at Qumran may reflect earlier practices, which developed into later practices involving simple-type cases and Type 1 slips.57 The “genizah” depositional model, which allows for an accumulation of materials extending over a long period of time, is particularly well suited to explain how both older and newer types of tefillin might be found in one and the same cave, although the other depositional hypotheses remain possible if we assume that the older types remained in use in tandem with the newer types at the point when material was deposited in the caves. Another possibility is that the multiplicity of tefillin types represents differing practices contemporaneous to one another. It is possible that different religio-ideological groups followed different tefillin practices, and that each tefillin type reflects a distinct group. It is even conceivable that tefillin may have been hidden away in caves near Qumran precisely because these exemplars reflected practices viewed as heterodox, and hence unacceptable for ritual use, by those who deposited them in the caves. We may imagine that neophytes joining the Qumran community would have brought with them the tefillin that they were accustomed to wearing prior to joining the community. After arrival at Qumran, and upon learning that such tefillin were invalid in the eyes of the community, these tefillin might have been permanently hidden away as “genizah.” Such a proposal would help explain the variety of tefillin types among the various Qumran caves and especially within “Cave 4Q” and Cave 8Q. It must be stressed that although the later rabbis seem to have adopted the simple-type tefillin case form as well as the Type 1 tefillin slip, this development in and of itself is

56  See above, n. 41. 57  Although Milik (DJD 6:47) claimed that the “Essene”-type tefillin slips at Qumran lasted from the middle of the second century BCE until the Great Revolt, while the “Pharisee”-type tefillin appeared only from the beginning of the first century CE, it is unclear on what grounds this assessment was made. For previous attempts to explain these two types of tefillin slips on the basis of chronological development rather than sectarian differences, see: Lawrence H. Schiffman, Review of Qumrân Grotte 4.II, by Roland de Vaux and Józef T. Milik, JAOS 100 (1980): 171; Nakman (“Contents,” 37–40).

Distribution Of Tefillin Finds Among The Judean Desert Caves

insufficient to allow us to assign either of these types to any particular Second Temple group.58 Finally, it is possible that the various practices in evidence at Qumran were all viewed as equally valid by members of the Qumran community, with all of these practices followed by different members of the same group at one and the same time. We are reminded that divergent types of tefillin cases and tefillin texts are not the only examples of diversity at Qumran; the texts found in the Qumran caves are remarkably varied in terms of textual character and scribal practices, including differing scribal conventions relating to orthography and morphology, choice of Aramaic script vs. Paleo-Hebrew script, varying methods of writing divine names, and more.59 While some of these practices might have been more closely associated with one group than another, the fact that all of these divergent practices were found side by side in the caves near Qumran may indicate that members of the Qumran community were rather open to such

58   Contra Milik (see above, n. 33), who assigned Type 1 tefillin slips to the Pharisees and Type 2 slips to the Essenes. Tov (“Tefillin,” 54*) assigned Type 1 tefillin slips found in “Cave 4Q” to “Pharisaic circles” and Type 2 slips in this cave to “the Qumranites;” with regard to tefillin from the other caves, Tov wrote: “[. . .] the background of the tefillin from caves 1 and 8 as well as from an undetermined cave is not clear” (“Tefillin,” 54*). 59  For a recent overview of the variety in textual character among the so-called Biblical texts at Qumran, see: Emanuel Tov, “The Biblical Texts from the Judean Desert: An Overview and Analysis of the Published Texts,” in The Bible as Book: The Hebrew Bible and the Judaean Desert Discoveries (ed. E. D. Herbert and E. Tov; New Castle, DE: Oak Knoll Press, 2002), 139–66. For an overview and in-depth analysis of divergent scribal practices among the Qumran manuscripts, see: Tov, Scribal Practices.

173

variety.60 Charlotte Hempel has recently argued that the liberal attitude toward scriptural texts at Qumran is paralleled by a similarly relaxed approach evidenced in the various manuscripts of the Community Rule.61 That such a surprisingly pluralistic approach toward cherished texts might have applied also to ritual objects such as tefillin is an intriguing possibility to consider.62

60  Tov’s comment on the remarkable variety in textual character at Qumran is most germane to the issue at hand: “Whether we assume that all the aforementioned texts were written at Qumran or that only some were written there while others were brought from elsewhere, the coexistence of the different categories of texts in the Qumran caves is noteworthy. The fact that different texts were found in the same caves probably reflects textual plurality for the period between the third century BCE and the first century CE [. . .] Since there is no evidence concerning the circumstances of the depositing of the scrolls in the caves or concerning their possibly different status within the Qumran sect, no solid conclusions can be drawn about the approach of the Qumranites towards the biblical text, but it is safe to say that they paid no special attention to textual differences such as those described here.” (Tov, “Biblical Texts,” 156). 61  Charlotte Hempel, The Qumran Rule Texts in Context: Collected Studies (TSAJ 154; Tübingen: Mohr Siebeck, 2013), 279–82, 336–37. 62   Contra Norman Golb, who considered such a possibility out of the question: “It defies logic, however, to believe that a small and radical sect, whether of Essenes or others, who were according to the standard theory highly restrictive and formal in their religious legislation and practice, would have allowed members to be so inconsistent with one another in carrying out a religious law that has been considered sacrosanct among practicing Jews for well over two millennia” (Norman Golb, Who Wrote the Dead Sea Scrolls? The Search for the Secret of Qumran [New York: Scribner, 1995], 103).

Part 4 Chronology, Functions, Connections

∵

CHAPTER 11

When and Why Were Caves Near Qumran and in the Judaean Desert Used? Mladen Popović This article deals with the history of the use of the caves— the chronology of the caves—and takes both material and textual remains into account. A comparative and regional approach that focuses on the Judaean Desert and the Dead Sea region is the starting point of my investigation; the caves near Qumran are an integral part of the Judaean Desert and the Dead Sea region. With regard to the reasons for the caves’ use and its history, this approach highlights similarities as well as differences between caves near Qumran and caves at other sites in the Judaean Desert in which manuscripts were found.1 Although we should approach the caves also by themselves, this is not where our enquiries should end. The caves need to be understood in context, which is determined by the connections and networks through which people ended up in the caves. This may redirect our attention to a new approach from a material perspective to the settlement of Qumran, the caves in its vicinity, and the Dead Sea Scrolls. 1

Dates When Judaean Desert Caves Were Used

Human uses of the caves in the Judaean Desert have left traces that enable us to date when human activities in the caves took place. Given the available evidence, there are generally two sets of data that inform us regarding the chronology of the use of the caves. First, there are material remains, such as small finds and pottery. Second, there are manuscripts, which, of course, count as material remains as well.

small finds less so.2 There is also evidence that the same cave was used in several different periods. For example, the Abi’or cave near Jericho was used in the Roman period, but also in the Mamluk period (fourteenth century) and again in 1948 and 1967. The caves at Murabbaʿat and at Qumran (Caves 9Q and 11Q) indicate human activity in the Iron Age II and in the Hellenistic and Roman periods.3 The chronological range makes a simple point clear. Evidence from various Judaean Desert sites suggests that from a diachronic perspective we have to reckon with multiple moments of human activity in these caves, as well as depositions of manuscript collections at a single site. 1.2 Manuscripts The manuscripts found in Judaean Desert caves are another obvious example of human activity in these caves that provide evidence for when the caves were used. For purposes of dating, the manuscripts or texts can be approached as linguistic objects, i.e. what is in the text, and the texts can also be taken as physical objects, i.e. as material remains. 1.2.1 Texts as Linguistic Objects Concerning texts as linguistic objects, documentary texts from the Judaean Desert have internal datings that provide a terminus post quem for when those texts were taken to the caves. Some examples may illustrate the ­reference to specific dates. From Ketef Jericho (Wadi alMafjar) come:

1.1 Material Remains The material remains indicate the general period in which caves in the Judaean Desert were used. This ranges from the Chalcolithic period, to the Iron Age, the Persian, Hellenistic, Roman, Byzantine, and Islamic periods. Pottery and coins provide the best general indications,

– Jer7, sale of date crop, if the reconstructed first line as a reference to the third year of emperor Domitian is correct then this papyrus dates to 84 CE – Jer13, 116 CE, if remains of the date formula are correctly reconstructed as referring to the eighteenth year of emperor Trajan – Jer16, 1 May 128 CE, also mentioning emperor Hadrian

1  For more discussion, details and references, see also Mladen Popović, “Qumran as Scroll Storehouse in Times of Crisis? A Comparative Perspective on Judaean Desert Manuscript Collections,” JSJ 43 (2012): 551–94; idem, “The Manuscript Collections: An Overview,” in T&T Clark Companion to the Dead Sea Scrolls (ed. G. J. Brooke and C. Hempel; London: T&T Clark, forthcoming).

2  See also the paper by Dennis Mizzi in this volume. 3  Popović, “Qumran as Scroll Storehouse,” 558–59; cf. Roland de Vaux, “Archéologie,” in Les petites grottes’ de Qumrân: Exploration de la ­falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 7, 8, 10, 12, 16, 24–25 for some Iron Age II pottery from four more non-manuscript caves.

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_013

178 From Murabbaʿat come: – Mur18, a debt acknowledgement from 55/56 CE, written in Tsoba – Mur24, a deed of lease in Hebrew from 134 CE – Mur30, a deed of sale of property in Hebrew from 134 CE – Mur114, a recognition of debt between two Roman ­soldiers from 141 or 171 CE – Mur117, mentioning emperor Commodus (180–192 CE) – Mur169, a receipt in Arabic from 928/939 From Sdeir comes Sdeir2, a document dated to the 6th of Adar in the third year of the Bar Kokhba Revolt, 135 CE. And of course many examples come from Naḥal Ḥever that are internally dated, and fall right up to and during the Bar Kokhba Revolt. Some texts from the Babatha archive date to the late first century. For example, regarding the Greek Babatha papyri, the oldest dates to 94 CE and the latest to 132 CE. These documentary texts provide clear evidence for a date after which a cave was used. And these internally dated manuscripts also provide a chronological indication for other documentary texts of which the date is no longer extant, because documentary texts were sometimes found together in bundles. For example, the thirty-five legal papyri written in Greek, Aramaic and Nabatean-Aramaic of the Babatha archive were found in one of the waterskins in locus 61 in the Cave of the Letters (Cave 5/6) in Naḥal Ḥever. The material from Naḥal Ḥever, for example, also illustrates how the evidence may provide a more specific date as well as an indication for the duration of the caves’ use. Different from Murabbaʿat and Qumran, for example, there is no evidence that the caves at Naḥal Ḥever were used in a prior period. All material remains date to the early second century CE. Thus, the evidence from Naḥal Ḥever indicates that the caves started to be used during the Bar Kokhba Revolt. The end date of this use can also be inferred from the archaeological and historical evidence. As is well known, at Naḥal Ḥever, two Roman camps were situated right above the two manuscript caves, 5/6 and 8. The skeletal remains inside both caves–nineteen from Cave 5/6 and more than forty from Cave 8, including a number of children–illustrate the horrible end of the refugees there. They probably died of starvation and thirst.4 This must have happened before or around the end of the Bar Kokhba Revolt.

4  Popović, “Qumran as Scroll Storehouse,” 555–56. See also the evidence discussed there for the Abi’or cave at Ketef Jericho.

Popović

It is possible to even be more specific with regard to people using the caves. If we take evidence from the Cave of the Letters (Cave 5/6) into consideration, it seems that at least some refugees could not have arrived there before November 134 CE. Beneath the water-skin with the purse containing Babatha’s archive (locus 61) another leather pouch was found beneath which were the papyri from Eliezer bar Shemuel’s archive that seemed to have fallen from the pouch. The latest manuscripts of Eliezer’s archive date to Kislev 134 (P. Yadin 45 and 46). This means that the Babatha archive could not have been placed there before November 134. Depending on the end date of the Bar Kokhba Revolt, one thus has at least something of a concrete starting point for the possible duration of their stay.5 1.2.2

Texts as Physical Objects: C14 and Palaeography The manuscripts are, obviously, physical objects. With regard to dating, these material remains provide evidence in the form of C14 and in the form of palaeography. Regarding C14, two dating samples for the Dead Sea Scrolls were conducted in the 1990s in Zürich and Tucson.6 However, it seems that not all castor oil contamination can be expected to have been removed for those samples. The radiocarbon dates previously reported cannot be guaranteed to be correct; some are likely younger, by an unknown amount, than the manuscripts’ true age.7 A new attempt to C14 date the Dead Sea Scrolls is necessary in order to validate the earlier samples in the 1990s.8 Palaeography is another feature of manuscripts as physical objects that provides clues as to the date of the manuscripts and therefore a general indication of the 5  Popović, “Qumran as Scroll Storehouse,” 560–61. However, we do not know whether these refugees were among those who died in the cave or whether they left earlier and if so how much earlier. 6  Georges Bonani et al., “Radiocarbon Dating of the Dead Sea Scrolls,” ʿAtiqot 20 (1991): 27–32; Georges Bonani et al., “Radiocarbon Dating of Fourteen Dead Sea Scrolls,” Radiocarbon 34 (1992): 843–49; A. J. Timothy Jull et al., “Radiocarbon Dating of Scrolls and Linen Fragments from the Judean Desert,” Radiocarbon 37 (1995): 11–19. 7  Kaare L. Rasmussen et al., “The Effects of Possible Contamination on the Radiocarbon Dating of the Dead Sea Scrolls I: Castor Oil,” Radiocarbon 43 (2001): 127–32; Kaare L. Rasmussen et al., “Reply to Israel Carmi (2002): ‘Are the 14C Dates of the Dead Sea Scrolls Affected by Castor Oil Contamination?’ ” Radiocarbon 45 (2003): 497–99. 8  Kaare L. Rasmussen et al., “The Effects of Possible Contamination on the Radiocarbon Dating of the Dead Sea Scrolls II: Empirical Methods to Remove Castor Oil and Suggestions for Redating,” Radiocarbon 51 (2009): 1005–22.

WHY WERE CAVES NEAR QUMRAN AND IN THE JUDAEAN DESERT USED ?

date a cave was used. If we take the common range at Qumran, the oldest manuscripts, in archaic script, are dated to the latter half of the third century BCE, while the latest ­manuscripts, in late Herodian script, to the middle of the first century CE. Most manuscript copies are dated to the first century BCE. It is worthwhile to ask again after the foundation of the chronological framework for Dead Sea Scrolls palaeography. The relative chronology (say the Cross model) seems sensible enough.9 However, one may ask what it means to put the upper-limit of archaic script at 250 BCE or late Herodian around the mid-first century CE, let alone the finer distinctions between late Hasmonaean and early Herodian. Why 250 BCE? And why the mid-first century CE? Focusing on the lower limit, the mid-first century CE, this undoubtedly was informed by historical considerations regarding the Jewish-Roman conflict in 66–70. However, one also needs to take into consideration material from other sites that are dated to the second century CE. Because of the relative chronology of the manuscripts, this seems important to take into account when considering the Dead Sea Scrolls manuscripts dated palaeographically to the late Herodian period in the first century CE. For example, from Murabbaʿat comes one documentary text, Mur30 (from 134 CE), which was subjected to C14.10 More significant in terms of palaeographic comparison are the literary texts dated to the early second century CE (Mur1 and Mur88).11 Given the relative chronology of the palaeographic data it seems sensible to bridge the first century BCE material and the second century CE material with palaeographic hands attributed to the first century CE, for Qumran and elsewhere. In passing, I mention that the palaeographic distinction between late Herodian and post-Herodian needs to be re-examined, questioning the presupposed historical divide of the first revolt. Needless to say, there remains a lack of a firm absolute chronological basis for the finer distinctions of palaeographic taxonomy. Of course, palaeography only may provide a very rough indication of a cave’s use, and primarily in terms of the date after which manuscripts ended up at a particular spot. This is readily apparent from the manuscript caves near Qumran: the third and second century BCE palaeographic dates for manuscripts do not imply the deposi9  See also the paper by Emile Puech in this volume. 10  Bonani et al., “Radiocarbon Dating of the Dead Sea Scrolls;” Bonani et al., “Radiocarbon Dating of Fourteen Dead Sea Scrolls.” 11  Józef T. Milik, “Textes hébreux et araméens,” in Les grottes de Murabba‘ât (DJD 2; ed. P. Benoit, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1961), 75.

179

tion dates at the caves for these manuscripts (see also the discussion below regarding the pottery from the marl and limestone manuscript caves near Qumran). As obvious as this may seem, it does raise the question regarding the history of the caves’ use. Were they used at one or two or more occasions, or were they used continuously? Before answering these questions, let us look at other evidence from the Judaean Desert. When it comes to the manuscript evidence for the use of caves, there are two diachronic perspectives to be taken in consideration. First, there was more than one occasion at which texts were brought to a site. This is obvious when it comes to the Arabic manuscripts from Murabbaʿat. These clearly do not belong to the same collection as the earlier manuscripts, dating from a much later time. But when the chronological range is more limited, more than one scenario may explain the evidence. The second diachronic perspective is that older texts were brought along when people went to the caves. While some of the literary texts from Murabbaʿat date to the early second century CE (Mur1 and Mur88), others date earlier, such as Mur2 to the period before the first revolt and Mur6, a nonbiblical text, to the first century BCE (late Hasmonaean or Herodian).12 These texts may have “arrived” in the caves when refugees during the first revolt (Mur2) hid there, they may have come there earlier (Mur6), or refugees brought all of them to the caves during the Bar Kokhba Revolt. All three scenarios are possible, since the caves in Murabbaʿat were used in more than one period. Intriguing evidence is not a manuscript (leather or papyrus) but an ostracon. Mur72 is a legal text in Aramaic that mentions Masada and dates to the late second/early first century BCE. The time span would indicate a different deposition date for this ostracon from that of the first or second revolt documents, but it is not readily evident to what historical context it is related. Would it be possible that an ostracon such as this was kept for several generations and much later brought along? Or is that unlikely, considering the material not being leather or papyrus, which would presumably indicate the value attributed to a text? While various scenarios are possible to explain the Murabbaʿat evidence, it seems that only one scenario is possible for the Naḥal Ḥever evidence. If indeed these caves were not used prior to the Bar Kokhba Revolt (see above), then it is clear that manuscripts of an older date were deposited at the site at that time and not at an ­earlier point in time more contemporary to the manuscripts’ 12  Milik, in DJD 2:78, 86.

180 palaeographic date (5/6ḤevNuma; 5/6ḤevPsalms; 8Ḥev 1/8ḤevXIIgr; 8Ḥev 2). As I have argued before, leather and papyrus manuscripts from the Judaean Desert had a useful life the duration of which was similar to what we know from elsewhere in the ancient Mediterranean world.13 Returning to the Dead Sea Scrolls from the caves near Qumran, from the evidence from other sites in the Judaean Desert two scenarios may be inferred: (1) manuscript deposits occurred more than once over the course of time at the same site; (2) older manuscripts were brought along and deposited at a moment in time later than the manuscripts’ palaeographic age. 1.3 Material Remains and Manuscripts Material evidence such as pottery and small finds on the one hand and manuscripts and ostraca on the other hand are not easily aligned in a chronological and historical sense. Regarding the manuscripts from the caves near Qumran, our thinking on this issue is to some extent confounded by the issue of the single or multiple character of the collection. Are the manuscripts the remains of multiple collections from a single group that were deposited at different times or those of multiple communities at different geographic locations that were deposited at Qumran at more or less the same time? Was the formation of these collections suddenly halted by external factors such as the first-century CE conflict with Rome or were the scrolls brought from various collections at various moments to Qumran?14 Considering texts as evidence, linguistic and physical, for the history of the use of Judaean Desert caves the simple and obvious point is that individual manuscripts are older than the use of particular caves, i.e. at Qumran thirdcentury manuscripts were not deposited there in the third century, etc., and comparative evidence from Murabbaʿat and Naḥal Ḥever dated to the Bar Kokhba Revolt indicates that older manuscripts were brought along by the refugees when they hid in those caves. Regarding Naḥal Ḥever, the belonging of literary texts to earlier deposits is unlikely in light of the caves’ history of use. Different from perhaps ostraca (see above the discussion regarding Mur72), literary manuscripts that are palaeographically dated earlier need not be regarded as chronologically or historically older, in the sense of having been deposited in the caves at an earlier date. In the case of Naḥal Ḥever it is evident that we are also dealing with synchronically multiple caches of manu13  Popović, “Qumran as Scroll Storehouse,” 562–64. 14  Popović, “Qumran as Scroll Storehouse,” 578–83, 592–93.

Popović

script collections in a single cave. The Cave of the Letters (Cave 5/6) included four private archives (Babatha, Eliezer bar Shemuel, Salome Komaise and the Bar Kokhba letters). There is no archaeologically established relationship between the literary texts from this cave (5/6ḤevNuma; 5/6ḤevPsalms) and these documentary caches. The literary texts either belonged together with the documentary texts of one of these archives and their presumed owner(s) or had arrived with someone else at the same time, as the Naḥal Ḥever caves were not in use prior to the Bar Kokhba Revolt.15 How should the material evidence, especially the pottery, from caves near Qumran be understood with regard to the history of the caves’ use? The pottery includes types from the first century BCE and the first century CE; this applies particularly to marl caves, less so to caves in the limestone.16 Does this imply a continuous use, extending over decades, or should we think of separate occasions sometime during the first century BCE and the first century CE? What evidence is there for long-term, continuous use, and how exactly should that be understood? How is such evidence different from evidence for multiple moments of short-term use over a longer period? These issues cannot be approached from a chronological perspective only. We also need to reflect on the purposes for which Judaean Desert caves were used. 2

Character of Judaean Desert Caves and Duration of Their Use

That Judaean Desert caves were used at multiple and very diverse moments in history, say Chalcolithic, Iron Age, and Roman periods, is an obvious and not very exciting observation. These are historically unrelated events of human activity; the temporal distance is too long to presuppose a direct connection. What is more challenging is to reflect on the evidence for human activity over a short time-span, say the first century BCE and the first century CE and to ask whether such moments of activity are historically related events and how we can determine that. This is directly relevant for the evidence in the caves near Qumran, since the chronological range of human activity extends from the first century BCE until the first century CE. The purpose of the caves’ use defines the character of their use and also its duration. Regarding caves in the Judaean Desert with manuscript caches, comparative 15  Popović, “Qumran as Scroll Storehouse,” 565–70. 16  See also Mizzi’s paper in this volume, especially footnote 69.

WHY WERE CAVES NEAR QUMRAN AND IN THE JUDAEAN DESERT USED ?

evidence indicates that the purpose for which caves were used was determined by violence and conflict, forcing people to seek refuge in the caves. The human remains found at various sites serve as dramatic testimony for this, and they indicate a short-term but intensive usage.17 Understanding the purpose and character of ancient human activity around and in these caves, it is important to take into consideration both caves where manuscripts were found and caves where that has not been the case. Many more caves were discovered that illustrate human presence and activity but no textual remains. For caves in the Qumran area it is debated whether the archaeological evidence indicates habitation in terms of temporary refuge or long-term dwelling.18 Besides purposes of hiding in times of conflict and violence, scholars have mentioned other purposes. For example, seasonal visits have been suggested for caves near Qumran, sometimes in relation to presumed liturgical highlights during the year such as a covenantal renewal ceremony. Another use has been ascribed to shepherds who herded their animals in the area. These, however, seem rather short-term uses of the caves. The purpose of hiding may represent the longest and most intense use, which may have lasted perhaps weeks or months if the evidence from Na�al Ḥever discussed above is something to go on. Limiting myself to the Qumran area,19 almost thirty additional caves in the Qumran vicinity indicate human presence contemporary with the settlement and the ­manuscript caves.20 This spatial and temporal context 17  Popović, “Qumran as Scroll Storehouse,” 554–57. 18  Joseph Patrich, “Khirbet Qumran in Light of New Archaeological Explorations in the Qumran Caves,” in Methods of Investigation of the Dead Sea Scrolls and the Khirbet Qumran Site: Present Realities and Future Prospects (ANYAS 722; ed. M. O. Wise et al.; New York: New York Academy of Sciences, 1994), 73–95; idem, “Did Extra-Mural Dwelling Quarters Exist at Qumran?” in The Dead Sea Scrolls Fifty Years After their Discovery: Proceedings of the Jerusalem Congress, July 20–25, 1997 (ed. L. H. Schiffman, E. Tov, and J. C. VanderKam; Jerusalem: Israel Exploration Society/Shrine of the Book, 2000), 720–27; Magen Broshi and Hanan Eshel, “Residential Caves at Qumran,” DSD 6 (1999): 328–48. Regarding Cave 11Q, cf. Florentino García Martínez, “Cave 11 in Context,” in The Dead Sea Scrolls: Texts and Context (STDJ 90; ed. C. Hempel; Leiden: Brill, 2010), 199–209; Daniel Stökl Ben Ezra, “Further Reflections on Caves 1 and 11: A Response to Florentino García Martínez,” in Hempel, The Dead Sea Scrolls: Texts and Context, 211–23. 19  See also Dennis Mizzi’s paper in this volume for conceptual reflections. 20  Cf. de Vaux, in DJD 3; Patrich, “Khirbet Qumran;” Broshi and Eshel, “Residential Caves.”

181

raises the question whether and if so how the different activities might have been interrelated: in Cave 29 the remains of storage or so-called scroll jars were found but no manuscripts, whereas in Cave 5Q manuscripts were found but no pottery. Different contexts should caution against easy alignments of various constellations of material and manuscript remains in Judaean Desert caves. Moreover, considering the manuscript caves, not all of the caves are the same: Caves 1Q, 4Q and 11Q are evidently different from Caves 7Q–10Q. Scholars often suggest that Caves 7Q–10Q were used as living quarters by single inhabitants, based on material and manuscript evidence or a combination thereof. For example, the date pits and other fruit in Caves 8Q–10Q would indicate them having been used as dwelling places, while the reed mat on the floor in Cave 10 reinforces that assessment for that cave. For Cave 8Q, scholars point to the presence of a mezuzah as evidence for its habitation. But material and manuscript evidence from, for example, Murabbaʿat indicates those caves to have been used by more than one person and for temporary refuge, not for long-time dwelling. The presence of a possible mezuzah (Mur5) matches the evidence from Qumran Cave 8Q. It is also important to note what was not found in the caves near Qumran, both those that yielded manuscripts as well as those that did not. No coins were found.21 This apparent absence of coins from the caves in the vicinity of Qumran is significant in light of some of the other caves where manuscripts turned up and where also coins as well as other objects were found (e.g., Murabbaʿat and Naḥal Ḥever; although Masada is not a cave or a complex of caves it seems somewhat comparable in the sense of having been a place of refuge for families affected by revolt or some of whose members were involved in revolt). At the caves near Qumran also no human remains were found. No skeletal remains were found in the caves at Murabbaʿat either; however, the pottery, domestic items such as combs, the wooden and stone spindle whorls and the remains of several sandals, including two for children, point to the presence of families there–at least at the time of the Bar Kokhba Revolt, but possibly also during the first revolt.22 While Qumran and Murabbaʿat are similar in the 21  De Vaux, in DJD 3:17, 35; Patrich, “Khirbet Qumran,” 92, reports one Roman coin from the first or second century in FQ37, but it comes from the terrace, not from the cave; Broshi and Eshel, “Residential Caves,” report the find of various coins from the areas between the caves but not in the caves. 22  Hanan Eshel, “Documents of the First Jewish Revolt from the Judean Desert,” in The First Jewish Revolt: Archaeology, History and Ideology (ed. A. M. Berlin and J. A. Overman; London:

182 absence of skeletal remains, the small finds assemblage is different.23 Murabbaʿat is more in line with other Judaean Desert caves, indicating the presence of refugees. We can perhaps be more specific, distinguishing between evidence from the first revolt and the Bar Kokhba Revolt. Most hiding complexes discovered thus far in archaeological excavations seem to date to the Bar Kokhba Revolt, but there is also evidence that a few were used earlier, during the first Jewish revolt.24 From a numismatic perspective, it seems that caves in the northern Judaean Desert were not used as places of refuge during or shortly after the first revolt, although they were used as such in earlier and later episodes of upheaval in the region. The 1993 Israel Antiquities Authority’s surveys and excavations of caves in the northern Judaean Desert called “Operation Scroll” turned up a disproportionately low number of coins minted between 6 and 67 CE, as well as a number of revolt coins from only four contexts. This paucity of coins is suggested to indicate that caves were not used much for refuge in this period.25 Does this apply to the caves near Qumran, arguing against a first century Routledge, 2002), 157–63; James S. McLaren, “Going to War against Rome: The Motivation of the Jewish Rebels,” in The Jewish Revolt against Rome: Interdisciplinary Perspectives (JSJSup 153; ed. M. Popović; Leiden: Brill, 2011), 129–53. For the presence of three earlier intertwined family archives, see Michael O. Wise, “Murabbaʿāt and the First Jewish Revolt,” in Pesher Naḥum: Texts and Studies in Jewish History and Literature from Antiquity through the Middle Ages Presented to Norman (Naḥum) Golb (SAOC 66; ed. J. L. Kraemer and M. G. Wechsler; Chicago: Oriental Institute, University of Chicago, 2012), 293–315. 23  See Popović, “Qumran as Scroll Storehouse,” and also Dennis Mizzi’s paper in this volume. 24  Yoram Tsafrir and Boaz Zissu, “A Hiding Complex of the Second Temple Period and the Time of the Bar Kokhba Revolt at ʿAin‘Arrub in the Hebron Hills,” in The Roman and Byzantine Near East (JRASS 49; ed. J. H. Humphrey; Portsmouth, RI: Journal of Roman Archaeology, 2002), 3:7–36; Amos Kloner and Boaz Zissu, “Hiding Complexes in Judaea: An Archaeological and Geographical Update on the Area of the Bar Kokhba Revolt,” in The Bar Kokhba War Reconsidered: New Perspectives on the Second Jewish Revolt against Rome (TSAJ 100; ed. P. Schäfer; Tübingen: Mohr Siebeck, 2003), 181–216. See also Yuval Shahar, “The Underground Hideouts in Galilee and Their Historical Meaning,” in Peter Schäfer, The Bar Kokhba War Reconsidered, 217–40, who distinguishes between two types of hideouts: (1) artificial caverns under the surface level of ancient settlements and (2) natural cave refuges outside settled areas. 25  Donald T. Ariel, “The Coins from the Surveys and Excavations of Caves in the Northern Judean Desert,” ʿAtiqot 41/2 (2002): 281–304 (284, 295–96). See also Mladen Popović, “Roman Book Destruction in Qumran Cave 4 and the Roman Destruction of Khirbet Qumran Revisited,” in Qumran und die Archäologie:

Popović

CE deposit of the manuscripts? Or might the general absence of coins and small finds in the caves near Qumran indicate the purpose for which caves near Qumran were used at the time, namely not for personal refuge but, with regard to the manuscript caves, for hiding manuscripts? There is no need to understand the evidence from the manuscripts caves near Qumran as implying continuous, long-term use over a century. The caves may have been visited and used at various occasions during that period, but it is hard to see how a continuous use over that period is archaeologically indicated. When it comes to chronology and the use of Judaean Desert caves there is evidence that suggests long time networks and connections in the region. The ostracon from Murabbaʿat that mentions Masada (Mur72) is paralleled by a documentary text from the first revolt: Mur19 was written during year six of the first revolt at Masada and then probably taken to one of the Murabbaʿat caves.26 This may suggest some form of tradition or knowledge about these caves being a particularly good hiding place. Where was such a tradition “at home,” at Masada or wider in the region? How did people end up at particular caves? 3

Caves in Context

We must look at the caves and their use in context, because caves are not ordinary dwelling places, such as houses. The extraordinary living conditions are indicated by, for example, the refugee circumstances. Limiting myself to caves with material remains and manuscript finds I will reflect on some possible scenarios for how people ended up, with manuscripts, at particular caves. The Masada-Murabbaʿat connection has already been referred to on the basis of Mur72 and Mur19. This connection may be due to local traditions, although pure chance cannot be ruled out of course. In any case, from Masada it seems people found their way to Murabbaʿat at different moments in the first century BCE and the first century CE. The Babatha and Salome Komaise archives provide another scenario for the second century CE. Both women came from the village of Maoza, to the south-east of the Dead Sea. How did they end up in Naḥal Ḥever to the west of the Dead Sea? The link is probably much more local. Babatha’s second husband came from En Gedi. The connection to Naḥal Ḥever ran via En Gedi. Salome Komaise’s family may have fled together with Babatha’s from Maoza Texte und Kontexte (WUNT 278; ed. J. Frey, C. Claussen, and N. Kessler; Tübingen: Mohr Siebeck, 2011), 239–91 (271–72). 26  Milik, in DJD 2:104–9.

WHY WERE CAVES NEAR QUMRAN AND IN THE JUDAEAN DESERT USED ?

to the En Gedi region. In En Gedi local people (her second husband’s network) may have helped these families to the Naḥal Ḥever caves. Locality and local connections seem to matter when it comes to why people ended up at particular caves. If we turn to the manuscript caves near Qumran it seems that the site of Qumran presents itself as an obvious local connection. But, as in the case of Babatha and Salome Komaise, the Dead Sea Scrolls (all or some?) may have come from elsewhere in Judaea. Scrolls were possibly sent to Qumran for safekeeping.27 Although not all scroll caves seem equally secure, there is no hard evidence that one of these eleven specifically was disturbed in antiquity,28 indicating that the scrolls indeed were kept safe. Why Qumran? It may very well have been a combination of factors, for example, the area as well as the site of Qumran itself. The large cemetery may be indicative of the site’s special character. The Qumran cemetery has been debated from many perspectives, such as demography, gender, ideas about the afterlife, the type of burial, and the period of use. The type of burial itself is not extraordinary. The shaft burial type of tomb has been found at other sites and in other periods: Ein el-Ghuweir, Hiam el-Sagha, Beth Zafafa and some others in Jerusalem and the big Nabatean cemetery in Khirbet Qazone. The type of burial in itself was not 27  Popović, “Qumran as Scroll Storehouse.” 28  Popović, “Roman Book Destruction in Qumran Cave 4.”

183

s­ pecial, but the large number of tombs and the cemetery’s location in the immediate vicinity of the site is significant. The settlement of Qumran may exhibit certain structural similarities with fortified farmsteads elsewhere in Judaea, but those other sites did not have large cemeteries directly next to them. The large size of the cemetery may indicate the attraction and possibly central role of Qumran within local and regional connections, comparable to the enormous cemetery of Khirbet Qazone that served the needs of several villages in the area. The remains of wood and other clues for secondary burial in the Qumran cemetery are possible indications that people brought their dead from elsewhere to Qumran. Something similar, i.e. a significant position of Qumran in a larger network of connections, may apply to the Dead Sea Scrolls that were found in the surrounding caves. This suggests that the site of Qumran in combination with the nearby caves and the Dead Sea Scrolls represents a fascinating mixture of rural and regional material culture on the one hand and urban, high literary culture on the other hand.29

29  See Mladen Popović, “The Ancient ‘Library’ of Qumran between Urban and Rural Culture,” in The Dead Scrolls at Qumran and the Concept of a Library (STDJ 116; eds. S. White Crawford and C. Wassén; Leiden: Brill, 2015), 155–167.

CHAPTER 12

The Connection between the Site of Qumran and the Scroll Caves in Light of the Ceramic Evidence Jodi Magness In November–December 1951, G. Lankester Harding (the chief inspector of antiquities in Jordan) and Roland de Vaux of the École Biblique in Jerusalem conducted their first season of excavations at Qumran. They excavated three rooms (L1, L2, L4) in the southwest corner and two rooms (L5, L6) in the northeast corner of the main building, and made a small sounding at the entrance to the site (L7). Despite the limited scale of their excavations, Harding and de Vaux noticed that “sunk into the floor of one of the rooms was a jar identical with most of those found in the Scrolls cave [Cave 1Q] . . . We thus, even in the small area so far excavated, have a direct connection with the Scrolls . . .”1 Lying on the floor next to the jar was a coin dated ca. 10 CE. Harding and de Vaux noted that the same types of cooking pots and oil lamps found in Cave 1Q were represented at the site. On this basis they were able to date these pottery types to the first century BCE and first century CE.2 The finds of the 1951 season prompted Harding and de Vaux to undertake large-scale excavations at Qumran from 1953–56.3 De Vaux concluded that Qumran was inhabited by the same sectarian group—who he identified as the Essenes— which deposited the scrolls in the nearby caves.4 De Vaux’s 1  Gerald L. Harding, “Khirbet Qumran and Wady Murabaʿat: Fresh Light on the Dead Sea Scrolls and New Manuscript Discoveries in Jordan,” PEQ 84 (1952): 104–9 (105). 2  Harding, “Khirbet Qumran and Wady Murabaʿat,” 105. 3  For preliminary reports see Roland de Vaux, “Fouille au Khirbet Qumrân: Rapport préliminaire,” RB 60 (1953): 83–106; idem, “Fouilles au Khirbet Qumrân: Rapport préliminaire sur la deuxième campagne,” RB 61 (1954): 206–36; idem, “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e, et 5e campagnes,” RB 63 (1956): 533–77. For a synthetic overview of the excavations, see idem, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973). For de Vaux’s field notes and excavation photographs, see Jean-Baptiste Humbert and Alain Chambon, eds., Fouilles de Khirbet Qumrân et de Aïn Feshkha. Vol. 1: Album de photographies; Répertoire du fonds photographique; Synthèse des notes de chantier du Père Roland de Vaux (NTOA.SA 1; Fribourg/Göttingen: Éditiones Universitaires/Vandenhoeck & Ruprecht, 1994). 4  De Vaux, Archaeology and the Dead Sea Scrolls, 128–38.

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_014

conclusion has been accepted by most scholars, based on the inclusion of sectarian works among the scrolls and the archaeological connection between the caves and the site indicated by the same types of pottery. In this paper, I review the ceramic evidence and show that not only are the same types of pottery found in the scroll caves and the settlement, but these vessels are made of the same clays (or pastes). In other words, the pottery from the scroll caves and the site shares the same morphology and chemical composition. I begin by reviewing the claim that Qumran was a pottery manufacturing center, not a sectarian settlement. 1

Was Qumran a Pottery Manufacturing Center?

Based on their 1993–2004 excavations at Qumran, Yitzhak Magen and Yuval Peleg identify Qumran as a pottery production center rather than a sectarian settlement.5 This claim is based on their “discovery” that one pool (L71, the largest stepped pool at Qumran), which de Vaux had partly excavated and which they finished clearing was filled with a thick layer of sediment consisting of clay-like silt (under a layer of refuse).6 Magen and Peleg provide no other new information to support their claim and do not even reassess previously published finds from Qumran, including (and especially) the potters’ workshop and kilns and the ceramic assemblage. They also fail to cite any scientific support for their claim, such as analysis of the “clay deposits” from L71 and the pottery from Qumran. If 5  Yitzhak Magen and Yuval Peleg, The Qumran Excavations 1993–2004: Preliminary Report (JSP 6; Jerusalem: Staff Officer of Archaeology, Civil Administration of Judea and Samaria/ Israel Antiquities Authority, 2007); also see Yitzhak Magen and Yuval Peleg, “Back to Qumran: Ten Years of Excavation and Research, 1993–2004” in Qumran, the Site of the Dead Sea Scrolls: Archaeological Interpretations and Debates; Proceedings of a Conference Held at Brown University, November 17–19, 2002 (STDJ 57, ed. K. Galor, J.-B. Humbert, and J. Zangenberg; Leiden: Brill, 2006), 55–113. 6  Magen and Peleg, The Qumran Excavations 1993–2004, 5, 13–14; Magen and Peleg, “Back to Qumran,” 66–68.

THE CONNECTION BETWEEN THE SITE OF QUMRAN & THE SCROLL CAVES

Magen and Peleg are correct, the sediment from L71 and the pottery vessels from Qumran should be made of the same clay. Instead, they describe the sediment in L71 as “high-quality potters’ clay” without providing any basis for this identification. Even if Magen and Peleg are correct in claiming that the sediment is potters’ clay, this would prove only that the sediment that washed in with the flood waters happened to contain clay, not that the pools were designed for use in a pottery production center.7 Of course, no one denies that pottery was manufactured at Qumran—a fact attested by the presence of the potters’ workshop. However, Magen and Peleg claim that this workshop produced pottery not for the needs of a sectarian community with strict regulations governing the observance of purity laws but for export and trade.8 If Magen and Peleg are correct, we would expect to find vessel types characteristic of Qumran—and particularly cylindrical jars—widely distributed around Judea (see below). In fact, Magen and Peleg’s identification of Qumran as a pottery production center flies in the face of all common sense. Why establish a pottery production center in a spot that lacks perennial sources of water and has scarce supplies of fuel for the kilns? Magen and Peleg apparently sense this problem, suggesting that the heat and aridity at Qumran would have made it possible to produce unfired earthenware vessels. However, they do not cite a single example of such an unfired vessel from Qumran, despite the fact that the arid conditions that preserved the scrolls in the caves should also have preserved such vessels, as at Masada where unfired dung vessels were found.9 Not only would it have been expensive to produce pottery at Qumran, but the cost of export would have been prohibitive. The cheapest and easiest means of transport has always been by water. Overland transport is expensive and difficult—especially in the case of ceramic vessels, which in antiquity had to be transported in bundles on pack animals and had a high rate of breakage. Because of this, usually only fine wares or amphoras (the contents of which were valuable) were transported long distances 7  Magen and Peleg, The Qumran Excavations 1993–2004, 13, 38, “posit that the main purpose of the entire complex water supply system, with its channels and large pools, was to provide potters’ clay” (13). 8  Magen and Peleg, The Qumran Excavations 1993–2004, 49–54, 58–61. 9  Magen and Peleg, The Qumran Excavations 1993–2004, 51–52. For the dung vessels from Masada, see Rachel Bar Nathan, Masada VII: The Yigael Yadin Excavations 1963–1965; Final Reports; The Pottery of Masada (Jerusalem: Israel Exploration Society/Hebrew University, 2006), 238–43.

185

overland, and at a high cost.10 The pottery produced at Qumran was undecorated and utilitarian, not fine ware, which means that it would not have been worth transporting in quantities overland. If Magen and Peleg are correct, we would expect to find Qumran products at contemporary sites along the shores of the Dead Sea that were accessible by boat. But these sites (including Ein Feshkha, Ein Gedi, Ein Boqeq, and Ein ez-Zara) have not yielded typical Qumran ceramic types such as cylindrical jars and bowl-shaped lids, and even ovoid jars—which apparently are a regional type—are rare elsewhere (see below). Magen and Peleg question the identification of the large pools at the site as miqva⁠ʾot and the assumption that the sect produced its own pottery because of concerns with ritual purity: “The issue is not whether there were ritual baths in Qumran; rather, it is whether the large reservoirs were also ritual baths.”11 They erroneously state that, “so far no pottery workshop from the Second Temple period has been found and fully excavated, so that we are unable to compare Qumran to any other contemporaneous site.”12 In 1993, David Adan-Bayewitz published a monograph on the potters’ workshop at Kfar Hananya in Israel’s Galilee, which was active from the early Roman to Byzantine periods.13 Potters’ workshops of the Roman period have been discovered elsewhere around the country, including a major workshop at the site of Givʿat Ram (or Binyanei Ha⁡ʾuma) in Jerusalem.14 None of these is mentioned by Magen and Peleg. 10  See, for example, David P. S. Peacock and David F. Williams, Amphorae and the Roman Economy: An Introductory Guide (London: Longman, 1982), 63–64; David Adan-Bayewitz, Common Pottery in Roman Galilee: A Study of Local Trade (BISNELC; Ramat-Gan: Bar Ilan University, 1993), 211–19. 11  Magen and Peleg, The Qumran Excavations 1993–2004, 37. 12  Magen and Peleg, The Qumran Excavations 1993–2004, 49. 13  Adan-Bayewitz, Common Pottery in Roman Galilee. 14  See Benny Arubas and Haim Goldfus, “The Kilnworks of the Tenth Legion Fretensis,” in The Roman and Byzantine Near East: Some Recent Archaeological Research (JRASS 14; ed. J. H. Humphrey; Portsmouth, RI: Journal of Roman Archaeology, 1995): 95–107; Benny Arubas and Haim Goldfus, eds., Excavations on the Site of the Jerusalem International Convention Center (Binyanei Ha`uma): A Settlement of the Late First to Second Temple Period, the Tenth Legion’s Kilnworks, and a Byzantine Monastic Complex;The Pottery and Other Small Finds (JRASS 60; Portsmouth, RI: Journal of Roman Archaeology, 2005). For a Late Roman-Byzantine potters’ workshop at Horvat ‘Uza in Galilee, see Amnon Ben Tor, “Excavations at Horbat ‘Usa,” ʿAtiqot 3 (1966): 1–24 [Hebrew]; Nimrod Getzov et al., Horbat ‘Uza: The 1991 Excavations. Vol. 2: The Late Periods (IAA Reports 42; Jerusalem: Israel Antiquities Authority, 2009).

186 In his study of the Kfar Hananya pottery (which includes scientific and typological analyses, a consideration of the relevant literary sources [primarily rabbinic literature], and a discussion of trade and distribution patterns), Adan-Bayewitz notes that, “The decline in the frequency of Kfar Hananya pottery with distance (the “friction effect” of distance) seems to reflect the high cost of overland transport.”15 For this reason, competing local wares were manufactured in the Golan instead of importing the Kfar Hananya products. The potters’ workshop at Binyanei Ha⁡ʾuma was active during the late Second Temple period (second century BCE to first century CE) and the time of Aelia Capitolina (second century CE), when it was the kiln works of the Tenth Roman Legion. Final reports on the pottery from Binyanei Ha⁡ʾuma were published in 2005, and a number of preliminary reports on this site appeared in print since the time of Michael Avi-Yonah’s 1949 excavations.16 Andrea Berlin, who published the final report on the pottery of the late Second Temple period from Binyanei Ha⁡ʾuma, provides an instructive contrast with the situation at Qumran: “There are a number of natural advantages to the site at the convention center. The most important is accessibility to good potting clay. The clay with which all of the production was made derives from the Motza Formation, which lay only a few kilometers away, along the Joppa road. Also vital are close sources of water and fuel for the firing. There is water at the N base of the ridge, fed by a branch of Na�al Soreq, and it is likely that the slopes were covered with vegetation. A fourth advantage is the site’s position along the road.”17 The features characteristic of Qumran—including the large number (and sizes) of stepped pools, animal bone deposits, and large adjacent cemetery—are unattested at other potters’ workshop sites, including Binyanei Ha⁡ʾuma and Kfar Hananya. The industrial area of the first century CE at Binyanei Ha⁡ʾuma was provided with a deep cistern. At Binyanei Ha⁡ʾuma and Kfar Hananya, the clay was quarried and brought to the workshop from nearby sources, not collected from sediment washed into stepped pools. To conclude, not only is there no positive evidence to support Magen and Peleg’s claim, but it is contradicted by comparisons with contemporary potters’ workshops around Palestine.

15  Adan-Bayewitz, Common Pottery in Roman Galilee, 213. 16  See Arubas and Goldfus, “The Kilnworks of the Tenth Legion Fretensis.” 17  Andrea Berlin, “Pottery and Pottery Production in the Second Temple Period,” in Arubas and Goldfus, Excavations on the Site of the Jerusalem Convention Center, 29–60 (51).

Magness

2

The Pottery Types from the Site of Qumran and the Caves

The connection between the site of Qumran and the scroll caves is indicated by the location of Caves 5Q and 7Q–10Q at the southern end of the marl terrace on which the settlement sits, which means they were accessible only by way of the site.18 In addition, the same pottery types typical of the first century BCE and first century CE are found in the settlement and in the scroll caves.19 These include bowl-shaped lids, ovoid jars, and cylindrical jars, the last represented by dozens of specimens at the site and in the surrounding caves but virtually unattested elsewhere.20 In Cave 1Q, de Vaux found the remains of at least fifty jars, all but two of them cylindrical.21 A large quantity of broken jars and lids was mixed with the collapse of Cave 3Q’s ceiling, including six cylindrical jars catalogued by de Vaux.22 The pottery from Cave 4Q, which was very fragmentary, included a dozen more or less complete jars, most of them ovoid.23 One cylindrical jar is illustrated from Cave 6Q, and de Vaux mentions three ovoid jars from Cave 7Q, two of which are illustrated (one of them inscribed “Roma”).24 The upper part of a cylindrical jar is illustrated from Cave 8Q, and the uncatalogued pottery includes fragments of three more jars, apparently cylindrical.25 Additional jars from Cave 8Q are mentioned in de Vaux’s field notes.26 Cave 10Q contained “sherds of Qumran type,” but no jars are listed.27 Pottery of Qumran type is also reported from Cave 11Q, but without specific references to jars.28 18  See de Vaux, Archaeology and the Dead Sea Scrolls, 96; on p. 100 he notes that the scrolls in Caves 7–10 were deposited before the caves eroded or collapsed. 19  As noted by de Vaux, Archaeology and the Dead Sea Scrolls, 102–3. 20  See Jodi Magness, Debating Qumran: Collected Essays on Its Archaeology (ISACR 4; Leuven: Peeters, 1994), 151–55. 21  Roland de Vaux, “La poterie,” in Qumran Cave 1 (DJD 1; ed. D. Barthélemy and J. T. Milik; Oxford: Clarendon Press, 1955), 8–13 (see 8); Figs. 2: 10–12; 3: 9–11. 22  Roland de Vaux, “Archéologie,” in Les ‘petites grottes’ de Qumrân: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 3–36 (see 7–8); Fig. 2:3. 23   Roland de Vaux, “Archéologie,” in Qumrân Grotte 4.II: I. Archéologie; II. Tefillin, Mezuzot et Targums (4Q128–4Q157) (DJD 6; ed. R. de Vaux and J. T. Milik; Oxford: Clarendon Press, 1977), 15–20 (see 15); Figs. 5: 1–3; 6: 12. 24  De Vaux, in DJD 3:10; Pl. V: 26–2 (Cave 6); 30; Fig. 6:5, 12 (Cave 7). 25  De Vaux, in DJD 3:30–31; Fig. 6:7. 26  Humbert and Chambon, Fouilles de Khirbet Qumrân, 343. 27  De Vaux, in DJD 3:31. 28  Humbert and Chambon, Fouilles de Khirbet Qumrân, 344.

THE CONNECTION BETWEEN THE SITE OF QUMRAN & THE SCROLL CAVES

No ovoid or cylindrical jars have been found in Jerusalem and no examples of the “classic” cylindrical jars that are so common at Qumran are represented at Jericho. Only one jar at Jericho is identified as this type by Rachel Bar Nathan (her Type J-SJ2B), but the lower half of the body is missing and the two large loop handles are unparalleled on “classic” cylindrical jars at Qumran.29 A small variant of cylindrical jars (Bar Nathan’s Type J-SJ2C) is represented at Jericho by only one complete specimen and “several fragments,” all of which were found in Herodian period contexts in the industrial zone.30 The rest of the related jars from Jericho are ovoid jars (her Types J-SJ2A and J-SJ2D, a small variant of the ovoid jars), a local type that appeared during the Hasmonean period.31 Ovoid jars are found at other sites in the Dead Sea region but are rare, with only one specimen each from Ein Feshkha, Ein Boqeq, and Ein ez-Zara.32 Outside Jericho, one cylindrical jar was reportedly found in a second century CE tomb at Abila in Jordan, although nowhere is it illustrated.33 Yitzhak Magen discovered cylindrical jars (which he calls “Qumran Jars”) in the late Second Temple period agricultural settlement at Qalandiya, north of Jerusalem. The only complete specimen represented is a small cylindrical jar that corresponds with Bar Nathan’s type J-SJ2C.34 Only rim fragments are 29   Rachel Bar Nathan, Hasmonean and Herodian Palaces at Jericho: Final Reports of the 1973–1987 Excavation; 3; The Pottery (Jerusalem: Israel Exploration Society/Hebrew University, 2002), 24–25; Pl. 2 no. 8. 30  Bar Nathan, Hasmonean and Herodian Palaces at Jericho, 26, Pl. 2 no. 9. 31  Bar Nathan, Hasmonean and Herodian Palaces at Jericho, 24–26. 32  For Ein Feshkha, see Roland de Vaux, “Fouilles de Feshkha,” RB 66 (1959): Fig. 1:4; on p. 252, he states that the ovoid jar comes from a Period III context at Ein Feshkha but represents the same type found in Period II contexts at Qumran. For Ein Boqeq, see Moshe Fischer, Mordechai Gichon, and Oren Tal, ʿEn Boqeq: Excavations in an Oasis on the Dead Sea. Vol. 2: The Officina, An Early Roman Building on the Dead Sea Shore (Mainz: Philipp von Zabern, 2000), Fig. 2.7: 1. For Ein ez-Zara, see Christa Clamer, Fouilles archéologiques de ʿAïn ez-Zâra/Callirrhoé: Villégiature hérodienne (BAH 147; Beirut: Institut français d’archéologie du Proche-Orient, 1997), Pl. 12: 15. 33  Farah S. Ma⁠ʾayeh, “Recent Archaeological Discoveries in Jordan,” ADAJ 4–5 (1960): 116, “The most interesting finds were that of an inkwell and a cylindrical jar which are closely paralleled by similar objects discovered at Qumran.” 34   Yitzhak Magen, “Qalandiya—A Second Temple-Period Viticulture and Wine-Manufacturing Agricultural Settlement,” in The Land of Benjamin (JSP 3; ed. Yitzhak Magen et al.; Jerusalem: Staff Officer of Archaeology, Civil Administration of Judea and Samaria/Israel Antiquities Authority, 2004), 85, Pl. 3:7.

187

preserved of the other jars.35 However, one vessel illustrated by Magen (Pl. 7:18) is a juglet with a tall, narrow neck, and the loop handles and profile of another vessel (a jar in Pl. 5:8) indicate that it too does not represent an ovoid or cylindrical jar. The remaining four jars illustrated by Magen might be cylindrical or ovoid jars, although it is difficult to be sure on the basis of the rim profiles alone, since one of the characteristics of this type is a broad, flat or concave base. For example, another “cylindrical” jar published by Magen has a button base.36 Bar Nathan published examples of ovoid and cylindrical jars from Masada.37 Seven jars were found, represented mostly by fragments (including one jar represented by a single rim fragment). Of these, one jar is ovoid (M-SJ17A) and two appear to be a small variant of cylindrical jars (M-SJ17E nos. 83–84). In my opinion, the fact that all of these jars come from “Zealot occupation” contexts supports Yigael Yadin’s suggestion that the rebels atop Masada were joined by refugees from Qumran after the site was destroyed by the Romans in 68 CE.38 3

Yellin et al.’s INAA Analysis

Petrographic analysis and Instrumental Neutron Activation Analysis (INAA) have been employed in attempts to clarify the connection between the pottery from Qumran and the scroll caves. Petrography is the study of thin sections of pottery or fired clay which compares visible mineralogical inclusions including temper. INAA is a nuclear method to determine the chemical composition of clay or pottery, with groupings based mainly on the trace elements present.39

35  Magen, “Qalandiya,” Pls. 4:19; 5:8; 7:16, 17, 18; 10:22. 36  Magen, “Qalandiya,” 84; Pl. 3:8. 37  Bar Nathan, The Pottery of Masada, 67–72. 38  See Yigael Yadin, Masada: Herod’s Fortress and the Zealots’ Last Stand (New York: Random House, 1966), 174; Emanuel Tov, “A Qumran Origin for the Masada Non-biblical Texts?” DSD 7 (2000): 57–73. 39  See Jan Gunneweg, “Archaeology and Archaeometry at Qumran,” in Khirbet Qumrân et ‘Aïn Feskha. Vol. 2 : Études d’anthropologie, de physique et de chimie (NTOA.SA 3; ed. J.-B. Humbert and J. Gunneweg; Fribourg/Göttingen: Éditions Universitaires/ Vandenhoeck & Ruprecht, 2003), XVIII; Jan Gunneweg and Marta Balla, “Neutron Activation Analysis, Scroll Jars and Common Ware,” in Humbert and Gunneweg, Khirbet Qumrân et ‘Aïn Feskha, 5; Jacek Michniewicz, Qumran and Jericho Pottery: A Petrographic and Chemical Provenance Study (Seria Geologia 20; Poznan: Wydawnictwo Naukowe UAM, 2009), 21.

188 The first results of INAA of Qumran pottery were published in a 2001 BASOR article by Joseph Yellin, Magen Broshi, and Hanan Eshel.40 They examined 31 samples from Qumran and eight samples from Ein el-Ghuweir.41 Most of the Qumran pottery comes from L48 (the stepped pool with the earthquake crack), L50 (a small stepped pool on the north side of L48), L52 (east of L48),42 L89 (the pantry), and L96 (an open courtyard on the southwest side of the settlement). Yellin et al. also analyzed an inkwell from L30 (the “scriptorium”), a jar inscribed Johanan Hatla from L34 (which was sunk into a basin in the southeast corner of the courtyard of the main building), an inscribed bag-shaped jar from Cave 6Q, and an ovoid jar inscribed “Roma” from Cave 7Q.43 No rationale is provided for the sampling strategy, that is, there is no explanation of why these particular vessels from these loci were selected for analysis. Yellin et al. identified two largely overlapping yet distinct chemical groups with some outliers: a Qumran group and a Jerusalem group.44 The Qumran group has a much larger root-mean-square deviation (standard deviation) than the Jerusalem group, which the authors attribute to the inclusion of cooking wares and the small sizes of some of the samples.45 Most of the Qumran pottery belongs to the group which they assume is local to Qumran (based on the presence of a potters’ workshop), although they did not analyze clays from Qumran and therefore had no chemical evidence for this identifica-

40   Joseph Yellin, Magen Broshi, and Hanan Eshel, “Pottery of Qumran and Ein Ghuweir: The First Chemical Exploration of Provenience,” BASOR 321 (2001): 65–78. 41   Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 65. 42  L52 was identified by de Vaux, Archaeology and the Dead Sea Scrolls, 7, as “a washing-place with a stone basin and a large sump.” 43  Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 66–68. Marcello Fidanzio points out that de Vaux’s catalogue for Cave 6Q lists only one uninscribed jar (a cylindrical jar), which was discovered by Bedouin; see de Vaux, in DJD 3:10; Pl. V:26–2. Fidanzio believes that the inscribed jar actually is one of two which came from a cave ca. 100 meters south of Cave 6Q (both acquired from an antiquities dealer), and that this jar was subsequently mislabeled 6Q1 (Alain Chambon’s handwriting). See Józef T. Milik, “Deux jarres inscrites provenant d’une grotte de Qumrân,” in Baillet, Milik, and de Vaux, Les ‘petites grottes’, 37–41. I am grateful to Fidanzio for sharing with me these observations. 44  Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 69–70. 45  Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 70, 72.

Magness

tion.46 The Jerusalem group includes the inkwell and the inscribed jars from L34 and Caves 6Q and 7Q,47 as well as the pottery analyzed from Ein el-Ghuweir. According to Yellin et al., the composition of the vessels in this group “agrees with the reference group representing Jerusalem and matching the Motza marl (Jerusalem clay) . . .”48 They add: “we find it quite odd that all four cylindrical jars . . . so typical to Qumran, should have come from Jerusalem”— but the only three jars illustrated in publications consist of a bag-shaped jar, an ovoid jar (inscribed Roma) (from Cave 7Q), and a wide version of a cylindrical jar (inscribed Johanan Hatla, from L34).49 There is no breakdown of the chemical composition of each sample except for Fe (Iron) and Sc (Scandium) content; instead the authors provide only mean readings.50 4

Gunneweg and Balla’s INAA Analysis

In 2003, Jan Gunneweg and Marta Balla published an analysis of approximately 200 samples of Qumran pottery, using INAA.51 Most of the sampled pottery comes from Qumran (from 53 different loci) and the surrounding caves (among them scroll caves 1Q, 3Q, 4Q, 6Q, 7Q, 8Q), but also includes pottery from other sites in the region (Jericho, Jerusalem, Hebron, Ein ez-Zara, Cypros, Masada, and Ein Feshkha), as well as kiln linings, oven covers, and clay balls from Qumran, and puddle clay and mud from Qumran, the Dead Sea, Hebron, and Jerusalem.52 In the same volume, Jacek Michniewicz and Miroslaw Krzyśko published the results of a petrographic analysis of fifty jars and five other vessels (saucers, bowls, and a pot) from Qumran and the surrounding caves, which they compared with clay samples from sites around Judea.53 Gunneweg and Balla criticize Yellin et al. because the average standard deviation (almost twenty percent) of 46   Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 73. 47  For the jar from Cave 6Q, see n. 43 above. 48  Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 75; also see 76, Fig. 7. 49   Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 77. 50  Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 72–76; the only exception is the Roma jar, the chemical composition of which is provided in Table 7 (p. 76). 51  Gunneweg and Balla, “Neutron Activation Analysis.” 52  Gunneweg and Balla, “Neutron Activation Analysis,” 7–8. 53  Jacek Michniewicz and Miroslaw Krzyśko, “The Provenance of Scroll Jars in the Light of Archaeometric Investigations,” in Humbert and Gunneweg, Khirbet Qumrân et ‘Aïn Feskha, 59–99 (see especially 62–66).

THE CONNECTION BETWEEN THE SITE OF QUMRAN & THE SCROLL CAVES

their Qumran group is three to four times higher than that of their Jerusalem group, which calls into question the validity of the Qumran group.54 They note that Yellin et al.’s assumption that their Qumran group is made from local clay is based not on chemical evidence but on the presence of the potters’ workshop.55 However, Gunneweg and Balla also failed to locate the source of the clay of their local (Qumran) pottery: “To my recollection, over the past 28 years, we have only been successful three times in matching readily made pottery with clay from a specific clay source, and we have refrained from giving it another try this time, although a few (seven) clay samples were included to clarify previous INAA results from sites around Qumran.”56 In fact, Michniewicz and Krzyśko conclude that, “The Qumran jars under investigation were made of a raw material which is not present in the vicinity of the Qumran site.”57 Before continuing with the question of the source of the clay, let us review the five chemical groups (I–V) identified by Gunneweg and Balla. Nearly all of the samples belong to three of the five groups, in approximately equal numbers: I (44), II (41), and III (41). Because the largest number of samples belongs to Group I, Gunneweg and Balla assume it is local clay, although no source has been found. According to Gunneweg and Balla, the similar chemical composition of the kiln linings, oven covers, clay balls, Qumran puddle mud, and Dead Sea mud indicates that Group I is made of clay local to Qumran. At the same time, they acknowledge that the kiln linings, Qumran puddle mud, stucco from the “scriptorium table,” and some of the oven covers and jar stoppers have a much higher calcium content than the Group I pottery (averages of 24 percent versus 7.8 percent), which they attribute to “dilution” of the clay.58 Gunneweg and Balla also note that some of the oven covers, jar stoppers, and clay balls have a low calcium content, which means that “both the high and low calcium-bearing clay were [sic!] found and used and local to Qumran.”59 The Johanan Hatla jar from L34 and the Roma jar from Cave 7Q, which Yellin et al. identified as made of Jerusalem clay, belong to Gunneweg and Balla’s Group I, which they

54  Gunneweg, “Archaeology and Archaeometry at Qumran,” XVII; Gunneweg and Balla, “Neutron Activation Analysis,” 5. 55  Gunneweg, “Archaeology and Archaeometry at Qumran,” XVII. 56  Gunneweg, “Archaeology and Archaeometry at Qumran,” XVII. 57  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 76. 58   Gunneweg and Balla, “Neutron Activation Analysis,” 8. For dilution see Hans Mommsen, “Short Note: Provenancing of Pottery—The Need for an Integrated Approach?” Archaeometry 46/2 (2004): 269. 59  Gunneweg and Balla, “Neutron Activation Analysis,” 9.

189

claim is made of clay local to Qumran.60 With the exception of the Roma jar, Yellin et al. do not list all of the trace elements in each vessel they analyzed, but only the amounts of Iron (Fe) and Scandium (Sc).61 A comparison between the numbers produced by the two sets of tests reveals significant differences, which are especially striking in the case of the Roma jar. The Johanan Hatla jar from L34: Fe (%): Yellin et. al. 3.81 versus Gunneweg and Balla 4.46 ± 0.05 Sc (ppm): Yellin et al. 16.35 versus Gunneweg and Balla 16.9 ± 0.2 The Roma jar from Cave 7Q:62 Ce (ppm): Yellin et al. 44.46 versus Gunneweg and Balla 9.09 ± 1.5 Co (ppm): Yellin et al. 11.71 versus Gunneweg and Balla 49 ± 2 Cs (ppm): Yellin et al. 5.29 versus Gunneweg and Balla 11.1 ± 0.3 Eu (ppm): Yellin et al. 1.03 versus Gunneweg and Balla 108 ± 2 Fe (%): Yellin et al. 3.32 versus Gunneweg and Balla 1.02 ± 0.07 Hf (ppm): Yellin et al. 4.22 versus Gunneweg and Balla 3.79 ± 0.04 La (ppm): Yellin et al. 23.33 versus Gunneweg and Balla 2.09 ± 0.10 Lu (ppm): Yellin et al. 0.33 versus Gunneweg and Balla 22.3 ± 0.2 Na (%): Yellin et al. 0.36 versus Gunneweg and Balla 7±1 Nd (ppm): Yellin et al. 17.77 versus Gunneweg and Balla 0.41 ± 0.01 Rb (ppm): Yellin et al. 90.00 versus Gunneweg and Balla 16 ± 4

60  For the Johanan Hatla jar, see Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 66 (their number 21), with Gunneweg and Balla, “Neutron Activation Analysis,” 10–11, 77 (their number 121). For the Roma jar, see Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 66, 75–76 (their number 24), with Gunneweg and Balla, “Neutron Activation Analysis,” 11–13 (their number 134). 61  Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 75; in other tables they provide the mean readings per group. 62  For the chemical composition of the Roma jar, see Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 76, Table 7 (their number 24), with Gunneweg and Balla, “Neutron Activation Analysis,” 48 (their number 134).

190

Magness

Sc (ppm): Yellin et al. 15.22 versus Gunneweg Balla 0.5 ± 0.05 Sm (ppm): Yellin et al. 4.22 versus Gunneweg Balla 15.4 ± 0.1 Ta (ppm): Yellin et al. 0.72 versus Gunneweg Balla 5.07 ± 0.01 Th (ppm): Yellin et al. 6.19 versus Gunneweg Balla (no reading) U (ppm): Yellin et al. 3.48 versus Gunneweg Balla 6.9 ± 0.2 Yb (ppm): Yellin et al. 2.02 versus Gunneweg Balla 3.6 ± 0.2

and and and and and and

Gunneweg and Balla’s Group I includes several complete or nearly complete vessels from the potters’ workshop (among them cylindrical jars), which suggests that Group I was indeed manufactured at Qumran (including L44 nos. 251, 268, 284; L45a no. 127; L59 no. 106; L84 no. 131).63 The fact that Group I includes one cylindrical jar from Cave 6Q (no. 231), one ovoid jar from Cave 7Q (the Roma jar), and two cylindrical jars from Cave 8Q (nos. 132, 186) attests to the connection between these caves and between these caves and the site.64 As Sidnie Crawford recently highlighted, Caves 7Q–10Q in the marl terrace were residential caves and therefore were part of the Qumran settlement. These caves yielded small numbers of scrolls which apparently were for individual use, in contrast to the larger collections in Cave 4Q and the caves in the limestone cliffs (which she believes were deposited over a relatively short period of time).65 The Johanan Hatla jar and a complete juglet from L114 (no. 185) indicate that the clay used to manufacture vessels belonging to Group I was in use prior to the site’s destruction in 31 BCE (when L34 went out of use) and ca. 9/8 BCE (when the pantry in L114 was destroyed).66 A complete bowl (no. 273) belonging to Group I from the upper level of L8 (the staircase in the southwest corner of the tower) indicates that this clay was still in use in the first century CE.67 63  Gunneweg and Balla, “Neutron Activation Analysis,” 10–13; see de Vaux, Archaeology and the Dead Sea Scrolls, 7. 64  See Gunneweg and Balla, “Neutron Activation Analysis,” 11, Table 3. 65  Sidnie White Crawford, “Qumran: Caves, Scrolls, and Buildings,” in A Teacher for All Generations: Essays in Honor of James C. VanderKam (JSJSup 153; ed. E. F. Mason et al.; Leiden: Brill, 2011), 1:259–60, 265, 272. 66  See Gunneweg and Balla, “Neutron Activation Analysis,” 11, Table 3. For these destruction dates, see Magness, Archaeology of Qumran, 66–68, 125–26. 67  See Gunneweg and Balla, “Neutron Activation Analysis,” 11, Table 3. For L8 see Humbert and Chambon, Fouilles de Khirbet Qumrân, 294–95.

Group II has high potassium (K) and low sodium (Na) values that are characteristic of Motza Formation clay, and no statistical match was found with clay local to Qumran (although as we have seen no local clay source has been located). Therefore, Gunneweg and Balla identify Group II as clay from Jerusalem or Beit ʿUmmar (near Hebron). They are unable to determine if the raw clay was brought to Qumran from a distance (which they consider unlikely) or the vessels were manufactured elsewhere and transported in finished form to Qumran.68 The vessels belonging to Group II include three cylindrical jars (nos. 118, 119, 238) which were embedded in a recess in L13 (a room to the south of the tower which contained a staircase); an ovoid jar (no. 242) from L61 (the potters’ workshop); two bowls, an ovoid jar, and a wheel made (“Herodian”) oil lamp from the pantry in L114 (nos. 179, 188, 244, 194); and a pseudo-Nabataean bowl fragment from Ein Feshkha (no. AF 203).69 A complete funnel (no. 276) from the Period II level in L86/89 indicates that in the first century CE, vessels were being manufactured of Group II clay.70 Group II pottery from the scroll caves includes one cylindrical jar from Cave 1Q; a bowl-shaped lid from Cave 3Q; one cylindrical jar, one “store jar,” and two bowl-shaped lids from Cave 7Q; and seven cylindrical jars, one bowlshaped lid, and one bowl from Cave 8Q.71 Scholars had already noted that the distribution of distinctly sectarian works indicates a connection between the various scroll caves, with copies of pesharim from Caves 1Q and 3Q, copies of the Community Rule from Caves 1Q, 5Q, and 11Q, and copies of the Damascus Document from Caves 5Q and 6Q (in addition, non-sectarian works that advocate a solar calendar were found in Caves 1Q, 2Q, and 6Q [Enoch], Caves 1Q, 2Q, 3Q, and 11Q [Jubilees]).72 The identification of scrolls from Caves 1Q, 2Q, 3Q, 4Q, 6Q, and 11Q that were written by the same scribal hand also links these caves.73 The distribution of Group II pottery confirms the connection between Caves 1Q, 3Q, 7Q, and 8Q, and the connection between these caves and the site. Group III’s chemical composition matches pottery from the Hasmonean and Herodian palaces at Jericho, which led Gunneweg and Balla to conclude that either “Jericho potters made this ware that was then shipped to Qumran, or that Qumran potters took the same clay as the Jericho people did and forged their own style. However, the former suggestion that local Jericho potters made these ves68  Gunneweg and Balla, “Neutron Activation Analysis,” 14. 69  Gunneweg and Balla, “Neutron Activation Analysis,” 13–18. 70  Gunneweg and Balla, “Neutron Activation Analysis,” 14. 71  Gunneweg and Balla, “Neutron Activation Analysis,” 14. 72  White Crawford, “Qumran: Caves, Scrolls, and Buildings,” 266. 73  White Crawford, “Qumran: Caves, Scrolls, and Buildings,” 267.

THE CONNECTION BETWEEN THE SITE OF QUMRAN & THE SCROLL CAVES

sels is much more likely because of the uniformity and the workmanship of these wares.”74 The Group III pottery includes a painted bowl (no. 204) from L61 (the potters’ workshop).75 It appears that all of the vessels analyzed from Period Ib contexts in L86/89 (the pantry) belong to Group III, including seven cups and bowls (nos. 205, 252, 253, 258, 259, 260, 261) and one table jug (114).76 Two of these vessels (nos. 252 and 261) were analyzed by Yellin et al. and are identified as their local Qumran ware.77 Group III also includes one cylindrical jar and one “store jar” from Cave 1Q, one cylindrical jar from Cave 3Q, and an imitation Hellenistic oil lamp from Cave 1Q (a type characteristic of Herod’s reign that is also found at Masada and Jericho), which demonstrates the connection between Caves 1Q and 3Q and between these caves and the site.78 Other lamps of this type from L66 (the potters’ workshop) (no. 290) and L130 (nos. 291–292) also belong to Group III.79 Group III vessels appear to date to the first century BCE, including the dishes from the pantry in L86/89, which was destroyed by the earthquake of 31 BCE, and the imitation Hellenistic oil lamps from L130 which are characteristic of the reign of Herod the Great. This suggests that Group III clay was in use mainly and perhaps only in Period Ib (first century BCE), whereas Group I includes vessels of the first century BCE and first century CE. Based on the published evidence, Group II clay might have been used only in the first century CE. Only eight vessels analyzed from Qumran (including one ovoid jar) belong to Chemical Group IV, which is identified as originating in the area of Edom/Nabataea.80 Another ovoid jar, two bowls, and a juglet belong to Chemical Group V, which is identified as coming from Jericho.81 No matches were found for four cooking pots, two of which come from L130.82

74  Gunneweg and Balla, “Neutron Activation Analysis,” 18. 75   Incorrectly described as a “pseudo-Nabataean bowl” by Gunneweg and Balla, “Neutron Activation Analysis,” 18. 76  Gunneweg and Balla, “Neutron Activation Analysis,” 18–19, 40. 77  Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 69, Table 1, nos. 19–20. 78  Gunneweg and Balla, “Neutron Activation Analysis,” 19, 23. For the distribution and dating of oil lamps of this type, see Bar Nathan, Hasmonean and Herodian Palaces at Jericho, 110–12 (Type J-LP3A1). 79  Gunneweg and Balla, “Neutron Activation Analysis,” 23. 80  Gunneweg and Balla, “Neutron Activation Analysis,” 19–22. 81  Gunneweg and Balla, “Neutron Activation Analysis,” 22. 82  Gunneweg and Balla, “Neutron Activation Analysis,” 22.

5

191

Michniewicz and Krzyśko’s Petrographic Analysis

The identification of Yellin et al.’s Qumran Group with Gunneweg and Balla’s Jericho Group (III), and Yellin et al.’s Jerusalem Group with Gunneweg and Balla’s Qumran Group (I) is complicated further by the results of the petrographic analysis conducted by Michniewicz and Krzyśko, which differ from those of Gunneweg and Balla.83 Among Michniewicz and Krzyśko’s observations are the following: 1)

“A great majority of the ceramic fragments under investigation and the clay samples taken in the field form a geochemically homogenous set.”84 After separating the pottery and clay samples into subsets, Michniewicz and Krzyśko found that clays taken from the Qumran aqueduct are the most distant from the Qumran pottery in composition: “Closest to each other are groups of jars and ceramics from the settlement [and caves].”85 The chemical composition of the Qumran pottery is closer to clay from the region of Jerusalem and Hebron than el-Jib.86 2) Clay taken from Na�al Arugot (near Ein Gedi) and Motza Formation clay are geochemically similar, but differences in composition were revealed by ­microscopic examination, indicating that “the interpretation of chemical analyses devoid of supplementary petrographic investigations could be misleading.”87 3) The Qumran pottery is made of raw material not found in the vicinity of the site of Qumran, including sediments of the Lisan Formation and clays from sediments of several wadis along the Dead Sea, some of which are very close to Qumran.88 4) The Qumran pottery was made by the same technique from similar but not homogenous raw materials, consisting of non-silty clay with an admixture of about ten percent quartz sand. Approximately half of the pottery analyzed contained preserved shale particles.89 5) Most of the pottery probably was made from Motza Formation clays. The carbonate content indicates that the clay was taken from the upper layers of the 83  Michniewicz and Krzyśko, “The Provenance of Scroll Jars.” 84  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 69. 85  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 74. 86  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 74. 87  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 75. 88  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 76. 89  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 76.

192

6) 7)

Magness

formation or that carbonate sand was added on purpose as temper.90 Petrographic differences indicate that the clays were not obtained from the same location at the same time.91 Petrographic and geochemical features indicate that the pottery from the caves and the site are made of the same material.92

Michniewicz and Krzyśko note that “the clayey materials from the region of Judaea are highly homogeneous from the geochemical point of view; hence, no precise determination of the provenance of local pottery is possible (e.g. it is not possible to differentiate the ceramics made from the clays of the Motza Formation taken in the vicinity of Hebron from the clays of the same formation taken in the vicinity of Jerusalem).”93 Yellin et al. made a similar observation: “the rare-earth patterns for the non-Jerusalem pottery and the Jerusalem pottery are similar, as would be expected on geochemical grounds because of the similarity of the geochemical history of the region . . . This similarity suggests that the non-Jerusalem composition could not have been imported from a distant and different geological region.”94

6

Michniewicz’s Petrographic and Chemical Analysis

In 2009, Michniewicz published a book-length study in which he highlights the limitations of INAA, noting that various factors often make it impossible to determine the precise provenance (clay source) of pottery.95 He criticizes Gunneweg and Balla for drawing conclusions that are not supported by the reference data or statistical computation, and notes that their claim that a large quantity of Qumran pottery was produced in Jericho is based on the analysis of only four pottery specimens from Jericho.96 Michniewicz concludes that “the opposing conclusions drawn by the two teams [Gunneweg and Balla versus Michniewicz and Krzyśko] result from different interpretations of chemical data rather than from differences in the NAA data obtained.”97 Michniewicz conducted a new petrographic and chemical analysis of 127 pottery samples, mostly from Qumran and Jericho, with a small number from Ein ezZara and Khirbet Mazin.98 Nearly all of the samples from Qumran and Jericho were made of one of three kinds of raw clay, described by Michniewicz as Petrographic

Table 12.1 Correlation of analyses performed on the same ceramic vessels by more than one team Reference to publication by de Vaux

Yellin et al.’s analysis

Gunneweg and Balla’s analysis

“Fouilles de Khirbet Qumrân,” Fig. 2:6 (shallow bowl #1591) “Fouilles de Khirbet Qumrân,” Fig. 2:11 (cup or bowl #1601) Archaeology and the Dead Sea Scrolls, Pl. XV:b (center) (the Johanan Hatla jar #621) “Archéologie,” Fig. 6:5 (the Roma jar #7Q6) “Archéologie,” Fig. 2:1 (cylindrical jar #8–11) “Fouilles de Khirbet Qumrân,” Fig. 2:8 (cup or goblet #1587)

Qumran Group (#19)

Group 3 (Jericho Group) (S-261) Group 3 (Jericho Group) (S-252) Group 1 (Qumran Group) (S-121)

Qumran Group (#20) Jerusalem Group (#21) Jerusalem Group (#24)

90  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 76. 91  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 76. 92  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 76. 93  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 76. 94   Yellin, Broshi, and Eshel, “Pottery of Qumran and Ein Ghuweir,” 73.

Group 1 (Qumran Group) (C-7 134) Group 1 (Qumran Group) (C-8 186) Group 3 (Jericho Group) (S-258)

Michniewicz’s analysis

Group 3 (not local to Qumran) Group 4

95  Michniewicz, Qumran and Jericho Pottery, 24. 96  Michniewicz, Qumran and Jericho Pottery, 26. 97  Michniewicz, Qumran and Jericho Pottery, 27. 98  Michniewicz, Qumran and Jericho Pottery, 27.

THE CONNECTION BETWEEN THE SITE OF QUMRAN & THE SCROLL CAVES

Groups I, II, and III.99 Petrographic Group I is characterized by slightly silty, foraminiferous clay (marly clay with variable amounts of fine limestone grains). It was used to manufacture bowls, storage jars, cylindrical jars, jugs, and oil lamps.100 Petrographic Groups II and III are similar in chemical composition and consist of rich clay but differ in tempers, with Group II containing pure calcareous sand and Group III containing quartz or quartzcarbonate sand. Both groups appear to be made of Motza Formation clay.101 Petrographic Groups II and III were used to manufacture most of the cylindrical jars as well as bowls, goblets, kraters, and pitchers. Storage jars made of this clay have also been found at Khirbet Mazin and Ein ez-Zara.102 A fourth kind of clay—Petrographic Group IV—is a silty, ferruginous terra rosa soil used mainly for cooking pots.103 Most of the Qumran pottery sampled by Michniewicz is unpublished and consists of eighteen oil lamps and sixtytwo other vessels or vessel fragments.104 The only four published pieces are two oil lamps from Trench A and L52 (nos. 211 and 941), a bowl (“goblet”) from L89 (no. 1587), and a cylindrical jar from Cave GQ8 (8–11/1). The lamp from L52 (no. 941) is a Hellenistic type characteristic of Herod’s reign that is represented by numerous specimens from L130. It belongs to Michniewicz’s Petrographic Group I.105 The other lamps of this type belong to Groups I and III, including two specimens (nos. 5085 and 5087) which were analyzed by Gunneweg and Balla and assigned to their Group III (their Jericho group).106 The oil lamp from Trench A (no. 211), which Michniewicz describes as “Herodian” but appears to be Hellenistic, belongs to his Petrographic Group II.107 The bowl (“goblet”) from the pantry in L89 belongs to Michniewicz’s Group IV (he notes that petrographically this bowl belongs to Group IV, but chemically is close to Group I).108 This bowl was ana99  Michniewicz, Qumran and Jericho Pottery, 85–90, 117, 135–36. 100  Michniewicz, Qumran and Jericho Pottery, 135–36. 101  Michniewicz, Qumran and Jericho Pottery, 137; for possible sources of Petrographic Group I see 141. 102  Michniewicz, Qumran and Jericho Pottery, 136–37. 103  Michniewicz, Qumran and Jericho Pottery, 138. 104  Michniewicz, Qumran and Jericho Pottery, 37–38, 85–86. 105  Michniewicz, Qumran and Jericho Pottery, 37, Table 4 no. 941; see de Vaux, “Fouilles au Khirbet Qumrân,” Fig. 3:16. For specimens from L130 see de Vaux, “Fouilles de Khirbet Qumrân,” Fig. 1:1–4. 106  Michniewicz, Qumran and Jericho Pottery, 37–38; see Gunneweg and Balla, “Neutron Activation Analysis,” 23. 107  Michniewicz, Qumran and Jericho Pottery, 37; see de Vaux, “Fouilles au Khirbet Qumrân,” Fig. 2:15. 108  Michniewicz, Qumran and Jericho Pottery, 85, 98; de Vaux, “Fouilles de Khirbet Qumrân,” Fig. 2:8.

193

lyzed by Gunneweg and Balla and was assigned to their Group III (Jericho group).109 Finally, the cylindrical jar from Cave GQ8 belongs to Michniewicz’s Group III; it too was analyzed by Gunneweg and Balla and was assigned to their Group I (Qumran group).110 Michniewicz concludes that because nearly all of the pottery from Jericho and Qumran is made from the same three varieties of raw material (clays), there is no basis for distinguishing vessels made at one site versus the other, contrary to Gunneweg and Balla.111 Furthermore, Petrographic Groups II and III consist of clays that do not come from Wadi Qumran and were not deposited in the form of suspension, as indicated by the abundant coarse fragments of shale (which are too heavy to be transported by water), the absence of microfossils (which are present in the Wadi Qumran deposit), and a different chemical composition. Petrographic Group I is more marly than the Wadi Qumran deposits and contains different trace elements. These data contradict Yitzhak Magen and Yuval Peleg’s claim that the Qumran pottery was manufactured from clay-rich silt that the stepped pools were designed to collect.112 7

The Results of the Various Analyses

The INAA and petrographic analyses reviewed above attest to a connection between the pottery from the various scroll caves and between these caves and the Qumran settlement. Nearly all of the pottery appears to have been manufactured at Qumran of clay obtained from two or three unknown sources, with an admixture of different tempers. As Ronald Bishop and James Blackman state, “as a first approximation, the pottery found to be in abundance at a site is more likely to be of ‘local’ production than is the pottery that is sparsely represented. If the abundant pottery of a given archaeological period is found to have a characteristic composition that is also present in the abundant pottery of earlier and successive periods, the assumption of locally available resources is strengthened. Note that here no actual source material is analyzed directly but is inferentially reflected by the ­composition of

109  Gunneweg and Balla, “Neutron Activation Analysis,” 19 no. 258. 110  Michniewicz, Qumran and Jericho Pottery, 86; see de Vaux, in DJD 3 Fig. 2:1; Gunneweg and Balla, “Neutron Activation Analysis,” 11, Table 3, no. C-8 186. 111  Michniewicz, Qumran and Jericho Pottery, 118, 141. 112  Michniewicz, Qumran and Jericho Pottery, 138–39, 141.

194 the artifacts.”113 Other scholars have remarked on the futility of trying to identify clay source(s), and instead recommend focusing on the composition of the paste.114 Although ancient potters usually obtained their clay within a radius of 3–4 kilometers from the manufacturing site, ethnographic studies have documented cases of clay procurement from distances of up to thirty kilometers or more.115 Michniewicz and Krzyśko describe a contemporary potter in Jericho who makes vessels of clay that he brings from el-Jib (north of Jerusalem), a distance of some 25 kilometers.116 Another possibility to consider is itinerant potters. Bill Sillar describes itinerant potters in the Andean highlands who travel with pre-prepared clay to lower lying communities, where they manufacture pots in exchange for local resources (traditionally the clay was transported using llamas). Itinerant potters typically manufacture larger vessels, which are difficult to transport. Sillar notes that itinerant potters in the Andes often form fictive kinship ties with at least one member of the local community, which helps facilitate the exchange of pottery for local resources.117 This raises the possibility that some of the Qumran pottery was manufactured by itinerant potters who could have been members of the sect, and either brought their own clay or mined it from sources along the way. The large number of vessels belonging to Group III at Qumran—including the dining dishes from the pantry in L86/89—contradicts Gunneweg and Balla’s claim that this pottery was manufactured in Jericho. Instead, it appears that potters at Jericho and Qumran obtained their clay from some of the same sources. These sources presumably were located somewhere between Jericho and Qumran, which are about twenty kilometers apart. 113   Ronald L. Bishop and M. James Blackman, “Instrumental Neutron Activation Analysis of Archaeological Ceramics: Scale and Interpretation,” Accounts of Chemical Research 35 (2002): 604. 114  See Mommsen, “Short Note,” 268. 115  Kevin J. Vaughn and Hector Neff, “Tracing the Clay Source of Nasca Polychrome Pottery: Results from a Preliminary Raw Material Survey,” JARS 31 (2004): 1583; Bill Sillar, Shaping Culture, Making Pots and Constructing Households; An Ethnoarchaeological Study of Pottery Production, Trade and Use in the Andes (BAR.IS 883; Oxford: Archaeopress, 2000), 91–92. 116  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 75. 117  Sillar, Shaping Culture, 92. For itinerant potters on Cyprus, see Gloria A. London, “On Fig Leaves, Itinerant Potters, and Pottery Production Locations in Cyprus,” in Cross-craft and Crosscultural Interactions in Ceramics (ed. P. E. McGovern, M. D. Notis, and W. D. Kingery; Westerville, OH: American Ceramic Society, 1989), 72.

Magness

This may be one reason why Yellin et al. identified vessels of this ware (Gunneweg and Balla’s Group III) as local to Qumran. The variability of clay utilized by potters in a single workshop is a well-documented phenomenon. Sometimes a workshop uses different pastes to produce different types of wares.118 Analyses of cooking ware produced in a contemporary center in Spain indicated a high chemical variability not only among the different vessels but even in the paste of a single pot, “to an extent which chemical compositions from one ‘workshop,’ or even one ‘pot,’ could be classified as having different provenances.”119 These results contradict the common assumption that pottery produced by the same potters in the same workshop using the same technology and local clays should exhibit chemical homogeneity.120 The authors of this study also call into question the notion that a sample taken for analysis is representative of the chemical composition of that pot.121 This may help explain the chemical variations in the samples of the Johanan Hatla and Roma jars taken by Yellin et al. and by Gunneweg and Balla. It also means there is no need to assume that Yellin et al.’s Qumran and Jerusalem groups, and Gunneweg and Balla’s Groups I–III were produced anywhere other than Qumran. To conclude: Petrographic and chemical analyses indicate that the pottery from the site of Qumran and the scroll caves is made of the same clays and pastes, confirming the connection suggested by typological and morphological similarities. Most of this pottery is made of clay from at least three unknown sources in the region of the Judean Mountains and Dead Sea, but not silt from Wadi Qumran. The potters at Jericho appear to have obtained their clay from some of the same sources. Finally, the conflicting conclusions published about the sources of the Qumran pottery illustrate the subjectivity involved in scientific analyses, which are the result of a process of selection and interpretation.

118  Mommsen, “Short Note,” 268. 119   Jaume Buxeda i Garrigós, Miguel A. Cau Ontiveros, and Vassilis Kilikoglou, “Chemical Variability in Clays and Pottery from a Traditional Cooking Pot Production Village: Testing Assumptions in Pereruela,” Archaeometry 45/1 (2003): 1. 120  Buxeda i Garrigós, Cau Ontiveros, and Kilikoglou, “Chemical Variability in Clays and Pottery,” 2. 121  Buxeda i Garrigós, Cau Ontiveros, and Kilikoglou, “Chemical Variability in Clays and Pottery,” 15.

CHAPTER 13

The Functions of the Caves and the Settlement of Qumran: Reflections on a New Chapter of Qumran Research Jürgen K. Zangenberg Being the (almost) last speaker in this conference I feel obliged to emphasize that I do not intend to summarize the results of my colleagues speaking before me, nor do I wish to utter the “last, final words” on a topic that is by far not fully researched. I am not going to do more than express my own limited insights and preliminary reflections on this fascinating matter.1 1

Initial Reflections

Let me begin with some exegesis of the title “The Functions of the Caves and the Settlement of Qumran” under which Jodi Magness’ and my paper have been jointly scheduled. As an exegete trained to carefully read texts, I sense certain statements in the title that—unconsciously or openly— contain a number of assumptions. I would like to address three of them, which I consider somewhat symptomatic for the previous and ongoing discussion. In this context I also wish to name a few tasks I see ahead of us on the way to intensified research on the Qumran caves: a)

First of all, it is remarkable that the title “The Functions of the Caves and the Settlement of Qumran” places the caves first and only then mentions the site of Qumran. I hope this sequence is not only rhetoric, but expresses a hermeneutical principle. Starting from the caves is certainly to be welcomed, since it leaves trodden paths and opens the perspective for alternative scenarios. Often enough, the settlement and interpretive models about its use and inhabitants have been used as hermeneutical lens to inter-

1   This article is the expanded version of a paper delivered on February 21 at the international conference “The History of the Caves of Qumran” in Lugano. I deliberately kept the oral form of this paper as far as possible to preserve its experimental, preliminary and introductory character. I wish to express my gratitude to professor Marcello Fidanzio and the Faculty of Theology of Lugano for inviting me to the very fruitful and inspiring conference, as well as to the many colleagues who shared their opinions with me about my paper.

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_015

b)

pret the caves, too, in the assumption that the caves needed Qumran to be properly explained. If Qumran was Essene, as the hypothesis shared by many ­colleagues claim, then the caves must stand in some sort of genuine relationship with the Essene inhabitants of the site: as hiding places, storage spaces, living quaters. But what happens if we turn this sequence around and first examine the caves before their interpretation is determined by their supposed relationship with the site of Qumran? I therefore think that the first task ahead of us is to fully understand that the caves have their own history apart from, before and after Qumran, including potential overlaps between the two. This brings me to my second observation. I am happy to see that the title speaks of “functions” in the plural. Nobody has ever claimed that all caves had one and the same function during their entire lives, either in relation to the settlement or in their own respect. Roland de Vaux already emphasized the great physical varieties of these grottoes, and their diversity regarding the finds from them. Some grottoes are no more than cracks in the limestone and unsuitable for habitation, others are human-made and spacious; some caves contained material from several periods, others hardly anything or nothing. This observation was confirmed by several presenters during the conference, too. It is therefore reasonable to assume that the caves had various functions, not all of which may have had anything to do with Qumran. The focus on the cave-Qumran relationship sometimes prevented scholars to think in longer trajectories when considering the caves. But functions may have changed overtime, and some caves even had different functions for different groups at one and the same moment. It, therefore, seems to me that we more than ever before need to talk about the caves in the plural and avoid presupposing that they could be subsumed under one and the same roof—be it Qumran or not. And we need to accept that every cave may have had not only one, but various and varying functions during its lifetime.

196 c)

Zangenberg

The third observation is a direct consequence of the previous two. I am somewhat confused by the expression “the caves” in the title and feel obliged to ask: Which caves do we need to consider to understand the phenomenon? Where should we draw the line, which caves should we include and which ones exclude? Are we only to look at what people sometimes call “scroll caves” because their relationship to Qumran is thought to be particularly obvious?2 Often enough, the answer given to this question was an affirmative “yes!,” so people concentrated on 1Q–11Q when considering the relationship of “the caves” with Qumran and saw the caves without scrolls as odd kids on the block. But de Vaux already knew that the “scroll caves” were the exceptions in a much larger number of grottos whose own relationship with Qumran is far less conspicuous. Or should we concentrate on those caves in which fragments of “scroll jars” were found, or simply add them to the “scroll-caves” without further reason, thereby silently circumventing the problem that “scroll caves” and caves with “scroll jars” or fragments of them are not necessarily identical, because some “scroll-caves” did not contain “scroll-jars” and not all “scroll jars” were in fact found in “scroll caves”? So what type of finds does make a cave “Qumran-relevant” and another not? What criteria do tie the “Qumran-relevant” caves together with each other and at the same time distinguish them from “less relevant” others? To avoid circular arguments, we consequently need to change our perspective and move away from the “exceptions” 1Q–11Q (de Vaux), and instead turn to the variety of caves found in the entire region along the northwestern shore of the Dead Sea.

A brief summary of the methods and results of cave explorations and finds, therefore, seems appropriate.

2  See, e.g., Sidnie White Crawford, “Qumran: Caves, Scrolls, and Buildings,” in A Teacher for All Generations: Essays in Honor of James C. VanderKam (JSJSup 153; ed. E. F. Mason et al; Leiden: Brill, 2011), 1:253–73 or—with a rather different direction—Stephen J. Pfann, “Reassessing the Judaean Desert Caves: Libraries, Archives, Genizas and Hiding Places,” BAIAS 25 (2007): 147–70.

2

History of Exploration

The surroundings of Qumran, including the caves, have been explored in three major steps, each following its own goals and agendas. 2.1 Roland de Vaux and Colleagues It took a good 5 years after the discovery of the first scrolls in the winter 1946/47 in 1Q and the start of excavations at Khirbet Qumran in 1951 through Roland de Vaux, that the caves found the full attention of research. Only in 1952 bedouin diggers and archaeologists became intangled in what could be called the famous “scroll race,” moving from one cave to the other hunting for scrolls. 2.1.1

Caves in the Limestone Cliffs

a) Triggered by illicit bedouin excavations in 1Q3 and 2Q,4 Roland de Vaux between March 10 and 29, 1952 conducted a systematic survey of the cliffs west of Qumran in an 3  Excavation between Feb 15 and March 5, 1949; a brief report on the excavation is available in Roland de Vaux, “La grotte des manuscriptes hébreux,” RB 56 (1949): 586–609; Gerald L. Harding, “Introductory, the Discovery, the Excavation, Minor Finds,” in Qumran Cave 1 (DJD 1; ed. D. Barthélemy and J. T. Milik; Oxford: Clarendon Press, 1955), 6–7 as well as Roland de Vaux on the pottery, in DJD 1:8–17; Stephen J. Pfann, “Sites in the Judaean Desert Where Texts Have Been Found,” in Companion Volume to the Dead Sea Scrolls Microfiche edition (2nd rev. ed.; ed. E. Tov and S. J. Pfann; Leiden: Brill 1995), 109–19; Jean-Baptiste Humbert and Alain Chambon, eds., The Excavations of Khirbet Qumran and Ain Feshkha: Synthesis of Roland de Vaux’s Field Notes (NTOA.SA 1B; trans. and rev. Stephen J. Pfann; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 96; Weston W. Fields, The Dead Sea Scrolls: A Full History. Vol. 1: 1947–1960 (Leiden: Brill, 2009), 94–95 and 110–13 emphasizes the general “lack of scientific control” and discusses the possibility that there had in fact been two “caves 1” and not only one from which the 1Q-texts had come. Yizhar Hirschfeld, Qumran in Context: Reassessing the Archaeological Evidence (Peabody, MA: Hendrickson, 2004), 38 reproduces Harding’s map of 1Q (DJD 1: Fig. 1) and indicates that 1Q might already have been searched in antiquity (Qumran in Context, 39). Apart from pottery and manuscript fragments, 1Q also produced “a quantity of olive stones, date stones, palm fibre, and small and large pieces of wood were recovered,” as well as two fragments of a wooden comb (thus Harding, in DJD 1:7). No indication is given in what stratigraphic relationship these finds might have been to the scrolls. 4  Excavated between March 12–14, 1952; report by Roland de Vaux, “Archéologie,” in Les ‘petites grottes’ de Qumrân: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 9 (Cave 19 = 2Q); cf. Pfann, “Sites in the Judean Desert,” 111; Fields, The Dead Sea Scrolls, 132–33.

The Functions Of The Caves And The Settlement Of Qumran

area stretching 4km to the south and 4km to the north. Soundings in 230 caves were “infructueux,”5 only 40 caves in the limestone cliffs (only about a seventh, ca. 13%) contained archaeological material, 26 of which included pottery identical to that of 1Q and Khirbet Qumran and other objects ranging from the Chalcolithic to the Islamic periods.6 Though mainly intended to secure new texts,7 de Vaux’ cave explorations were indeed groundbreaking: For the first time, the caves had been made a topic of ­systematic research, and reports have been published in due course. b) During this first race along the cliffs in March 1952, de Vaux and his teams only discovered one single scroll cave (3Q) before the Bedouin,8 and unfortunately missed 11Q which was discovered by bedouin only in early February 1956.9 Apart from 11Q de Vaux and his teams missed a number of other very interesting caves. When de Vaux 5  De Vaux, DJD 3:4. 6   Reports by de Vaux, DJD 3:3–36; see also Roland de Vaux, “Exploration de la région de Qumrân: Rapport préliminaire,” RB 60 (1953): 540–61; William L. Reed, “The Qumran Caves Expedition of March, 1952,” BASOR 135 (1954): 8–13; Humbert and Chambon, Excavation of Khirbet Qumran, 65–66 with finds list. For comments on the different nomenclatures in de Vaux’ publications and for a finds summary see Mireille Bélis, “Révision commentée des différents systèmes de numération,” in Khirbet Qumran et ‘Aïn Feshkha. Vol. 2: Études d’anthropologie, de physique et de chimie (NTOA.SA 3; ed. J-B. Humbert and J. Gunneweg; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 409–15. On the survey see also Pfann, “Sites in the Judean Desert,” 112; Lena Cansdale, Qumran and the Essenes: A Re-evaluation of the Evidence (TSAJ 60; Tübingen: Mohr Siebeck, 1997), 171; Fields, The Dead Sea Scrolls, 132–33. The exact amount of examined caves differs between Reed, Pfann and Fields, the latter mentioning 225 caves surveyed, 37 “further investigated, and 25 were found to contain shards of pottery similar to those discovered at Qumran, as well as remnants of cloth similar to the scroll wrappings found in Cave 1” (Fields, The Dead Sea Scrolls, 132). 7  Cf. the comment by Dennis Walker and Robert H. Eisenman, “The 1990 Survey of Qumran Caves,” QC 2 (1992): 45: “The first survey of the cliffs near Qumran was primarily a search for more Dead Sea Scrolls.” 8  Excavated between March 14–25, 1952 and reported by de Vaux, in DJD 3:7–8 (Cave 8 = 3Q) and 26; Humbert and Chambon, Excavations of Khirbet Qumran, 66 with finds list; cf. Reed, “The Qumran Caves Expedition,” 10–12; Pfann, “Sites in the Judean Desert,” 112; Fields, The Dead Sea Scrolls, 134–38. 9  Excavated between February 26 and March 14, 1956 and published in Roland de Vaux, “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e et 5e campagnes,” RB 63 (1956): 573–74; Humbert and Chambon, Excavations of the Khirbet Qumran, 67–68 with finds list; also see de Vaux’s short note in DJD 3:13; cf. Pfann, “Sites in the Judean Desert,” 113; Fields, The Dead Sea Scrolls, 299–303, 343.

197

explored 11Q in February 1956 he realized that bedouin had continued excavating several caves de Vaux had already examined in 1952. He “found two caves of interest side by side almost opposite Cave 3, but the Bedouin had cleared out any manuscripts that might have been there, and had left behind only pottery contemporary with Qumran, and bits of fabric and leather.”10 c) Two other caves are of particular interest. As leader of one of de Vaux’ survey teams, Józef T. Milik discovered two quite remarkable caves during his extensive surveys between March 10 and 29, 1952. The first one (Cave 12) was located on March 12, 1952 in the limestone cliffs north of Qumran. It was not habitable inside, but people apparently camped outside in the entrace area (abri) where they had stored provisions. The pottery was neatly put up against the side of the rock, comprising storage jars (including at least one cylindrical specimen of the “scroll jar”-type), “four pots as jar covers, one bowl and an exceptionally well preserved palm mat” that was found covering the jars.11 As no scrolls were found, Cave 12 unfortunately did not make any larger impact on the discussion. Shortly thereafter, one of Milik’s team members found Cave 29.12 Squeezing himself through a 2m long, narrow tunnel, Milik reached a round, inner chamber of ca. 3m diameter where he found “a very large number of scroll jars, unbroken, neatly arranged, and to his disappointment, empty.”13 The jars (7 of them cylindrical) had been piled up against the wall of the grotto with 16 lids stacked up one atop each other next to them. In addition to that, de Vaux records a bowl, a lamp and fragments of at least 6 more storage jars. But the actual assemblage might have even been larger, since Milik had apparently only cleared some of the soft surface sediment to removethe visible 10  De Vaux, “Fouilles de Khirbet Qumrân,” 573–74; Fields, The Dead Sea Scrolls, 299. 11  Zdzislaw J. Kapera and Robert Feather, Doyen of the Dead Sea Scrolls: An In-Depth Biography of Józef Tadeusz Milik (1922–2006) (Qumranica Mogilanensia 17; Cracow/Mogilany: Enigma Press, 2011): 27–28f. The discovery date mentioned in Kapera and Feather (February 12, 1952) is in conflict with de Vaux, in DJD 3:8, Pl. VI, IV (Cave 12). 12  Fields, The Dead Sea Scrolls, 136–37 (137) on the basis of an interview with Józef T. Milik in 2004. It should be interesting to explore if any documentation was made by Milik and if some of it might have been preserved. Because Cave 29 did not contain scrolls when found in 1952, it is not mentioned by Pfann, “Sites in the Judean Desert,” but see de Vaux, in DJD 3:11. Pl. IV; Humbert and Chambon, Excavations of the Khirbet Qumran, 71; Kapera and Feather, Doyen of the Dead Sea Scrolls, 27–29. I am grateful to my colleague Dr. Zdzislaw J. Kapera (Cracow) for supplying me with relevant literature on this cave. 13  Fields, The Dead Sea Scrolls, 137.

198 jars and lids, but neither systematically excavated the accumulated sediment nor documented the entire cave or the layers he encountered. So we do not know what other objects might have been stored besides the jars, what other evidence of human use there might have been in the cave, such as a fireplace or remains of food stuffs. Instead, the empty storage jars with accompanying lids immediately reminded Milik of the famous 8th century CE report by patriarch Timothy I. about text discoveries near Jericho, so he dubbed this cave “Timothy Cave.” Milik obviously thought that the cylindrical jars once contained scrolls which had later been removed, but there is no positive proof for that.14 Unfortunately, Milik did not explore any alternative theories. De Vaux, in any case, rejected the possibility of habitation without giving any reasons.15 Since the “Timothy Cave” and Cave 12 were no “scroll caves” in the proper sense despite their large numbers of “jars,” they were largely ignored in subsequent discussions. Here clearly a chance was missed, and we see how problematic it is to separate “scroll caves” from “non-scroll caves.” With its domed central chamber and its narrow, protected entrance Cave 29 might very well have been used as supply depot for locals or as hiding place for a group of refugees in times of danger—or both. We will never know. 2.1.2 Caves in the Marl Terrace Proper archaeological examination of the caves located in the edge of the marl terrace at Qumran only started on September 22, 1952 and lasted until the 29th (so half a year after the cliff expedition), after bedouin had been clearing out more than half of the sediments in 4Q, removed the “majority of the fragments” and begun selling them via Kando.16 This is somewhat surprising when we consider that excavations had already been going on in Qumran on 14  Not a single scrap of manuscript was recorded in Cave 29 which might have been left had scrolls been removed 700 years after deposition. De Vaux, in DJD 3:11 did not accept Milik’s interpretation; see also Kapera and Feather, Doyen of the Dead Sea Scrolls, 28. 15   On that problem see Mladen Popović, “Qumran as Scroll Storehouse in Times of Crisis? A Comparative Perspective on Judaean Desert Manuscript Collections,” JSJ 43 (2012): 551–94 (560 n. 25). 16  On 4Q see Roland de Vaux, “Archéologie,” in Qumrân Grotte 4.II: I. Archéologie; II. Tefillin, Mezuzot et Targums (4Q128–4Q157) (DJD 6; ed. R. de Vaux and J. T. Milik; Oxford: Clarendon Press, 1977), 3–22 (9–13); Humbert and Chambon, Excavations of Khirbet Qumran, 66, 96 with (incomplete?) list of finds; Fields, The Dead Sea Scrolls, 138–54 (144); Dennis Walker, “Qumran Cave 4 and the 1952 Survey,” QC 3 (1993): 39.

Zangenberg

top of the very same marl terrace since 1951. We do not know exactly for how long the bedouin had been able to work undisturbed in 4Q before de Vaux and his team intervened, but the bedouin expedition must have been a highly effective and large enterprise, at times comprising up to 100 volunteers.17 The loss of information, in any case, was dramatic and very likely can never be remedied. By taking out thousands of text fragments without any documentation, the diggers deprived the finds of their proper archaeological context, mixed up the stratigraphy right down to the lower levels of the accumulation in the cave, and caused a lot of damage to scroll and non-scroll material. It is even doubtful if all the material the Ta‘amireh had taken from 4Q during their work has eventually reached the desks of scholars.18 Fortunately enough, de Vaux’s team was still able to explore “the lower layers of the cave and a small underground chamber (. . .), and they discovered the original entrance,”19 though everything had to be conducted under the “deadly combination” of great haste with minimal documentation.20 Reading these comments, one is struck by the discrepancy between what we really can know about the character and use of 4Q, and the many detailed hypotheses that have been based on the fragmentary data from it ever since, e.g. on how the scrolls might have been stored in 4Q and who might have lived there. Triggered by the 4Q calamity, a similar race with the bedouin ensued like in March 1952 with equally limited success for the archaeologists. Again, the bedouin were faster and more effective, but the archaeologists at least succeeded in September 1952 to be the first to arrive at 5Q.21 At the same time, however, they had to accept that much more prolific 6Q had already in early Spetember 1952 been found by bedouin.22 Caves 7Q–10Q at last were discovered by the archaeologists, too, during a systematic exploration of possibly collapsed caves at the southern tip of the marl terrace during the 4th season between February 16 and March 2, 1955.23 But all these caves had 17  Fields, The Dead Sea Scrolls, 139–40. 18  De Vaux, in DJD 6:3–5 plus 537 n. 14 and 16; Fields, The Dead Sea Scrolls, 144. 19  Fields, The Dead Sea Scrolls, 141. 20  Hirschfeld, Qumran in Context, 39; excavation report de Vaux, in DJD 6:3–22. 21  Fields, The Dead Sea Scrolls, 125, 142; Pfann, “Sites in the Judean Desert,” 112; see the report de Vaux, in DJD 3:26. 22  Fields, The Dead Sea Scrolls, 142; Pfann, “Sites in the Judean Desert,” 112; report de Vaux, in DJD 3:26. 23  Fields, The Dead Sea Scrolls, 252; Pfann, “Sites in the Judean Desert,” 113; reports on 6Q de Vaux, in DJD 3:26; on 7Q de Vaux, in DJD 3:27–30; on 8Q de Vaux, in DJD 3:30–31; on 9Q de Vaux,

The Functions Of The Caves And The Settlement Of Qumran

been considerably damaged by erosion, many others must already have been washed down into the Wadi by floods,24 and there were again some grottos that the archaeologists completely missed.25 2.1.3 Evaluating Roland de Vaux’ Interpretation The huge efforts undertaken by Roland de Vaux and many other colleagues between 1949 and 1956 resulted in a couple of hypotheses that have deeply influenced our way of interpreting the caves up to this very day. a) First of all, de Vaux very importantly noted the physical difference of the many caves he examined and saw the relationship between physical appearance and potential uses. “All chambers hollowed out in the marl terrace give the impression of having been designed as dwelling places and the objects found in them, vases for domestic use, date-stones, scraps of leather, rags, ropes, and a mat, prove that they were in fact inhabited.”26 To realize that human dwelling and object deposition may be two different issues wasa an important step forward. Regarding the caves in the limestone cliffs, de Vaux further differentiates between caves with wide mouths and high roofs that may occasionally have served a shelters for shepherds with their flocks (3Q and 11Q), low, narrow and uneven grottos that are in doubt of ever having been used as dwelling places (1Q, 2Q), and the majority of the limestone caves, namely crevices, depressions in the rock or fissures that all were too small to inhabit, but suitable for storing provisions and hiding valuables.27 b) Though realizing the caves’ different physical appearances de Vaux took a very important methodological step. Whenever de Vaux found pottery from the periods attested at the Qumran site in a cave, he saw this cave as part of the same large Essene network. This step had tremendous consequences: De Vaux was now able to combine both scroll and non-scroll caves, and “the” caves with the site. More than the scrolls, that suffered from the in DJD 3:31; on 10Q de Vaux, in DJD 3:31; on all of them Humbert and Chambon, Excavations of the Khirbet Qumran, 67 with lists of finds per cave. 24  See also Magen Broshi and Hanan Eshel, “How and Where Did the Qumranites Live?” in The Provo International Conference on the Dead Sea Scrolls: Technological Innovations, New Texts and Reformulated Issues (STDJ 30; ed. D. W. Parry and E. Ulrich; Leiden: Brill, 1998), 266–73 (270–71). 25  On Broshi and Eshel see 2.2.3. below. 26  See Roland de Vaux, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973), 55. 27  De Vaux, Archaeology, 55–56.

199

disadvantage of mostly having been found by non-specialists and never in the site itself, pottery proved to be the prime archaeological Leitfossil in Qumran cave research, because it established a firm link between caves and site.28 In a famous, often quoted passage of his “Archaeology and the Dead Sea Scrolls” de Vaux writes: “The use of the caves is not merely contemporaneous with Khirbet Qumran; it is organically connected with its history. The fact that it begins and ends at the same points in time as the communal occupation of the Khirbeh is significant in itself. But more than this, Khirbet Qumran lies at the centre of the area throughout which the caves are scattered, while the caves in the cliffs are grouped most closely in the vicinity of Khirbet Qumran, and none of the caves in the terrace lies more than a hundred metres from the ruins. The pottery is identical to that of the Khirbeh. The same pastes have been used and the same forms recur here, particularly in the case of the many cylindrical jars, which, apart from a single exception, are not / found outside the area of Qumran. The indications are, therefore, that all the ceramic material was manufactured in the same place, the workshop which was in the use at Khirbet Qumran during Periods Ib and II.”29 c) Finally, another problem is presented by the fact that de Vaux in the end did not clearly specify how far the area of “cave/site-sectarians” actually extended. After seeing Ras Feshkha, the limits of his 1952 surveys, as “geographical limits of the region’s occupation,” he was forced to revise that assumption after the discovery of Khirbet Mazin, but never fully integrated these finds into his overall concept.30 How “close” does a cave have to be to Qumran to qualify as “sectarian”? The notion of “vicinity”—frequently referred to above all in the case of the marl caves—remains vague and arbitrary. After all, Qumran was thought to be the isolated home of an apocalyptic community with very restricted contacts to the outside—if at all. Dennis Walker and in the meantime also Joan E. Taylor and Shimon Gibson have convincingly qualified this assumption by a careful study of tracks and paths leading to and from Khirbet Qumran.31 According to them, Qumran was “moderately connected locally” and 28   Like, e.g., White Crawford, “Qumran: Caves, Scrolls, and Buildings.” 29  De Vaux, Archaeology, 54–55. 30  See the important remarks by Dennis Walker, “Notes on Qumran Archaeology: The Geographical Context of the Caves and Tracks,” QC 3 (1993): 93–100 (96–99). 31  Walker, “Notes on Qumran Archaeology;” Joan E. Taylor and Shimon Gibson, “Qumran Connected: The Qumran Pass and Paths of the North-western Dead Sea,” in Qumran und die Archäologie: Texte und Kontexte (WUNT 278; ed. J. Frey,

200 integrated into a Late Iron Age network of settlements and paths: “These paths were used also in the Hasmonean and Early Roman period when some Iron Age sites were resettled.” As “traffic cannot have been very significant or else the paths would have been better developed,” Qumran certainly was not in the center of things. But the conclusion is inevitable that Qumran was accessible and also accessed by various tracks to the south, north and west, and by sea routes along the western shore of the Dead Sea and across it.32 Taylor’s and Gibson’s observations are not only important because they show that no caves nor the Qumran site were closed off from towns and cities outside the region (though some may not be visible without insiders’ knowledge), but also that caves and site were components of a complex regional framework. We will come back to this aspect below when discussing the results of the “Caves in the Northern Judaean Desert” expedition.33 The regional approach seems to have changed our interpetation of the pottery, too, at least until recently. At the beginning of Qumran research, de Vaux’ scenario had virtually become canonical.34 De Vaux saw Qumran as center of all the surrounding caves, may they contain scrolls or not. The sheer fact that the ceramics from the caves and the site apparently were contemporaneous and looked the same, plus the geographical closeness of the caves to the site were considered sufficient reason to assume that all pots were “made in Qumran”—and even by the same people. In the meantime, previous petrographic research by Marta Balla and Jan Gunneweg,35 as well as typological C. Claussen, and N. Kessler; Tübingen: Mohr Siebeck, 2011), 163–95. 32  Shimon Gibson and Joan E. Taylor, “Roads and Passes Round Qumran,” PEQ 140 (2008): 225–27; Taylor and Gibson, “Qumran Connected;” see the excellent recent book by Joan E. Taylor, The Essenes, the Scrolls, and the Dead Sea (Oxford: Oxford University Press, 2012) who, nevertheless, remains sceptical about a too farreaching integration of Qumran into the regional framework. 33  See below 2.3. 34  De Vaux, Archaeology, 55. White Crawford, “Qumran: Caves, Scrolls, and Buildings,” e.g., criticizes what she sees as “antideVaux” interpretations by a renewed combination of scrollcaves and text readings. 35  Jan Gunneweg and Marta Balla, “Neutron Activation Analysis: Scroll Jars and Common Ware,” in Humbert and Gunneweg, Khirbet Qumrân et ‘Aïn Feskha, 4–46; Jan Gunneweg, “EngediQumran-Schoeyen and Allegro Jars: Four Similar Looking Dead Sea Jars that All Have a Different Provenience: Appearances Provide Sometimes the Wrong Information,” in Outdoor Qumran and the Dead Sea: Its impact on the Indoor Bio- and Material Cultures at Qumran and the Judean Desert Manuscripts;

Zangenberg

studies of ceramics from Masada, Qumran and the winter palaces at Jericho by Rachel Bar Nathan36 have substantially corrected this scenario. They left no doubt that Qumran both supplied this market with common ware, as well as it was a recipient of products from other factories in the same regional market. Since the number of samples is still limited and the statistical basis is quite narrow, it is difficult to make any statements on quantitative percentages of “outside” relative to “onsite” pottery at Qumran, about possible specialization of production sites on particular types (such as, e.g., Qumran on “scroll jars”) or how much pottery reached the northwestern Dead Sea region from further away. What is clear, however, is that we are dealing with a region that was supplied from different angles. Not even the famous “scrolls jars” are exclusively made in Qumran as Jan Gunneweg was able to find out on the basis of NAA.37 But now, as Jodi Magness’ conference paper shows, things seem much less clear anymore.38 We will see how the discussion develops, as soon as more and broader sampling and analysis of pottery from the Qumran site and caves are carried out. Ultimately, it will need to be reexamined if and how far “Qumran pottery” in reality was “regional pottery,” used and partly produced in Qumran but not exclusively connected to it. The consequences of how this question is answered will be significant: If ceramics cannot help us anymore to identify “Qumranites” over against others, then potential “Qumranites” would be everywhere—or there are no proper “Qumranites” left at all, only inhabitants of the same region present and active in various contexts including site and caves. In that case, not the caves were “organically connected” to the “history” Proceedings of the Joint Hebrew University and COST Action D-42 Cultural Heritage Workshop held at the Hebrew University of Jerusalem in May 25–26, 2010 (ed. J. Gunneweg and C. Greenblatt; Jerusalem: Hebrew University, 2012) at http://micro5.mscc.huji .ac.il/~msjan/aeqsj.pdf. 36  Rachel Bar Nathan, Hasmonean and Herodian Palaces at Jericho: Final Reports of the 1973–1987 Excavations. Vol. 3: The Pottery (Jerusalem: Israel Exploration Society/Hebrew University, 2002), 23–27, 202–3; idem, “Qumran and the Hasmonean and Herodian Winter Palaces of Jericho: The Implication of the Pottery Finds on the Interpretation of the Settlement at Qumran,” in Qumran, the Site of the Dead Sea Scrolls: Archaeological Interpretations and Debates; Proceedings of a Conference Held at Brown University, November 17–19, 2002 (STDJ 57; ed. K. Galor, J.-B. Humbert, and J. Zangenberg; Leiden: Brill, 2006), 264–77 (esp. 275–77); Rachel Bar Nathan, Masada VII: The Yigael Yadin Excavations 1963–1965; Final Reports; The Pottery of Masada (Jerusalem: Israel Exploration Society/Hebrew University, 2006), 71–72, 375–77. 37  Gunneweg, “Engedi-Qumran-Schoeyen.” 38  See Jodi Magness’ contribution in this volume.

The Functions Of The Caves And The Settlement Of Qumran

of Qumran, as de Vaux thought, but instead caves and site were both “organically connected” to the region and its history and economy. If Magness’ analysis is correct, however, and the Qumran pottery was in fact a strictly local product, this picture would be quite different. In any case, the potential existence of several local production sites (now disputed), and the exchange of goods document Qumran’s interconnectedness with the region.39 Be that as it may, much more attention should be directed to understanding Qumran’s economic infrastructure, and the role the caves played in it. Nevertheless, the value of these efforts for our discussion on the function of the caves remains limited: none of the caves was properly excavated or sufficiently documented. We do have a few plans, but by far not from all “important” caves, and hardly any stratigraphic section drawings. De Vaux’ verbal descriptions of the 40 grottos with traces of habitation in the authoritative report in DJD 3 are generally quite brief. The same also counts for the description of finds, above all the pottery. When reading de Vaux’ ceramic report one quickly realizes that—very much in line with contemporary practice—only complete vessels were catalogued, fragments after a short identification (“genre Qumrân”) frequently discarded. How much was missed or remained unrecorded? Of course, the main types are carefully documented by drawings and plates and described on the basis of the then available evidence.40 But is this report complete? In addition to that, de Vaux does provide information about where vessels were found only for selected caves and only in summary form, making it tremendously difficult to recognize assemblages of different types or their relationship to the surrounding space and other non-ceramic objects. Such information, however, would be necessary to reconstruct activities in which the pottery was used (e.g., cooking, storage), and on that basis enable scholars to draw conclusions regarding possible functions of a given cave.

39  Jan Gunneweg and Marta Balla, “Was the Qumran Settlement a Mere Pottery Production Center? What Instrumental Neutron Activation Revealed,” in Holistic Qumran: Trans-Disciplinary Research of Qumran and the Dead Sea Scrolls; Proceedings of the NIAS-Lorentz Center Qumran Workshop 21–25 April 2008 (STDJ 87; ed. J. Gunneweg, A. Adriaens and J. Dik; Leiden: Brill, 2010), 39–61 state that no pottery with the “local fingerprint of Qumran” was exported to any site outside the settlement. Pottery with this fingerprint dates from the 40s BCE to the end of the site. But there is material in Qumran that came from outside. 40  See de Vaux’s report on “Céramique du genre Qumrân” in DJD 3:13–15 and Figs. 2–5.

201

2.1.4 Addendum: The “Christmas Cave” Only 7 km south of Qumran in the mouth of Na�al Kidron, John Allegro, allegedly on Christmas Day 1960, found a large cave with abundant traces of human habitation from between the Chalcolithic period to the Bar Kokhba Revolt, including much material from the end of the Second Temple Period. The cave has been forgotten and the finds that were stored in the Rockefeller Museum were overlooked or erroneously associated with Qumran. Only in 2003, Mireille Bélis recognized the importance of this cave when she was studying textile fragments labelled “QCC” (“Qumran Christmas Cave”). In 2007 the cave was relocated by Roi Porat, Hanan Eshel and Amos Frumkin, excavated and documented. Reports were published in a 2010 Hebrew University conference volume.41 On the basis of ceramic samples from the Christmas Cave analyzed by Marta Balla with INAA it appears that the inhabitants of Christmas Cave had no connection with the contemporaneous Qumranites.42 This picture was confirmed by the analysis of the textiles.43 2.2 Explorations in the 1980s and 1990s De Vaux’s conviction that both the caves and the site were inhabited by the same people inspired him to ponder about the function of the caves for running the site. As the site did not show much evidence of living quarters among all the “communal” and industrial elements, it seemed reasonable to assume that at least some of the caves served as dwellings for the Qumranites. Consequently, de Vaux concluded that some of the “estimated 200 members of the ‘monastic’ community of Qumran had lived in caves.”44 But what does “living” actually mean in terms of archaeology? Was it permanent or temporary, and how can we distinguish that in the archaeological record? 41  See the reports in Session I of Gunneweg and Greenblatt, Outdoor Qumran at http://micro5.mscc.huji.ac.il/~msjan/qum ranproceedings2010.html. 42  Marta Balla, “The Search for Interregional Contacts between Different Groups of Jews Living in the Dead Sea Area during the Early Roman Period on the Basis of Pottery,” in Gunneweg and Greenblatt, Outdoor Qumran at http://micro5.mscc.huji .ac.il/~msjan/balla.pdf. 43  Orit Shamir and Na⁠ʾama Sukenik, “Qumran Textiles and the Clothing of the Inhabitants of Qumran,” in Gunneweg and Greenblatt, Outdoor Qumran at http://micro5.mscc.huji.ac.il/ ~msjan/orit.pdf. While clothes from Qumran were made of undied linen without decoration, textiles from the Christmas Cave were coloured and made of different fabrics like others from refugee caves (might it be possible to date them to the time of Bar Kokhba instead of the end of the Second Temple period?). 44  De Vaux, Archaeology, 86; Cansdale, Qumran and the Essenes, 171.

202 A good example how difficult that distinction is are the five wooden poles from GQ17, apparently remains of an awning—while Roland de Vaux sees them as evidence for prolonged habitation, Joseph Patrich interprets them as temporary.45 And are we always speaking of the same groups being active at a particular given time? Lena Cansdale, e.g., claims that pottery from the last Qumran period “may have been used by the persons who investigated and prepared the caves and later hid the scrolls in them. (. . .) Caves were the obvious hiding places for precious possessions, and as Qumran could supply suitable cylindrical jars to protest the precious scrolls, the caves nearby were selected as repositories.”46 But is this really so obvious? If the pottery should need to be dropped as characteristic of an isolated Qumran, the fact that “the same” pottery was found in caves and on the site, would not indicate anymore that only Qumranites were responsible for their deposition. Has anybody ever checked that for the ceramics from the caves? The 1980s and 1990s brought about a renewed interest in the archaeology of Khirbet Qumran and its surroundings and the function of the caves in connection to the site. Research by Joseph Patrich on the one hand and Magen Broshi and Hanan Eshel on the other, but also by Robert Eisenman and Dennis Walker mark important steps in that direction. Though at first sight dealing with de Vaux’ old question where the Qumranites lived, Patrich and Broshi/Eshel—perhaps unwillingly—demonstrated the deadlocks into which one is lead by always using the Qumran site as hermeneutical key for the caves. Eisenman and Walker’s expedition, however, had a much broader scope. 2.2.1 Joseph Patrich In 1983 Joseph Patrich proposed “a systematic survey and excavations of caves in the Judean desert.”47 In the winter of 1984 and 1985 Patrich reexamined the caves explored by de Vaux in 1952 in order to decide, “which caves could be still further explored by means of archaeological excavations.” It proved difficult to relocate many of de Vaux’ caves. The apparent lack of a network of paths led Patrich 45  Baillet and Milik, in DJD 3: Pl. VII.3; Joseph Patrich, “Khirbet Qumran in Light of New Archaeological Explorations in the Qumran Caves,” in Methods of Investigation of the Dead Sea Scrolls and the Khirbet Qumran Site: Present Realities and Future Prospects (ANYAS 722; ed. M. O. Wise et al.; New York: New York Academy of Sciences, 1994), 73–96 (93). 46  Cansdale, Qumran and the Essenes, 173. 47  Patrich, “Khirbet Qumran,” 74–75.

Zangenberg

“to the inescapable conclusion that all the members of the community lived inside Khirbet Qumran.”48 Apart from reexcavating 3Q and 11Q, Patrich also discovered a number of hitherto unknown caves: a)

Cave 24, a “large and habitable” cave 50 m north of 11Q with clear evidence of several phases (Neolithic, Chalcolithic and Roman). The cave was carefully excavated and described, but without plan. Several fireplaces in various spots were located inside the cave, plus a “goat dung layer”—the first time such data were ever recorded! Finds comprised “cylindrical jars” associated with bag shaped ones, juglets, cooling pots and a mutilated lamp. “Other finds include date pits, pieces of ropes, basketry and cloth, a bronze needle with a square section, and a bronze ear- or nose ring.”49 All these finds date to the late Herodian and early post-70 period and may testify the presence of shepherds (goat dung!) and later refugees (defaced lamp!)—if these two groups were not even identical.50 b) In 1988 and 1991 Patrich excavated Cave 13 (“Balsam Oil Cave,” perhaps Cave 2 of de Vaux’ 1952 survey?).51 This cave was apparently occupied by refugees around 68 CE, “bringing with them jars with dates for food, and hiding behind the huge rocks” (pottery mostly Qumran Ib and little II).52 According to Patrich, the users were no sectarians and the cave did not function as permanent dwelling. c) On the terrace in front of the large cave FQ37 (1 km south of Qumran, identical to Cave 37 of de Vaux’ 1952 expedition?),53 Patrich found in 1989 a nail, two arrowheads, a fragment of a wooden plate, a Roman bronze coin with defaced portrait on the obverse, fragments of four glass plates, fragments of two stone vessels and many potsherds from various types 48  Patrich, “Khirbet Qumran,” 75, already questioned by Walker, “Notes on Qumran Archaeology” on the basis of intensive surveys; fully corrected by, e.g., Taylor and Gibson, “Qumran Connected.” 49  Patrich, “Khirbet Qumran,” 90. 50  Patrich, “Khirbet Qumran,” 90 speaks of “inhabitants that found refuge at the site at ca. 70 CE or later.” 51  Joseph Patrich and Benni Arubas, “A Juglet Containing Balsam Oil (?) from a Cave near Qumran,” IEJ 39 (1989): 43–59 on first season 1988; Patrich, “Khirbet Qumran,” 91 on the first and a second 1991 season. 52  Unfortunately, it remains unclear what type of jars Patrich found (bag-shaped or cylindrical) apart from the bag shaped one on p. 87 Fig. 13. 53  On Cave 37 see de Vaux’ report in DJD 3:12.

The Functions Of The Caves And The Settlement Of Qumran

of jars, cooking pots, juglets and bowls associated with a mud floor mixed with palm fronds.54 While the arrowheads in Patrich’s opinion speak in favour of “a place of refuge in emergency,” the more permanent floor “points to a dwelling place of a more prolonged nature in time of peace” used by a Qumranite living in ascetic seclusion.55 Patrich’s suggestion is, though intriguing, above all owed to the intention to search for extra-mural living quarters of Qumranites who were seen as Essenes. Patrich in any case acknowledges that the evidence from FQ37 is difficult to reconcile with alleged “Qumranite living quarters,” but—being under the influence of the monastic Qumran paradigm—still tries to integrate FQ37 into the “Qumran-Essene-model.” d) In addition to these caves Patrich’s team examined grottoes of a very different kind: places accessible only by means of rope and ladders, or enlarged natural caves. Comparable to hideouts in the Galilee, these caves apparently served as refuge during the 1st and 2nd revolt. Patrich calls them “hideouts prepared in advance by villagers living on the fringes of the desert, for times of emergency.”56 e) Finally, Patrich reexamined 3Q and 11Q in 1986 and 1988 respectively. It turned out that the ceiling of 3Q had collapsed “many years before the jars with the scrolls were hidden in the cave. During periods Ib and II at Qumran the shape of this cave was very much like that at the start of our excavation.”57 For Patrich, 3Q was rather a genizah, “not a good place to choose, if the idea was to find a hiding place for an active library.” In addition to that, Patrich provided the first accurate plan of 3Q and 11Q. 11Q had already entirely been cleared by de Vaux, no new finds were reported.58 Summarizing his results, Patrich concludes that “the caves in the limestone cliffs, as a whole, did not serve as habitations for the members of the Dead Sea Sect, but rather as stores and hiding places,” and “in the entire area around Qumran there are no remains of supposed encampments of tents and huts.” Consequently, “(t)he dwelling quarters should be sought inside the wall of Khirbet Qumran, 54  Patrich, “Khirbet Qumran,” 91–93. 55  Patrich, “Khirbet Qumran,” 92–93. 56  Patrich, “Khirbet Qumran,” 73–74. He also mentions Byzantine cave hermitages (“Khirbet Qumran,” 74). 57  Patrich, “Khirbet Qumran,” 77. 58  Patrich, “Khirbet Qumran,” 76, 80 Fig. 2 on 3Q, 77,90 and 82 Fig. 5 on 11Q.

203

mainly on the upper story.”59 Be that as it may, Patrich’s observations indeed mark a major step forward, since they not only record important detail, but also introduce a notion of diachrony and differentiation of local uses in one and the same cave. Though still operating within the traditional Qumran-Essene paradigm and focussed on the narrow question of where the Qumranites lived, Patrich widens the perspective beyond de Vaux by differentiating between various groups of cave dwellers. But without plans and a more detailed documentation of layers and locations it remains very difficult to decide what finds should be assigned to the “more permanent” and which to the “emergency occupation.” 2.2.2

Dennis Walker and Robert Eisenman 1988–1992 A much larger number of caves in the vicinity of Qumran was searched by three expeditions headed by Robert Eisenman (California State University, Long Beach) in the years 1988–1992.60 Eisenman motivated these activities by three observations: First, caves continued to be found even in de Vaux’ old survey area, “including 8 more scrollbearing caves near Qumran;”61 second, though Pesach Bar Adon excavated sites in the area south of de Vaux’s survey activity (e.g. En el-Ghuweir), no attempts were made to examine cliffs nor caves; and third, the apparent lack of habitational space in Qumran and En Feshkha raised the question of where the Qumranites lived. The largest expedition was launched in January 1990 and covered a 24 km range between 5 km north of Qumran to Wadi Murabba‘at 20 km to the south of it. 137 “proper” caves were noted,62 85 of which showed traces of human habitation from the Chalcolithic period to the modern era. 59   Patrich, “Khirbet Qumran,” 93; Cansdale, Qumran and the Essenes, 170–73 (172). See already de Vaux, Archaeology, 56–57: “It is justifiable to conclude that the buildings were designed to house the central services and to provide lodgings for the administrators and guardians of the group, but that the majority of its members had their living quarters outside.” 60  I thank my colleague Zdzislaw J. Kapera for directing my attention to this expedition and relevant literature, see e.g. Walker and Eisenman, “The 1990 Survey;” Walker, “Notes on Qumran Archaeology.” 61  Walker and Eisenman, “The 1990 Survey,” 46. 62  Walker and Eisenman, “The 1990 Survey,” 47: “By ‘proper’ cave, we mean a cave of sufficient size to have been used by humans” and does not include the ca. 200 deep holes or crevices (“blind caves”) that the expedition also located. In “The 1988–1992 California State University Dead Sea Walking Cave Survey and Radar Groundscan of the Qumran Cliffs,” QC 9 (2000): 123–30, Robert Eisenman speaks of “some 485 caves and/or depressions” in total (125).

204 South of Murabba‘at the density of caves drops, probably due to different geological conditions. Unlike de Vaux’s surveys that were basically only dealing with the caves themselves and their contents, Eisenman had a much wider perspective: “But our search through the hills and cliffs yielded far more than caves. Painstakingly built aquaeducts high in almost inaccessible areas, a dressed stone cave entrance, several ancient gravesites, traces of an ancient road, the presence of datable potsherds, and more enabled us to piece together a picture of an amazing area that is yet to be fully understood.”63 While the “number of caves per kilometer is highest adjacent to and just north of Qumran,” Eisenman and Walker say, “Second Temple potsherds were fairly evenly distributed the length of the survey.”64 Despite the “high density areas” with a higher percentage of “type 1 caves,” there were many quite isolated caves, too.65 The ceramic evidence from the caves fits well with “the archaeological evidence from the ruins in the area. Together they suggest a bustling and widespread population during the turbulent Second Temple period,” closely connecting sites as En el-Ghuweir, Qumran and En Feshkha.66 In a way, Eisenman’s surveys, perhaps for the first time, demonstrated and documented that Qumran together with the caves were part of a large and complex environment. Unfortunately, the results of these activities were never properly published, so the dates of the finds remain unknown to us, as well as the contents of each cave. Today, Eisenman’s surveys are largely ignored in research. 2.2.3 Magen Broshi and Hanan Eshel Triggered by the debate about alternative interpretations of the Qumran settlement in the late 1980s and 1990s and in an attempt to defend the traditional theory, Magen Broshi and Hanan Eshel—convinced by the general correctness of the “Qumran-Essene-Model”—reviewed de Vaux’ publications and examined the marl terrace and the limestone cliffs in the vicinity of Qumran to get a clearer picture of the character of habitation and the location of the living quarters of the Qumranite.67 In 1995 and 1996, so only shortly after Patrich’s explorations, Broshi and Eshel “examined seven caves in the marl plateau north west of the community center,” in two of which the floor could be

63  Walker and Eisenman, “The 1990 Survey,” 47. 64  Walker and Eisenman, “The 1990 Survey,” 48. 65  Walker and Eisenman, “The 1990 Survey,” 48. 66  Walker and Eisenman, “The 1990 Survey,” 49. 67  Broshi and Eshel, “How and Where;” Magen Broshi and Hanan Eshel, “Residential Caves at Qumran,” DSD 6 (1999): 328–48.

Zangenberg

reached (C and F).68 None of them had been explored by de Vaux, but they were severely looted and disturbed by bedouin in the meantime. a) According to Broshi and Eshel the marl terrace caves were ideal for living, most probably “started as natural cavities” and were later artificially enlarged,69 8Q likely even was a workshop.70 According to them, most Qumranites lived as “troglodytes” in marl terrace, not in limestone cliffs that were beyond “sabbath limit.”71 Jodi Magness supports that: “(F)ew if any of the members of the community lived (that is, slept) inside the settlement,” domestic pottery in caves indicates that “some of the caves were inhabited, because no one carries cooking pots, dining dishes, and storage jars to a cave unless they are living there! Other members of the community must have lived in tents and huts around the site. Some of this habitation could have been seasonal.”72 But here, too, opinion is divided, and Yizhar Hirschfeld rejects the notion that marl caves were pleasant living quarters and rather sees them as “a last resort.”73 The conflicting results and conclusions of both authors lead into a deadlock. Though supplying valuable new information on old and new caves, they come up with conflicting answers for their common question of where the Qumranites lived. How could we distinguish a “permanent” from a “prolonged” or a “temporary” occupation? Perhaps there is no “either / or.” Patrich is certainly right when he writes: “A habitat of prolonged occupation should leave evident architectural remains. It should have been strong enough to withstand wind storms and winter rains, scant as they are in the Qumran region. One would also expect to find indications of stone removal, water supply installations, and a network of trails among the dwellings themselves and between them and Qumran. No such remains exist in the entire area.”74 But now we know

68  Broshi and Eshel, “How and Where,” 271. 69  Broshi and Eshel, “How and Where,” 271. 70  Broshi and Eshel, “How and Where,” 271. 71  See especially in Broshi and Eshel, “Residential Caves,” but cf. the critique by Joseph Patrich, “Did Extra-Mural Dwelling Quarters Exist at Qumran?” in The Dead Sea Scrolls Fifty Years after their Discovery: Proceedings of the Jerusalem Congress, July 20–25, 1997 (ed. L. H. Schiffman, E. Tov, and J. C. VanderKam; Jerusalem: Israel Exploration Society/Shrine of the Book, 2000), 720–27; Hirschfeld, Qumran in Context, 43. 72  Jodi Magness, The Archaeology of Qumran and the Dead Sea Scrolls (Grand Rapids, MI: Eerdmans, 2002), 70. 73  Hirschfeld, Qumran in Context, 43. 74  Patrich, “Extra-Mural Dwelling Quarters,” 721–22.

The Functions Of The Caves And The Settlement Of Qumran

that there was a network of paths,75 and people may have gone to the caves with different aims. The only type of settlement that really qualifies for Patrich’s criteria about permanent are Qumran itself, En Feshkha, et-Turabe en Mazin. It is easy to imagine that these settlements were inhabited by a number of permanent occupants, but these people were not the only persons who lived in and from them. For a couple of caves, our data in fact do seem to suggest that other people lived in them permanently or at least for longer stays: “In addition to remains of storage or so-called scroll jars, particularly from Caves 1 and 3 (but also from Cave 29 where no manuscripts were found), excavators found the remains of household pottery such as jugs, plates, bowls and even a few cooking pots in various quantities, except for Cave 5 where no pottery or objects were found. (. . .) The pottery in Cave 7, the date pits and remains of other fruit in Caves 8–10, four lids, one small plate, one fragmentary lamp, fragments of four jars, one bowl, a piece of a sandal sole end three phylactery cases in Cave 8, a few pottery shards and three pieces of rope in Cave 9, and a partially preserved reed mat, a few pottery shards and a lamp with a paired nozzle on the floor of Cave 10, may indicate that they were used as dwelling places.”76 One could also add Cave 12 and FQ37 here. Though it is difficult to distinguish repeated stays (perhaps related to seasons) from permanent occupation, it is at least evident that caves with indications of longer habitation are not confined to the “scroll caves.” But this is not the entire picture. The sheer variety of caves and their users and uses is certainly too diverse to expect a general solution. Apart from more permanent inhabitants, nomadic or semi-nomadic pastoralists were also presentin the semi-arid region around Qumran: “(S)cattered sherds found in a desert area with sparse architectural reamains or none at all will generally be interpreted as indicating the emphemeral stay of a shepherd or a passer-by.”77 These sherpherds may have pitched up a tent or lit a fire only for short stays in front of a cave, but still may have left supplies hidden inside them. Here, too, the distinction between “permanent” and “short-stay” is not useful.

75  Taylor and Gibson, “Qumran Connected.” 76  Popović, “Qumran as Scroll Storehouse,” 559–60, see also 561. “In addition, some of the natural caves in the limestone cliffs that did not yield manuscript finds were deemed to have been ­habitable during the period that the Qumran settlement was in use (. . .); see, e.g., Caves 10, 31, 32, 37, 39 and 40 (de Vaux, in DJD 3:8, 11, 12)” Popović, “Qumran as Scroll Storehouse,” 560 n. 23. 77  Patrich, “Extra-Mural Dwelling Quarters,” 724.

205

In all this economic network of caves, paths, pastures and plantations the Qumran site very likely played a crucial role with its facilities (potter’s kiln, storage, workshops, cemetery) and attracted very different groups.78 Many more people were living around Qumran than only “Qumran inhabitants.” Caves in the Northern Judaean Desert (CNJD) Explorations (“Operation Scroll”) A final attempt to systematically survey the eastern edge of the Judean Desert for caves was made by the “Caves in the Northern Judaean Desert” project during two months following November 15, 1993.79 Ca. 165 of the 495 cavesites surveyed in 15 zones stretching from Niran north of Jericho along the Dead Sea to just south of Mitzpe Shalem showed human presence (i.e. a considerably higher percentage than recorded by de Vaux), even remains of various types of constructions were found in 65 caves. The expedition followed several goals. Apart from inventorizing old ones and finding and exploring as many new caves as possible with or without written material, the p ­ roject— confirming to a more recent awareness about questions of regionality—was also interested in the infrastructure between the caves. Consequently, paths and tracks were recorded, tying these caves into the regional landscape instead of only seeing them in relation to their alleged “center” Qumran. The CNJD surveys are the most comprehensive and systematic explorations of caves along the western shores of the Dead Sea so far, have accumulated many new data and opened many new insights. Not only did CNJD discover a number of scrolls and blank scroll material in caves of area V and X in the lime stone cliffs, they also found new caves in the marl terrace south of Qumran (region XIV). Relevant for our purpose are especially regions X, XI and 2.3

78  See Jürgen K. Zangenberg, “Opening Up Our View: Khirbet Qumran in a Regional Perspective,” in Religion and Society in Roman Palestine: Old Questions, New Approaches (ed. D. R. Edwards; London: Routledge, 2004), 170–87; Hirschfeld, Qumran in Context, 45; David Stacey, “Seasonal Industries at Qumran,” BAIAS 26 (2008): 7–29. 79  Michel Itah, Yoni Kam, and Ronny Ben Haim, “Region X: Survey and Excavations of Caves along the Fault Escarpment South of Almog Junction and West of Qalya,” ʿAtiqot 41/1 (2002): 175–86 [Hebrew] and ʿAtiqot 41/2 (2002):169–76; Yuval Baruch, Gabriel Mazor, and Deborah Sandhaus, “Region XI: Survey and Excavations of Caves along the Fault Escarpment above Horbat Qumran,” ʿAtiqot 41/1 (2002): 189–98 [Hebrew] and ʿAtiqot 41/2 (2002): 177–83; Nir Tal and Gershon Oron, “Region XIII: Survey and Excavations of Caves along the Fault Escarpment South of Qumran,” ʿAtiqot 41/2 (2002): 187–93.

206

Zangenberg

XII to the north, west and south of Qumran. Several of the region XI caves closest to Qumran clearly showed evidence of habitation dating to the Early Roman period, but there was no real difference in density of ER cave occupation compared to other regions. CNJD again confirmed de Vaux’ statements about the diversity of cave forms and locations. The most important result, however, is to realize that so many more caves had existed around Qumran than de Vaux and Patrich and Broshi/Eshel had already found. But despite these important results, CNDJ also suffers from a number of shortcomings and problems. The published reports are often very summary. More often than not, e.g., they mention “pottery vessels” without broader specification as to the types or exact location which makes it very difficult to reconstruct the character of habitation? Often material was found disturbed by previous excavations (both by de Vaux or bedouin) or out of context dumped in front of caves. Excavations were often small, no find assemblages are really shown in situ, very few section drawings are provided, so that stratigraphic relations and the character of habitation remains unclear: temporary (refugees), seasonal (herdsmen), permanent, with animals (“fence” built of field stones in XI/6, XI/10— but from what period?) or without? Can we distinguish different groups of refugees from 1st and 2nd century CE? Except textiles and basketry,80 little information is provided about other finds, which is unfortunate, since, e.g., animal bones could have provided indications for food consumption in the caves. 3

Discussion of the Evidence

Let me summarize. First of all, a lot depends on from where people start asking their questions from and what they want to see or intend to show. The narrower such an initial question is, the more likely the answer will correspond to the question and invite vicious circles lurk around the methodological corner. The relationship between caves and Qumran goes beyond only specific events in the history of the region. The deposition of manuscript material and how this all might add up to a “library” is not the proper way to understand “the caves” as an archaeological phenomenon, it is only one event in the history of some of them. Archaeologically, therefore, it is not justified to put 80  Orit Shamir and Alisa Baginski, “The Later Textiles, Basketry and Cordage from Caves in the Judaean Desert,” ʿAtiqot 41/2 (2002): 241–55; Tamar Schick, “The Early Basketry and Textiles from Caves in the Northern Judaean Desert,” ʿAtiqot 41/2 (2002): 223–39.

the emphasis on a pre-selected number of caves. We need to refocus, move away from individual caves, and look at the region within which all these caves functioned with or without Qumran. A comprehensive view on the region can help us understand topographical relationships between various features, and a careful analysis of the shape and content of individual caves will show us their various uses through time, all adding up to a complex reconstruction of the physical and historical environment. Archaeologically, it is all about how and with what to contextualize both caves and the Qumran site. What we need is a systematic inventorization (description, mapping, interpretation) of all explored caves and a diachronical analysis of their use as part of a habitation history of the region. The Lugano-conference has already made important steps towards that goal. Nevertheless, I need to add two caveats for the current discussion: 3.1 Postdepositional Processes It is very difficult to reconstruct a situation in the past (e.g., the deposition of scrolls) from just looking at its condition at the moment of discovery. The reason for that are “post-depositional processes.” Caves continuously change over time, and what is in them changes with them. So, when archaeologists enter a cave they encounter remains originating from different periods and a situation that has resulted from the combined effect of very different causes. Processes that have developed over time are found congelated into a seemingly static situation having lost its diachronic component. Natural (water, climate, erosion, collapse, sedimentation), animal-made (worms, bat dung) and human (looting, destruction, moving inside cave, taking away from cave) causes have determined how a cave looks like when archaeologists first see it. The extent to which a cave was affected by these factors varies considerably and often largely remains unknown to us. Though natural, therefore, post-depositional processes are not predictable and do not result in the preservation of a representative segment of subsequent deposits, but a random one. Methodologically, therefore, these processes are very difficult to roll back to a particular moment of time, to reconstruct it and speculate about how and why this particular situation came about. Despite all popular confidence in the conserving power of caves, it therefore seems quite naive to me to assume that a cave could encapsulate the past unchanged and that the situation in a cave would at times of its discovery just be the same as it had been left in the past.81 Each cave is—to use a methaphor—a mul81  Hartmut Stegemann, Die Essener, Qumran, Johannes der Täufer und Jesus: Ein Sachbuch (4th ed.; Freiburg/Basel/Wien: Herder, 1994), 98—just to name an example—is overstating

The Functions Of The Caves And The Settlement Of Qumran

tilayered palimpsest written by many individual hands, of which a part unknown to us has been lost. The fact that “we do not know how much we do not know” is the most difficult part of the methodological challenge.82 3.2 The Lack of (Published) Documentation A second caveat needs to be added. Cave archaeology is not only technically very difficult, it also demands a lot from those who document archaeological activities. Only careful stratigraphic excavation coupled with meticulous documentation will help us reconstruct the history of transformations that a cave had to undergo. As we all know, archaeological material that is not recorded and published is irrelevant. Unfortunately, not many of the caves from the surroundings of Qumran meet such high standards. The reasons are evident and can, of course, not be blamed on the individual explorers. But the consequence of the fact should the evidence: “Als die Beduinen—zu Beginn des Jahres 1947— die Höhle 1Q entdeckten, müssen sie darin alles noch ganz unberührt genauso vorgefunden haben, wie die Qumran-Siedler es 68 n. Chr. hinterlassen hatten. Nur war mehr als die Hälfte der versteckten Schriftrollen im Laufe der Zeit schon weitgehend verrottet.” That is impossible. 82  See Jürgen K. Zangenberg, “Zwischen Zufall und Einzigartigkeit: Bemerkungen zur jüngsten Diskussion über die Funktion von Khirbet Qumran und die Rolle einiger ausgewählter archäologischer Befunde,” in Frey, Claussen, and Kessler, Qumran und die Archäologie, 121–46 (126–31). The obvious fact that there is a discrepancy of unknown size between what we have preserved and what might have been deposited in antiquity is only slowly making an impact on Qumran cave research. Sidnie White Crawford, e.g., does not mention the problem at all in her recent study (“Qumran, Caves, Scrolls, and Buildings”). Joan E. Taylor, “Buried Manuscripts and Empty Tombs: The Qumran Genizah Theory Revisited,” in ‘Go Out and Study the Land’ ( Judges 18:2): Archaeological, Historical and Textual Studies in Honor of Hanan Eshel (JSJSup 148; ed. A. M. Maeir, J. Magness, and L. H. Schiffman; Leiden: Brill, 2011), 269–315 (esp. 298–305) acknowledges that “the possible extent of the original scrolls corpus buried in caves was huge, since caves may have been substantially cleared in antiquity (. . .) as well as been damaged so that scrolls were consumed by rats” (305–6). That is certainly true, but Taylor still seems to take for granted that the present evidence offers a sufficient basis to characterize the “collection” as “Essene” and hypothesize about its deposition—and the character of the scroll caves. Popović explicitly accepts Taylor’s statement about the originally much higher number of manuscripts, but announces a study on the “volume of space needed for storing the scrolls” (“Qumran as Scroll Storehouse,” 587). To accomplish that, one would not only “need to consider the number of scrolls, the different sizes of the scrolls and the presumed diameter when rolled up” (“Qumran as Scroll Storehouse,” 587); one would above all also need information about everything lost to us to be able to add this to what we think we still have available.

207

not be ignored either. Even if a few caves were recently excavated with some proper stratigraphic control, most of them were subjected to not more than brief soundings, i.e. limited cleaning activities mostly to recover scrolls or other objects. Many other caves were only superficially examined, because they either were difficult to access due to fallen rocks or other reasons, or because working conditions in dust, heat or lack of fresh air became unbearable,83 or because they had been looted previously so that an excavation did not seem feasible anymore. Often enough, archaeologists did not even properly record the situation encountered when the cave was discovered, not to speak of data and observations relevant for reconstructing the situation 2000 years ago. So often there is no reliable description, which could be taken as starting point to unravel the history of how a cave and its contents might have developed over time and have produced the situation that modern explorers encountered. Furthermore, it is not enough to state that pottery ­contemporaneous to Khirbet Qumran Ib or II was found in a given cave. Often, only “cylindrical jars” are explicitly mentioned, largely because of their alleged relationship with scrolls, the rest of the pottery, however, is rarely specified. But in order to understand the character and chronology of habitation in the caves, we must know, e.g., where this pottery was located in the cave, with what other finds it was associated (bones, tools, a fireplace?) or in what relationship it was with material above and below it. Very rarely do we get this sort of information, however, but without it we only pretend to know more than we actually do. 3.3 Consequences Saying all this should not be taken as cheap criticism against the colleagues who have conducted these archaeological activities under often very hard conditions. It is meant to warn us against drawing over-confident conclusions from material that cannot support them. It should be immediately evident, however, that such insecurity regarding the circumstances of deposition and post-­depositional processes has serious consequences not only for our interpretation of the fragmented scrolls,84 but also for determining the functions that the caves might have had in a region that was more than just the “­surroundings of Qumran.” A few examples might illustrate this: a)

There is still no clear indication of which cave was inhabitable and which not, and what sort of habitation we can assume in the first category. A consensus

83  See e.g. Patrich, “Khirbet Qumran,” 76. 84  See Zangenberg, “Zwischen Zufall und Einzigartigkeit,” 125–31.

208

Zangenberg

exists perhaps that limestone caves like 1Q, 2Q, 3Q, 6Q and 11Q were less suitable for long term occupation while that was the main purpose of the caves in the marl terrace. “Die Höhlen im Kalksteinkliff sind als Wohnhöhlen auf lange Frist ungeeignet, da der Boden zum Schlafen zu geneigt ist, sie zumeist nicht ausreichend belüftet sind und auch die Wasser- und Nahrungsmittelzufuhr schwierig ist.”85 A closer look, however, reveals that this is certainly not a general rule, but very much dependent upon the conditions of each individual cave. We also should avoid too narrow notions of “permanent” ­versus “temporary” habitation. Both may be relative. A good example is 11Q which is rather unique due to its many diverse finds: “The finds included a small pick, a chisel, and a knife, all of iron. The amount of pottery was small, but typical of Qumran, including a jar and two lids; there were in addition pieces of linen, basketry, ropes and several inscribed parchments.”86 Of all the metal tools, however, only the notorious hatchet from 11Q has found any attention as alleged evidence of Essene toilet practices.87 The question is, if this piece should be separately interpreted apart from its context. The fact that the hatchet was found together with other tools rather suggests that it belonged to a tool kit or a depot concealed by local inhabitants whose Essene identity is far from proven. Sometimes, an entrance of a cave was used for habitation like in an abri. In that respect, the evidence from the Qumran region caves should be compared to what was found in Murabba‘at.88 4Q, e.g., might have been used by the inhabitants before the deposition, but this does not

85  Thus Daniel Stökl Ben Ezra, “Wie viele Bibliotheken gab es in Qumran?” in Frey, Claussen, and Kessler, Qumran und die Archäologie, 327–46 (339) and Patrich, “Caves;” different de Vaux, Archaeology, 56–57 and de Vaux, in DJD 3:34 for caves 3Q and 11Q. 86  De Vaux, “Fouilles de Khirbet Qumrân,” 573–74; Patrich, “Khirbet Qumran,” 77; see also Popović, “Qumran as Scroll Storehouse,” 562. 87  Roland de Vaux, “Une hachette essénienne?” VT 9 (1959): 399– 407; cf. Magness, Archaeology of Qumran, 105–13. 88   Popović, “Qumran as Scroll Storehouse,” 555 refers to Murabba‘at : “The pottery, domestic items such as combs, the wooden and stone spindle whorls and the remains of several sandals, including two for children, point to the presence of families there;” see Roland de Vaux, “La période romaine,” in Les grottes de Murabbaʿât (DJD 2; ed. P. Benoit, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1961), 29–48.

b)

c)

necessarily imply that the scrolls deposited there were used by the inhabitants. Who lived in the caves? It seems beyond doubt to me that the area around Qumran was inhabited by different groups of people and was used temporarily or occasionally by perhaps even more. Different kinds of occupation existed parallel to each other. All groups had a particular role to fulfill, and perhaps even people changed from one kind of habitation to another. They all, in one way or the other, must have left traces in the archaeological record. It seems reasonable to assume that Qumran played a role for more people than only its inhabitants. I do not see any reason to separate the Qumran inhabitants from regional non-Qumran inhabitants. In any case, Qumran contains facilities that may have served the local population as well as its inhabitants: large storage rooms, space for animals, pottery kiln, date wine press, installations to repair metal tools, water and a large cemetry. We should therefore not see Qumran as center of a network of scroll-caves, but as sitting on the “top” “at the northern end of a fertile, active shoreline region.”89 And the scrolls? De Vaux’ equation of the troglodytes with the Qumranites and their identification with the Essenes triggered various detailed scenarios of how the scrolls were hidden and by whom. I am very sceptical about these scenarios. We still lack a full documentation and publication of the pottery excavated by de Vaux in both site and caves, and cannot fully compare the material with the wider regional context. Daniel Stökl Ben Ezra, for example, attempts to bolster his phasing of manuscript deposits, above all, on the presence or absence of particular jar types.90 Without proper documentation of what was really found in relevant caves such hypotheses are built on insufficient evidence.

In the end, “(we) do not know whether at one time or another all of the scrolls found in the caves were housed at the Qumran settlement.”91 It is very likely that some of the caves on the marl terrace might in one way or the other 89  Walker, “Notes on Qumran Archaeology,” 99. 90  Daniel Stökl Ben Ezra, “Old Caves and Young Caves: A Statistical Reevaluation of a Qumran Consensus,” DSD 14 (2007): 313–33; idem, “Wie viele Bibliotheken”; cf. Popović, “Qumran as Scroll Storehouse,” 579–83 on Stökl Ben Ezra as well as on other theories of “layered scroll deposition.” 91  Popović, “Qumran as Scroll Storehouse,” 586; different, e.g., White Crawford, “Qumran, Caves, Scrolls, and Buildings” 273.

The Functions Of The Caves And The Settlement Of Qumran

have belonged to the Qumran complex on top: as storage place for goods, as workshops or as dwellings.92 Some ­contained “scroll jars,” some did not, some texts, others not. Perhaps some caves had different purposes at the same time, others might have changed their function(s) over time. Since the present composition of the textual conglomeration is at least as much a product of 2000 years of postdepositional processes as of purposeful collection and formation, it is moot to speculate about the Essene character of the texts and in what sense we might have to do with a “library.”93 At one point or another, be it in steps, stages or waves, some people have decided to hide their valuable manuscripts in caves that had already been 92  In agree with Taylor, “Buried Manuscripts and Empty Tombs,” 305: “(T)he artificial marl terrace Caves 4Q–5Q, 7Q–10Q are connected with the occupation area of Qumran,” but in possibly very different ways. I do not see, however, that scrolls were buried in “empty tombs” (Taylor, “Buried Manuscripts and Empty Tombs,” 288–91). 93  Zangenberg, “Zwischen Zufall und Einzigartigkeit;” see also Popović, “Qumran as Scroll Storehouse,” 585–94.

209

in use for other purposes. It is also likely that some of the marl caves were artificially adjusted, enlarged or otherwise shaped to fulfil their job (4Q being a particularly good candidate). But none of the caves seems to have been made especially for storing manuscripts. Why these people chose those caves that we now call “scroll caves” 1Q to 11Q we do not know. The fact that the region “known as a good hiding place” might have been sufficient,94 though this is perhaps more true for the some of the caves in the cliffs than e.g. for 4Q. But whoever came to hide scrolls certainly had other options apart from 1Q to 11Q in the region as well, options that other manuscript-owners used for their own purposes as the collections at Masada, Na�al Ḥever and Murabba‘at demonstrate. Qumran itself might have offered additional space (4Q!) and assistance to hide what needed to be hidden. To me, the most likely historical background scenario for the deposition of the scrolls near Qumran still are the catastrophic events in the wake of the First Jewish War.

94  Popović, “Qumran as Scroll Storehouse,” 593.

Part 5 Short Papers

∵

CHAPTER 14

The Inscriptional Evidence from Qumran and its Relationship to the Cave 4Q Documents Sidnie White Crawford Ostraca, broken sherds of pottery, are the ubiquitous archaeological remains from ancient Israel and Judah. As such, they serve as the most important diagnostic indicator for the dating of the levels of a tel. They also served as “scratch paper” for the ancients, leaving an important record of everyday activities, particularly regarding trade and education, as well as onomastic lists for various ethnic groups. The inscriptions from Khirbet Qumran has now been fully published, thanks to the work of André Lemaire,1 affording a unique opportunity to study an inscriptional corpus in conjunction with a corpus of literary scrolls that were untouched after antiquity. I am convinced, by a plethora of archaeological evidence, that the scrolls found in the eleven caves in the vicinity of Khirbet Qumran are part of the archaeological landscape of Qumran,2 and it is thus legitimate, and in fact necessary, to study the scrolls and other archaeological remains from the caves in conjunction with the archaeological remains from the buildings of Qumran. Because of the massive fire that swept over Qumran in the destruction of 68,3 very little organic material survives from the khirbeh. Thus no scroll material was found. Therefore we must rely on the ostraca found at the site to give us any information we can glean about writing activities in the buildings. That writing activity took place there is certain, given the presence of inkwells, but of what 1  André Lemaire, “Inscriptions du khirbeh, des grottes et Aïn Feshkha,” in Khirbet Qumrân et ʿAïn Feshkha. Vol. 2: Études d’anthropologie, de physique et de chimie (NTOA.SA 3; ed. J.-B. Humbert and J. Gunneweg; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 341–88. The information which follows is taken from Lemaire’s publication, unless otherwise noted. 2  For discussion, see Jodi Magness, The Archaeology of Qumran and the Dead Sea Scrolls (Grand Rapids, MI: Eerdmans, 2002), 43–44, and Sidnie White Crawford, “Qumran: Caves, Scrolls, and Buildings,” in A Teacher for All Generations: Essays in Honor of James C. VanderKam (JSJSup 153; ed. E. F. Mason et al.; Leiden: Brill, 2011), 1:253–74. 3  Roland de Vaux, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed; London: Oxford University Press for the British Academy, 1973), 36, and Magness, Archaeology of Qumran, 61.

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_016

type and for what purpose? I will attempt to answer that question in the first part of this paper; in the second I will investigate whether any connections can be drawn with the scroll collection from Cave 4Q. The corpus of inscriptions found at Qumran was either on ceramic remains, both on intact vessels and on pieces of vessels, or on stone seals or weights. These inscriptions were incised into the clay before firing, or scratched, written, or painted on after firing. If the inscription was written or painted on after firing, it was done either with black carbon ink or red ochre.4 The inscriptions found in the khirbeh are written in Hebrew/Aramaic (often it is difficult to determine if the language is Hebrew or Aramaic), Greek, and Latin in the form of Roman numerals. 49 Hebrew/Aramaic inscriptions were discovered (13 are marked by Lemaire as questionable), seven in Greek (one marked as questionable), one in Latin, two with Hebrew/Aramaic letters and Roman numerals, and four in which the language is unidentified. In addition, two inscriptions in paleo-Hebrew from the Iron 2 period, including a lmlk stamp seal, were discovered. Finally, there were six graffiti that contained only figures such as lines. The caves yielded the following inscriptions: one in Heb/ar from Cave 4Q, one Heb/ar and one Greek inscription from Cave 6Q, one Heb/ar, one Greek, and one uncertain inscription from Cave 7Q, one inscription in Heb/ar from Cave 8Q, and one Heb/ar from Cave 10Q. No inscriptions were found in Caves 1Q–3Q, 5Q, 9Q and 11Q. The fact that inscriptions were found in three languages, Hebrew, Aramaic and Greek, indicates some degree of trilingualism in the Qumran settlement. However, it will be important in this regard to study the find sites, the stratigraphic dates, their chemical group (if available) and the genres of the inscriptions for clues about distribution 4  Gunneweg and Balla note that red ochre was found in locus 2 in the main building, indicating that it is possible that the red painted inscriptions were made at Qumran. Jan Gunneweg and Marta Balla, “Possible Connection Between the Inscriptions on Pottery, the Ostraca and Scrolls Found in the Caves,” in Humbert and Gunneweg, Khirbet Qumrân et ʿAïn Feshkha, 389–96 (391).

214 and use of languages.5 Therefore I will now turn to the find sites. When studying the locus numbers associated with the inscriptions, it is evident that inscriptions were found scattered in every area of the site, including the main building, the southeast annex that contained the pottery kiln (including the two sherds found by James Strange in 1996 along the base of the eastern wall),6 and in the western industrial complex. At least one inscription was found in the following loci: Trench A, West Trench, 4, 8, 10A, 15, 23, 27, 28, 29, 30, 34, 35, 38, 39, 54, 59, 61, 63, 73, 78, 84, east of locus 84, 89, 110, 111, 116, 121, 124, 125, 129, 130, 135, and 143. There were, however, loci where more than one inscription was found, which is of interest because it may indicate a concentration of writing activity in those areas. These loci fall into three groups. In the main building, more than one inscription was found in loci 30, 34, 35, and 39. Locus 30, is the room identified by de Vaux as “the scriptorium;” three inscriptions were found there.7 However, one of these, a Greek stone seal (KhQ439), comes from Period 3 (marked as questionable; see below), after the community settlement was destroyed. Two inscriptions were found in Locus 34, a “small room” in the southeast corner of the main building in Period 2. Locus 35, also with two inscriptions, is a room next to Locus 34 in the southeast quadrant. Locus 39, where two inscriptions were found, is a “square chamber” in the northeast quadrant of the main building. Thus, no real pattern emerges for the inscriptions found in the main building. The loci of the southeastern annex where more than one inscription was found are 61 (three inscriptions), 84 (two, including the inscription found to the east) and 143 (two). Loci 61 and 84 are workrooms associated with the pottery kilns, while locus 143 is a “small shelter” at the furthest southern end of the southeast annex. The associ5  See Appendix 1 for a chart detailing the object number, language, locus and stratigraphy for each inscription. 6  James F. Strange, “The 1996 Excavations at Qumran and the Context of the New Hebrew Ostracon,” in Qumran, the Site of the Dead Sea Scrolls: Archaeological Interpretations and Debates; Proceedings of a Conference Held at Brown University, November 17–19, 2002 (STDJ 57; ed. K. Galor, J.-B. Humbert, and J. Zangenberg; Leiden: Brill, 2006), 41–54. 7  The loci numbers and their descriptions are taken from JeanBaptiste Humbert and Alain Chambon, eds., Fouilles de Khirbet Qumrân et de Aïn Feshkha. Vol. 1: Album de photographies; Répertoire du fonds photographiques; Synthèse des notes de chantier du Père Roland de Vaux (NTOA.SA 1; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 1994), which transcribes de Vaux’s field notes.

WHITE CRAWFORD

ation with the pottery kilns is natural, since there presumably would have been more available sherds there than elsewhere in the settlement; thus no significance should be attached to these find sites. The western industrial complex had five loci where more than one inscription was found: 110 (the large round cistern; four inscriptions), 111 (three inscriptions), 124 (which de Vaux described as “a solid mass of masonry to support the building of the storeroom to the west of the western outbuilding;” seven or eight inscriptions),8 129 (two) and 130 (two). A total of 19 or 20 inscriptions were found in this area.9 This may seem at first glance to be an important indicator of writing activity in the western industrial complex, but it must be noted that for the two loci where the most inscriptions were found, 110 (four) and 124 (eight), the pottery sherds were either fill that had drifted into the round cistern (l. 110), or were deliberately used to reinforce a wall (l. 124). Thus these sherds most likely came from elsewhere around the site and give no information about writing activity in this area.10 Taken together, then, the find sites of the inscriptions give no indication of a concentration of writing anywhere in the ruins. The dates of the inscriptions are quite difficult to determine, except within broad parameters. None of them contain any kind of date formula. De Vaux registered two of the inscriptions as belonging to the Iron II period, three from Period 1,11 30 to Period 2, three to Period 3, and one to Periods 2 or 3. Six are noted as surface finds, while ten are from unstratified contexts. The majority of the inscriptions, therefore, come from Period 2, which is unsurprising,

8  Following the translation of Stephen J. Pfann: Jean-Baptiste Humbert and Alain Chambon, eds., The Excavations of Khirbet Qumran and Ain Feshkha: Synthesis of Roland de Vaux’s Field Notes (NTOA.SA 1B; trans. and rev. Stephen J. Pfann; Fribourg/ Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 6. 9  See also Gunneweg and Balla, “Possible Connection,” 393, who have different counts. 10   Contra Gunneweg and Balla, “Possible Connection,” 393, who argue that it is unlikely that this is in fact a dump. However, de Vaux does not describe Locus 124 as a dump; his notes read “Massif de fondation pour soutenir les magasins à l’ouest des dépendances, construits au bord abrupt du wadi.” Humbert and Chambon, Fouilles de Khirbet Qumrân, 14. 11  See also Frank M. Cross and Esther Eshel, “Ostraca from Khirbet Qumran,” in Qumran Cave 4.XXVI: Cryptic Texts, ed. S. J. Pfann; Miscellanea, Part 1, ed. P. S. Alexander et al. (DJD 36; Oxford: Clarendon Press, 2000), 497–514 (509), where Cross dates KhQ 3 (= KhQ161) to Period 1.

INSCRIPTIONAL EVIDENCE FROM QUMRAN & THE CAVE 4Q DOCUMENTS

since Period 2 was the last major habitation of the site.12 Eight of the thirty, however, are from the fill in Locus 124, which is a reuse. Thus their actual date is indeterminate. If we subtract these eight from Period 2 and add them to the surface/unstratified group, 24 are now of indeterminate date, while 22 can be securely located in Period 2. The Period 2 inscriptions, however, still outweigh those of the other periods by a significant margin. A selection of the sherds were subjected to Instrumental Neutron Activation Analysis to determine the locality of the clays from which the vessels were made.13 There are four chemical groups as identified by Gunneweg and Balla: Chemical Group 1- local to Qumran. Five of the ostraca tested belonged to this group: KhQ386, 621, 2176, 2556 and 7Q6. Four of these were in Heb/ar; one had ten lines in ink. None were in Greek or Latin. Chemical Group 2- clay from the Hebron/Motsa ­formation.14 Nine ostraca were made from this clay: KhQ 425, 425c, 1313, 1401, 2109, 2125, 2507, 2553, and 2554. Eight of these inscriptions were Heb/ar, while one bore a Latin number. Chemical Group 3- Jericho. Fourteen ostraca were made of clay that Gunneweg and Balla identify as coming from Jericho (but see below): KhQ 387, 426, 461, 635, 680, 681/691, 711, 979, 1110, 1650, 2587, 2609, and 3759. Seven of these were Heb/ar, four were Greek, one was uncertain, and two had lines or incisions. Chemical Group 4- Edom (Nabatea). Two sherds, KhQ 2416 and 2417, both from locus 130 and with the same type of inscription, Heb/ar letters plus Roman numerals, were made of this geographically distant clay. 12  De Vaux notes that only the north wing of the main building was reoccupied during Period 3 (de Vaux, Archaeology and the Dead Sea Scrolls, 63–64). See also Joan E. Taylor, “Khirbet Qumran in Period III,” in Galor, Humbert, and Zangenberg, Qumran, the Site of the Dead Sea Scrolls, 133–46. 13  Jan Gunneweg and Marta Balla, “Neutron Activation Analysis: Scroll Jars and Common Ware,” in Humbert and Gunneweg, Khirbet Qumrân et ʿAïn Feshkha. Vol 2: Études d’anthropologie, de physique et de chimie, 3–54. Jodi Magness gives a concise definition of INAA: “a nuclear method to determine the chemical composition of clay or pottery, with groupings based mainly on the trace elements present.” Jodi Magness, in the present volume. I would like to thank Professor Magness for sharing her paper with me prior to publication. 14  The Motsa Formation is the most well-known and widely utilized clay formation in Judea (it is still in use at the present day). See Jacek Michniewicz and Miroslaw Krzyśko, “The Provenance of Scroll Jars in the Light of Archaeometric Investigations,” in Humbert and Gunneweg, Khirbet Qumrân et ʿAïn Feshkha, 59–100.

215

Gunneweg and Balla make the assumption that those sherds from chemical groups 2–4 were imported into Qumran after firing.15 This would mean that any sherds with incised inscriptions, made on the clay before firing, would not contain information (e.g. names) directly pertaining to the Qumran site, although those with inked or painted inscriptions could have been reused at Qumran. It would seem, therefore, that some separations based on the origination of the clays can be made among the ostraca. However, as Michniewicz and Krzyśko observe in their petrographic analysis in the same volume, it is “difficult to determine deposit provenance on the microregional scale.”16 According to them, “the clayey materials from the region of Judea are highly homogenous from the geochemical point of view,” so that “no precise determination of the provenance of local pottery is possible.”17 In other words, the distinction between pottery made from “Jericho” clay and pottery made from “Qumran” clay, which Gunneweg and Balla attempt to draw, is probably too fine to be realistic, given the close geographical proximity of Qumran and Jericho.18 Further, Michniewicz and Krzyśko observe that clays close to Qumran were not used to manufacture pottery at Qumran, so the “Qumran designation is misleading.”19 Thus, it seems safest to assume that the clays used in Groups 1, 2 and 3 came from the general area of Judea, with a preference for the good clay of the Motsa Formation. Therefore, INAA analysis does not contribute very much to our knowledge of the provenance of the inscriptions. The exception is the two sherds from Chemical Group 4, the Edom/Nabatea group, which we may assume were imported. Finally, we must look at the genres of these inscriptions. These are often difficult to ascertain, since the inscriptions are often very broken, and only a letter or two remains. However, some inscriptions can be identified by type, or at least by a descriptive category. I will begin with those inscriptions which seem to have something to do with measurements, trade, or economic transactions. Four inscriptions, KhQ2416, 2417, 2538 and 6Q1, may indicate a weight, measurement, or quantity. They are all found on pitchers or storage jars. Two (2416 15  Gunneweg and Balla, “Possible Connection,” 391–92. 16  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 60. 17  Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 76. 18  See also Magness’s conclusions in this same volume. 19  Michniewicz and Krzyśko tested clays taken from Qumran deposits from the coast of the Dead Sea, from the Wadi Qumran estuary from deposits underneath Cave 4, and from above the aqueduct. None of this clay was found in the ceramics subjected to petrographic analysis. Michniewicz and Krzyśko, “The Provenance of Scroll Jars,” 62, 74, 76.

216 and 2417) combine Heb/ar letters with Roman numerals.20 Four inscriptions, KhQ461, 2554, 2261c and 4Q1, seem to have to do with trade or import/export. Two of these, 2554 and 2261c, are identical and found in the same locus (l. 124; note, however, that this is fill reinforcing a wall). Five inscriptions (KhQ386, 680, 1095, 2115 and 3759) are simple lines or incisions, which may indicate counting or calculation. Finally, there is one deed of gift or sale, the famous KhQOstracon 1 found in the 1996 Strange excavation.21 Another group of inscriptions contains marks of identity. Three seals were discovered, two in Greek (KhQ439 and 2145), and one with a rosette and what appears to be two paleo-Hebrew letters (KhQ2208).22 Their function is transparent. Less obvious in function are the inscriptions that just contained a single name. There are nine of these, all in Heb/ar, and all bearing a Jewish name common in the late Second Temple period: KhQ621 (Yohanan Hatla), 935 (ŠLY), 1313 (Honiah), 1416 (Meshullam), 1650 (Eleazar), 2108 (Pinhas), 2125 (Samuel?), 2136 (ŠWH), and 2507 (Eliezrah). None of the names repeat, although there is one Eleazar and one Eliezrah. Of the eight, five (621, 935, 1650, 2136 and 2507) were found either on intact vessels or remains of once intact vessels, which may indicate ownership of some kind. In this last group one might wish to include the “rûma’ ” or “rôma’ ” inscriptions, KhQ681/691 and 7Q6, if “rûma’ ” is a proper name and not a geographic indicator.23 20  Recall that these two ostraca are most likely imported into Qumran. 21  Published by Cross and Eshel, “Ostraca From Khirbet Qumran,” 497–507. Two subsequent articles, by Ada Yardeni and Emile Puech, dispute some of the readings of Cross and Eshel. They famously read the word ‫ ליחד‬in line 8, thereby connecting the deed to the ‫ יחד‬referred to many times in the Community Rule. Yardeni and Puech both read instead ‫( א]חר וכול אילן‬with variants). All agree, however, that it is a deed transferring property. Ada Yardeni, “A Draft of a Deed on an Ostracon from Khirbet Qumrân,” IEJ 47 (1997): 233–37; Emile Puech, “L’ostracon de Khirbet Qumrân (KhQ1996/1) et une vente de terrain à Jéricho, témoin de l’occupation essénienne à Qumrân,” in Flores Florentino: Dead Sea Scrolls and Other Early Jewish Studies in Honour of Florentino García Martínez (JSJSup 122; ed. A. Hilhorst, E. Puech, and E. J. C. Tigchelaar; Leiden: Brill, 2007), 1–29. 22  KhQ439 and 2208 are dated to Period 3. KhQ2145 is unstratified. 23  De Vaux noted that the name is attested in Nabatean and Palmyrene, and may be related to ‫ רמי‬at Elephantine. He suggests that the Greek name ‛Pούμας, frequent at Dura Europos, is a transcription of the Aramaic form. Roland de Vaux, “Archéologie,” in Les ‘petites grottes’ de Qumrân: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 30.

WHITE CRAWFORD

A second group of ostraca containing names appears to be concerned with trade or import/export. Four contain the Aramaic word ‫לאפך‬, “to return,” which seems to indicate trade. One vessel, KhQ2553, has two names, Shimon bar Yonatan and Mattatay bar Yosef. KhQ2554 and KhQ2661c have the same inscription: ‫שנים ולי לאפך‬, “two, to PN?, to return.” Lemaire suggests that the PN could be “Julia.”24 However, one can also understand it as “and to me.” 4Q1 has the name Yair. Again, these are common Jewish names of the period. The two ostraca from Chemical Group 4 (KhQ2416 and 2417), the Edom/ Nabatean group, each has a name, GRP and Zimri, followed by Roman numerals. These vessels were almost certainly imported. Finally, KhQOstracon 1, the deed of gift or sale, contains in lines 2, 9 and 14 the name “Honi, son of[,” in line 3 “Eleazar son of Nahmani,” and in lines 10 and 14 “Hisday, slave of Honi.”25 Two of these names, Honi/Honiah and Eleazar, are found in other inscriptions (KhQ1313 and KhQ1650 respectively), but it is impossible to tell if they are the same people. Finally, there is a Greek seal, KhQ439, possibly from Period 3, bearing the name “Joseph.” Since none of the names, unless one accepts Lemaire’s reading of “Julia” in KhQ2554 and 2661c, repeat, it is impossible to draw any conclusions as to whether any of these names belonged to inhabitants of the khirbeh, although that remains a possibility. Four inscriptions are single letter inscriptions, all on intact vessels. Three are Heb/ar, while one is probably Greek. KhQ1110 has the Hebrew letter ‫ש‬, while KhQ1403 and 8Q10 have the letter ‫ט‬. This ‫ ט‬may stand for ‫טהר‬, “pure,” or ‫טמא‬, “impure.” 6Q2 has a Μ. Finally, there are three ostraca that have been identified as student exercises. Two, KhQ16126 and KhQ2289, are Hebrew abecedaries. KhQ2207, a stone plaque with five lines of writing, has an unknown content. That is, the letters are random, rarely forming any recognizable word. Further, the letters are roughly formed and irregular. Lemaire suggested that this was “probably an incomplete apprentice scribe’s exercise.”27 To summarize thus far, the inscriptional evidence from the khirbeh and the caves allows us to make the following 24  Lemaire, “Inscriptions du khirbeh,” 368. 25  Cross and Eshel, “Ostraca From Khirbet Qumran,” 497–507. Neither Yardeni nor Puech read “Hisday, slave of Honi.” 26  KhQ161 was also published by Eshel as KhQOstracon 3 (Cross and Eshel, “Ostraca From Khirbet Qumran,” 509–14). According to Eshel, the script is “transition to Herodion,” that is, mid-first century BCE (509). 27  Lemaire, “Inscriptions du khirbeh,” 360.

INSCRIPTIONAL EVIDENCE FROM QUMRAN & THE CAVE 4Q DOCUMENTS

statements. 1. Some kind of scribal training was going on, given the presence of scribal exercises. 2. Trade into Qumran from outside the settlement is also indicated. 3. Hebrew and Aramaic were the common languages of the settlement, along with an ability to at least decipher Greek and Roman numerals. Next let us turn to some manuscript evidence from Cave 4, which may provide us with some connections to the inscriptional evidence. We begin with perhaps the most pertinent connection, the presence of scribal exercise fragments in Cave 4. 4Q234, 341, and 360 have all been identified as scribal exercises. 4Q234, Exercitium Calami A, written in a first century BCE script and penned in various directions, may contain a quotation of Gen 27:19–21, but this identification is uncertain, since only two words are extant.28 4Q360, Exercitium Calami B, in an early Herodian bookhand, is also penned in various directions. It contains the name Menahem, repeated three times.29 Although this is in the realm of speculation, it seems possible that here we have the name of the scribe himself. 4Q341, Exercitium Calami C,30 is clearly a writing exercise. Lines 1–3 contain a series of letters, sometimes alphabetical. Lines 4–7 and 9 contain some personal names: Magnus, Malchiah, Mephiboshet, Gaddi, Dalluy, Hyrcanus, Vanni and Zakariel. In addition, there is a name written in the right-hand margin, Omriel(?). The name Mephiboshet has an obvious biblical reference (2 Sam 9), and Hyrcanus may refer to the Hasmonean John Hyrcanus. However, the most likely explanation for the list of names is that they are the names of individuals known to the scribe, who practiced his craft by writing the names of his friends. Why were these small, seeming worthless exercises placed in Cave 4Q? 4Q234 may contain a scriptural text, but otherwise these small scraps are not even literary. I would suggest that they are part of the material from the khirbeh that was thrust helter-skelter into Cave 4Q in anticipation of the Roman attack.31 In any case, the pres28  Ada Yardeni, “234. 4QExercitium Calami A,” in DJD 36:185–86. 29  Ada Yardeni, “360. 4QExercitium Calami B,” in DJD 36:297. 30   Olim 4QTherapeia, 4QList of Proper Names. Joseph Naveh, “341. 4QExercitium Calami C,” in DJD 36:291–93. 31  Stökl Ben Ezra has suggested that Cave 4 material from the first century BCE and earlier may have been hidden there prior to 9–8 BCE. Daniel Stökl Ben Ezra, “Old Caves and Young Caves: A Statistical Reevaluation of a Qumran Consensus,” DSD 14 (2007): 313–33. However, the only record I have found concerning the state of the excavated materials from Cave 4 is that of Cross from 1958, who remarks, “I was struck with the fact that the relatively small quantity of fragments from the deepest levels of the cave nevertheless represented a fair cross section of the whole

217

ence of scribal exercises in both the inscriptional corpus and the Cave 4Q corpus ties the two together. The second group of Cave 4Q manuscripts that has connections to the inscriptional evidence is the documentary texts, i.e. deeds, accounts, etc. (4Q342–359). These manuscripts are more controversial than the scribal exercises, since, while the Bedouin claimed they came from Cave 4Q, Yardeni has shown that at least one (4Q347) and possibly more actually came from Wadi Seiyal.32 On the other hand, Lange and Mittmann-Richert have argued that some of them did in fact come from Cave 4Q.33 To err on the side of caution, I will discuss only those documents with paleographic and/or C14 dates that fall within Periods 1 and 2 at Qumran. 4Q342, an early first century CE Aramaic letter, contains the names Judah (l. 3), Eleazar (l. 3), and Elishua (l. 4). 4Q343, a Nabatean letter from the middle of the first century BCE, contains the name Shimon in line 13, and possibly an unknown proper name in line 14. 4Q345, an Aramaic or Hebrew deed of the middle-late first century BCE, contains the name Yeshua on the recto upper line 6, the name Hoshayah son of[ on verso line 20, and Ishmael son of Shimon on verso line 21 (note the appearance of the name Shimon in 4Q343). 4Q346, an Aramaic deed of sale from the late first century BCE, contains the name Shimon (again!) on line 3, and Manasseh on line 6. 4Q348, a Hebrew deed of the middle to late Herodian period, contains the following names: Menahem . . . son of Eleazar (upper, line 1); Shimon (upper, line 5); Yehohanan son of Yehosef (upper, line 9); Yehosef, Mattatyah son of Shimon and Eleazar (lower, line 14); Hanan and Eleazar deposit in the cave, which suggests . . . that the manuscripts may have been in great disorder when originally abandoned in the cave. The paucity of sherds in the cave certainly indicates that the scrolls of Cave IV were not left stored away in jars.” Frank M. Cross, The Ancient Library of Qumran and Modern Biblical Studies (Garden City, NY: Doubleday, 1958), 27 n. 32. This suggests that the majority of the Cave 4 scrolls were placed in the cave quickly, prior to the Roman attack. 32  Ada Yardeni, “Documentary Texts Alleged to be from Qumran Cave 4,” in Aramaic, Hebrew and Greek Documentary Texts from Naḥal Ḥever and Other Sites: With an Appendix Containing Alleged Qumran Texts (The Seiyâl Collection II) (DJD 27; ed. H. Cotton and A. Yardeni; Oxford: Clarendon Press, 1997), 284. 33  See Armin Lange and Ulrike Mittmann-Richert, “Annotated List of the Texts from the Judaean Desert Classified by Content and Genre,” in The Texts from the Judaean Desert: Indices and an Introduction to the Discoveries in the Judaean Desert Series (DJD 39; ed. E. Tov; Oxford: Clarendon Press, 2002), 115–64 (144). The documents they list as coming from Cave 4 are 4Q352, 4Q352a, 4Q353, 4Q354, 4Q355, 4Q356, 4Q357, 4Q358, 4Q351, 4Q350, 4Q346, and 4Q345.

218 son of Shimon, son of Honi (lower, line 15); Manasseh (lower, line 17); and Shimon from the Timber Market, a district in Jerusalem (lower, line 18). The repetition of the names in this group, Eleazar (342, 348), Shimon (343, 345, 346, 348), and Manasseh (346, 348), suggests a personal archive of some kind, perhaps belonging to Shimon, whose name appears five or six times. Further, there are evident family relationships among the names: Shimon is listed as the father of Ishmael, Mattatyah and Eleazar, which also suggests a personal archive. Some of these names also occur on inscriptions from the khirbeh: Honi/ Honiah (KhQOstracon 1, KhQ1313), Eleazar (KhQ1650, KhQOstracon 1), Yehohanan/Yohanan (KhQ621) Yehosef/ Yosef (4Q1), and Shimon (4Q1). Further, in 4Q348 Shimon is identified as the son of Honi, a name that appears in KhQOstracon 1. It is important to emphasize that none of these names necessarily refers to the same individual, especially since these are common Jewish names of the period. However, the last example, that of Shimon son of Honi (4Q348), gives the best possibility for a connection with an inscription, placing Shimon, the owner (?) of the personal archive found in Cave 4Q into a familial relationship with Honi, the giver of the gift in KhQOstracon 1. The ostracon and the deed date paleographically to the same period, which strengthens the connection. Although there are several accounts listed among the 4Q fragments, only two seem certain to have originated at Qumran. The first is 4Q350, Account gr, which is an opsisthograph of 4Q460, Narrative Work and Prayer.34 Yardeni suggests that since the account is in Greek it was written by non-Jews (Roman soldiers?) during the partial occupation in Period 3. This is certainly possible but not certain. The second, 4Q355, Account C ar or heb, is written on the verso of 4Q324, Mishmarot C. Its content is illegible. The most that can be said, then, is that some accounts have surfaced in Cave 4Q which resemble the accounts or figures in the inscriptional corpus.

34  Yardeni, “Documentary,” 294.

WHITE CRAWFORD

Finally, I would like to compare the names found in 4Q477, Rebukes Reported by the Overseer (late Herodian script), with the names listed above.35 The names on 4Q477 are as follows: Yohanan ben Ar[ (frg. 2 ii 3), Hananiah Notos (frg. 2 ii 5), and Hananiah ben Shim[on] (frg. 2 ii 9). The names on this list are without argument members of the Yahad (frg. 2 ii 6). Again, we find some overlap with the names already listed: Yohanan on KhQ621, and Shimon on 4Q2553 and especially 4Q343, 345, 346 and 348. However the Shimon in 4Q477 is the father of Hananiah, not mentioned in the 4Q archive. These two names are so common that it does not seem wise to make any overt connections.36 1 Conclusions The inscriptional evidence from Khirbet Qumran and the manuscripts from Cave 4 do exhibit some connections that add to the many pieces of evidence that tie the ruins of Qumran to the caves. The first, and most important, is the presence of scribal exercises in both places. These exercises, which are unlikely in the extreme to have been brought to Qumran from the outside, demonstrate that scribal activity and training was taking place at Qumran itself. The second is the presence of documentary texts and accounts in the khirbeh and in the caves, evidence of commercial activity of some kind, including the sale/ exchange of property, at Qumran. The third is the onomasticon listed above, consisting of a common repertoire of Jewish names from the Second Temple period. However, there is some recurrence of names that tie certain documents together, especially the personal archive of Shimon from Cave 4Q with KhQOstracon 1, both mentioning a certain Honi. On the basis of this evidence we can therefore make the claim that there are modest written connections between the khirbeh and the caves.

35  Esther Eshel, “477. 4QRebukes Reported by the Overseer,” in DJD 36:474–84. 36  See also Eshel, “477. 4QRebukes,” 480.

219

INSCRIPTIONAL EVIDENCE FROM QUMRAN & THE CAVE 4Q DOCUMENTS Table 14.1 Table of inscriptional remains from Khirbet Qumran KhQ Number

Language

Locus

Stratigraphy

Other

 161a  192  386  387  425  426  427  439  461  498  498bis  572  621  635  680  681  691  682  701  711  734  935  979 1095 1110 1235 1236 1264 1313 1401 1403 1416 1650 2208 2108 2109 2215 2124 2125 2136 2145 2176 2207 2289 2416 2417

Heb/ar Greek

Trench A   8  23  27  10A  15  28  30  29  30  30   9A  34  35 West trench  35  37  34  39  38  39  28  54  59  61  66  73  63  61  84  61 East of 84  89 111 110 110 111 110 116 110 121 125 129 135 130 130

Disordered/Level 1 Period 2 or 3 Period 2 Period 2 Period 2 Period 3 Period 1? Period 3? Disordered Period 2 Period 2 Disordered Period 1 Period 2

abecedary

Greek Heb/ar? Heb/ar Heb/ar? Greek Heb/ar Heb/ar Heb/ar? Heb/ar? Aramaic Greek Heb/ar? Heb/ar Heb/ar? Heb/ar Lat? Heb/ar? Heb/ar Greek Heb/ar? Paleo-Heb Paleo-Heb Heb/ar? Heb/ar? Latin Heb/ar? Heb/ar Heb/ar Paleo-Heb letters? Heb/ar Heb/ar Greek Heb/ar Heb/ar? Greek? Heb/ar Heb? Heb/ar Heb/ar Heb/ar

Surface Surface Period 2 Surface No stratigraphy Period 2 Period 2 Period 2 Period 2? Period 2 Iron II Iron II No stratigraphy Period 2 End of Period 2 Period 2 No stratigraphy Period 2 Period 3 Fill Fill Period 2 Fill Period 2 Fill Surface Surface Surface Period 2 Period 2

graffito

Stone seal

Whole jar graffito Join with 691 Join with 681

Whole jar

Intact bowl lmlk stamp

Whole jar Whole jar

Stone seal

Stone weight; graffito

stone

220

WHITE CRAWFORD

(cont.) KhQ Number

Language

Locus

2507 2538 2539 2553 2554 2556 2557 2563 2575 2587 2609 2661c 3759 4037 5167 Ostracon 1

Heb/ar Heb/ar Heb/ar Aramaic Heb Heb? Heb/ar Heb/ar?

Greek? Heb/ar Heb

Ostracon 2

Heb

 4QA  4Q1  6Q1  7Q1  7Q6  7Q7  8Q10 10Q1

Heb/ar Heb/ar Heb/ar

North of Cave 1Q 124 124 124 124 124 124 124  30 143  78 124   4 111 ? Eastern side of the marl terrace wall Eastern side of the marl terrace wall Cave 4Q Cave 4Q Cave 6Q Cave 7Q Cave 7Q Cave 7Q Cave 8Q Cave 10Q

a Also KhQOstracon 3.

Heb? Latin? Greek? Heb

Heb/ar Greek Heb Heb/ar

Stratigraphy

Period 2? Period 2? Period 2? Period 2? Period 2? Period 2? Period 2? Earlier than Period 2 Earlier than Period 2 Period 2 Period 2? No stratigraphy Period 2

Other

graffito Bowl

graffito

30–68 CE 30–68 CE

Whole jar; letters and figures Whole jar Whole jar

CHAPTER 15

The Coins of Khirbet Qumran from the Digs of Roland De Vaux: Returning to Henri Seyrig and Augustus Spijkerman Bruno Callegher In October 2007, after the lectio magistralis of the academic year 2007–2008 at the Studium Biblicum Franciscanum [SBF],1 Michele Piccirillo (1944–2008) [MP]2 showed me three coin containers in his office. These were plastic sheets with a varying number of small pockets.3 They were separated by cards marked out in boxes, on which, page after page, beneath each coin, was a number in pencil that was completely baffling. There was no link between those numbers and any other documentation because of a lack of references. The three files were marked, on the inside, with the initials K.Q. + FESCH [Khirbet Qumran and Ain Feshkha], but also contained evidence that the set had almost certainly not been prepared like that by Roland de Vaux (1903–1971)[RDV]4 and had been tampered with over the decades: specimens had been removed by someone who had access or wished to examine the finds, as this paper will report. Piccirillo, then Director of both the Archeological Museum and the Coin Cabinet of the SBF, invited me to examine the dossier “coins from the Qumran digs” because he was puzzled by some studies at that time newly published on the coins.5 He repeated his invitation in February 2008, but also on that occasion I did not accept because I had other research dossiers to complete. Nonetheless, in the meantime a rapid search in the SBF library had clarified that what I had seen was the

answer to the communis opinio that the single coin finds from KQ had been lost since 1960, almost certainly due to the negligence of Augustus Spijkerman [AS] (1920–1973).6 As the finding of coins traceable to the first digs at that site would have contributed to a more secure chronology and also to determining the function of the site, I asked permission to inform colleagues at the Coin Cabinet of the Israel Antiquities Authority [IAA] about the “rediscovery” of the KQ coins, who at once showed great interest in the unexpected recovery.7 Nevertheless, in the next three years I did not follow up this research. In 2010 the then Dean of the SBF, Claudio Bottini, requested two collaborations only apparently unrelated. The first was to provisionally arrange the papers of AS, collected by MP who, having taken over as Director of the Archeological Museum of the Studium, had kept them. The second was to organize the identification of the whole coin cabinet, left without a keeper, but in need of the specimens of the historical collection being separated from those collected by the SBF archaeologists at the digs in various areas of Palestine and Jordan. Working contemporarily on the Spijkerman archive and in the coin cabinet proved to be exceedingly fruitful. The papers of the AS numismatist included letters to many numismatic and archeological colleagues, for example to Henri Seyrig (1895–1973) [HS],8 Leo Kadman

1  “Cronaca” of Studium Biblicum Franciscanum. 2  Giovanni Claudio Bottini, “Michele Piccirillo (1944–2008) francescano della Terra Santa e archeologo,” LA 58 (2008): 479–500; Giovanni Claudio Bottini, “Ricordo di Michele Piccirillo,” in Notiziario Studium Biblicum Franciscanum; Anno Accademico 2008– 2009 (2009): 8–24; Giovanni Claudio Bottini and Massimo Luca, eds., Michele Piccirillo francescano archeologo tra scienza e provvidenza (SBF Museum 15; Milano: Edizioni Terra Santa, 2010). 3  The containers are of the type Koloman Harder-Made in Austria, while the sheets are marked Secrian-Milano. 4  The handwriting of the numbering, is in fact, not RDV’s. So far, despite various attempts, it has not been possible to trace the author of the numbering: AS [Augustus Spijkerman] and MP can also be excluded. 5  I prefer to be prudent and not offer any citation because this would be wholly self-referential as it could be neither confirmed nor denied.

6  The memory of AS, academic and director of the Museum, so far relies only on Bellarmino Bagatti, “P. Augusto Spijkerman (1920– 1973),” LA 23 (1973): 387–90; Virgilio Corbo, “Prefazione,” in Augustus Spijkerman, Cafarnao, Vol. 3: Catalogo delle monete della città (SBF. CMa 19; Jerusalem: Franciscan Printing Press, 1975), 10. 7  I saw Robert Kool and Donald Ariel photographic documentation in the IAA. 8  George Le Rider, “Henri Seyrig 1895–1973,” RSN 52 (1973): 167–71; Jean-Claude Cheynet, Cécile Morrisson, and Werner Seibt, eds., Sceaux byzantins de la collection Henri Seyrig: Catalogue raisonné (Paris: Bibliothèque Nationale, 1991), 5–8. The archives of the scholar are deposited in the Cabinet des Médailles of the Bibliothèque Nationale in Paris; the library, instead, was split between three institutes at Neuchâtel (I thank Prof. Hédi Dridi, Director of the Institute of Archaeology at Neuchâtel for the information on the Henri Seyrig library).

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_017

222 (1895–1963),9 Leo Mildenberg (1913–2001),10 Herbert Cahn (1915–2002),11 but most importantly there were mentions of the work of classification of the coins dug up at KQ. In fact, his file referred to the cataloguing of a part of the hoard of tetradrachms discovered at KQ, here documented by the letter of receipt of the materials by the National Archeological Museum of Amman12 and by around a hundred black and white photographs of the tetradrachms examined, still in their original developer’s pouches, marked Qumran.13 Equally conclusive was the exchange of letters with RDV concerning the financial correspondent of Neue Zürcher Zeitung of Zurich, H. Heymann, who was persistently requesting information on the presence of a coin minted at Ascalon “of the year 72/73,” countermarked with the stamp of the X Legio.14 The obvious diligence of the Franciscan in identifying the single coin finds of that site also emerged, using a classification method known to the scientific community. However, any concrete result of his efforts was missing from the archive, for example any text or notes, the existence of which, at 9  Ernest W. Klimowsky, “Leo Kadman: In Memoriam,” INJ 2 (1964): 3–6. 10  For an initial approach to the numismatic interests of this scholar, first and foremost the coinage of Palestine, see Leo Mildenberg, Ulrich Hübner, and Ernst Axel Knauf, eds., Vestigia Leonis: Studien zur antiken Numismatik Israels, Palästinas und der östlichen Mittelmeerwelt (NTOA 36; Fribourg/Göttingen: Éditions Universitaires/Vandehoeck & Ruprecht, 1998); Dan Barag, ed., Studies in Memory of Leo Mildenberg, INJ 14 (2000–2002). 11  Numismatist and “antiques dealer” for many years, commemorated by Ute Wartenberg, “Herbert A. Cahn, 1915–2002,” American Numismatic Society Magazine 1/2 (2002): 33; “Herbert A. Cahn, 1915–2002,” The Celator 16/9 (2002): 32; cf. also Hans Voegtli, “Bibliographie Herbert A. Cahn, 1975–1984,” in Festschrift Herbert A. Cahn: Zum 70. Geburtstag gewidmet und herausgegeben vom Circulus Numismaticus Basiliensis (ed. B. Schärli and H. Voegtli; Basel: Circulus Numismaticus Basilienis, 1985), xiii–xv. Other information at www.dictionaryofarthistorians.org/cahnh.htm. 12   S BF [ASBF]. AS Trust Archive, A. Spijkerman. Corrispondenza 1956–1973 † 23.06.73 [fasc. 8]. The original is held in the EBAF Archive [AEBAF]. KQ Trust with a paper signed by Adnan Hadidi dated 19 November 1960, from which it is learned that the restitution took place on two different occasions: the first time for 406 specimens, the second for the missing two, for a total of 408 coins. 13   A SBF. AS Trust, A. Spijkerman [fasc. 6] The handwritten note is by Augustus Spijkerman. 14  The letters held in ASBF. AS Trust, A. Spijkerman. Corrispondenza 1956–1973 † 23.06.73 [fasc. 8] integrate with those, more numerous, in the AEBAF. KQ Trust. This subject is referred to by Roland de Vaux, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973): 40, n. 1.

CALLEGHER

least at that stage, could only be guessed. As mentioned, the studies relative to that site had so far always considered the single coins as lost, although citing or describing them. Nobody knew of the AS papers held at the Studium and in general limited themselves to reporting a generic existence of “Spijkerman’s identification cards” without any archival reference, without mention of their number, contents, nor any place of conservation, all vital information for retracing that research itinerary.15 In addition, the epithet “Spijkerman’s cards,” led to the assumption that the cards would have to have been compiled/edited by AS himself, but without giving reasons for the attribution. There was therefore a need to widen the search, obviously starting from the archive and the coins kept in the three containers, for other paper documents/manuscripts useful for reconstructing the relationships between the various individuals involved in the examination of those coins in the period between the years of the first discovery and the involvement of the Franciscan numismatist. Despite the dossier being reopened in 2008, still in 2010 no one at the SBF was aware of the existence of the “cards” attributed to Spijkerman, even less where they had ended up. The memories of AS’s involvement were by now distant and vague, no more than “heard it said that.” There was therefore much surprise when, having begun the identifications in the coin cabinet, I informed the personnel of the Studium and Archeological Museum that I had found the three files shown to me three years earlier by MP,16 which had been deposited there following the post mortem census of his papers and the finds kept in his office, but most of all that they had to be linked to the notes and unpublished materials of AS.17 I consequently hypothesized that other documentation relating both to AS and the coins of KQ was to be found at the École Biblique et Archéologique Française in Jerusalem [EBAF]. For the research, the references of 18 June 1979 were 15   Kennet Lönnqvist and Minna Lönnqvist, “The Numismatic Chronology of Qumran: Fact and Fiction,” NC 166 (2006): 121–65 (121): “Spijkerman’s identification cards are the source for the Qumran coin lists, but most of the attributions of the bronze coins cannot now be checked.” There is no archival reference. 16  There are good reasons for supposing that only MP knew about this information on the coins from the RDV Khirbet QumranDigs. The containers were probably previously kept in his office. 17  This can be confirmed by Claudio Bottini and by the current director of the Museum, Eugenio Alliata. They were unaware that the coins of Khirbet Qumran were known and held by MP. In fact, those three containers were deposited in the coin cabinet together with many others that held thousands of specimens found by the archaeologist over three decades of archaeological activity.

THE COINS OF KHIRBET QUMRAN FROM DE VAUX ’ S DIGS

essential since the publication of the Revue Biblique18 predated the study of Marcia Sharabani: Ultérieurement, cet inventaire [that edited by Henri Seyrig, cf. infra] fut révisé par le Père A. Spijkerman. Quelques années après sa mort (23 juin 1973), son manuscrit fut retrouvé par le Père M. Piccirillo au Couvent de la Flagellation, siège du Studium Biblique des Pères Franciscains à Jérusalem. Le Père M. Piccirillo transmit ce manuscrit en décembre 1977 à M.elle Marcia Sharabani,19 conservatrice au service des Antiquités, qui voulut bien accepter de publier un lot incomplet de 153 monnaies demeurées au Musée Rockefeller.20 However, more than fifty years had passed since Spijkerman’s involvement and all the protagonists of the digs in the early 1950s were dead. The dossier was definitively reopened in June 2011.21 The EBAF archive, consulted on more than one occasion, the last and most profitable between 16–22 July 2014,22 proved surprising. What was “found” met expectations: the dig diaries of the 1951–1958 18  Robert Donceel, “Reprise des travaux de publication des fouilles au Khirbet Qumrân,” RB 99 (1992): 557–73 (559) note n. 8, indicates P. Tournay as writer of that note. 19   A EBAF. KQ Trust: a note handwritten by Robert Donceel reads: “1 des deux chemises passeés au musée per le p. Tournay (à la demande du père Benoit), en 1977 (cf. RB 1980: 274: présentation de l’article de M.me Sharabani). Obtenu du père Piccirillo / Récupéré par Humbert et Donceel/ le 3/3/1988]/ Tous les doc. en provenant ont été marqué O [rouge] en haut à dr.] / [autre chemise, la 2me: X] À ne pas rendre / Il s’agit de papiers d’étude et de notes. Les listes (Qumran) font double usage avec celles dont copie a été exécutée par Sharabani (4/3/88). En autre, ceci concerne a grande partie Feshkha. Donné au musée en 1977 par erreur.” Therefore, in 1977 Piccirillo handed over to the Rockefeller Museum for reasons of scientific research: AS’s manuscript regarding the revision of the hoard [hereinafter: AS’s List.1] and the catalogue of single finds at Qumran [hereinafter: AS’s List.2] and Ain Feshkha. 20   Editorial introduction in Marcia Sharabani, “Monnaies de Qumrân au Musée Rockefeller de Jérusalem,” RB 87 (1980): 274– 84 (274). AEBAF. KQ Trust: handwritten note by R. Donceel that confirms this timing, ut supra. 21  For the EBAF, authorization by the then Director Hervé Ponsot and the archaeologist Jean-Baptiste Humbert; for the SBF by Eugenio Alliata, Director of the Archeological Museum of the SBF. In my archive copy of the letter dated 28 June 2011 requesting authorization, then given verbally without written formalities. 22  This date is obviously later than that of my lecture in Lugano, 21 February 2014 on the occasion of the International Conference: “The History of the Caves of Qumran,” 20–21 February 2014.

223 campaign were consultable both in the redaction/notes of RDV and in the typewritten version (catalogues), plus a substantial documentation of loose pages, manuscripts and letters all concerning the coinage of the 1951–1958 digs at KQ. There was a metal box containing typed cards with handwritten monograms, integrations and corrections, certainly attributable to the Dutch numismatist as his handwriting was recognizable. On the identification label, also this written by AS, was “K. Q. monnaies 1954– 1955” “Fiches 1956–1958.” A cursory appraisal led immediately to his classification of the single finds, in point of fact unpublished in that, if the long list/catalogue published in 200623 refers to this work, it is not supported by the necessary references card by card and the transcription, if it is a transcription, does not reveal the meticulous identification by the numismatist of the Studium. An initial synopsis of the materials from the SBFAS and EBAF archives, but most of all Jean-Baptiste Humbert’s personal memory threw unexpected light on the entire numismatic dossier of the RDV KQ-Digs.24 It emerged, for example, that the handwritten classification of the hoard of 561 tetradrachms by HS [Seyrig’s List], first cited by Sharabani,25 really did exist. However, even if it was the property of the EBAF, it had been on temporary deposit for some years at the Coin Cabinet of the IAA. The destiny was also gradually clarified of AS’s contributions to the study of the hoard from Locus 120 and the identification of the single finds. MP, with the intention of reuniting everything pertinent to KQ, from the coins to the handwritten studies, had transferred AS’s List.1 to the Rockefeller Museum and, on that occasion, also AS’s List.2 with other papers of his. Of these, however, the Studium had lost trace and nobody, in 2011, knew about the documentary dispersal of AS’s work. 23  Lönnqvist and Lönnqvist, “The Numismatic Chronology,” 146: “The order of the coins presented here is that of the loci in which the coins were originally found. de Vaux’s and Spijkerman’s identifications of Jewish coins were made according to Reifenberg 1947, abbreviated here R. [Adolph Reifenberg, Ancient Jewish Coins, Jerusalem 1947].” 24  Some information on the succession of interventions on the KQ numismatic dossier may be read in Lönnqvist and Lönnqvist, “The Numismatic Chronology,” but it clarifies neither the amount, nor the content and not even the availability of the documents to which it refers, which seems rather desultory within a research procedure claiming to be rigorous. 25  “H. Seyrig en fit l’inventarire provisoire en mai 1955. Cet inventaire d’une vingtaine de pages contient une section analytique avec les dates des monnaies (sicles et demi-sicles tyriens, tétradrachmes et didrachmes séleucides, deniers romains et un classement général par ordre chronologique.” Editorial introduction in Sharabani, “Monnaies de Qumrân,” 274.

224

CALLEGHER

Furthermore, even if intellectual property of the Studium, a very conspicuous part of AS’s numismatic dossier, recovered from the EBAF in 1988, was then deposited at the IAA together with Seyrig’s List.26 This was the state of the research between June 2011 and summer 2012: the documentary panorama for the coins from the RDV KQ-digs had become decidedly less confused, enough to induce me to accept—even if with some misgivings—Piccirillo’s invitation: to re-examine the entire dossier and publish the coins from those long ago digs.27 As mentioned, however, it was the research conducted between 16 and 22 July 2014 that unraveled the events during the intricate affair of the “coin finds” of the KQ area.28 1

The Dossier Found and Reassembled

This paper gives a first summary report prior to the publishing of the collection: coins found, manuscripts [Seyrig’s List; AS’s List.1; AS’s List.2], cards [metal box], various papers at the ASBF and AEBAF. For clarity, it is worth referring to some new archive acquisitions and presenting them according to the composition of the coin finds of KQ. In fact, in numismatic terms they can be divided in two groups: a hoard of 561 silver coins (the vast majority tetradrachms) and the single finds collected in the various loci and strata of the site (circa 690 coins, almost all of copper or copper alloy).29 26  I thank the Director of the SBF for the authorization to study the entire AS archive, even if at present not held at the Studium. 27   The invitation to participate in the Colloque Henri Seyrig (Paris 10–11 October 2013) organized by the Accademies des Inscriptions et Belles Lettre contributed to this decision: cf. Bruno Callegher, “Henri Seyrig, Auguste Spijkerman et les institutions de recherches à Jérusalem,” Syria, forthcoming. 28  This archival research integrates and clarifies what was presented in the lecture at the International Conference: “The History of the Caves of Qumran,” Lugano 20–21 February 2014. 29   For the hoard: Sharabani, “Monnaies de Qumrân;” Aida S. Arif, A Treasury of Classical and Islamic Coins: The Collection of Amman Museum (London: A. Probsthain, 1986), J 8110; Jean-Baptiste Humbert and Alain Chambon, eds., Fouilles de Khirbet Qumrân et de Aïn Feshkha. Vol. 1: Album de photographies; Repertoire du fonds photographique; Synthèse des notes de chantier du Père Roland de Vaux (NTOA.SA 1; Fribourg/ Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 1994): 329–30, then translated in Roland de Vaux and Ferdinand Rohrhirsch, Die Ausgrabungen von Qumran und En Feschcha: IA: Die Grabungstagebücher (NTOA.SA 1A; trans. and notes F. Rohrhirsch and B. Hofmeir; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 1996), 119–31 (120);

Many people have applied themselves to these two groups, essential for establishing the function of the site and the chronology in relation to the other archaeological finds, suggesting interpretations, but always irrespective

Lönnqvist and Lönnqvist, “The Numismatic Chronology,” 138–40; Kennet Lönnqvist, The Report of the Amman Lots of the Qumran Silver Coin Hoards: New Chronological Aspects of the Silver Coin Hoard Evidence from Khirbet Qumran at the Dead Sea (Amman: National Press, 2007). References to the single coin finds in Roland de Vaux, “Fouilles au Khirbet Qumrân: Rapport préliminaire,” RB 60 (1953): 83–106; idem, “Fouilles au Khirbet Qumrân: Rapport préliminaire sur la deuxième campagne,” RB 61 (1954): 206–36; idem, “Fouilles au Khirbet Qumrân: Rapport préliminaire sur le 3e, 4e et 5e campagnes,” RB 63 (1956): 533–77. RDV’s numismatic data, undoubtedly integrated by contributions of AS, were first picked up in: Ernest-Marie Laperrousaz, Qoumrân: L’établissement essénien des bords de la Mer Morte; Histoire et archéologie du site (Paris: A. & J. Picard, 1976), 151–54, in part. n. 5: “Les listes données dans la Revue Biblique, LXI, 1954, p. 230 et LXIII, 1956, p. 565 étaient provisoires, les monnaies n’ayant pas été toutes nettoyées et d­ éterminées. Préparant le catalogue définitif, le R. P. A. Spijkerman, O.F.M., a notablement complété ces listes et a changé un certain nombre d’attributions en faisant des meilleures lectures etc.;” Robert Donceel and Pauline Donceel-Voûte, “The Archaeology of Khirbet Qumran” in Methods of Investigations of the Dead Sea Scrolls and the Khirbet Qumran Site: Present Realities and Future Prospects (ANYAS 722; ed. M. O. Wise et al.; New York: The New York Academy of Science, 1994), 1–38 (3–6); Jerzy Ciecieląg, “Coins from the So-Called Essene Settlements on the Dead Sea Shores,” in Mogilany 1995: Papers on the Dead Sea Scrolls Offered in Memory of Aleksy Klawek (Qumranica Mogilanensia 15; ed. Z. J. Kapera; Cracow: Enigma Press, 1998), 105–15; de Vaux and Rohrhirsch, Die Ausgrabungen, 119–31; Robert D. Leonard, “Numismatic Evidence for the Dating of Qumran: A Review Article,” QC 7 (1997): 225–34, nn. 3, 4; Jean-Baptiste Humbert and Alain Chambon, eds., The Excavations of Khirbet Qumran and Ain Feshkha: Synthesis of Roland de Vaux’s Field Notes (NTOA.SA 1B; trans. and rev. Stephen J. Pfann; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003); Catherine M. Murphy, Wealth in the Dead Sea Scrolls and in the Qumran Community (STDJ 40; Leiden: Brill, 2002), 305–17; Hanan Eshel and Magen Broshi, “Excavations at Qumran, Summer of 2001,” IEJ 53 (2003): 61–73; Lönnqvist and Lönnqvist, “The Numismatic Chronology.” For the finds from the post RDV dig campaigns, not considered here as they are not strictly pertinent to the subject of the paper, Ya⁠ʾakov Meshorer, “The coins from Qumran,” INJ 15 (2006): 19–23; Yoav Farhi and Randall Price, “The Numismatic Finds from the Qumran Plateau Excavations 2004–2006, and 2008 Seasons,” DSD 17 (2010): 210–25; Donald T. Ariel, “Coins from Excavations at Qumran,” coin report unpublished, updated 2.11.2010: I thank my colleague for permission to consult this.

225

THE COINS OF KHIRBET QUMRAN FROM DE VAUX ’ S DIGS

of the realien, the coins, considered irretrievable,30 and the archives. 1.1 The Hoard from Locus 120 On 21 March 1955, two hoards were found in Locus 120 at KQ (they are recorded in the dig diaries with the numbers 2543, 2545): “À droite de la porte, en entrant, en creusant sous le sol supérieur, on trouve deux petits pots remplis de monnaies tyriennes en argent.”31 The next day “un troisième tésor est recueilli contre le mur nord, dans une cruchette dont le col est trop étroit pour introduire les piéces ou la cruchette était déjà cassée” (assigned the no. 2547);32 all three had been concealed at “un niveau inférieur” than the “sol du locus,” which means that they were in the same stratigraphic position. A few weeks later, the archaeologist RDV who was in charge of the dig, chose HS for their identification. He was the only possible choice, as he was at that time the most competent numismatist and interested in the Hellenistic coinage, especially Seleucid, of the whole Near East. Thus, in the month of May, HS tackled the finds, maintaining a rigorous distinction and indicating each of the three parts as Lot A, B, C, the same as those attributed by RDV. The scholar, doubtlessly through first-hand experience of finds and the examination of coin hoards, had guessed that this was a concealment distributed in three different points, but of the same hoard of coins, as the terracotta 30  Donceel and Donceel-Voûte, “The Archaeology,” 4 wrote: “Those he was able to read [AS], after treatment finally numbered 691 items. We have been able to read no more than 196 (2/7ths) of these, which we photographed, weighed, and described. Of the still missing coins no trace is to be found at the École Biblique or at the Rockefeller Museum.” It can therefore be deduced that the two authors must have seen at least a part of the coins kept at the SBF; a synopsis would be desirable between the photographic material in their possession (ut scripserunt) and what has now been recovered. The number they indicated, ca. 196 specimens, compared with the 424 specimens found on 14 July 2014 in the coin cabinet of the SBF (196+424=620), would give ca. missing 70 coins for the re-composition of the collection. This was followed by Lönnqvist and Lönnqvist, “The Numismatic Chronology,” 121, n. 3, that provides a preliminary and nullifying fact: “Information on the coins varies, but they are not at the École Biblique et Archéologique Française, Jerusalem, the Israel Antiquities Authority, Jerusalem, or the Studium Biblicum Franciscanum, Jerusalem.” 31  Humbert and Chambon, Fouilles de Khirbet Qumrân, 329–30. At the AEBAF. KQ Trust, Photo Files: 450/Pl. XXXVII, nos. 13.595– 13.599 show coins and ceramic containers of Lots A–B found in Locus 120, as they were shortly after discovery. 32  I have not verified the existence of Photo Files for Lot C at the AEBAF. KQ Trust.

containers were of the same type and the Locus was rather cramped, probably with only one entrance.33 Of that collaboration there remains a letter he wrote to RDV, dated 12 May 1955, which I reproduce here because of its pertinence and the profile of the scholar that emerges. Mon révérend Père, de retour à Beyrouth, j’ai jeté un coup d’œil sur mes listes de trésor de Qumran, et les ai collactionnées avec mes listes de monnaies, empruntées aux principales collections. Le trésor de Qumran, par un bon nombre de variétés nouvelles, complète notablement notre connaissance de monnayage tyrien pour la période qui précède immédiatement l’ère chrétienne ; en outre il nous informe très heureusement de ce qu’était alors la circulation monétaire en Palestine – sans que l’on puisse dire qu’aucune des variétés en question présente par elle-même un intérêt particulier. Pour ce qui est de la date d’enfouissement, la pièce décisive est le tétradrachme – jusqu’ici unique – de l’an 118 de Tyr (9/8 av. J.-C.). Mais cette date est seulement approximative, car l’an 118 est suivi d’une lacune insolite dans les émissions : on ne connait jusqu’ici aucune monnaie d’argent des années 119 à 122, ni 124, cependant que 123 et 125 ne me sont connues chacune que par un seul exemplaire. Ce n’est qu’en 126, que l’atelier reprend des émissions abondantes, qui seraient certainement représentées dans le trésor si celui-ci eut été enfoui après cette date. L’enfouissement a donc dû se produire vers 126 au plus tard (1 av. J. C./1 ap. J.-C.). – Je suis d’autant moins enclin à voir dans ces faits une illusion, résultant du hasard des trouvailles, que plusieurs pièces de votre trésor ont justement été contremarquées en 118, et ce, sans nul doute, à Tyr. Cette pratique trahit elle aussi les difficultés de la trésorerie : au lieu de battre des espèces nouvelles, la ville a remis en circulation, tout en les validant par une estampille, des vielles espèces royales qui n’auraient sans doute plus eu cours légal à Tyr sans ce poinçon – bien qu’elles dussent encore circuler en fait dans le public. – Je chercherai, quand j’en aurai le loisir, si la raison de ces difficultés peut encore être conjecturée. Les listes un peu hâtives que je viens d’établir contiennent visiblement quelques erreurs de détail, sans quoi je vous en enverrais une copie. Si elles peuvent vous rendre service, je vous les remettrai après les avoir révisées à mon passage à Jérusalem 33  Humbert and Chambon, Fouilles de Khirbet Qumrân, 96.

226

CALLEGHER

l’automne prochain. Entre nous soit dit, le classement de Harding est fort sujet à caution, et il y aurait intérêt à ce que l’inventaire que vous publierez soit aussi correct qu’on peut le demander, devant des exemplaires souvent très usés, et nettoyés avec grande énergie. Laissez-moi saisir cette occasion de vous remercier encore, vous et les autres Pères de S. Etienne, de l’accueil si aimable que je viens de recevoir. Ce m’est toujours une joie, de retourner dans votre admirable maison, où ne lient des souvenirs qui me tiennent tant à cœur. Veuillez croire, je vous prie, à mon très sincère attachement. Henri Seyrig. From contents of the letter (almost a micro treatise on numismatics in which crucial subjects are tackled such as the interruption of issues and the putting back on the market of countermarked specimens, the difference between legal tender and circulating currency), it can be deduced that HS examined the hoard after something similar had already been done by Lankester Harding (1901–1979),34 that the specimens had been very heavily restored/cleaned, that many showed visible signs of the wear and tear caused by circulation, but most especially that the notes made in Jerusalem formed a work in progress, to be organized prior to the delivery of the definitive classification fixed for the next trip to Jerusalem in autumn 1955. It is therefore possible that the Seyrig’s List consultable today is the outcome of the definitive elaboration and that the preparatory papers are held in the HS Trust at the Cabinet des Médailles—National Library in Paris.35 As to the manuscript available in Jerusalem, the previously mentioned Seyrig’s List, it is clear and exhaustive. It consists of 30 pages, divided by the author in two sections, the first from pages 1 to 12, the second from pages 1 to 18, but with the addition of a complementary numbering from 13 to 30 to provide continuity. On pages 1 to 12 he recorded, respecting the division of the three groups,36 the various denominations specifying the issuing authorities, monograms and Semitic letters, 34  Gerald Lankester Harding in that period was Director of the Department of Antiquities of Jordan-Amman and represented the Antiquities Service during the digs at KQ: Humbert and Chambon, Fouilles de Khirbet Qumrân, 404; see also Michael Macdonald, “A Bibliography of Gerald Lankester Harding,” ADAJ 24 (1980): 8–12. 35   Non vidi. I thank the current Conservator of the Cabinet des Médailles, Frédérique Duyrat, for the indication of documents in the cited trust regarding this hoard of tetradrachms- didrachms. 36  On the top left of the first sheet, HS wrote: “Trésor de Qumran. Inventaire des trois lots, établi à Jérusalem par moi, mai 1955.” With critical scrupulousness, he uses the first person singular.

the attributions to Tiro but with non-identifiable authorities. On page 5 he classified six Roman denarii (and only six). Pages 13 to 30 (also 1–18) reconstruct the set ordered by denominations and monograms/mint signs and group the similar specimens present in the three lots. On page 27 numerous coins are listed with “Dates illisibles et impossibles à restituer” or “Pièces complètement illisibles” but always with the reference to “tétradrachmes/didrachmes.” Pages 28–29 reunite the “Monnaies séleucides (variétés non ment. [tionnées] par Newell,37 soulignées en rouge,” while on the last page he classified the six Roman denarii accompanied by the citation of the list in Sydenham, at that time the standard work for the Roman Republican coinage.38 The position of the find in the Locus, its stratigraphy, the type of containers and the chronological coherence of the set were all elements that suggested to the scholar that it was a single hoard. The annotations reported on the manuscript also allow it to be deduced that for HS the definition “illisible” indicated, at that moment and in the state of conservation of the coins, the impossibility of identifying some essential data (monograms, Semitic letters, legends) but not the monetary type. He also verified the numerical coherence of the three groups crosschecking it with the dig diaries. The first lot was of 222 coins (Lot A), the second 186 coins (Lot B), the third 153 coins (Lot C). These figures differ slightly from what is found in the literature, where 223 coins are attributed to Lot A and 185 coins to Lot B.39 M. Sharabani gave the first account of what happened prior to his study:40 “Subsequently, the hoards [rectius: the lots] were mixed together, and what the original hoard contained can only be determined according to the preliminary list compiled by the late H. Seyrig.” Perhaps relying on it being a single hoard, RDV united the three parts to then divide them in two segments: the first of 153 coins was delivered to the Rockefeller Museum,41 the second with the remaining 408 coins went to the Amman 37  Edward T. Newell, The Coinage of the Eastern Seleucid Mints: From Seleucus I to Antiochus III (Numismatic Studies 1; New York: American Numismatic Society, 1938); idem, The Coinage of the Western Seleucid Mints: From Seleucus I to Antiochus III (Numismatic Studies 4; New York: American Numismatic Society, 1941). 38   Edward A. Sydenham, The Coinage of the Roman Republic (London: Spink & Son, 1952). 39  Lönnqvist and Lönnqvist, “The Numismatic Chronology,” 122. 40  Sharabani, “Monnaies de Qumrân,” 275. 41  The analogy between the 153 coins of Lot C with those delivered to the Rockefeller Museum, then published by Sharabani, it is not sufficient to hypothesize that Lots A and B were instead delivered to the Museum in Amman.

THE COINS OF KHIRBET QUMRAN FROM DE VAUX ’ S DIGS

Museum. It was without doubt a controversial decision and the basis of the current difficult reconstruction of the set. However, Seyrig’s notes and list existed. Shortly after, in 1960, AS attempted to recompose the set.42 I limit myself to mentioning that for the hoard of Locus 120 he not only examined the holdings in the Rockefeller Museum, but requested the autopsy of the collection, perhaps travelling initially to Amman, certainly arranging that as many of the specimens as possible should come to the SBF Museum. In fact, on 19 November 1960, Adnan Hadidi, Keeper of the Museum in Amman, delivered to him in Jerusalem, at that time in Jordan, a part of the coins held in that museum: “87 shekels de Tyr, 17 tétradrachmes de Démétrius II, 33 tétradrachmes d’Antiochus VII, 53 sicles séleucides, 6 deniers consulaires romains.”43 Of the AS numismatist’s classification much of whose method derives from Seyrig’s List, AS’s List.1 survives and many tens of photographs, each one marked with the lot it belongs to and an identification number.44 This 50-page manuscript of his which, as mentioned, passed first of all to the Rockefeller Museum, then to the EBAF and lastly to the IAA, records the following data in tabular form: progressive number, distinction between hoard A/B/C, chronology relative to the Tiro dating system, conventional BCE chronology, monograms, Semitic letters, metrological data (diameter, weight, mint), place of conservation of the coin: A (= Amman), J (= Jerusalem) and comments. Some of the lines are only partly complete because they refer to coins recorded by HS, but not examined directly by AS, i.e. the part of the hoard that did not arrive from Amman. After having pointed out the numerical correspondence with Seyrig’s List, the numismatist of the SBF summed up his work on an extraordinarily important page from which an essential fact emerges: that both, at a distance of time, classified six and only six Roman denarii, also agreeing on the identification (Photo 1). Some years later, these two protagonists having died, Sharabani could use the two classifications to publish the holdings at the Rockefeller Museum: “A comparison with Seyrig’s list shows that most 42  Sharabani, “Monnaies de Qumrân,” 275: “In 1960, the late Father A. Spijkerman studied the 402 coins [the number differs from the 408 consigned] he found in Amman and the 153 coins in Jerusalem.” 43  This is written in the letter of transfer dated Amman 19/11/1960, held in the ASBF. AS Trust, A. Spijkerman. Corrispondenza 1956– 1973 † 23.06.73 [fasc. 8]. As mentioned, the original document that explains part of this affair is held at the AEBAF. KQ Trust. 44  I have not done any research at the National Archaeological Museum of Amman, but it is at least possible that documentation is held there which AS attached to the coins when re-­ consigning them after finishing the examination in Jerusalem.

227 of the shekels and half-shekels in Jerusalem belong to Hoard A, while the rest of Hoard A [---] was transferred to Amman.”45 But between the intervention of AS and that of Sharabani the holdings in Amman had been tampered with more than once, as well summarized in a report by an uncertain author,46 in a note by the numismatist Christian Augé in the 1980s47 and in a letter from Gus W. Van Beek of the National Museum of Natural HistorySmithsonian Institution of 8 April 1988, with which he sends to the EBAF the entire documentation relating to the loan of 50 specimens for the Exhibition on the Dead Sea Scrolls at Washington in December 1964.48 Of the 408 specimens, 351 were available in the 1980s at Amman: 50 had disappeared after the return from the exhibition in Washington,49 2 had passed to the Archaeological 45   Sharabani, “Monnaies de Qumrân,” 275. Lönnqvist and Lönnqvist, “The Numismatic Chronology,” 138, seem to interpret the holdings at the Rockefeller Museum of Jerusalem as the whole of Lot C: “Hoards A, B, C contained 223, 185 and 153 silver coins respectively, a total of 561, according to H. Seyrig. We have information on the exact composition and latest coins for only the Jerusalem lot of 153 coins, the only one of the three to have been published”. 46   A EBAF. KQ Trust, sheet signed Michel and post-dated to 15 November 1980 by Robert Donceel. It was probably starting from this report that Donceel suggested an investigation to reconstruct the intactness of the holdings in Amman and of which he writes in “Reprise des travaux,” 559–60. 47   A EBAF. KQ Trust. The notes probably date to 1984. Two passages are particularly relevant. The first: “Argent romain 12: 4 deniers républicains (2e et 1er s. a. J.-C.) / 8 dénier (et drachmes) impériaux: sans doute à exclure du trésor;” the second, a summary: “Les monnaies de la Cittadelle (349 en tout) sont groupées en 12 lots (d’importance inégale), dans des enveloppes: ces lots correspondent à 12 photos (le droit et le reverse). Le numéro d’inventaire commun est J 8110 (1 ò 12).” 48   A EBAF. KQ Trust. Letter to J.-B. Humbert of 8 April 1988: “I am enclosing copies of (1) the actual case and panel in which the coins were exhibited throughout the tour, (2a–d) letters pertaining to the last showings of the exhibition at the British Museum, (3) my letter to Fawzi Zeyaddin in 1981, and (4) my memory note based on a telephone call from State Department dating which a confidential telegram was rend to me containing this information: in the event that you can’t read my handwriting, it states: Wes Adams, Chargé d’(affairs) in Amman (cabled that the) Dead Sea Scrolls (loan material) arrived in perfect condition June 16 (1966). Open(ed) and o.k. in June 18. Finis !!!”. All these documents can be consulted at the AEBAF. KQ Trust and testify that all the coins had been returned and were dispersed after that date. 49  Nevertheless, Donceel, “Reprise des travaux,” 559, declares [but does not cite any documents] that “grâce à l’aide du personnel du cabinet numismatique du musée Rockefeller, nous avons pu retrouver les 50 monnaies du lot ‘Amman’ portées disparue

228 Museum of Jordan University in Amman, 3 were gifted to the president of the USA, Richard Nixon [who, in accordance with American law, had passed them on to the Smithsonian Institution], all trace of 2 has been lost. In 1985–1986 the Amman holdings were examined for the publication of a book, largely photographic. In it the distribution of the KQ material is not clear, but the photographs,50 especially two,51 are indispensable for any reexamination of the entire hoard and especially to define the incursions sustained since the early 1960s and ascertained by Christian Augé.52 Despite the existence and importance of HS’s List and AS’s List.1 being known, they were not used by the successive scholars, who tackled this numismatic documentation in both narrative53 and more strictly numismatic terms.54 Disregarding them was poor methodology.55 To acknowledge their precision and classification method, almost an anticipation of the criteria to which the Fundmünzen der Antike contemporarily (and independently) conformed,56 would have raised doubts, especially regarding the necessity to confirm or refute, but on documentary evidence, all the information given there. In other words, it would have been necessary to compare the coins and manuscripts, accept or refute the overall number in the three groups and especially the identification of the coins, of all the coins, carried out by two numismatists, to then demonstrate, again with reference to the realien and not merely allude to the fact that that the two scholars could not count but most of all that they had confused the monetary series. Since in the recent

au retour d’une tournée au USA puis en Europe;” he also states that he had available the “listes autographes de H. Seyrig.” This should date to 1986–1988. 50  Arif, A Treasury, 192–205. 51  Arif, A Treasury, 204–5. 52  This reconstruction and the incursions are reported firstly in Donceel and Donceel-Voûte, “The Archaeology,” 4, n. 12 and later, with new data, by Lönnqvist, The Report of the Amman Lots, 3, n. 5, but without precise references to the documentation in the EBAF archive. 53  Donceel and Donceel-Voûte, “The Archaeology.” 54  Lönnqvist and Lönnqvist, “The Numismatic Chronology,” 138– 40; Lönnqvist, The Report of the Amman Lots. 55  Lönnqvist, The Report of the Amman Lots, 3: “H. Seyrig’s list is the only source to the original composition of the hoards (not accessible for this study).” The preceding literature, e.g. Sharabani, “Monnaies de Qumrân,” also mentioned the existence of an analogous work by AS. 56  Hans Gebhart et al., “Bemerkungen zur kritischen Neuaufnahme der Fundmünzen der römischen Zeit in Deutschland,” JNG 7 (1956): 9–76.

CALLEGHER

reconstruction of the hoard of KQ-Locus 120,57 one of the tenets is considered to be the documented presence (had it really eluded HS and AS?) of “Roman denarii (12) and one Nabatean silver coin,” the study of the last two plates in the photographic volume of 1986 would have been useful at least in raising doubts and connecting what was reproduced there both to Seyrig’s List and AS’s List.1, and to the reports of the intrusions into the hoard over the years.58 In the meantime, in fact, the Roman denarii from six had become twelve and a Nabatean silver coin had been added, which would also have eluded the first examination,59 and as the author-editor declares: “There are neither six Republican denarii presently in the three published hoards but only four (at least only four exist in Amman); all the four Roman Republican denarii are very early indeed and represent the earliest specimens recovered in whole area of Syria-Palestine.”60 The Amman Museum photographs61 would therefore have clarified that the four coins, Roman Republican denarii, were perfectly compatible with the four denarii reported in HS’s list: Sydenham 417= J 8110 Ob.12/Rev. 12: n. 18 = RRC 274/1;62 Sydenham 661= J 8110 Ob.11/Rev. 11: n. 29 = RRC 340/1; Sydenham 540a = J 8110 Ob.11/Rev. 11: n. 30 = RRC 302/1; Sydenham 723= J 8110 Ob.11/Rev. 11: n. 28 = RRC 350A/2. Of the list edited by the archaeologist-numismatist of Beirut only two denarii were missing: Sydenham 455=RRC 233/1 and Sydenham 1181=RRC 517/2,63 an absence explainable, lectior facilior, by the reported intrusions after 1960.64 Also the observation that these denarii “represent the earliest specimens recovered in whole area of Syria-Palestine” would have required a supplementary literature search to place them within the context of the monetary circulation in that period. In fact, again in the same volume, in the photographs assigned the number J 8109, the coin no. 17 is a Republican denarius

57  Lönnqvist, The Report of the Amman Lots 58  Arif, A Treasury, 202–5. 59  Lönnqvist, The Report of the Amman Lots, 9: “The components in the Amman hoard are: the Seleucid tetra-and didrachms (90), the Tyrian tetra-and didrachms (251) and the Roman denarii (12) and one Nabatean silver coin.” 60  Lönnqvist, The Report of the Amman Lots, 5. 61  Arif, A Treasury, 202–5, with reference to no. J 8110. 62   R RC = Michael H. Crawford, Roman Republican Coinage (Cambridge: Cambridge University Press, 1974). 63  My classification, based only on photographs, requires verification with the real coins. 64  Donald T. Ariel and Jean-Philippe Fontanille, The Coins of Herod: A Modern Analysis and Die Classification (AJEC 79; Leiden: Brill, 2012), 34, Pl. 2: these denarii are kept at IAA.

THE COINS OF KHIRBET QUMRAN FROM DE VAUX ’ S DIGS

of P. Laeca (110–109 BC) = RRC 301/1.65 Other Roman Republican denarii, what’s more found in an area close to that of KQ, had been known since 1958: Sydenham 595 = RRC 336 (C. Allius Bala); Sydenham 976 = RRC 463 (Mn. Cordius Rufus); Sydenham 1216, 1217, 1221, 1227, 1330, 1243, 1244 = RRC 544/14, 15, 18, 23, 26, 36, 37 (M. Antonius).66 In summary, the collation of the HS and AS manuscripts with the photographs held in the SBF archive (Photo 2), with the dossier in the EBAF archive and with the references deducible from the book of photographs published in 1986 makes the interpretations of the hoard dubious, as proposed in its most recent revision, but interesting from the numismatic point of view.67 1.2 The Single Finds The second segment of the KQ coin finds, the single finds, recovered in the various areas and strata during the 1951–1956 campaign have an even more obscure history. Whoever wrote this would have to immediately inform the scientific community that the EBAF archive holds substantial documentation signed by RDV on the identification of these coins, subdivided by Loci, chronology, issuing authorities and with specific attention to the presence of specimens of Herod the Great. It was probably compiled during the dig campaign for an initial chronological guidance, using the Reifenberg catalogue. These papers include a “Statistique monnaies/époque et essai de graphique avant fin de fouilles,”68 produced prior 65  Arif, A Treasury, 150–51. 66  Józef T. Milik and Henri Seyrig, “Trésor monétaire de Murabbaʿât,” RN 6/1 (1958): 11–26 (23, nn. 1–9). For other finds, probably successive, after Michael van Lassen, “The Coinage Pattern of Palestine in the First and Second Centuries A.D.,” NCirc 79/6 (1971): 240–41, see other data from Hebron: Coin Hoards (ed. P. J. Martin; London: Royal Numismatic Society, 1975), 1:38, n. 90, picked up in Kevin Butcher, Coinage in Roman Syria: Northern Syria, 64 BC–AD 253 (London: RNS, 2004), 273, n. 28; from Hebron, but without a precise location: Butcher, Coinage in Roman Syria, 274, n. 33a; from Beit Guvrim/Eleutheropolis: Butcher, Coinage in Roman Syria, 273, n. 30. The acceptance of the Roman system of accounting, in particular of the denarius, was handled by Christopher Howgego, “The Relationship of the Issar to de Denar in Rabbinic Literature,” INJ 8 (1984–1985): 59–64. 67  Lönnqvist, The Report of the Amman Lots. See also the review by Jean-Baptiste Humbert, “Activités de l’École Archéologique Française de Jérusalem. L’archéologie de Qumrân. Le dossier numismatique,” École Biblique et Archéologique Française de Jérusalem. Rapport d’Activités 2007–2008, Jerusalem: EBAF archives: 48–57 (53–56). 68   A EBAF. KQ Trust: 23 handwritten sheets of different sizes. It is possible that RDV consulted a colleague in this phase, but as will be explained, the support of AS can be excluded.

229 to the first published information in 1954: “Environ 250 monnaies proviennent de presque tous les loci de tous les niveaux. Elles étaient si oxydées que, sauf de rares exceptions, elle ne sont devenues lisibles, qu’après un long traitement chimique.”69 A second contribution two years later is also of interest, in which RDV listed the coins found during the 1954–1956 digs. In this, the coins are divided according to the chronology of the site proposed by RDV himself, explainable only by the fact that he needed the numismatic data to date the strata in a sort of automatic correspondence between chronology of the money and that of the archaeological stratum. The scholar seems to consider neither the difference between period of issue and time of its being lost, survival of the oldest specimens as they could fill some role within an accounting system in response to a specific demand for coins or values not or no longer represented by the circulating currency, nor wellknown phenomena such as monetary residuality, operating in particular in phases of transition or change of the denominations. The coins were thus grouped in Période I (from John Hyrcanus to Herod the Great), Période II (from Herod Archelaus to 68/69, third year of the first AntiRoman Revolt), Période III (issues slightly after 67/68).70 Sometime later AS took over or was delegated. I have not been able to reconstruct with the help of the archives the exact time of the entrusting of the analytical study of these coins to AS, but this happened at different times to those commonly accepted, probably from 1958, when a collaboration began with Jean Starcky (1909–1988)71 of the EBAF to save a hoard that had appeared on the antiquarian market in Jerusalem, which they later published 69  De Vaux, “Fouilles au Khirbet Qumrân: Rapport préliminaire,” 229–30. 70  This was a method widespread among all archaeologists of the period, so it does not seem right to make a negative evaluation. In this regard, for historical reasons, the statements by K. and M. Lönnqvist cannot be agreed with, in “The Numismatic Chronology,” 121, where they declare: “However, the lack of final excavation reports, the many misinterpretations of the coinage in the preliminary reports by the original excavator and later by non-archaeologists (and also archaeologists) and non-numismatists, and the dispersal and disappearance of most of the bronze coinage have hampered the production of a comprehensive numismatic synthesis, which is attempted here for the first time.” Said in this way it is not clear to whom the authors refer: to de Vaux? To Henry Seyrig? To Augustus Spijkerman? In the 1950s, the qualification of archaeologist and numismatist for scholars, many of whom had interrupted their studies to serve in the war, cannot be referred to today’s criteria, which have been defined very slowly over decades. 71  Emile Puech, “In memoriam. L’Abbé Jean Starcky (1909–1988),” RevQ 53 (1989–1990): 3–6.

230

CALLEGHER

together.72 The accreditation perhaps came about in this way and in a short time the Franciscan became the reference numismatist for the archaeologists of the EBAF, who assigned him to study all the coin finds: hoards and single coins. However, it was his collaboration that generated confusion over time—of which he was certainly unaware. In order to clarify this some documentary evidence is proposed below.73 First of all, it is not true that the coins were handed over to the Franciscans, in particular to AS, after every dig: Pendant les campagnes des fouilles, les monnaies enregistrées sur le site étaient déposées chaque semaine au Studium Biblicum Franciscanum (SBF) de Jérusalem puisque c’est là que le P. Spijkerman s’employait à les décrasser et à en faire une première lecture. Les collectes étaient chaque année déposées par le SBF au siège des Antiquités, le Palestine Archaeological Museum (Rockefeller Foundation), lequel faisait fonction de succursale pour Jérusalem du Service des Antiquités de Jordanie. Le SBF ne conserve que les quelques dizaines de monnaies issues de l’ultime campagne de 1956 à Aïn Feshkha. Il faut accepter que les monnaies ont disparu de Musée Rockefeller [---].74 If this had happened, RDV would have mentioned it in his report of 1956, where he lists and thanks various collaborators.75 AS only arrived in Jerusalem in August 1954 and his venture into numismatics was gradual, certainly not during 1955, perhaps only from 1957–1958, when the digs at KQ were over.76 One of RDV’s pages would indicate the year 1959,77 followed by a greater involvement from 1960

72   A SBF, Spijkerman Trust, fasc. 10, sheet IV, 1; Augustus Spijkerman and Jean Starcky, “Un nouveau lot de monnaies palestiniennes,” RB 65 (1958): 568–84, Pls. XVIII–XX. 73  Donceel and Donceel-Voûte, “The Archaeology of Khirbet Qumran,” n. 15; Lönnqvist and Lönnqvist, “The Numismatic Chronology,” 121; Lönnqvist, The Report of the Amman Lots, 3. 74  Humbert, Activités, 54. Also in this case I can testify that this fichier exists, although I do not know whether it is the original or a photocopy, at the Israel Antiquities Authority, according to information from D. Ariel, whom I thank for his kindness. 75  De Vaux, “ Fouilles au Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e et 5e campagnes,” 533–34. 76  The biographical-scientific story of AS is retraced, with unpublished documents from the archive, in Bruno Callegher, “Note su Augustus Spijkerman numismatico (1920–1973),” LA 64 (2014): 615–47. 77   A EBAF. KQ Trust: “Monnaies/Communication Spijkerman 25/11/1959.”

onwards.78 The first two contributions of AS could be the “Provisional Catalogue of the coins found at Kh. Qumran” in which are identified, with references to the numbers of the dig diary and the Reifenberg catalogue,79 the Seleucid, Asmonean, Nabatean coins and of the local mints starting from Ascalon80 and the summary regarding the presence of coins of Herod the Great in the 1951–1955 digs.81 What remains of this work, as well as the previously-mentioned preparatory papers, is the 26-page manuscript [AS’s List.2], a sheet with the prudent caption: “Lot of 18 coins, said to have been found at Kh. Qumran” and probably the cards in the metal box that, if he did not type them, he certainly thoroughly examined and completed with signed comments. In AS’s List.2, in tabular form, the coins are listedclassified in a concise way, with a division that follows the years of the digs and reference to the numbers in the dig diary: “KH.Qumran. 1951: 102–124,” “KH.Qumran. 1953: 147–912,” “KH.Qumran. 1954: 972–1667,” “KH.Qumran. 1955: 2001–2650,” “KH.Qumran. 1956: 2666–2672.” Each of these tables reports: a progressive number, the number of the dig register, a bibliographical reference relative to the coin, the position in the site/Locus, the dating BC–AD and a column for any notes. The preliminary cross-check between the manuscripts of RDV and AS, the bound catalogues of the dig,82 the cards and the preserved coins have confirmed that they are two types of interconnected documentation and therefore corroborate one another. Thus the first coin in the first container held at the SBF bears the number 147. It is an AE minted in the second year of the first Anti-Roman revolt (67 CE).83 It corresponds to the coin in AS’s List.2-digs 1953, to the coin described on the card with the same number in the metal box and to 78   A EBAF. KQ Trust: “Monnaies de Khirbet Qumrân / Identifications au 28/2/1960/ R. P. Spijkerman,” followed by “Spijkerman 2 mars 1960 Monnaies de Qumrân.” These, like almost the entire numismatic dossier, were checked by R. Donceel who wrote many notes in pencil on the originals. 79  Adolf Reifenberg, Ancient Jewish Coins, (Jerusalem: Rubin Mass, 1947) was at that time the reference catalogue. 80   A EBAF. KQ Trust: 4 signed sheets of AS. 81   A EBAF. KQ Trust: 2 signed sheets of AS. 82   A EBAF. KQ Trust: typewritten and bound files, entitled “Fouilles de Khirbet Qumrân”. 83  Ya⁠ʾakov Meshorer, Sylloge Nummorum Graecorum: The Collection of the American Numismatic Society; Part 6; Palestine-South Arabia (New York: American Numismatic Society, 1981), nos. 427 onwards; Ya⁠ʾakov Meshorer, Gabriela Bijovsky, and Wolfgang Fischer-Bossert, eds., Coins of the Holy Land: The Abraham and Marian Sofaer Collection at the American Numismatic Society and the Israel Museum (Ancient Coins in North American Collections 8, 2 vols.; New York: American Numismatic Society, 2012), 273–74.

THE COINS OF KHIRBET QUMRAN FROM DE VAUX ’ S DIGS

the citation in the list published in 1996.84 The successive n. 155 indicates a prutah of Alexander Jannaeus85 and also corresponds to the other three sources.86 There is analogous coherence between number of the coins, identification and literature sources for no. 2396 of the second container (various coins of Alexander Jannaeus) and for nos. 164, 174, 176, 177, 183, 184, 186, 199, 200, 212, 272–275, 290, 292, 309, 330, 331, all prutot of Jannaeus.87 To continue with the account of the “loss” of the single finds of KQ, two international appeals are surprising, the first in 1992 the second in 1994. In the first it is declared that their number [---] fut fixé à 691. De ces dernières [---] nous avons pu retrouver 196 monnaies; nous les avons photographiées, décrites, et pesées. Comme aucune trace des monnaies manquantes n’a été vue à l’École biblique ou au musée Rockefeller, nous avons écrit à des cabinets numismatiques qui, au Proche-Orient et en Europe paraissaient susceptibles de les avoir recueillies. Nous saisissons l’occasion de cet article pour lancer un appel à quiconque pourrait nos informer.88 This very extraordinary international appeal was repeated in 1994: We take this opportunity to appeal to anyone who might possess information on what happened to these coins.89 The perplexity can be deduced from a consultation of the archives. The first document is found in the first file containing the coins of KQ. It is a letter dated May 1988 by Robert Donceel who signed for the removal of 10 coins.90 The second is a letter of 20 May 1989 in which Donceel declares the removal of 200 coins from the single finds at KQ for a more accurate weighing.91 In both cases the coins 84  Also verified in de Vaux and Rohrhirsch, Die Ausgrabungen, 126 “Erster Aufstand, (68–69 nr. Chr.), 7.0.” 85  Meshorer, Sylloge Nummorum Graecorum, nos. 56 onwards; Meshorer, Bijovsky, and Fischer-Bossert, Coins of the Holy Land, 249–55. 86  De Vaux and Rohrhirsch, Die Ausgrabungen, 122, 7.0. 87  De Vaux and Rohrhirsch, Die Ausgrabungen, 121. 65.0 “19 Münzen.” 88  Donceel, “Reprise des travaux,” 560. 89  Donceel and Donceel-Voûte, “The Archaeology,” 4. 90   A SBF. AS Trust. 91   A EBAF. KQ Trust. On EBAF headed notepaper, is written: “Il sottoscritto, professore all’Università di Lovanio (Louvain la Neuve, Belgio) riconosce di aver preso in con[segna] nr. 200 monete di Khirbet Qumran in deposito presso il Stidium (sic!) Biblicum

231 were never returned to their original position92 and their restitution, given the appeals of 1992 and 1994, would represent the necessary response. The cited documents and contradictions contained in the articles of the two amateurs (for example the claim that they had found, weighed, photographed and identified 196 coins, without a topographical reference) demonstrate that between 1980 and 1988 Donceel—Donceel-Voûte knew of the existence of the coins, knew where they were held and had been able to examine them. There is therefore a lack of credibility when the two wrote some years later without any reference to the SBF and especially to the cited note of MP: A more painful matter is that of the “excavation coins” [----] . . . Of the still missing coins no trace is to be found at the Ecole Biblique or at Rockefeller Museum; we have therefore written, with no result yet, to a series of numismatic collections which, in Europe and the near East, seemed likely to have acquired part of all of the 5/7 the still missing.93 Recently, twenty years after those appeals, a long list/catalogue of the single coins of KQ has been published. The authors, however, take their disappearance for granted, listing the identifications of RDV and AS94 without making any reference to their manuscripts nor to the cards in the metal box nor to the archival documentation in the EBAF and SBF and obviously to the actual coins.95

Franciscanum di Gerusalemme, nelle (sic) scopo di farne effet[tua]re un pesaggio accurato con attrezzatura specifica non disponib[ile] nel convento dei padri francescani. Fatto a Gerusalemme, li 20/5/1989.” 92  I personally verified the non-return of the 10 specimens to the SBF; for the other 200 coins, apart from my verification, this is proved by Michele Piccirillo’s signed annotation to the cited letter of 20 May 1989: “Les monnaies ne sont jamias rentrées chez nous [SBF] Jérusalem 24.3.92.” 93  Donceel and Donceel-Voûte, “The Archaeology,” 4–5; also doubts about the statement on p. 5, n. 15: “The archives of A. Spijkerman contain nothing that look the written results of his research.” 94  Lönnqvist and Lönnqvist, “The Numismatic Chronology,” 123, n. 14: “Spijkerman attributed most of these to Hyrcanus I”, 124, n. 20: “According to Spijkerman’s identification”, n. 22: “Coin n. 1318 is misattributed in the listing of 1994 [---]. Spijkerman’s description corresponds to TJC 367 [---]”. The statement on p. 121 is also unclear: “In 2003 the coin identification from Spijkerman’s cards were published in a new English edition of the Vaux’s field diary.” Which “cards” are referred to? To the manuscript, to the cards, to one of the two, and to which, or to both? 95  Lönnqvist and Lönnqvist, “The Numismatic Chronology,” 121–65, an interesting contribution also for the method and hypotheses formulated.

232

CALLEGHER

After this brief excursus it appears evident that the surviving documentation will be sufficient for a reliable study of the set. It is necessary to begin again from the direct protagonists, who were skilled at compiling information, identifications and materials—of course according to the methods and criteria of that time—but still retaining their documentary importance sixty years later. However, this does not detract from the value of previous works, which perhaps anticipated the impossibility of gaining contemporary access to the dossier, archive and coins. 2

Some Methodological Reflections

In this preliminary phase, and after having examined the actual coins in view of their upcoming publication, I feel it would be opportune to make some comments on methodology to clarify the reference scenario for old coins from archaeological contexts.96 They are first of all, perhaps nothing more than, a document of an economic nature. This irrefutable fact means that within archaeological contexts the relationships may be clarified between money, its use, its debt-paying function, any hoarding, and more in general the economy and therefore its contribution to defining the activities carried out in a site. By means of the seal/stamp, weight and alloy of a coin one can arrive quite rapidly at the year of issue and also who minted it and where, but this ease does not mean that the numismatic find can be considered as a sort of virgin canvas on which to project an interpretation influenced by today’s rationale and current economic-monetary systems. In antiquity, and therefore also in the centuries during which KQ would have been frequented (a lengthy and entirely indicative chronology is suggested: 2nd century BCE–3rd century CE), money and currency were radically different from the way we consider them today. We know from rabbinic sources and from that great historical source the Old Testament, that in the area populated by Jews, between Galilee and Judea, the unit of account was the shekel and that trade was regulated mostly in silver and sometimes in gold, weighed and of excellent alloy.97 Probably starting 96  Patrik J. Casey, Understanding Ancient Coins: An Introduction for Archaeologists and Historians (London: Batsford, 1986); Jean-Marc Doyen, “Cliométrie et numismatique contextuelle: compter et quantifier le passé? Petite histoire de la méthode (1960–2011),” JAN 1 (2011): 9–46. 97  Genesis 23:16; Exodus 38:24; Ezekial 45:12; Judges 9:4. The Old Testament also records other coin terms: the daric, for example in Ezra, Nehemiah and Chronicles, which suggest that it is being used as a weight rather than a specific gold coin.

from the 5th–4th century BCE in some mints of the territory of the Philistines silver became an effective currency following the Attic coin denominations and imitating the types;98 later, in Palestine and Samaria silver weighed in shekels became an effective currency but its value was so high that it also became necessary to mint submultiples to fit the prices of economic transactions;99 in the same period, however, it was mainly used by bureaucrats (in particular by the priests of the Temple) for the accounting of resources and relative values, i.e. of debits and credits (taxes, interest, rents, various duties).100 It should also be mentioned that this “silver coin” was widely hoarded, put aside as reserve funds—what we now define “capital wealth.” And the reason for this is that, while the assets and liabilities were calculated in silver (shekels and fractions) there was no reason to honour them with circulating currency: debts and loans could be paid with many available means, in money but also in other goods, to which a value could be attributed expressed in a unit of account to offset the various operations. Among the few who used silver coins regularly (but sometimes also as goods), were the merchants who often worked around the Temple and sometimes independently. They were also moneylenders: this obviously required book-keeping with stable prices to regulate debts and loans and protect the value and any interest. Very probably, starting from John Hyrcanus I (134–104 BCE) and up to the Jewish War (66–70 CE), in the territories at first Asmonean and then Herodian, coins were minted in copper alloy, named prutah in their base unit and aligned with the weight of the contemporary Seleucid system.101 They were fiduciary coinage in that their value and their acceptance in exchanges was assured by the issuing authority rather than by their weight and metal alloy, they were therefore overvalued with regard to the intrinsic 98  Haim Gitler, The Coinage of Philistia of the Fifth and Fourth Centuries BC: A Study of the Earliest Coins of Palestine (Collezioni Numismatiche: Materiali Pubblici e Privati; Milano: Ennerre, 2006); an effective summary in Meshorer, Bijovsky, and FischerBossert, Coins of the Holy Land, 112–24 [Gaza] (112–14). 99  Meshorer, Bijovsky, and Fischer-Bossert, Coins of the Holy Land, 79, 239–40. 100  Brooks Levy, “Later Tyrian Shekels: Dating the ‘Crude’ Issues; Reading the Controls,” in XIII Congreso Internacional de Numismatica, Madrid 2003: Actas-Proceedings- Actes I (ed. C. Alfaro, C. Marcos, and P. Otero; Madrid: CIN, 2006), 885–90; Meshorer, Bijovsky, and Fischer-Bossert, Coins of the Holy Land, 285–86; Ariel and Fontanille, The Coins of Herod, 41–42. 101  For the discussion on this denomination, see David Hendin, “The Metrology of Judean Small Bronze Coins,” AJN 21 (2009): 105–21 (106).

THE COINS OF KHIRBET QUMRAN FROM DE VAUX ’ S DIGS

value of their metal content. Indeed, within some series and denomination, the weight and metallic variations are notable because a predetermined number of pieces were obtained from a given amount of metal, so their metrology should not have had significant consequences in the circulation as it was a token value. These characteristics are common to many issues of the Asmonean and Herodian series and were very profitable for the state that paid with this type of money, whereas for the collection of taxes imposed the good silver coin, the shekel.102 It is also worth noting that the valuation/quotation of these single pieces was often determined by market circumstances (availability or scarcity of divisional coins) because these specimens, in point of fact fiduciary, lacked explicit indications of their face value, as happens with modern coins. Nevertheless their withdrawal from the market depended on monetary reforms or their capacity for insertion in an accounting system where those values, even if low, were a response (money supply) to a request for those divisional coins (money demand). If they continued to represent a given value within an accounting system (1, 1/2, 1/3, ¼ etc.), they could be accepted for long periods and even the minting of similar denominations could become less necessary. This is one of the reasons why the Asmonean, e.g. of Alexander Jannaeus, and Herodian coins circulated for a long time and were inserted in the Imperial Roman accounting system in relation to the smallest unit of that system: the quadrans. This is confirmed by at least two pieces of evidence: the enormous circulation lasting for centuries and documented by the coin finds in Judea and neighbouring territories, where the coins, especially Asmonean, are associated to much later specimens, and the state of conservation of the prutot, often very poor, demonstrating their prolonged use. In short, to understand something of coinage in the ancient world it is necessary to discard the hypothesis that this subject is simple and self-evident, keeping clearly in mind that money and currency would have corresponded, as the studies of Mitchell Innes clarified around a century ago.103 If the difference between money and currency is accepted as fundamental, we must in the first place think of money as a unit of account, a useful and practical means to stabilize and guarantee the value of both public and private contracts, debts and loans over time. These 102  Hendin, “The Metrology,” 119; Ariel and Fontanille, The Coins of Herod, 20–28. 103  Mitchell-Innes, “What is Money?” Banking Law Journal 30/5 (1913): 377–408; Mitchell Innes, “The Credit Theory of Money,” Banking Law Journal 31/1 (1914): 151–68.

233 could be paid within the stipulated times, in general not short-term, because debts, by their very nature involve temporary deferrals, thus we can deduce that if many documents regarding this region cite values or prices expressed in silver shekels, this does not signify that these refer to minted coins, to real shekels; there is in fact no reason to suppose a wide use of minted coins for transactions or debts involving large sums. In order to instead suppose a wide use of small coins or token coins, the only ones able to represent trade values/prices at the lowest level, we must necessarily remember that this could only occur in a strong state, with a lively currency supply and demand, both for state payments (army, gratuities and congiary)104 and to ensure debt-paying at the lowest level, with practical and useful coins for daily exchanges. In brief, in antiquity money was not abundant and most of all it is inconceivable that it was used frequently, if not in well-defined areas with characteristics that we could define as a monetized economy. In addition to this general introduction, the coin finds of an archaeological site need to be analyzed at least according to two quantitative parameters of an economic-numismatic nature. The first is the equation of Irving Fisher [M (monetary supply) × V (velocity of circulation) = P (prices) × Q (quantity of trade)], i.e. that the monetary stock used in a site depends on the volume of demand for a given product (ceramics, agricultural produce, handiwork, textiles, etc.) whose value can be monetized, a circumstance that influences both the volume and velocity of the exchanges of the monetary supply: the demand for and availability of a product in a site directly influences the number of payments and their amount. The second model of analysis, which calculates the percentage of loss of coins year by year, better known as the methodology to establish economic relations through statistical treatment of the data, using Annual Average Coin Loss (the system developed by Ravetz and Casey), has to be integrated with data regarding the fact that many issues cannot really be subdivided into years but only for the duration of a certain authority and that not a few coins, especially those in copper or copper alloy remained in use for a long time, well after their date of issue.105 It is only on the basis of these coordinates 104  For Herod’s reasons for minting coins, see Ariel and Fontanille, The Coins of Herod, 20–28. 105  For an example within this region, see Jane DeRose Evans, The Coins and the Hellenistic, Roman and Byzantine Economy of Palestine (The Joint Expedition to Caesarea Maritima: Excavation Reports; Boston: American School of Oriental Research, 2006), 6:63–64 which elaborates a Pearson’s Chi-Square Test. Relating to Qumran, K. and M. Lönnqvist have knowledge on some

234 and the composition of the monetary stock found that hypotheses can be made about the economic function of a site and its relationships with other places characterized by analogous numismatic data. As known, the single finds of KQ mostly date to between the 1st century BCE and 1st century CE: i.e. between Alexander Jannaeus and the Anti-Roman revolt of 66–70 CE, even if there are significant authentications of copper coins of the Anti-Roman revolt of 132–135 CE. Because the chronological range is well defined, little more than 150 years, the distribution of the issues according to the criterion of annual loss without doubt possesses an intrinsic logic for the numismatic documentation, even if many denominations were widespread and circulated for many decades (prutot of Alexander Jannaeus and Herod, small bronze coins of the Roman Procurators), for the simple reason that, inserted in a shared accounting system, all were accepted at their nominal value. It follows that Asmonean coins, of Herod and of the Procurators could be on the market together, at the same time, in the same place, probably with the same stable quotation because they were fiduciary. Returning now to the single coin finds of KQ. The great majority between the Asmoneans and the first revolt; their quantity is considerable, in the order of hundreds of pieces, almost all small coins but also a few tetradrachms. What may we deduce, if for now entirely hypothetically, on the basis of the methodological coordinates described above? That in this site, of quite a limited size (perhaps a village, certainly not a city in the sense that this definition could be given in that period),106 contrary to what happened in numerous other sites of the same size, here coins performed a clearly commercial function, debt-paying for all exchanges, from the biggest (a purchase with payment in silver coins: tetradrachms) to the smallest (a terracotta plate or other goods of modest value). The number of transactions had to be frequent enough to justify the hundreds of losses and failed recoveries, only partly attributable to the type of ground or floor (sand, mats, straw) where they had fallen while passing from one hand to another. Very briefly: for the numismatist and the historian of money the numismatic data of KQ do not differ significantly from those of other archaeological sites, on the contrary they show many analogies both qualitative

­ ethodological limits, “The Numismatic Chronology,” 130–35 m and Table 4 and 7. 106  Correctly, Lönnqvist and Lönnqvist, “The Numismatic Chronology,” 141–42 pay attention to the spatial distribution of the coins.

CALLEGHER

and quantitative.107 It may therefore be deduced that there was extensive trade with frequent exchanges and of different amounts, thus ruling out a closed economy, with external relations selected or regulated by some internal administrative authority. How can the presence of such a large number of small coins, lost and not later recovered be explained, if not in terms of frequency of exchanges at the lowest price levels? Establishing what the inhabitants might have traded, at not too high a unitary price, will be a task for the archaeologists.108 107  Some archaeological sites. MASADA: Ya⁠ʾakov Meshorer, “The Coins of Masada,” in Masada I: The Yigael-Yadin Excavations 1963–1965; Final Report; The Aramaic and Hebrew Ostraca and Jar Inscriptions (ed. Y. Yadin and J. Naveh; Jerusalem: Israel Exploration Society/ Hebrew University, 1989), 71–132 and Pl. 81–81; EnGedi: Leir Y. Rahmani, “The Coins,” ʿAtiqot 5 (1966): 52; Ya⁠ʾakov Meshorer, “The En-Gedi Hoard,” in Actes du 8ème Congrès International de Numismatique (ed. H. A. Cahn and G. Le Rider; Paris: Association Internationale des Numismates Professionnels, 1976), 1:11–112; Donald T. Ariel, “The Coins of En-Gedi” in En-Gedi Excavations I: Final Report (1961–1965) (ed. E. Stern; Jerusalem: Israel Exploration Society, 2007), 423–27; Gabriela Bijovsky, “The Coins,” in En-Gedi Excavations 2: Final Report (1996–2002) (ed. Y. Hirschfeld; Jerusalem: Israel Exploration Society/Hebrew University, 2007), 157–232; Jerusalem: John C. Lawrenz, “The Jewish Coins,” in Excavations in Jerusalem 1861–1967 (ed. A. Douglas Tushingham, 1; Toronto: Royal Museum Ontario, 1985), 1:156–66; Donald T. Ariel, “The Coins. J. M. (Givati Parking Lot) City of David,” ʿAtiqot 52 (2006): 71–88; Richard Reece et al., “Jerusalem: the Coins,” in Excavations by K. M. Kenyon in Jerusalem 1961–1967. Vol. 5: Discoveries in Hellenistic to Ottoman Jerusalem; Centenary Kathleen M. Kenyon 1906–1978 (LevantSup 7; ed. K. Prag; Oxford: Oxford Books, 2008), 411–31; Donald T. Ariel, “Coins,” in Jewish Quarter Excavations in the Old City of Jerusalem Conducted by Nahman Avigad, 1969–1982; 4:The Burnt House of Area B and Other Studies. Final Report (ed. H. Geva; Jerusalem: Israel Exploration Society, 2010), 236–47; Haim Gitler, “A Comparative Study of Numismatic Evidence from Excavations in Jerusalem,” LA 46 (1996): 317–62. Other sites, even if distant, have provided analogous numismatic documentation: Caesarea Maritima: DeRose Evans, The Coins, 116–22; MIGDAL (coins being studied by the author); GAMLA (I thank Danny Sion for the information). 108  For the economic function of the site, there are interesting hypotheses by Yizhar Hirschfeld, Qumran in Context: Reassessing the Archeological Evidence (Peabody, MA: Hendrickson, 2004); Jodi Magness, The Archaeology of Qumran and the Dead Sea Scroll (Grand Rapids, MI: Eerdmans, 2002); eadem, Debating Qumran: Collected Essays in its Archaeology (ISACR 4; Leuven: Peeters, 2004); Lönnqvist and Lönnqvist, “The Numismatic Chronology;” Lönnqvist, The Report; Jan Gunneweg and Marta Balla, “Was the Qumran Settlement a Mere Pottery Production Center? What Instrumental Neutron Activation Revealed,” in Holistic Qumran: Trans-Disciplinary Research of Qumran and the Dead Sea Scrolls; Proceedings of the NIAS-LORENTZ Center

THE COINS OF KHIRBET QUMRAN FROM DE VAUX ’ S DIGS

In this phase of the research, it can therefore be hypothesized that the KQ site was thoroughly monetized, in a continuous way and for quite a long period (1st century BCE–1st century CE). 3

Coins in the Caves

Following the same methodology, the previous observations or hypotheses can be extended to the absence or presence of coins minted prior to 67/68 CE in the caves investigated in the same year in which the KQ site was extensively excavated. The systematic surveys conducted in numerous caves along the desert escarpment that slopes down towards the Dead Sea (for example near Jerico and Wadi Murabbaʿat) have led to the finding of many coins, often in considerable quantities and of high intrinsic value (aurei and denarii), but in general referable to the events of 132–135 CE [Cave X/35 (Q3)109 or some residential caves near KQ,110 the el-Jai Cave,111 the Cave of the Sandal-West Jerico,112 and the Cave of Wadi el-Māckūk,113

Qumran Workshop 21–25 April 2008 (STDJ 87; ed. J. Gunneweg, A. Adriaens, and J. Dik; Leiden: Brill, 2010), 39–61; a very useful update, in this volume: Jürgen K. Zangerberg. 109  Donald T. Ariel, “The Numismatic Finds,” ʿAtiqot 51/2 (2002): 169–76 (172); Donald T. Ariel, “The Coins from the Survey and Excavations of Caves in the Northern Judean Desert,” ʿAtiqot 51/2 (2002): 281–305. 110  Magen Broshi and Hanan Eshel, “Residential Caves at Qumran,” in DSD 6/3 (1999): 328–48. 111  Hanan Eshel and Boaz Zissu, “Coins from the el-Jai Cave in Naḥal Mikhmash (Wadi Suweinit),” INJ 14 (2000–2002): 169–75. 112  Hanan Eshel and Boaz Zissu, “Roman Coins from the ‘Cave of the Sandal’ West Jericho,” INJ 13 (1994–1999): 70–77. 113  Hanan Eshel, “A Coin of Bar Kokhba from a Cave in Wadi el-Māckūk,” INJ 9 (1986–1987): 51–52.

235 of Wadi Murabbaʿat,114 of En-Gedi115 and of Mŭghâret Umm et Tûeimîn (Jerusalem West and the two hoards of Khirbat Wadi Ḥamam)116]. If in the archaeological strata of a cave only coins whose chronology ends in 67/68 CE are not found, what can we hypothesize or deduce? That up to that time they were not inhabited frequently or continuously, perhaps not at all or only sporadically, for very brief periods. This is because, in a context of a thoroughly monetized economy at all levels, as seems to have been the case at KQ, there are no logical elements to suppose a selective use of currency in some sites and not in others. If the caves in the escarpment around KQ—as revealed by the archaeological digs of RDV—have provided only rare and occasional coins, it can be supposed that they did not have a residential, and therefore functional use, similar to that of the principal site, which, from what has so far been ascertained, had a typically trading function, with an extensive and prolonged use of both silver and divisional currency.

114  Milik and Seyrig, “Trésor monétaire,” 11–26; Ephraim Stern, “The Wadi Murabbaʿat Caves,” NEAEHL 3:833–35. 115  Roi Porat, Hanan Eshel, and Amos Frumkin, “Finds from the Bar Kokhba revolt from two caves at En Gedi,” PEQ 139/1 (2007): 35–53. 116  Boaz Zissu et al., “Coins from the Bar Kokhba Revolt Hidden in Meḥarat Ha-Teḥomim (Mŭghâret Umm et Tûeimîn), Western Jerusalem Hills,” INJ 17 (2009–2010): 113–47; Boaz Zissu and David Hendin, “Further Remarks on Coins in Circulation during the Bar Kokhba War: Te’Omim Cave and Horvat ‘Ethri Coins Hoards,” in Judaea and Rome in Coins: 65 BCE–135 CE; Papers Presented at the International Conference hosted by Spink, 13th–14th September 2010 (ed. D. M. Jacobson and N. Kokkinos; London: Spink & Son, 2012), 215–28.

236

Figure 15.1

CALLEGHER

Final synthesis of Augustus Spijkerman’s revision of the Khirbet Qumran Hoard, f. 30 (recto). ©Archive of Studium Biblicum Franciscanum-Jerusalem.

THE COINS OF KHIRBET QUMRAN FROM DE VAUX ’ S DIGS

Figure 15.2

Small black-and-white photographs of the obverse and reverse of the silver coins (Khirbet Qumran Hoard), sent from the Archaeological Museum-Amman to father Augustus Spijkerman (November 1960). ©Archive of Studium Biblicum Franciscanum-Jerusalem.

237

CHAPTER 16

Dating the Scroll Deposits of the Qumran Caves: A Question of Evidence Gregory L. Doudna In the 1990s I raised the question of whether the scrolls in the caves of Qumran had been deposited as late as the first century CE. This was the first since the excavation of Qumran in 1951 that that had been questioned. I argued that the scroll deposits appeared to predate the first century CE in their entirety and had nothing to do with the First Revolt of 66–70 CE.1 This argument was joined by Ian Hutchesson (1999), Alan Crown (2005), Ian Young (2002, 2005, 2013), and David Stacey (2013).2 However nearly two decades later the questioning of the existence of First 1  Gregory L. Doudna, “Dating the Scrolls on the Basis of Radiocarbon Analysis,” in The Dead Sea Scrolls After Fifty Years: A Comprehensive Assessment (ed. P. Flint and J. C. VanderKam; Leiden: Brill, 1998), 1:458–64; idem, “Redating the Dead Sea Scrolls Found at Qumran: The Case for 63 BCE,” QC 8/4 (1999); idem, 4Q Pesher Nahum: A Critical Edition (JSPSup 35; Sheffield: Sheffield Academic Press, 2001), 698–705, 746–53; idem, “The Legacy of an Error in Archaeological Interpretation: the Dating of the Qumran Cave Scroll Deposits,” in Qumran, the Site of the Dead Sea Scrolls: Archaeological Interpretations and Debates; Proceedings of a Conference Held at Brown University, November 17–19, 2002 (STDJ 57; ed. K. Galor, J.-B. Humbert, and J. Zangenberg; Leiden: Brill, 2006), 147–57; idem, “The Sect of the Qumran Texts and its Leading Role in the Temple in Jerusalem During Much of the First Century BCE: Toward a New Framework for Understanding,” in Qumran Revisited: A Reassessment of the Archaeology of the Site and its Texts (BAR.IS 2520; ed. D. Stacey and G. L. Doudna; Oxford: Archaeopress, 2013), 107–19. 2  Ian Hutchesson, “63 BCE: A Revised Dating for the Depositation of the Dead Sea Scrolls,” QC 8 (1999): 177–94; Alan Crown, “An Alternative View of Qumran” in Samaritan, Hebrew and Aramaic Studies Presented to Professor Abraham Tal (ed. M. Bar Asher and M. Florentin; Jerusalem: Bialik Institute, 2005), 1–24; Ian Young, “The Stabilization of the Biblical Text in the Light of Qumran and Masada: A Challenge for Conventional Qumran Chronology?” DSD 9 (2002): 364–90; idem, “The Biblical Scrolls from Qumran and the Masoretic Text: A Statistical Approach,” in Feasts and Fasts: A Festsrift in Honour of Alan David Crown (Mandelbaum Studies in Judaica 11; ed. M. Dacy, J. Dowling, and S. Faigan; Sydney: Mandelbaum Publishing, 2005), 81–139; idem, “The Contrast Between the Qumran and Masada Biblical Scrolls in the Light of New Data: A Note in Light of the Alan Crown Festschrift,” in Keter Shem Tov: Collected Essays on the Dead Sea Scrolls in Memory of Alan Crown (ed. S. Tzoref and I. Young; Piscataway, NJ: Gorgias Press, 2013), 113–19; David Stacey, “A Reassessment of the Stratigraphy of Qumran,” in Stacey and Doudna, Qumran Revisited, 63.

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_018

Revolt scroll deposits remains regarded as an outlier argument, barely making a ripple in much mainstream discussion. When I consider the underlying facts and force of the argument this response puzzles me. Here I review this question, but first a preliminary point. In looking for a correct earlier context for the scroll deposits I focused on the end of what Qumran excavator Father Roland de Vaux called Qumran’s “Period Ib.” This focus involved regrettable mistakes on my part at earlier stages involving the dating of the end of Ib. In the 1990s de Vaux’s date for the end of Ib of 31 BCE had been widely and long accepted. Another analysis was that of ErnestMarie Laperrousaz who dated the end of Ib to 67–63 BCE. And a third date, 40–37 BCE, earlier held but abandoned by de Vaux, also retained adherents. The publication of the pottery of the Netzer excavations of Jericho by Rachel Bar Nathan in 2002 made it clear that all of those dates for the end of Qumran’s Period Ib were incorrect; they were all too early.3 Jodi Magness was right in 1995 and 1998 in dating the end of Ib later in the reign of Herod, toward the end of the first century BCE.4 The present argument embraces the corrected ending of Qumran’s “Period Ib” toward the end of the first century BCE. Yet fundamentally the argument is that the traditional dating of the scroll deposits of the caves of Qumran to as late as the time of the First Revolt is supported by neither evidence nor plausibility. This argument stands independently of whether a specific earlier context for the deposits is correctly identified, which is a distinct issue.

3  Rachel Bar Nathan, Hasmonean and Herodian Palaces at Jericho: Final Reports of the 1973–1987 Excavations. Vol. 3: The Pottery (Jerusalem: Israel Exploration Society/Hebrew University, 2002). 4  Jodi Magness, “The Chronology of the Settlement at Qumran in the Herodian Period,” DSD 2 (1995): 58–65; eadem, “The Chronology of Qumran, Ein Feshkha, and Ein El-Ghuweir,” in Mogilany 1995: Papers on the Dead Sea Scrolls Offered in Memory of Aleksy Klawek (Qumranica Mogilanensia 15; ed. Z. Kapera; Cracow: Enigma Press, 1998), 55–76.

239

DATING THE SCROLL DEPOSITS OF THE QUMRAN CAVES

1

A Drop in Flourishing Historical Allusions in the Qumran Texts to Flat-Line Zero in the Second Half of the First Century BCE Suggests the Scroll Deposits Occurred Earlier than Commonly Supposed

It is well recognized that there is no allusion in the Qumran texts to anything later than the second half of the first century BCE. The many allusions in the texts to figures, events, and contexts in the first century BCE drop at some point in the second half of the first century BCE to flat-line zero thereafter. In contrast to dozens of Qumran texts composed in the first century BCE, not a single text composed in the first century CE exists in the finds from the caves of Qumran. A 2003 study of Michael Wise remains the most comprehensive attempt to inventory the historical allusions in the Qumran texts.5 Wise counted what he defined as “first-order” allusions, and not “second order” allusions (allusions which depend on the correctness of a prior allusion identification), which Wise suggested would have increased—perhaps doubled—the numbers if that were done. Wise counted 6 allusions in the second century BCE, rising dramatically to 25 in the first century BCE ending at 37 BCE. Then, 0 for the final third of the first century BCE, 0 for the first century CE, 0 for second century CE, etc. Other studies have found this same pattern of distribution.6 Wise likened the absence of first-century CE allusions in the Qumran texts to the Sherlock Holmes dog which did not bark. Sherlock Holmes remarked in one of his cases that he was puzzled by “the curious incident of the dog in the night-time.” He was told, but the dog did nothing in the night-time. Holmes replied, “Yes, that was the curious incident.” As Wise noted, the complete absence 5  Michael O. Wise, “Dating the Teacher of Righteousness and the Floruit of his Movement,” JBL 122 (2003): 53–87. 6  Geza Vermes, “Historiographical Elements in the Qumran Writings: A Synopsis of the Textual Evidence,” JJS 58 (2007): 121–39; Kenneth Atkinson, “Representations of History in 4Q331 (4QpapHistorical Text C) 4Q332 (4QHistorical Text D), 4Q333 (4QHistorical Text E), and 4Q468e (4QHistorical Text F): An Annalistic Calendar Documenting Portentous Events?” DSD 14 (2007): 125–51. I differ from Wise in interpretation of several of the allusions; for the same allusion passages used by Wise I count 21 agreements and 10 differences (cf. Doudna, “The Sect of the Qumran Texts”). Retabulating Wise’s list according to my analyses of the same text passages my counts come to 6 for the second century BCE, 23 for the first century BCE ending ca. 30 BCE, 0 for the final third of the first century BCE, 0 for first century CE, 0 for second century CE, etc.—despite differences on individual datapoints, the exact same shape of the graph.

of newly composed texts throughout the first century CE from a sect which up to that point had been prolific in composing new texts is indeed a very curious thing. Scholars who hold to the First Revolt endpoint for the scrolls of the caves of Qumran have never satisfactorily explained this phenomenon. Why would a group stop composing new texts and switch to only copying old texts for nearly a century? What kind of sense does that make? Attempts to explain this odd behavior according to the First Revolt assumption have the ring of “just-so” ad hoc explanations. The apparent oddity is an artificial creation of the premise of the First Revolt endpoint for the texts. In fact the end of allusions in the Qumran texts after a certain point in the second half of the first century BCE—the “radio silence” from that point forward—in a database the size of the Qumran cave finds is one of the clearest signals that the scroll deposits were earlier than commonly supposed. An earlier dating of the scroll deposits themselves is the simplest way to account for the total silence in those texts after late first century BCE. 2

How Did the Notion of First Revolt Scroll Deposits at Qumran Come to Be Believed in the First Place?

After the Dead Sea Scrolls came to light in 1947 the latest scrolls were dated palaeographically to mid-first century BCE by Solomon Birnbaum and William F. Albright.7 After Qumran Cave 1Q was excavated in 1949 Albright assessed on ceramic typology grounds that the cylindrical, wide-mouthed “scroll jars” and two “hellenistic” lamps associated with the scrolls preceded a Romanization of pottery in Judea which started in the second half of the reign of Herod late in the first century BCE.8 Albright and the other archaeologists noted that a small quantity of later Roman period domestic pottery and lamps had also been found in Cave 1Q, but they considered that pottery intrusive and later, not associated with the scrolls and scroll jars. Albright considered the evidence conclusive: 7  Solomon Birnbaum, “The Dates of the Cave Scrolls,” BASOR 115 (1949): 20. 8  Albright: “[N]one of the scrolls [of Qumran Cave 1Q] can be later than about 50–25 BCE . . . It is a striking fact that we had reached the same approximate date [from palaeography] for the Habakkuk Scroll and its congeners (about the beginning of the Herodian Age, or a very little earlier) that is now imposed by the ceramic evidence” (William F. Albright, “Are the ‘Ain Feshkha Scrolls a Hoax?” JQR 40 [1949]: 47).

240 “We may rest assured . . . not a single piece [fragment of scroll] found in the cave is later than the early years of the Herodian Age at latest.”9 This changed with the excavation of Qumran of Nov.– Dec. 1951. An identical “scroll jar” was found installed in the floor of Qumran’s Locus 2. Other pottery in the same room was found identical to the domestic pottery in Cave 1Q. Coins from the first century CE were found in Locus 2. The coins of Locus 2 and neighboring loci ended at the time of the First Revolt and there had been a destruction of the site by fire at that time. It followed (so it seemed) that the scroll jars and scroll deposits of Cave 1Q were first century CE. In his introduction to Volume I of the Discoveries in the Judean Desert series which appeared in 1955, Jordanian director of antiquities and co-excavator of Qumran G. Lankester Harding stated definitively: “Excavation of the settlement at Khirbet Qumran has established beyond doubt that all of the material was deposited in these caves in the late first century AD.”10 Albright accepted the new findings, as did every known archaeologist and text scholar in the world apart from a few who argued the scrolls were later still. One of the most significant results of the archaeological redating of 1951, which had been announced despite it being in apparent disagreement with the palaeographic datings, was its effect on those very palaeographic datings. What happened is very simple to understand: based on the archaeological findings the palaeographic datings were adjusted to be in agreement. The typologically most developed formal hands among the Qumran texts were redefined with later dates than previously published, ending at the time of the First Revolt. In Jan.–Feb. 1952 more texts were found in caves at Wadi Murabba‘at, an early second-century CE Bar Kokhba Second Revolt site. Professionally-produced biblical scrolls identical in their consonantal texts to the medieval Masoretic Text were found written in scribal hands typologically developed beyond the latest formal hands among the Qumran finds. Therefore the Murabba‘at biblical texts—with their developed keraias and decorative flourishes of the letters—were defined as postdating 70 CE in their dates of scribal copying (instead of dated earlier, before the fall of the temple), based on the premise that the Qumran texts defined palaeographical development until ca. 70 CE. Frank Cross, Jr., became the key figure here. 9   William F. Albright, “The Dead Sea Scrolls of St. Mark’s Monastery,” BASOR 118 (1950): 6. 10  Gerald L. Harding, “Introductory, the Discovery, the Excavation, Minor Finds,” in Qumran Cave 1 (DJD 1; ed. D. Barthélemy and J. T. Milik; Oxford: Clarendon Press, 1955), 4.

DOUDNA

Cross reconstructed stages of typological development and assigned to each a successive series of date ranges, one after the other, ending for the Qumran texts at the time of the First Revolt. This culminated in Cross’s wellknown study of 1961 which has been the standard used for palaeographic dating ever since.11 On this basis all of the non-Qumran biblical scrolls then known (from Wadi Murabbaʿat and Naḥal Ḥever) were dated post-70 CE and their scribal hands named “post-Herodian formal,” in Cross’s 1961 reconstruction.12 To repeat and to be clear: the focus on the First Revolt as the date of the latest Qumran texts began as an archaeological correction in the dating of the scroll jars and scroll deposits in the caves. Then the palaeographic datings were secondarily adjusted to be in alignment. That was the specific direction of the logic. The archaeological ­correction of the scroll deposits to the First Revolt was the basis for fixing the palaeographic dates of the latest Qumran cave texts to the first century CE, and the dates of the biblical texts postdating the Qumran caves’ latest to later than 70 CE. 3

How Sound was the 1951 Archaeological Redating?

The 1950s’ archaeological redating of the scroll deposits in the caves of Qumran to the First Revolt was entirely based upon three foundational assumptions, all three deeply flawed. The first, which arose from the first season of excavations of Qumran in 1951, was that all non-Iron Age pottery at the site of Qumran was exclusively first century CE in date, from a single period of habitation in the first century CE that ended at the time of the First Revolt. That assumption was short-lived. Starting in the second excavation season at Qumran in 1953, the excavators were surprised to uncover evidence of extensive activity from the preceding century, the first century BCE. The room of Locus 2 with the scroll jar buried in the floor was now identified by de Vaux as having been built in the earlier period, in the first century BCE. De Vaux now explained that the floor of Locus 2 had been swept clean and reused by people of Period II in the first century CE. That was why, in de Vaux’s revised explanation all pottery found 11  Frank M. Cross, “The Oldest Manuscripts from Qumran” JBL 74 (1955): 147–72; idem, “The Development of the Jewish Scripts,” in The Bible and the Ancient Near East: Essays in Honor of William Foxwell Albright (ed. G. E. Wright; Garden City, NY: Doubleday, 1961), 133–202. 12  Cross, “The Development of the Jewish Scripts,” 174, 199, n. 136a.

DATING THE SCROLL DEPOSITS OF THE QUMRAN CAVES

in Locus 2 was first century CE, because those were the items in use at the time of the destruction of ca. 68 CE, even though the room and its floor had actually been built much earlier, in the first century BCE. What no one at the time seemed to consider is that the scroll jar buried in the floor of Locus 2, after it was installed, became at that point part of the floor, not loose above the floor. That scroll jar, if it was already there, would have been inherited along with the floor from the earlier period by the new people of Period II, even if all other older loose pottery on top of the floor was swept clean and removed in Period II as de Vaux reconstructed. That, and the what would later become identified as three scarce Antigonus Mattathias coins of 40–37 BCE (two certain and a third with a question mark, out of only six total Antigonus Mattathias coins found at the entire site) found close to the buried scroll jar in the floor of Locus 2—one under the floor directly against the jar—suggest that that jar could well have been installed in the floor of Locus 2 late first century BCE.13 At minimum that would appear to be a reasonable interpretation, such that any claim of a later or first-century CE date of installation of that jar in the Locus 2 floor would seem to call for evidence or argument to be accepted, as distinguished from uncritical reliance on the original publication of that jar’s date at a time when no one knew of the existence of a first-century BCE history of that locus. But it was that Locus 2 scroll jar which had been broadcast to the world as the evidence that the Qumran cave scroll jars were first century CE instead of first century BCE. That Locus 2 scroll jar, more than any other single reason, was why all of the scroll jars and scroll deposits in Cave 1Q had been redated later. Throughout his preliminary reports of the excavations of Qumran of 1951–1956 de Vaux maintained that not just that Locus 2 scroll jar but all scroll jars found both at the site and in the caves dated first century CE, “Period II.” It was only in 1959 that de Vaux stated for the first time, in his Schweich Lectures to the British Academy, that the scroll jars found in the caves of Qumran actually dated— de Vaux now said—to both first century BCE and first century CE instead of first CE exclusively.14 De Vaux gave no 13  The Antigonus Mattathias coins from Locus 2 are: no. 109 (“northwest corner against the jar buried under the paved floor”); no. 110 (?); and no. 111, in Jean-Baptiste Humbert and Alain Chambon, eds., The Excavations of Khirbet Qumran and Ain Feshkha: Synthesis of Roland de Vaux’s Field Notes (NTOA.SA 1B; trans. and rev. Stephen J. Pfann; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 12. 14  Roland de Vaux, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973), 54.

241 specifics then or later. Similarly in DJD 3 in 1962 de Vaux stated explicitly for the first time that some scroll jars from the buildings of Qumran were from the first century BCE, again without specifics such as which loci or scroll jars were affected.15 In other words—although there is no record that anyone at the time noticed the point—the principal argument underlying the archaeologists’ announcement to the world after 1951 that the cave scroll deposits had been redated “beyond doubt” to the time of the First Revolt had just explicitly disappeared, just like that, in 1962. Because: if some scroll jars in Qumran’s buildings were in use in the earlier time (as per de Vaux now, post-1962), and those jars were not exclusive to the first century CE, that logically removed certainty that any of the same kind of jars in the caves must be first century CE. For it could not now logically be excluded that all of the scroll jars in the caves represented activity from the earlier time of Qumran (since the scroll jars were now acknowledged to have been in use at that earlier time). The archaeological basis for asserting that any of the scroll jars or deposits of scrolls in the caves must date to as late as the middle or second half of the first century CE, was gone. With the first foundational assumption gone, the second and third ones took its place. The second foundational assumption was that one single group of people exclusively controlled and inhabited Qumran autonomously for an entire ca. 150–200 years from the start of the site until the First Revolt, continually through all of what de Vaux defined as three archaeological periods, Ia, Ib, and II, except for a break between Ib and II when the site was uninhabited. That, combined with a further assumption that scroll deposits in the caves would not have ended while people related to the scrolls continued at the site, functioned to prevent scholars from considering whether the deposits of scrolls in the caves might have been a phenomenon limited to some earlier subset of the site’s history, not necessarily ending at the First Revolt. As is well known, many archaeologists have rejected this second foundational assumption. That is, they reject a Ia/Ib or Ib/II continuity in people and function at Qumran (e.g. Bar Adon, Humbert, Hirschfeld, Magen, Peleg).16 15  Roland de Vaux, “Archéologie,” in Les ‘petites grottes’ de Qumrân: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 14. 16  Pesach Bar Adon, “The Hasmonean Fortresses and the Status of Khirbet Qumran,” EI 15 (1981): 349–52 [Hebrew], 65 english summary; Jean-Baptiste Humbert, “L’espace sacré à Qumran: Propositions pour l’archéologie,” RB 101 (1994): 161–214; Yizhar

242 Others have proposed a discontinuity in people inhabiting Qumran internal to Period II itself, in which earlier people in Period II were replaced by different people who continued until the end of Period II. The first to suggest this was de Vaux who proposed in 1954 that an earlier of two successive groups in II did the scroll deposits and then left the site, followed by a distinct second group at the site until the end of II.17 Going beyond de Vaux, Frederick F. Bruce and Stephen J. Pfann in the Encyclopedia Judaica hold that there exists sufficient archaeological (material) evidence indicating a difference in people between Ib and the First Revolt to justify the identification and naming of a new, distinct archaeological period representing a nonsectarian habitation at Qumran between Ib and the First Revolt, which they call “Period IIb.”18 Meanwhile, David Stacey argues that first-century BCE Qumran was linked to the Royal Estate at Jericho and used for water-intensive industrial processes which supplied the demands of the Estate; as the Royal Estate declined after the time of Herod and Archelaus so too did the demand for goods produced at Qumran. With the decline and abandonment of the Royal Estate in the first century CE there would have been an influx of new people to Qumran with a different impetus and purpose, in Stacey’s view.19 Each of these last-named (de Vaux, Bruce and Pfann, Stacey) suppose the latest inhabitants of first-century CE Qumran ending at the First Revolt postdated the major scroll deposits of Cave 1Q, Cave 4Q, etc. In this way a notion of discontinuity in principle between earlier scroll deposits and different, later, non-cave-scrolls-related first-century CE people and activity at the site before the destruction of the First Revolt has been present from the beginning in some de Vaux-aligned archaeological Hirschfeld, Qumran in Context: Reassessing the Archaeological Evidence (Peabody, MA: Hendrickson, 2004); Yitzhak Magen and Yuval Peleg, The Qumran Excavations 1993–2004: Preliminary Report (JSP 6; Jerusalem: Staff Officer of Archaeology, Civil Administration of Judea and Samaria/ Israel Antiquities Authority, 2007). 17  De Vaux: “On peut supposer que l’ensemble de celle-ci [la Communauté] avait en effet quitté les lieux, en mettant ses trésors à l’abri, mais qu’un parti d’hommes résolus avait tenté de défendre le bâtiment” (Roland de Vaux, “Fouilles au Khirbet Qumrân: Rapport préliminaire sur la deuxième campagne,” RB 61 [1954]: 234). 18  Frederick F. Bruce and Stephen J. Pfann, “Qumran,” EncJud 16:771. Bruce and Pfann state that Period IIb was only ca. 2–3 years in duration, but it is not clear what archaeological grounds prevent Bruce and Pfann’s IIb from having begun much earlier than 66 CE. 19  Stacey, “A Reassessment,” 63, 70.

DOUDNA

interpretations of Qumran (e.g. de Vaux), even though this point has not generally been appreciated. In any case, the variety of archaeologists’ conjectures concerning which of Qumran’s pre-70 CE occupants were and were not involved with the deposits of scrolls in the caves betrays the arbitrary nature of such reconstructions. The third foundational assumption concerned the interpretation of evidence that refugees or fugitives used caves around Qumran at the time of the First Revolt. De Vaux and others assumed that those people must have done the deposits of scrolls and scroll jars too, since both involved caves. But as has often been pointed out, there are countless examples of caves with relics of human activity from different eras of time unrelated to each other in the same cave. No good argument has been set forth for excluding this interpretation—so common in other caves—in the case of scroll-bearing caves of Qumran which also had domestic pottery from the time of the First Revolt (e.g. the scroll jars and scroll deposits date late first BCE; refugee activity dates ca. 60s CE/First Revolt). In sum, none of the three foundational assumptions which created the claim on archaeological grounds concerning First Revolt scroll deposits was valid. 4

Do Lamps in Cave 1Q Date the Scroll Deposits to the Time of Herod?

Two lamps found with the scroll jars and scrolls in Cave 1Q in 1949, called at the time “hellenistic” lamps, date narrowly to the time of Herod the Great. Thirtynine more of this type of lamp were found in the excavations of Qumran according to the latest numbers of the lamp study of Jolanta Młynarczyk of 2013.20 All of these lamps found at Qumran are from the late first century BCE, without exception. Several more of these lamps were found at Jericho and Masada from the time of Herod (none first century CE).21 On the basis of the distributions Młynarczyk concluded in agreement with earlier studies that all of the lamps of this type were manufactured at Qumran itself, therefore now called “Qumran lamps.” When these lamps of Qumran end, so do these lamps end everywhere, because no longer were these lamps being made at Qumran. The two “Qumran lamps” in Cave 1Q have no other apparent reason for being there than association with the deposits of scroll jars and scrolls. Certainly the early archaeologists assumed such an association. 20  Jolanta Młynarczyk, “Terracotta Oil Lamps from Qumran: The Typology,” RB 120 (2013): 99–133. 21  Bar Nathan, Hasmonean and Herodian Palaces, 110–12.

DATING THE SCROLL DEPOSITS OF THE QUMRAN CAVES

Two knife-pared, bow-spouted “Herodian” lamps as well as domestic pottery from the first century CE also were found in Cave 1Q. The trapezoid shape of and line markings on the “Herodian” lamp’s nozzle (of the one de Vaux published; de Vaux said the other, unpublished, one “appartenu au même type” as the published one [DJD 1, Fig. 3.1]) indicate those two lamps from Cave 1Q belong to Smith’s Type 2 class of “Herodian” lamps, believed to have started in Judea about 35 CE and the most common type of lamp in Judea at the time of the First Revolt.22 These two pairs of lamps in Cave 1Q—the pair of “Qumran” (“hellenistic”) lamps and the pair of “Herodian” lamps—are therefore decades apart in their datings and cannot be interpreted in terms of contemporaneous uses by the same people. Since the Smith Type 2 “Herodian” lamps are contemporary with the first-century CE domestic pottery in Cave 1Q, those Smith Type 2 Herodian lamps are well understood as associated with the domestic pottery of Cave 1Q. There is no good reason to suppose any additional scroll deposits in Cave 1Q three-quarters of a century later. The lamps in Cave 1Q reveal that the two classes of phenomena—scroll jars and scroll deposits on the one hand, and first-century CE refugees or fugitives’ fleeting uses of caves on the other—were significantly separated in time and not contemporary. 5

The Scroll Jars Do Not Prove the First Revolt Dating

The “scroll jars” in the buildings of Qumran were in use in the time of Herod (some continued in use later) according to a study I published in 2006 of datings of scroll jars found at the site based on de Vaux’s field notes and other available information.23 That study concluded that of 15 identified scroll jars from the buildings at Qumran, 7 appeared to be from late first BCE, an additional 3 (including the famous one of L2) probably were from late first BCE but with a question mark, and the remaining 5 were uncertain or there was no information.24 22  Młynarczyk, “Terracotta Oil Lamps,” 108–17. 23  Doudna, “Legacy of an Error,” 150–52, 155–57. 24  One of the items in that inventory may not be a scroll jar, changing the total to 14 and “uncertain” to 4. A Gunneweg and Balla reference in 2003 to a scroll jar in “the make-up of the tower glacis” likely refers to no. 939 of L17 which de Vaux published rather than to de Vaux’s “jar embedded in the reinforcement” of L17 identified as no. 794 in a footnote in the Pfann 2003 edition of de Vaux’s field notes. In this correction no. 939 changes from “Uncertain” to “Ib” for reasons given in the “Notes and

243 At the Lugano conference Jean-Baptiste Humbert kindly provided me with a draft discussion of the southeastern annex in which he argued that both ovoid and scroll jars in those loci date to first-century CE contexts. David Stacey has commented to me that Humbert’s interpretation of the southeastern annex jar datings seems convincing, based on an association in the same context with first-century CE Nabatean ware. However Stacey suggests the uses could be secondary: “Were [these large jars] being made at Qumran then? Perhaps. Or were they, perhaps, scavenged from nearby caves? If the population is now largely ‘refugee’ it would make sense for them to utilise perfectly good—and heavy—jars if they stumbled across them in one or other of the caves. It makes so much sense to me that people living in Qumran in mid 1st century CE—quite possibly newly arrived—would utilise perfectly good jars standing idly around in nearby caves.” At Jericho, Bar Nathan reported that the only example of a Qumran type of scroll jar (Jericho Type J-SJ2B) was found in a context dated to the time of Herod (none from first century CE). At Ein Feshkha, de Vaux noted that Ein Feshkha’s Period II of the first century CE corresponded exactly to Qumran’s Period II chronologically and in all of its kinds of small pottery, but that there were no scroll jars in Ein Feshkha’s Period II.25 At Masada, Bar Nathan did report scroll jars found in First Revolt “Zealot Occupation” contexts which Bar Nathan interpreted as having been brought to Masada at the time of the First Revolt.26 But in the opinion of David Stacey, who was a field supervisor in the Netzer excavations at Jericho and also worked on the Masada Yadin excavations, all of the scroll jars found at Masada are likely residual or reuses from the stores of Herod at Masada.27 Comments” for no. 794 applied now to no. 939 (Doudna, “Legacy of an Error,” 156). 25  De Vaux, Archaeology and the Dead Sea Scrolls, 64. 26  Rachel Bar Nathan, Masada VII: The Yigael Yadin Excavations 1963–1965; Final Reports; The Pottery of Masada (Jerusalem: Israel Exploration Society/Hebrew University, 2006), 67–72. 27  Doudna, “The Sect,” 113. I asked Stacey to elaborate on this point. Stacey returned the following response and I thank Stacey for permission to quote it here: “I was in the library a couple of days ago and took a look [again] at the Masada final reports regarding the scroll jars. There were sherds from 7 jars, claimed to be ‘complete’ on the basis that a rim and base fragment seemed to be from the same jar. No single jar was anywhere near actually complete. They were found in four locations in the Western Palace, and in various places in the casement wall: 8 rims and

244

DOUDNA

Based on the finds from these sites (Qumran, Jericho, Ein Feshkha, Masada), the scroll jars do not prove First Revolt scroll deposits. Instead it is very plausible that the time of Herod was the era of manufacture of these jars associated with the huge numbers of identical jars found in the caves around Qumran (over 150 by de Vaux’s count), and the scrolls deposited with those jars. 6

Today, the Most Common Claim of Evidence for the First Revolt Dating is Palaeographic Dates. However First-Century CE Palaeographic Dates Assigned to Qumran Cave Texts Provide No Basis for Confidence That Those Texts Were First Century CE

The issue in the palaeographic datings of the Qumran texts is the validity of the absolute dates in the palaeographic script charts. A first observation is that the small number of decades separating mid-first century CE from the time of Herod is barely greater than acknowledged 3 handles were found variously in Buildings 7 and 9, from a ‘dump’ in the storerooms, and from a ‘dump’ in the Northern Palace. It is impossible from what is published to identify the exact locations because each room was given only one locus number. Although each sherd should theoretically be assigned to a particular pottery basket which would give an indication of more precise location within each locus, these numbers are, not surprisingly as there must have been tens of thousands of individual baskets, not published. Four of the ‘complete’ vessels came from the Western Palace where, ‘if one judges from the building remains the Zealot presence . . . was minimal’ (Ehud Netzer, Masada III: The Yigael Yadin Excavations 1963–1965; Final Reports; The Buildings, Stratigraphy and Architecture [Jerusalem: Israel Exploration Society/Hebrew University, 1991], 633). One supposedly ‘complete’ jar came from Room 502 in the Western Palace, which was the largest storeroom found at Masada, and which continued as a storeroom in the Zealot period (Netzer, Masada III, 634). Also found in 502 was one complete sj [storage jar] type M-SJ5 (see table at Bar Nathan, Masada VII, 400). A number of sherds from this type of sj, inscribed with the names ‘Zenon’, ‘Shabyo’ or ‘Bar Yasson’ were found in both Herodian and Zealot contexts. Bar Nathan, from the evidence at Masada, ‘suggests that it was produced in the Herodian period’ and specifically states that ‘it seems that during the Zealot Occupation, some of the jars were moved to other localities, such as Storerooms 140 and 502 [emphasis Stacey’s], Room 526 in the Western Palace, Casemates 1103, 1141, 1197 and 1201 in the eastern and southeastern sections of the wall, and Tower Rooms 1110 and 1156’ (Bar Nathan, Masada VII, 49–50). If this is extremely likely for SJ5 then the same is equally true for SJ17 (scroll jar)” (David Stacey, email, 24 March 2014).

margin of error, but that is not the important point. The important point is the circularity in which scribal hands of texts from Qumran’s caves were defined after 1951 as dated as late as the first century CE because those defining the palaeographic sequences believed Qumran scroll deposits at the time of the First Revolt had been firmly established archaeologically. No information in the years since has materially altered this epistemological circularity. Radiocarbon dates on Qumran texts that have been done until now have not altered this picture.28 7

The Biblical Texts of the Caves of Qumran Predate Stabilization of the Hebrew Biblical Text Which Occurred before the First Revolt

About 25 biblical texts have been found at Judean Desert sites other than Qumran. As brought out many times by Emanuel Tov, the biblical texts found at the sites other than Qumran are all exact-Masoretic (MT) in their text type, in contrast to a well-known pluriformity of the Qumran biblical texts.29 Tov further emphasizes that even 28  Space limitation prevents discussion here of radiocarbon datings (see Doudna, “The Sect,” 108–112). Suffice it to say that the existence of first-century CE Qumran texts can be established through radiocarbon dating in really only one of two ways. First, by obtaining individual Qumran text radiocarbon dates whose calendar date possibilities after calibration entirely postdate first century BCE at 95% confidence and exclude (by rechecking) that those results are from statistical variation or a contaminated sample. Or second (requiring a greater number of datings and a higher order of statistical analysis), by obtaining enough repeated radiocarbon dates of Qumran texts whose calendar date possibilities after calibration entirely postdate first century BCE at the weaker but narrower 68% confidence interval, even if first century BCE is not excluded at 95% confidence in any of the individual datings. Neither of these criteria have been met in the existing radiocarbon data. A reasonable assessment of the existing radiocarbon data is that of J. van der Plicht and K. Rasmussen: “The dates suggest a possible range from the third century BC to the first century AD for texts from caves near Qumran, with a strong concentration of probable dates in the second or first century BC” (Johannes van der Plicht and Kaare L. Rasmussen, “Radiocarbon Dating and Qumran,” in Holistic Qumran: Trans-Disciplinary Research of Qumran and the Dead Sea Scrolls; Proceedings of the NIAS-Lorentz Center Qumran Workshop 21–25 April 2008 [STDJ 87; ed. J. Gunneweg, A. Adriaens, and J. Dik; Leiden: Brill, 2010], 111). 29  Emanuel Tov, “The Dead Sea Scrolls and the Textual History of the Masoretic Text,” in The Hebrew Bible in Light of the Dead Sea Scrolls (FRLANT 239; ed. N. Dávid et al.; Göttingen: Vandenhoeck & Ruprecht, 2012), 42–43, 46.

DATING THE SCROLL DEPOSITS OF THE QUMRAN CAVES

in the case of the “MT-aligned ” biblical texts of Qumran, none were “carefully copied” in exact agreement with MT in the way that is the case for all of the biblical texts from all other Judean Desert sites. This description was confirmed by studies of Ian Young of 2002, 2005, and 2013.30 Young compiled statistics of “words per variant” of biblical texts from Qumran and the other Judean Desert sites compared to the medieval Leningrad Codex. Young found, in agreement with Tov, that the Qumran biblical texts differ systematically from the biblical texts from the other Judean Desert sites. At least most, and it is possible all, of the 25 biblical texts from the non-Qumran Judean Desert sites are pre-First Revolt in their dates of scribal production. But whether it is all, or only most, every non-Qumran biblical text from before the First Revolt is carefully-copied exact-MT. This is the true textual reality of first-century CE Judea before the First Revolt, reflected in later rabbinic traditions of careful copying of biblical texts in the pre-destruction standing Second Temple. What implications does this have for the Qumran biblical texts? Some think of Qumran as sui generis, a sect operating in isolation from the rest of contemporary Judaism, cut off from the temple and operating independently in a world of its own. Within this interpretive matrix it is then imagined that the Qumran caves’ range and variety of biblical texts, and the substantial subset of non-carefully-copied MT-type biblical texts among that mix in the Qumran caves, were contemporary in the first century CE with the 100% carefully-copied exact-MT biblical texts of all of the other Judean Desert sites. Then at the First Revolt, so the theory goes, both Qumran’s and Masada’s biblical texts came to an end at the same time, even though they are systematically different in kind. It is supposed that the people of the scrolls of Qumran’s caves intentionally and systematically boycotted the carefullycopied exact-MT biblical texts of the temple or synagogues associated with the temple through the entirety of the first century CE prior to the First Revolt. The ad hoc and intrinsically unlikely nature of the preceding reconstruction seems not to be sufficiently appreciated. There is neither evidence nor reason to suppose that the collectors of the Qumran texts boycotted the carefully-copied biblical texts identical to MT because the MT tradition had become associated with a rival sect, or because the collectors of the Qumran texts objected to biblical texts being copied too accurately, or because such 30  References above n. 2.

245 texts were being copied in the temple. These notions, for which there is no support in the Qumran texts themselves nor in any known ancient external testimony, appear to have been invented for the sole purpose of explaining anomalies created by the assumption of the First Revolt endpoint for the Qumran texts. These ad hoc explanations are even more questionable in light of a sea change in scholarly perception in recent years. No longer do many Qumran text scholars regard it as sensible that the people of the Qumran texts were wholly separated from the temple in Jerusalem. And once that idea goes—the notion of long-term, continuous, generational separation from the Jerusalem temple by the people of the Qumran texts—the issue necessarily emerges of how to understand the biblical texts of the caves of Qumran all of which differ so systematically from the non-Qumran exact-MT biblical texts of pre-First Revolt first-century CE Judea. What is the solution? Ian Young gave the only explanation that makes sense: quite simply, the Qumran biblical texts all predate the biblical texts of the other Judean Desert sites. All of the texts of the caves of Qumran are from before the stabilization of the biblical text which took place before the First Revolt. In this light the stabilized biblical text of the rabbis after the First Revolt was simply a continuation of a stabilized biblical text already in place in Judea before the temple was destroyed. The watershed for the stabilization of the Hebrew biblical text becomes not the disaster of the destruction of the temple and then accidental survival and fixing of the MT type of the late first century CE as the rabbis reorganized, as commonly supposed, but rather an earlier intentional text stabilization accompanied by programmatic scribal practice reforms implemented in the new temple of Herod perhaps in the closing years of the first century BCE or early first century CE, affecting all Hebrew biblical texts emanating from the temple from that point forward, copies used by all Jews of first-century CE Judea irrespective of party, all of whom participated in the temple in Jerusalem. In such a perspective the scrolls of Qumran become representative of a textual spectrum of the Judea of second half of the first century BCE before the biblical text was stabilized. The era of interest for the date and circumstances of the scroll deposits of the Qumran caves, the stabilization of the biblical text, and the conceivable relationship between these two phenomena, becomes the reign of Herod toward the end of the first century BCE, rather than the First Revolt and destruction of the temple which was still decades in the future.

246 8

DOUDNA

Conclusion: Albright’s Original Interpretation of Cave 1Q of 1949 Ironically Appears to have been Correct After All, and the Date Correction Following 1951 was the Mistake, Not Vice Versa. The Scrolls of the Caves of Qumran in their Entirety Take their Rightful Place as the Remains of a Lost Textual World before the First Century CE

I started this paper with attention to the striking drop in historical allusions in the Qumran texts, from flourishing numbers in the first two-thirds of the first century BCE to flat-line zero thereafter. I closed with the systematic differences between the biblical texts of Qumran, none exactMT, and the biblical texts of the other Judean Desert sites, all exact-MT. Both of these descriptions, each independently established on the basis of aggregate, large-scale data analyses, are simply stunning in impact if one takes a step back and looks at them with fresh eyes. Like flashing beacons these are signals in the data that the First Revolt endpoint for the Qumran cave texts is not correct. When all is said and done, these two signals are the argument and the evidence for the earlier dating. The principal reason these signals have not registered in common scholarly consciousness seems to be the palaeographic dates which are assumed to establish the existence of first-century CE dates of text copies found in the caves. But the absolute dates of the “late Herodian formal” and “post-Herodian formal” scribal hands defined by Cross 1961 are derivative from the flawed 1951 archaeological redating of the scroll deposits discussed earlier. The flawed archaeological redating of 1951 provided the framework or template within which Cross labored to accurately reconstruct the development of the scribal hands. The absolute datings of the “Herodian formal” hands reconstructed by Cross no doubt are close to correct but

for this question which devolves to issues of small numbers of decades that is not good enough. It is no disrespect to Cross’s formidable study of 1961 if today there is some critical engagement or nuancing or departure from what sometimes seems to have become a scholarly doctrine of inerrancy concerning the absolute datings of Cross 1961. The scribal hands must be reassessed free of presupposition that the Qumran cave texts continued to the time of the First Revolt. A shift in understanding in which the dates of the latest formal hands in the Qumran caves are situated perhaps in the time of Herod will not create a gap in typological development in the first century CE. The gap is filled by “late Herodian formal” developing in the time of Herod and “post-Herodian formal” developing in the first half of the first century CE. Once this is realized, no longer will the saying of Matt. 5:18 referring to iotas and k­ eraias in the writing of scribes scrupulously copying the books of Moses with letter-perfect accuracy, and, alluding to the decorative keraias of the most developed formal hands, be regarded as anachronistic. Matt. 5:18 may become recognized as a realistic allusion to scribal practice and ideology before the destruction of the temple, yet postdating the latest texts of Qumran. In this picture the various waves of people who were at Qumran in the first century CE disturbed, opened, looted, and possibly read the scrolls they encountered in the caves by accident, thus accounting for the anciently disturbed conditions in the caves closest to the site, with remains of anciently-opened scrolls such as the torn-off leather tabs and strings left on the floors of Caves 4Q and 8Q. There has as yet been no positive evidence set forth that first-century CE people at Qumran added any new literary texts to the ones they encountered in the caves— texts which may have seemed to them, as to us, as if they were from another world and time.

CHAPTER 17

History of the “Qumran Caves” in the Iron Age in the Light of the Pottery Evidence Mariusz Burdajewicz Since the late ‘40s and the beginning of the ‘50s of the last century, the north-eastern part of the Judean Desert has been witnessing numerous archaeological surveys and works (Table 17.1; Fig. 17.1). The main, but not the only one, goal of various expeditions sent to the Dead Sea region, was the quest for more and more scrolls. On this occasion, apart from manuscripts, many other artefacts, like pottery and the so-called small finds,1 have come to light. Their chronology range in date from the Chalcolithic to the Arab periods. The aim of this short paper is to present, on the basis of pottery evidence, some observations concerning the use of the caves during the Iron Age II–III. The later part of the Iron Age II, roughly the 8th–7th/6th centuries BCE, corresponds with the earliest phase of the settlement at Khirbet Qumran itself, represented mainly by pottery finds2 and possibly by some architectural structures.3 Although this conference is focused on the so-called Qumran caves, we cannot exclude from our investigation caves which are more remote from the Wadi, and the ruins of Qumran itself. Methodological approach, which has been chosen, requires to perceive the “Qumran caves” in the context of many other caves which pierce the rocky escarpment of the Rift Valley west of the Dead Sea littoral, and not as an isolated geographical/cultural phenomenon. In March 1952 the Qumran Cave Expedition (hereafter QCE) from the École Biblique et Archéologique in

Jerusalem carried out a survey of about 270 sites situated to the north and south of the Qumran site.4 Of the 40 sites (almost exclusively caves or cavities), where the traces of human presence from various periods were found, only few of them yielded the finds pertaining to the Iron Age. One large bowl and one lamp came from caves GQ 27 and GQ 39 respectively.5 Fragments of two vessels dated to the Iron Age II are mentioned as coming from cave GQ 13, and a few pottery sherds, possibly dated to the same period, from cave GQ 6.6 The survey was completed in 1956, and some additional pottery fragments from the Iron Age were found in cave 11Q (fragments of jars, two lamps and rounded juglet)7 and in two caves, A and B, situated near Cave GQ 8 (=3Q).8 Of the ten artificial caves carved in the marl terrace, in the very close vicinity of the settlement, only one (9Q) yielded material evidence from the Iron Age. It is a single fragment of a bowl, burnished horizontally. The bowl, dated by de Vaux to the 8th–7th centuries BCE,9 represents the same type as bowl from cave 27 mentioned above (cf. Fig. 4:1).10 In a survey conducted by Joseph Patrich at the turn of 1984 and 1985, Cave XIII/2, which also may be the very same one as that surveyed in 1952 by the QCE (=Cave 37), was investigated. The excavation yielded an interesting assemblage of objects, including pottery, as well as two burials dated to the Iron Age.11

1  See in this volume, Dennis Mizzi. 2  Mariusz Burdajewicz, “La poterie de l’âge du Fer à Khirbet Qumrân,” (forthcoming). 3  Roland de Vaux, Archaeology and the Dead Sea Scrolls (The Schweich Lectures 1959; rev. ed.; London: Oxford University Press for the British Academy, 1973), 2, Pl. 3. However, de Vaux’s reconstruction of the Israelite building has been recently questioned (Jean-Baptiste Humbert, “Arguments en faveur d’une résidence pré-essénienne,” in Fouilles de Khirbet Qumrân et ‘Aïn Feshkha: Études d’anthropologie, de physique et de chimie [ed. J.-B. Humbert and J. Gunneweg; Fribourg/ Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003], 468). See also Yitzhak Magen and Yuval Peleg, The Qumran Excavations 1993–2004: Preliminary Report (JSP 6; Jerusalem: Staff Officer of Archaeology, Civil Administration of Judea and Samaria/ Israel Antiquities Authority, 2007), 376–80, 407, Fig. 4.

4  Maurice Baillet, Józef T. Milik, and Roland de Vaux, eds.; Les ‘petites grottes’ de Qumrân: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; Oxford: Clarendon Press, 1962). 5  Roland de Vaux, “Archéologie,” in DJD 3:10, 12, Figs. 5.1, 5.3; Pl. 7. 6  De Vaux, in DJD 3:7–8. 7  Roland de Vaux, “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e 4e et 5e campagnes,” RB 63 (1956): 574. 8  De Vaux, in DJD 3:13, Cave A: “Quelques tessons du Fer II;” Cave B: “peu de tessons (. . .), davantage du Fer II.” 9  De Vaux, in DJD 3:31. 10  I am very grateful to prof. Joan Taylor from the King’s College London, for her help in identification of this bowl. 11  Joseph Patrich, “Hideouts in the Judean Wilderness: Jewish Revolutionaries and Christian Ascetics Sought Shelter and

© koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_019

248

BURDAJEWICZ

During the same survey, Cave 13 has been identified to the north of Cave GQ 8 (=3Q). According to the excavators, it may have been the same cave which is marked as No. 2 in the French publication. Among the discovered finds there were rim fragments of two cooking pots, one bowl and bowl base, all dated in general to the Iron Age II.12 The survey “Operation Scroll,” carried out on behalf of the Israel Antiquities Authority in the Northern Judean Desert,13 brought to light evidence of human presence in about 165 caves.14 Artifacts, mainly pottery, dated to the Iron Age have been found in 16 caves, of which only nine are located in the southern part of the surveyed area (regions X–XIV). Among these caves, six yielded pottery dated in general to the Iron Age; from two caves came evidence from the Iron Age II, while in the other cave the finds were assigned to the Iron Age II–III period. Very interesting is Cave XII/49,15 which probably was excavated in the past by Roland de Vaux.16 From here comes the largest assemblage of the pottery dated to the Iron Age II–III. The following types are represented: jars, bowls, cooking pots and juglets.17 From Cave XII/56 rim fragments of two jugs dated to Iron Age are reported,18 and from Cave XIII/13 rim fragment of a large bowl.19 Also three caves, XI/15, XIV/55 and XIV/56, yielded fragments of Iron Age pottery; unfortunately, no details and illustrations have been published.20

Moving to the south, two sites should be mentioned. The first site is located in Wadi Murabba‘at, about 18 km south of Khirbet Qumran, where four caves were investigated by the expedition of the École Biblique et Archéologique. Caves 1, 2 and 3 contained an unspecified number of Iron Age pottery sherds. The published finds include fragments of two hole-mouth jars with two (or four?) handles, one large bowl and one juglet from Cave 2, and only one fragmentary bowl from Cave 1. Cave 3, apart from some Iron Age pottery sherds (unpublished), yielded also other small objects, which could be possibly dated to the Iron Age.21 Roland de Vaux also published a water container (decanter) which, already after excavations, was found nearby in a hole in the rock, and a very interesting seven-nozzled lamp, which in turn was purchased from Bedouins.22 The other site is the “Cave of the Pool,” which was examined in the framework of the Expedition to the Judean Desert on behalf of the Israel Exploration Society. The Iron Age pottery found in the cave is represented by fragments of a bowl, two cooking pots, lamp, three jugs (decanters) and juglet. This pottery assemblage is dated by the excavators to the late 7th century BCE.23

Protection in Cliffside Caves,” BAR 15/5 (1989): 32–42 (34); Robert Eisenman, “The 1988–1922 California State University Dead Sea Walking Cave Survey and Radar Ground Scan of the Qumran Cliffs,” QC 9 (December 2000): 123; Nir Tal and Gershon Oron, “Survey and Excavations of Caves along the Fault Escarpment South of Qumran,” ʿAtiqot 41/2 (2002): 189. 12  Joseph Patrich and Benny Arubas, “A Juglet containing Balsam Oil (?) from a Cave near Qumran,” IEJ 39 (1989): 46, Fig. 3:2–5; Joseph Patrich, “Qumran Caves,” ESI 13 (1995): 64; Michel Itah, Yoni Kam, and Ronny Ben Haim, “Survey and Excavations of Caves along the Fault Escarpment South of Almog Junction and West of Qalya,” ʿAtiqot 41/2 (2002): 169–76. 13  Lior Wexler, ed., “Surveys and Excavations of Caves in the Northern Judean Desert (CNJD)—1993,” Part 1 [Hebrew], Part 2 [English], ʿAtiqot 41 (2002). 14  Lior Wexler, “Preface,” ʿAtiqot 41, Part 2 (2002): v–xviii. 15  Fawzi Ibrahim, “Region XII: The Excavation of Cave XII/49,” ʿAtiqot 41/2 (2002): 185. 16  De Vaux, in DJD 3:10, Fig. 5:1. 17  Ibrahim, “Region XII: The Excavation of Cave XII/49,” 185, Fig. 2:1–13. 18  Fawzi Ibrahim, “The Excavation of Cave XII/56,” ʿAtiqot 41/2 (2002): 186, Fig. 1:1–2 [Hebrew]. 19  Tal and Oron, “Survey and Excavations,” 191, Fig. 8:2. 20  Yuval Baruch, Gabriel Mazor, and Debora Sandhaus, “Region XI: Survey and Excavations of Caves along the Fault Escarpment above Ḥorbat Qumran,” ʿAtiqot 41/2 (2002): 179; Uzi Dahari, “Region XIV: Survey and Excavations of Caves along the Fault

Pottery is one of the main means that we have at our disposal to investigate various topics concerning the history of the caves. However, the Iron Age material from the caves, with regard to the pottery finds, is not numerous, and the repertoire of ceramic types is quite limited. There is also no real Iron Age stratigraphy in the caves, which, in addition, were re-used and disturbed in the later periods. It means that any analysis of the caves’ chronology, history, etc., has to be based on comparative pottery research. Inevitably, we must refer to the better known and more abundant pottery assemblages, coming from other sites, particularly from Khirbet Qumran, which is the closest

1

Discussion of Pottery Types

Escarpment from Naḥal Kidron to Naḥal Deragot,” ʿAtiqot 41/2 (2002): 217; Wexler, “Preface,” Table 1. It seems that in Table 1, the Cave 56 is mistakenly assigned to Region XIII (there is no such cave on Map XIII, 219); Cave 56 is, however, located in Region XIV on Map XIV. 21  These finds include two iron knives and fragments of decorated bone plaque and iron shaft: Roland de Vaux, “Archéologie,” in Les grottes de Murabbaʿât (DJD 2; ed. P. Benoit, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1961), Figs. 10:8, 10, Pl. Vib: 4, 5; Fig. 12:6, Pl. Vib: 2; Fig. 11:19, Pl. Xia: 5. 22  De Vaux, in DJD 2:26–28, Fig. 6. 23  Nahman Avigad, “Expedition A—Nahal David,” IEJ 12 (1962): 174, Fig. 4, 1–10.

HISTORY OF THE “ QUMRAN CAVES ” IN THE IRON AGE

249

1.2 Holemouth Jars Of some interest are two holemouth jars found in Cave 2 in Murabba‘at (Fig. 17.2:3–4). They are characterized by wide mouth, a thick, oblong in section horizontal rim folded inwards which, in one case, has an external ridge; below the rim, a very short concave “neck” sharply widens downwards; one of the jars has rounded, the other carinated upper part of the body. Both jars had either two or four vertical handles drawn from the bottom of the “neck” to the body. Roland de Vaux cited vessels from Megiddo, 1.1 Necked Jars There are only two examples of necked jars, both coming Stratum III, Lachish, Stratum III, and Tell Beit Mirsim, from Cave XII/49 (Fig. 17.2:1–2). One of them has slightly Level A, as possible parallels.33 In our opinion the best parallels are Types 8A, B in inverted, thickened rim oblong in cross-section; the other one is thickened to the exterior and circular in cross-­ Gezer. The first one is dated to the 8th/7th century BCE, section. As only the rims and part of the slightly con- while the second one to the 7th–6th centuries BCE. Type cave necks are preserved, it is impossible to make a more 8A, although has parallels in the north, is typical southprecise statement as to their general shape and dating. ern form, which appears in sites of the Judean Hills, the Nevertheless, it can be said that such rims and necks are Shephelah and the northeast Negev.34 Related forms, common in the 7th–6th centuries BCE on the sites exca- dated to the late 7th/early 6th centuries BCE, are found in vated in the Judea and the Negev. At the end of the 6th cen- Kadesh Barnea, Stratum 2,35 and Horvat Qitmit.36 However, the jars from Wadi Murabbaʿat seem to be tury BCE they become less and less ­common.24 Parallels are known, among others, from Khirbet Qumran,25 En Gedi, quite early in this series due to their close similarity to a Stratum V,26 Jerusalem (Ophel),27 Arad, Stratum VII and four-handled pithos from Stratum IV in Beer Sheba, dated VI,28 Tell ‘Ira, Strata VII and VI,29 Beer Sheba, Stratum II,30 to the 10th/9th century BCE,37 and considered as the prototype of Type 8A in Gezer.38 Tell Masos, Area G,31 and Stratum, Qadesh Barnea 2.32 settlement site to the “Qumran caves,” and from the settlement in En Gedi (Tel Goren) located ca. 35 kilometers to the south. These are the only two settlements that taken together provide a good and rather complete picture of ceramics used in the region of the Dead Sea during the Iron Age II–III. Other important sites in Judea, such as Jerusalem for example, will also be taken into consideration in comparative study of pottery finds.

24  Miriam Aharoni and Yohanan Aharoni, “The Stratification of Judahite Sites in the 8th and 7th Centuries B.C.E.,” BASOR 224 (1976): 73–90. 25  Burdajewicz, “La poterie de l’âge du Fer.” 26  Irit Yezerski, “Pottery of Stratum V,” in En-Gedi Excavations I: Conducted by B. Mazar and I. Dunayevsky; Final Report (1961– 1965) (ed. E. Stern; Jerusalem: Israel Exploration Society, 2007), Pl. 9:5–10. 27  Eilat Mazar and Benjamin Mazar, Excavations in the South of the Temple Mount: The Ophel of Biblical Jerusalem (Qedem 29; Jerusalem: Institute of Archaeology, Hebrew University, 1989), Pl. 27:5. 28  Ze’ev Herzog et al., “The Israelite Fortress at Arad,” BASOR 254 (1984): Figs. 25:15; 29:11. 29  Liora Freud, “Pottery: Iron Age,” in Tel ʿIra: A Stronghold in the Biblical Negev (ed. I. Beit Arieh; Tel Aviv: Institute of Archaeology, Tel Aviv University, 1999), Stratum VII = Fig. 6.75:1; Stratum VI = Figs. 6.62:19; 6.91:13; 6.69:14; 6.101:6. 30  Yohanan Aharoni, ed., Beer-Sheba I: Excavations at Tel Beer-Sheba, 1969–1971 Seasons (PIA 2; Tel Aviv: Institute of Archaeology, Tel Aviv University, 1973), Pl. 73:4. 31  Orna Zimhoni, “The Pottery,” in Ergebnisse der Ausgrabungen auf der Hirbet el-Msas (Tel Masos) 1972–1975 (ADVP 6; ed. V. Fritz and A. Kempinski; Wiesbaden: Otto Harrassowitz, 1993), Vol. 1: Pl. 166:11, 14. 32  Hannah Bernick-Greenberg, “The Ceramic Assemblages and the Wheel-Made Pottery Typology,” in Excavations at Kadesh Barnea (Tell el-Qudeirat) 1976–1982 (IAA Reports 34; ed. R. Cohen and

1.3 Jugs and Juglets A complete jug and fragments of rims and necks from Cave 2 in Wadi Murabba‘at, Cave XII/49 and Cave of the Pool, belong to a large family of vessels widely known as decanters (Fig. 17.3:1–5). Characteristics of this type are everted simple rim, either rounded or triangular in crosssection, ridge in mid-height of the neck, handle drawn from the ridge to the carinated shoulders, and ring base. It is assumed that the vessel profile developed from a square to an elongated body.

H. Bernick-Greenberg; Jerusalem: Israel Antiquities Authority, 2007), Pls. 11.77:2. 33  De Vaux, in DJD 2:26 and references. 34  Seymour Gitin, Gezer III: A Ceramic Typology of the Late Iron II, Persian and Hellenistic Periods at Tell Gezer (Annual of the Nelson Glueck School of Biblical Archaeology 3; Jerusalem: Hebrew Union College, 1990), 128–30, Pls. 18; 1, 2; 26:31. 35  Bernick-Greenberg, “The Ceramic Assemblages,” Figs. 11.78:4; 11.127:20. 36  Liora Freud and Itzhak Beit Arieh, “Pottery,” in Horvat Qitmit: An Edomite Shrine in the Biblical Negev (MSSMNIA 11; ed. I. Beit Arieh; Tel Aviv: Institute of Archaeology, Tel Aviv University, 1995), Fig. 4.6:36. 37  Aharoni, Beer-Sheba I, Pl. 55, 20. 38  Gitin, Gezer III, 128.

250 Decanter is one of the most distinctive types of the late Iron Age ceramic repertoire. Such vessels have been discovered in many excavated sites both in the north and in the south. However, they were especially popular in the Judean sites and in the Negev. The type occurs from the 8th through the early 6th century BCE. Although it is assumed that decanters were used to serve mainly water, David Ussishkin admits that these jugs could have also been used as wine containers.39 Parallels to this type come from Khirbet Qumran,40 En Gedi, Stratum V,41 Jerusalem (Cave I,42 and Ophel43), Ramat Rahel, Strata VA et B,44 Lachish, Strata III–II,45 Arad, Strata VII–VI,46 Tell Masos, Area G,47 and Kadesh Barnea, Stratum 2.48 From Cave 2 in Murabbaʿat and Cave of the Pool in Naḥal David some fragments belong to small juglets (Fig. 17.3:8–10). Specifically, they represent a type of dipper juglet which is characterized by an elongated cylindrical body, neck concave in profile, ending with a flaring plain rim, a single handle attached to the rim and sloping shoulders, and a rounded base. This type of juglet is common in Judean sites since the 8th century BCE until the early 6th century BCE. Doubtlessly such vessels can be considered as one of the most diagnostic types of the 7th/6th century BCE ceramic horizon. Among the numerous dipper juglets excavated so far, we can cite these from En Gedi, Stratum V,49 Jerusalem

39  David Ussiskhin, “Excavations at Tel Lachish 1973–1978,” Tel Aviv 5 (1978): 88. 40  Burdajewicz, “La poterie de l’âge du Fer.” 41  Yezerski, “Pottery of Stratum V,” Pl. 8. 42  Itzhak Eshel, “Two Pottery Groups from Kenyon’s Excavations on the Eastern Slope of Ancient Jerusalem,” in Excavations by K. M. Kenyon in Jerusalem 1961–67 (BAMA 6; ed. I. Eshel and K. Prag; Oxford: Oxford University Press, 1995), Vol. 4:Fig. 25:5. 43  Mazar and Mazar, Excavations in the South of the Temple Mount, Pl. 28:5–8. 44  Yohanan Aharoni, Excavations at Ramat Rahel: Seasons 1961 and 1962 (Serie Archeologica 6; Rome: Università degli Studi, Centro Di Studi Semitici, 1964), Figs. 20:12; 35:4; idem, Excavations at Ramat Rahel: Seasons 1959 and 1960 (Serie Archeologica 2; Rome: Università Degli Studi, Centro Di Studi Semitici, 1962), Figs. 11:27, 28, 43. 45  Orna Zimhoni, “The Pottery of Levels III and II,” in The Reneved Archaeological Excavations at Lachish (1973–1994) (MSSMNIA 22; ed. D. Ussishkin; Tel Aviv: Emery and Claire Yass Publications in Archaeology, Tel Aviv University, 2004), Vol. 4:Fig. 26.51:1–5. 46  Herzog et al., “The Israelite Fortress,” Figs. 25:11; 29:8. 47  Zimhoni, “The Pottery,” Pl. 165:15, 16. 48  Bernick-Greenberg, “The Ceramic Assemblages,” Pl. 11.75:14. 49  Yezerski, “Pottery of Stratum V,” Pl. 6:13–16.

BURDAJEWICZ

(City of David, Stratum 12),50 Lachish, Stratum III,51 Tel Batash, Stratum II,52 Gezer, Strata VIA–VB/V/A,53 Arad, Strata X–VIII,54 Horvat ‘Uza, Stratum III.55 This type is also known from the sites excavated in the north.56 Two pottery fragments from Cave XII/49 and Cave of the Pool (Fig. 17.3:6–7) seem to belong to small juglets. However, they are too small to indicate their form and propose dating. Little can be said about two pottery fragments found in Cave XII/56 (Fig. 17.3:11–12). According to the excavator, the first one can be compared with a one-handle jug from En Gedi.57 The second fragment can be identified either as handless neck jug or miniature jar. Related early 6th century BCE forms are found in Jerusalem (Armenian Garden), Period IA.a58 and Ramat Rahel.59 1.4 Bowls Bowls (Fig. 17.4:1–9) are the most numerous among the pottery vessels from the caves. All of them belong to one general type which is characterized by thickened, folded out rim, which can be either thick or slender and elongated;60 the body can be either rounded or slightly 50   Alon De Groot and Donald T. Ariel, “Ceramic Report,” in Excavations at the City of David 1978–1985 Directed by Yigal Shiloh. Vol. 5: Extramural Areas (Qedem 40; ed. D. T. Ariel; Jerusalem: Institute of Archaeology, Hebrew University, 2000), Fig. 27:11. 51  Zimhoni, “The Pottery of Levels III and II,” Fig. 26.4; 13–14. 52  Amihai Mazar and Nava Panitz-Cohen, Timnah (Tel Batash) II: The Finds from the First Millenium BCE (Qedem 42; Jerusalem: Institute of Archaeology, Hebrew University, 2001), 124–25. 53  Gitin, Gezer III, Pls. 19:10–12, 14; 23:9. 54  Lily Singer-Avitz, “Arad: The Iron Age Pottery Assemblages,” Tel Aviv 29 (2002): 155. 55  Liora Freud, “Iron Age Pottery,” in Horvat ʿUza and Horvat Radum: Two Fortresses in the Biblical Negev (ed. I. Beit-Arieh; Tel Aviv: Institute of Archaeology, Tel Aviv University, 2007), Fig. 3.40:13–14. 56  Gitin, Gezer III, 156–61, with discussion and bibliography. 57  Ibrahim, “The Excavation of Cave XII/56,” 186. See also Benjamin Mazar, Moshe Dothan and Immanuel Dunayevsky, “En-Gedi: The First and Second Seasons of Excavations, 1961–1962,” ʿAtiqot 5 (1966): Fig. 31:11. 58  Arlotte Douglas Tushingham, Excavations in Jerusalem 1961–1967 (Toronto: Royal Ontario Museum, 1985), Vol. 1:Fig. 4:18; Mazar and Mazar, Excavations in the South of the Temple Mount, Fig. 1:28. 59  Aharoni, Excavations at Ramat Rahel: Seasons 1961 and 1962, Fig. 11:15. 60  For variants of rim shape of this type of bowls, see: Hendricus J. Franken, “The Twelve Pottery Classes,” in Excavations in Jerusalem 1961–1967: The Iron Age Extramural Quarter on the South-East Hill (BAMA 2; ed. H. J. Franken and M. L. Steiner; Oxford: Oxford University Press, 1990), 2:103, Figs. 6–18.

HISTORY OF THE “ QUMRAN CAVES ” IN THE IRON AGE

carinated; the base is flat, sometimes slightly concave. The bowls could have also had red/brown slip and horizontal burnishing. Such bowls are common throughout the country, both in the north and in the south, and they range in time from the end of the 10th century until the beginning of the 6th century BCE. For that reason they cannot serve as a precise chronological indicator. Such bowls are very common in Khirbet Qumran, where they constitute almost 60% of all vessels.61 In En Gedi, Stratum V, they correspond to Type B I.II.62 In the region of the Dead Sea and Judean Desert they are also known from Rujm el-Bahr, Qasr el-Yahud, ‘Aïn Ghuweir, ‘Ein et-Turaba.63 These bowls correspond to Franken’s Class 4 in Jerusalem (South-East Hill), where they occurred already in Phase 2 of the 9th/8th century, but the most popular were in Phases 7–8, dated to the 7th century BCE.64 Numerous examples were also found in Cave I, dated to between 698 and c.650 BCE.65 In Gezer Stratum VA, they correspond to types BL 50A–C (rounded) and BL 71 (slightly carinated) dated to the 7th–6th century BCE.66 They occur in Timnah as Type Bl 13 in Strata II–III,67 and in Lachish, Stratum II.68 Other numerous parallels can be found in the Negev: Arad X,69 Horvat ʿUza III,70 Tell ʿIra VII–VI,71 Beer Sheba II,72 and Tell Masos, Area G.73 1.5 Large Bowl/Krater One fragment from Cave XIII/13 (Fig. 17.4:10) belongs to a large, deep bowl with inverted rim thickened to the exterior. The profile of the body could have been either rounded or carinated. Usually either two or four vertical handles were attached to the rim and the body. The diameter of the mouth was always very large, between 30 and 61  Burdajewicz, “La poterie de l’âge du Fer.” 62  Yezerski, “Pottery of Stratum V,” 87–88. 63  Pesach Bar Adon, “Excavations in the Judean Desert,” ʿAtiqot 9 (1989) [Hebrew], 8, Fig. 9:1 (Rujm el-Bahr); 28, Fig. 2:1, 3 (Qasr el-Yahud); 38, Fig. 8:1–6 (ʿAïn Ghuweir); 45, Fig. 5:5–10 (ʿEin et-Turaba). 64  Franken, “The Twelve Pottery Classes,” 103–4, Figs. 6–16, 6–18. 65  Eshel, “Two Pottery Groups,” 41, Fig. 14, 1–22. 66  Gitin, Gezer III, 167–68, 195, Pl. 27:24, 28. 67  Mazar and Panitz-Cohen, Timnah (Tel Batash) II, 39–40. 68  Zimhoni, “The Pottery of Levels III and II,” Fig. 26.54:3–5, 10–12, 26.55:1–4, 24–28. 69  Herzog et al., “The Israelite Fortress,” Fig. 12:6. 70  Freud, “Iron Age Pottery,” Fig. 3.15:3. 71  Freud, “Pottery: Iron Age,” figs. 6.59:6–11; 6.61:3. 72  Aharoni, Beer-Sheba I, Pl. 59:63–71. 73  Zimhoni, “The Pottery,” Pl. 163:8–14.

251 40 cm. This type of vessel, used to prepare and/or serve food at the table, occurs already at the end of the 9th century BCE and its form, mainly the rim shape, evolves gradually in the next centuries. Its apogee in the Judean Hills, in the Negev and on the Shephelah, seems to fall on the 7th century BCE Parallels can be cited from En Gedi, Stratum V,74 ʿAïn Ghuweir,75 Jerusalem,76 Tell Beit Mirsim, Level A,77 Lachish, Strata III and II,78Arad VII,79 Kadesh Barnea, Stratum 2,80 and Tell Masos, Area G.81 1.6 Cooking Pots Among the cooking pots discovered in the caves, four distinct types can be distinguished. 1. Closed cooking pot with high flaring neck and rounded rim; in the lower part of the neck there is a sharp carinated ridge; globular body and two vertical loop-handles are drawn from rim to shoulders (Fig. 17.5:1). This type appears in Judean sites from the mid-7th century until the beginning of the 6th century BCE. Strange enough, not a single example was attested in Khirbet Qumran. However, this type was found in En Gedi, Stratum V.82 Other p ­ arallels were found in Ein et-Turaba,83 Jerusalem (City of David),84 Lachish, Level II,85 Tel Masos, Area G,86 Tel ʿIra, Stratum VI,87 Horvat ʿUza, Stratum III, and Horvat Radum.88 2. Wide mouth cooking pot with inverted neck/rim internally grooved or straight, with external ridge; two loop-handles attached at rim and shoulders; shallow 74  Yezerski, “Pottery of Stratum V,” Pl. 2:2–9. 75  Bar Adon, Excavations, 38, Fig. 8:9. 76  Franken, “The Twelve Pottery Classes,” 116–17, Fig. 6–50. 77   William F. Albright, The Excavations at Tell Beit Mirsim in Palestine. Vol. 1: The Pottery of the First Three Campaings (AASOR 21–22; New Haven: American Schools of Oriental Research, 1932), Pl. 20:7–8, 10–12. 78   Zimhoni, “The Pottery of Levels III and II,” Fig. 26.54:13, 26.56:7–11. 79  Herzog et al., “The Israelite Fortress,” Fig. 25:6. 80  Bernick-Greenberg, “The Ceramic Assemblages,” Fig. 11.72:11. 81  Zimhoni, “The Pottery,” Pls. 163:17, 18; 164:1–6. 82  Yezerski, “Pottery of Stratum V,” Pl. 4:6–14. 83  Bar Adon, Excavations, Fig. 6:3–5. 84  Yagel Shiloh, “A Group of Hebrew Bullae from the City of David,” IEJ 36 (1986): Fig. 6:16. 85  Zimhoni, “The Pottery of Levels III and II,” Fig. 26.53:5–6; 26.55:17; 26.56:14. 86  Zimhoni, “The Pottery,” Pl. 165:1–4. 87  Freud, “Pottery: Iron Age,” Figs. 6.91; 4–5; 6.100:12–13. 88  Freud, “Iron Age Pottery,” Figs. 3.17:3; 3.18:11; 3.35:6–9 (Horwat ‘Uza); 16.4:10, 17 (Horwat Radum).

252 g­ lobular (?) body (Fig. 17.5:2–3). This type appeared at the end of the 8th century and continued into the 7th century BCE. Close parallels are found in Jerusalem, Cave I,89 Tell es-Nasbeh,90 Beer Sheba, Stratum II,91 Arad, Level 8.92 3. Wide mouth cooking pot with everted, horizontally grooved rim; the body can be either globular or squat bulging; two loop-handle were drawn from the rim to the sloping shoulder (Fig. 17.5:4). This type appeared already in the 8th century BCE, and became the commonest type of cooking pots at the end of the Iron Age. Parallels are found in Jerusalem (City of David),93 Lachish, Stratum III,94 Arad, Strata X–VIII.95 Related examples are found also in Gezer, Stratum VA (7th–6th centuries BCE).96 4. Wide mouth cooking pot with ridged-stepped rim inclined inwards; body profile can be biconical or rounded; two loop-handle attached at rim and shoulder (Fig. 17.5:5–8). This is a typical Judean cooking pot of the 8th and early 7th centuries BCE. However, this type is very rare in the north. Parallels are found in Arad, Strata X–VIII (second half of the 9th and the 8th centuries BCE,97 Beer Sheba, Stratum III (9th/8th century BCE),98 Tell Beit Mirsim A2,99 Gezer, Strata VIA and VB/VA (mid-8th century BC–7th century BCE).100 1.7 Lamps All lamps found in Caves 11Q, GQ39 and Cave of the Pool (Fig. 17.6:1–4) belong to one general type. It is characterized by a high and thick flat base, bowl-shaped body, everted rim, which may be either horizontal or oblique, rounded or tapered; the rim is pinched or folded to form a nozzle. Such lamps were very common both in the north and in the south during the Late Iron Age, that is in the 7th—beginning of the 6th centuries BCE. It is a com89  Eshel, “Two Pottery Groups,” Fig. 18:4–11. 90  Joseph C. Wampler, Tell en-Nasbeh: Excavated under the Direction of the late William Frederic Badè; II. The Pottery (Berkeley/ New Haven: Palestine Institute of Pacific School of Religion/ American Schools of Oriental Research, 1947), Pls. 48, 48. 91  Aharoni, Beer-Sheba I, Pl. 61:86. 92  Herzog et al., “The Israelite Fortress,” Fig. 22:7. 93  De Groot and Ariel, “Ceramic Report,” Figs. 16:19–20; 19:27; 25:15–16. 94  Zimhoni, “The Pottery of Levels III and II,” Fig. 26.4:8. 95  Singer-Avitz, “Arad: The Iron Age,” 110–214. 96  Gitin, Gezer III, 219–21, Pl. 27:11–14. 97  Aharoni and Aharoni, “The Stratification,” Fig. 3:3. 98  Aharoni, Beer-Sheba I, Pl. 56:13. 99  Albright, The Excavations at Tell Beit Mirsim, Vol. 1:Pl. 56:3; idem, The Excavations at Tell Beit Mirsim. Vol. 3: The Iron Age (AASOR 21–22; New Haven: Yale University Press, 1943), Pl. 19.1, 4. 100  Gitin, Gezer III, 69, 217–19, Pls. 22:2–3, 5; 24:14.

BURDAJEWICZ

monly accepted opinion that these lamps developed from a very similar type with a low base which occurs in the 8th century BCE, or even earlier.101 In such a case, lamp 11Q–8, with somewhat lower base, would represent a transitional type between early and late type of lamps. We should note that this lamp was found in the same cave as the high-base lamp 11Q–5. This fact may indicate that either there were two distinct phases of human presence in the cave, or that lamp 11Q–8 was in continuous use for a long time and eventually was used simultaneously with the late type. Since lamps of this type are well known, and many of them excavated in Judean sites have been widely published, we will limit ourselves to a few parallels from the following sites: En Gedi, Stratum V,102 Jerusalem (the Armenian Garden and Ophel),103 Ramat Rahel Stratum V,104 Lachish, Stratum II,105 Timnah (Tel Batash) Stratum II,106 Arad, Strata VII–VI,107 Tell Masos, Area G,108 Kadesh Barnea, Stratum 2.109 For lamp 11Q–8 good parallels are 8th century BCE lamps from Arad, Strata IX–VIII and Lachish, Stratum III.110 Seven-nozzled lamp with a disc base from Wadi Murabba‘at site is of particular interest (Fig. 17.6:5). It is the only lamp of this type so far known from the discussed area. Even though such lamps occur at several sites, both in the south and the north of the Palestine, they are not very common. Probably the earliest example of seven-nozzled lamp comes from the Middle Bronze IIA–B cult place at Nahariyya to the north of Akko.111 This type becomes more popular in the 10th century BCE, or even slightly earlier.112 In the north they were discovered at the following sites: Hazor, Strata IV–V in Area B, dated to the 8th/7th century 101  Juan M. Tebes, “The Pottery Assemblage of Jerusalem’s NeoBabylonian Destruction Level: A Review and Discussion,” Antiguo Oriente 9 (2011): 303. 102  Yezerski, “Pottery of Stratum V,” Pl. 11:6–11. 103  Arlotte Douglas Tushingham, Excavations in Jerusalem 1961–1967 (Toronto: Royal Ontario Museum, 1985), Vol. 1:Fig. 4:18; Mazar and Mazar, Excavations in the South of the Temple Mount, Pl. 8:13. 104  Aharoni, Excavations at Ramat Rahel: Seasons 1959 and 1960, figs. 11:36–38; 27:8; 28:50. 105   Zimhoni, “The Pottery of Levels III and II,” Figs. 26.54:17; 26.55:18–20. 106  Mazar and Panitz-Cohen, Timnah, 133–34. 107  Herzog et al., “The Israelite Fortress,” Figs. 25:14; 29:10. 108  Zimhoni, “The Pottery,” Pl. 166:16–19. 109  Bernick-Greenberg, “The Ceramic Assemblages,” Pls. 11.78:12; 11.83:7; 11.115:17. 110  Herzog et al., “The Israelite Fortress,” figs. 18:13; 22:17; Zimhoni, “The Pottery of Levels III and II,” 1794, Fig. 26.5:7. 111   Moshe Dothan, “The Excavations at Nahariya: Preliminary Report (Seasons 1954/55),” IEJ 6 (1956): 19, Fig. 4. 112  Albright, The Excavations at Tell Beit Mirsim 1, 70–71.

253

HISTORY OF THE “ QUMRAN CAVES ” IN THE IRON AGE

BCE,113 Tell el-Fara’ah (North) in the 7th century context (Stratum VIIe),114 Megiddo, Stratum III,115 and Beth Shean, Upper Level V (on the tell,)116 and the Northern Cemetery (Tomb 66 A–C).117 In the south, this type occurs in Tell Beit Mirsim, stratum B, Beth Shemesh IIb, Tell en-Nasbeh.118 Outside Palestine, a few examples of this type of lamp are known from Cyprus. In Amathus a seven-nozzled lamp with a flat thin base, was found in a bothros, containing material dated to the Cypro-Archaic I Period, i.e. 750–600 BCE.119 Two other examples were found in a sanctuary in Limassol-Komissariato, dated to the CyproArchaic period, 750–475 BCE. According to the excavator, the lamps may be connected with some cultic context, possibly with strong Phoenician influence.120 Another seven-nozzled lamp of unknown provenance is in the collection of the Cyprus Museum in Nicosia.121 In the light of the comparative evidence presented above, the presence of this type of lamp in Murabba‘at is rather surprising, and does not match the whole picture of the ceramics from the area of the Dead Sea and the Judean Desert. The pattern of distribution of seven-nozzled lamps suggests that they were more common in the northern sites than in the southern ones. It should also be stressed that the lamp in question was not found in situ by de Vaux’s team, but was purchased from the Bedouins. In 113  Yigael Yadin et al., Hazor. Vol. 1: An Account of the First Season of Excavations, 1955 (James A. Rothschild Expedition at Hazor; Jerusalem: Magnes Press, 1958), Pl. 71:25. 114  Alain Chambon, Tell el-Far’a. Vol. 1: L’âge du Fer (Mémoire 31; Paris: Éditions Recherche sur les Civilisations, 1984), 68, Pl. 59, 21. 115  Robert S. Lamon and Geoffrey M. Shipton, Megiddo I: Seasons of 1925–34, Strata I–V (Oriental Institute Publication 42; Chicago: University of Chicago Press, 1939), 171, Pl. 37, 16. 116  Eliezer D. Oren, The Northern Cemetery of Beth Shan (Leiden: Brill, 1973), 110, Fig. 42b, 19–20. 117  Frances W. James, The Iron Age at Beth Shan: A Study of Levels VI–IV (Philadelphia: University Museum, University of Pennsylvania, 1966), 126, 4. 118  Albright, The Excavations at Tell Beit Mirsim, 1:70–71, Pl. 23, 3; Elihu Grant and George E. Wright, Ain Shems Excavations V (Text) (Biblical and Kindred Studies 8; Haverford: Haverford College, 1939), 140; Wampler, Tell en-Nasbeh, Pl. 71:1652. 119   Karin Nys, “Terre cuite,” in Amathonte VI: Le sanctuaire d’Aphrodite des origines au début de l’époque imperial (Études chypriotes 17, ed. S. Fourrier and A. Hermary; Athènes/Paris: École française/De Boccard, 2006), 93, Fig. 407, Pl. 18:6. 120  Vassos Karageorghis, Two Cypriotes Sanctuaries of the End of the Cypro-Archaic Period (Rome: Consiglio Nazionale delle Ricerche, 1977), 54, nos 57–58, Fig. 18, Pl. 23 and remarks on page 62. 121  Therese Oziol, Les lampes du musée de Chypre: Salamine de Chypre VII (Paris: De Boccard, 1977), 26–27, no. 39, Pls. 3, 54.

fact, we know nothing certain about its actual origin. All these facts make its presence in such remote place more than suspicious. 2 Summary To sum up, only a small numbers of caves, where traces of human presence have been found, yielded the material evidence pertaining in general outline to the Iron Age II–III, that is from the 8th throughout the early 6th centuries BCE. Since, as we have already mentioned, in the majority of these caves there was no real stratigraphy, and the finds from various periods were often mixed up, the dating of the caves’ occupation cannot be precisely determined. Typological comparison which could be made with the pottery discovered in two important settlement sites, namely Khirbet Qumran and En Gedi, permits to narrow the date of some of the finds to the 7th or even to the late 7th–early 6th century BCE. It does not mean of course that all these caves were occupied in the same time. We can only safely say that over a period of about two centuries, some of these caves have been visited and/ or even inhabited. As for the function of the caves, it seems that each one of them should be treated as a separate, individual case study. After all, we must not forget that the natural conditions in the caves were not favorable to permanent residence.122 Probably most of them served as temporary shelters in case of a threat or hiding-places for all kinds of “outlaws.” Notable is the rarity of vessels for storage and for food preparation. There were also no other types of table vessels, like plates and shallow bowls with simple or everted rims, which were common in Khirbet Qumran,123 En Gedi and many other late Iron Age sites in Judean Hills. Apparently, daily life needs of the caves’ visitors were very basic. Nevertheless, some of the caves could have been indeed used as places of permanent or at least seasonal occupation. For example, Cave 2 in Wadi Murabba‘at, apart from bowls, jugs and a lamp (?), contained also two heavy jars used rather for storage than for transportation. Moreover, other small finds, including personal objects, were also present in these caves,124 which suggests a period of shorter or longer occupation, rather than a single, overnight visit. 122  See Jean-Baptiste Humbert in this volume. 123   Burdajewicz, “La poterie;” Yezerski, “Pottery of Stratum V,” Pls. 1:1–8; 3:–5. 124  See above, n. 20.

254 However, due to the paucity of finds from the period in question, far reaching conclusions regarding the usage of the caves tend to be risky. It seems that such is the case of the artificial caves cut in the marl terrace. According to Frank Moore Cross, these caves, judging by their “shape and floor plan,” were originally “typical” Iron Age tombs made by the inhabitants of the Qumran village. Later, in the early Roman period, the caves were reused as places of habitation by the Essenes.125 However, this is rather personal opinion expressed in an interview than actual scientific hypothesis based on solid grounds. It is enough to recall that we do not know the original shape of these caves since they were damaged in the course of time both by natural conditions and human activity. More plausible, but not to be proven at this moment, it the idea that these caves have been cut as places of habitation by the first settlers in the Iron Age II, before they, or their successors, eventually settled down on the plateau. Finally, it should be noted that combined results of all surveys conducted thus far in this area clearly indicate that the majority of the caves (in total 13) used within the Iron Age II–III periods are located in the very north-­ western part of the Dead Sea region. It is a marked cluster of caves in close proximity to the settlement on the plateau (Fig. 17.1). Further to the south, there are only three caves

125  Hershel Shanks, Frank Moore Cross: Conversations with a Bible Scholar (Washington DC: Biblical Archaeology Society, 1994), 114–15.

BURDAJEWICZ

with Iron Age pottery, XIII/13/1–5, XIV/55 and XIV/56, scattered within a certain distance from each other. The next marked cluster of four caves is located as far as Wadi Murabba‘at, in some distance from the Dead Sea shore. It is an intriguing fact that in the vicinity of En Gedi, which beside Qumran is another large settlement on the western shore of the Dead Sea, only one cave, namely the Cave of the Pool in Naḥal David, yielded some Iron Age pottery finds. Moreover, in excavator’s opinion this cave was not occupied for an extended period of time. When we take into consideration the Iron Age pattern of cave-occupation in the escarpment along the Dead Sea littoral, the contrast between the southern part near En Gedi and the northern part near Khirbet Qumran is striking and cannot be accidental. It seems also unlikely, that there were no contacts between the “users” of the caves and the inhabitants of the Qumran village. Unfortunately, the archaeological sources at our disposal today are still too scarce to determine precisely the mutual relationships between the villagers and the cavemen. Doubtlessly, this is an issue requiring further in-depth studies. Meanwhile, we may temporarily state that, with regard to the Iron Age period at least, the northern cluster of caves seems to really deserve the name “Qumran caves.”

255

HISTORY OF THE “ QUMRAN CAVES ” IN THE IRON AGE Table 17.1

Juxtaposition of caves with the Iron Age pottery

Caves with Iron Age Pottery Finds No. QCE Patrich CNJD survey

1

GQ 2

13

2 3

Murabbaʿat expedition

EJD

Period

References

X/35

Iron II

X/47 XI/15

Iron Iron II

Patrich and Arubas, “A Juglet containing Balsam Oil,” Fig. 3:4–5; Itah, Kam and Ben Haim, “Survey and Excavations,” 171 Wexler, “Preface,” CNJD 2, Table 1 Baruch, Mazor and Sandhaus, “Survey and Excavations,” 179 de Vaux, “Archéologie,” 7 de Vaux, Roland, “Fouilles de Khirbet Qumrân. Rapport préliminaire,” 574 de Vaux, “Archéologie,” 13

4 5

GQ 6? 11Q

Iron Iron II

6 7 8 9 10

A B GQ 13 9Q GQ 27

Iron Iron Iron Iron II Iron II–III

11

XII/56

12

GQ 37 FQ37

13 14

GQ 39

15 16

XII/49

XIII/2 XIII/13/1–5 XIV/55 XIV/56*

17 18

Cave 1 Cave 2

19 20

Cave 3

de Vaux, “Archéologie,” 8 de Vaux, “Archéologie,” 31 de Vaux, “Archéologie,” 10, Fig. 5:1; Ibrahim, “The Excavation of Cave XII/49,” 185, Fig. 2:1–13 Iron Ibrahim, “The Excavation of Cave XII/56,” 186, Fig. 1:1, 2 Iron Tal and Oron, “Survey and Excavations of Caves,” 189 Iron de Vaux, “Archéologie,”, 12, Fig. 5:3, pl. VII Iron Tal and Oron, “Survey and Excavations,” 191, Fig. 8:2 Iron Dahari, “Survey and Excavations,” 217 Iron Dahari, “Survey and Excavations,” 217; Wexler, “Preface,” CNJD 2, Table 17.1 from the middle de Vaux, “L’archéologie,” 9, 26–28, Fig. 6:5 of the 8th–end de Vaux, “L’archéologie,” 11, 26–28, Fig. 6:3, of the 7th BCE 4, 6, 7 de Vaux, “L’archéologie,” 13, 26–28 Cave of Avigad, “The Expedition to the Judean the Pool Desert,” 174, Fig. 4

256

Figure 17.1

BURDAJEWICZ

Location of caves with the Iron Age Pottery (after de Vaux, “Archéologie,” Fig. 1; Bar Adon, Excavations in the Judean Desert, Fig. 1; Wexler, ed. Surveys and Excavations of Caves in the Northern Judean Desert.

HISTORY OF THE “ QUMRAN CAVES ” IN THE IRON AGE

Figure 17.2

Jars: 1–2 = Cave XII/49 (after Ibrahim, “The Excavation of Cave XII/49,” Fig. 2: 1–2); 3–4 = Cave 2 in Wadi Murabba⁡ʿat (after de Vaux, “L’archéologie,” Fig. 6: 3, 6).

Figure 17.3

Jugs and juglets: 1 = Cave 2 in Wadi Murabba⁡ʿat (after de Vaux, “L’archéologie, ” Fig. 6: 1); 2–3 = Cave of the Pool (after Avigad, “The Expedition to the Judean Desert,” Fig. 4: 4, 6); 4-6 = XII/49 (after Ibrahim, “The Excavation of Cave XII/49,” Fig. 2: 10–12); 7 = Cave of the Pool (after Avigad, “The Expedition to the Judean Desert,” Fig. 4: 5); 8 = Cave 2 in Murabba⁡ʿat (after de Vaux, “L’archéologie, ” Fig. 6: 4); 9–10 = Cave of the Pool (after Avigad, “The Expedition to the Judean Desert,” Fig. 4: 3, 10); 11–12 = Cave XII/56 (after Ibrahim, “The Excavation of Cave XII/56,” Fig. 1: 1–2).

257

258

Figure 17.4

BURDAJEWICZ

Bowls: 1–4 = Cave XII/49 (after de Vaux, “Archéologie,” Fig. 5: 1; Ibrahim, “The Excavation of Cave XII/49,” Fig. 2: 3–5); 5 = Cave of the Pool (after Avigad, “The Expedition to the Judean Desert,” Fig. 4: 1); 6–7 = Cave 13 (after Patrich and Arubas, “A Juglet containing Balsam Oil (?) from a Cave near Qumran,” Fig. 3: 4–5); 8–9 = Caves 1 and 2 in Wadi Murabba⁡ʿat (after de Vaux, “L’archéologie, ” Fig. 6: 5, 7); 10 = Cave XIII/13 (after Tal and Oron, “Survey and Excavations,” Fig. 8: 2).

HISTORY OF THE “ QUMRAN CAVES ” IN THE IRON AGE

Figure 17.5

Cooking pots: 1–2 = Cave of the Pool (after Avigad, “The Expedition to the Judean Desert,” Fig. 4: 7–9); 3-6 = Cave XII/49 (after Ibrahim, “The Excavation of Cave XII/49,” Fig. 2: 6–9); 7–8 = Cave 13 (after Patrich and Arubas, “A Juglet containing Balsam Oil (?) from a Cave near Qumran,” Fig. 3: 2–3). 

Figure 17.6

Lamps: 1–2 = 11Q; 3 = 9Q (Burdajewicz, “La poterie de l’âge du Fer à Khirbet Qumrân,”); 4 = Cave of the Pool (after Avigad, “The Expedition to the Judean Desert,” Fig. 4: 2); 5 = Wadi Murabbaʿat (after de Vaux, “L’archéologie, ” Fig. 6: 2).

259

Appendix

∵

CHAPTER 18

Finds from the Qumran Caves: Roland de Vaux’s Inventory of the Excavations (1949–1956) Marcello Fidanzio and Jean-Baptiste Humbert As a complement to the papers presented at the Lugano Conference, here we publish Roland de Vaux’s inventory of the materials found in the Qumran caves (1949–1956) kept in the archives of the École Biblique et Archéologique Française de Jérusalem (EBAF).1 The inventory is composed of three successive records: – HCI, a handwritten card index of the pottery from cave 1Q, 1949 excavation. – TCI, a typewritten card index of the materials from the other caves, excavated between 1952 and 1956. – TB, a typewritten book2 of the same materials as the TCI, prepared at a later date.

The Nature of the Documentation

De Vaux published preliminary reports on the finds of the Qumran caves in Revue Biblique3 and final reports in the DJD series.4 The inventory kept at the EBAF was an internal working tool, made by de Vaux’s team for the study of the materials5 and later used for publication. 1   We would like to thank Francesca Romana Fiano and Maria Conconi for their feedback on this paper. 2  In the TCI each card represents one object; in the TB one or more objects were inventoried on one page until the page was full. The pages are numbered in succession (1–75) and are hardbound. 3  Roland de Vaux, “La grotte des manuscrits hébreux,” RB 56 (1949): 586–609; idem, “Exploration de la Région de Qumrân: Rapport préliminaire,” RB 60 (1953): 540–61; idem, “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e et 5e campagnes,” RB 63 (1956): 533–77. 4   Roland de Vaux, “La poterie,” in Qumran Cave 1 (DJD 1; ed. D. Barthélemy and J. T. Milik; Oxford: Clarendon Press, 1955), 8–17; idem, “Archéologie,” in Les ‘petites grottes’ de Qumrân: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre (DJD 3; ed. M. Baillet, J. T. Milik, and R. de Vaux; Oxford: Clarendon Press, 1962), 3–36; idem, “Archéologie,” in Qumrân Grotte 4.II: I. Archéologie; II. Tefillin, Mezuzot et Targums (4Q128–4Q157) (DJD 6; ed. R. de Vaux and J. T. Milik; Oxford: Clarendon Press, 1977), 3–22. The final report about the cave 11Q is still unpublished. 5  E.g., the first bowl-shaped lid of the first cave (Q5) is at first drawn upside down, that is as a bowl, then correctly described as a lid on the right side of the same card. © koninklijke brill nv, leiden, ���7 | doi ��.��63/9789004316508_020

Some observations allow us to understand better the nature of this documentation. First, we can ask why only the data in the TCI was inserted in the TB and not that of the other card index, the HCI. From studying the TB it is clear that it was made after the 1956 excavation (on the basis of the TCI, see below): at that time the pottery of cave 1Q (inventoried in the HCI) had already been published in its final form in DJD 1 and the data from the HCI were not reported again in the later internal working tool. The data in the inventory are not identical to the data published in the reports. In the DJD volumes the typology is more developed than in the inventory. Some objects that are recorded individually in the inventory (with their own description, drawing, and photograph) are grouped by type with other objects in DJD. In a few particular cases an object in DJD is classified under a different type with respect to the inventory. There are also instances where some items not registered in the inventory were published in DJD, and a few items were inventoried but not published (see below). The inventory therefore does not represent de Vaux’s last word on the archaeological materials, but simply a work-in-progress. When approaching this documentation it is necessary to take into account the time gap that separates us from de Vaux’s work. For example the study of pottery (which constitutes the largest portion of the inventoried material) had not yet been codified in a rigorous way in terms of its methods of analysis, and the terminology for classification had not yet been standardized.6

What Can We Learn from Studying the Inventory?

The inventory contains materials discovered in cave 11Q, which still remain unpublished. Furthermore, there are a few items excluded from the final reports, as well as other unpublished information on published items, such as 6  Roland de Vaux’s study of the pottery found at Qumran became the basis of one of the first major studies of the pottery from the region. Cf. Paul W. Lapp, Palestinian Ceramic Chronology 200 B.C.–70 A.D. (New Haven: American Schools of Oriental Research, 1961).

264

FIDANZIO AND HUMBERT

state of preservation (i.e., intact, complete, incomplete or fragment), measurements (already published only for the pottery from cave 1Q), contextual location of some objects inside the caves, photographs, and some drawings of nonceramic objects. Finally, by comparing the inventory data with the publications it is possible to illustrate de Vaux’s progress in the study of the material of the caves, and to clarify details which would otherwise be difficult to understand.

The Handwritten Card Index (Hci)—Cave 1Q (see pp. 281ff.) The HCI takes into account only the pottery of the first cave. In 1949 de Vaux and the EBAF were associated with the excavation of Gerald Lankester Harding, of the Jordan Department of Antiquities. The Palestine Archaeological Museum also took part in the excavation. In the preliminary report published in RB, de Vaux discussed only the pottery.7 Likewise, in the archaeological section of DJD 1 he authored only the chapter on the pottery (unlike what will happen for the publications dealing with the other caves8),

Figure 18.1

while the archaeological excavation and the small finds were presented by Harding. The linen textiles were published by Grace M. Crowfoot. The card index of the pottery from cave 1Q is handwritten (HCI): a comparison with the published texts on the same finds (see Table 18.1) allows one to familiarize oneself with the handwriting, and understand it. Most of the cards are in the same hand, but notations in other handwriting are also present. The work was clearly carried out by a team. The HCI records the following information (Fig. 18.1): – on the top left: Qumran Cave.9 – on the top right: Q-n°.10 On the top right, the initial Q followed by a number was the nomenclature used when the first cave with the manuscripts was found (cave 1Q). It was thought to

Handwritten Card Index (HCI): Q6.

7  In the second part of the preliminary report he offered some information on the manuscript fragments. 8  In 1952, after the discovery by the Bedouins of the second cave with manuscripts, the Department of Antiquities of Jordan was not able to take part in the exploration of the rock cliffs, which was then entrusted to the École Biblique et Archéologique Française, the American School of Oriental Research and the Palestine

Archaeological Museum. The Department of Antiquities of Jordan afterwards took part in the excavations of caves 4Q–11Q, A and B. 9  Here it is written in English while the rest of the card is in French. The Jordan Department of Antiquities was directed by the British archaeologist Gerald L. Harding. 10  n° = number.

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956)

be the only one,11 as is clear from the writing on the top left, in the singular and without any further details. The number following Q in the top right is a serial number used to inventory the pottery discovered in this cave.12 – on the left part of the cards:13 the drawing of the object at a scale of 1:2 or 1:4. Not all objects are drawn; some cards (Q22–Q33) refer to other objects which are identified as being of the same type and for which a drawing exists.14 Such 11  The cards regarding the jars (1Q40–1Q42, 1Q45–1Q53) report the number 1 before the Q, while the other cards present only the letter Q without any preceding number. We know that at least two cards of the HCI, and in all probability also the other ones, were made before 17/2/1951 (on this date two objects were sold, as written on the cards for Q13 and Q46). One can ask how a card made before that date, at a time when the existence of other caves with finds was not yet known, could use the nomenclature 1Q. The simplest conclusion is that the number had been written at a later time; this seems evident in at least the cases of cards 1Q49 and 1Q50. On two cards (1Q40 and 1Q42), under the nomenclature, a reference is given to DJD 1, which must have been added after the publication of the volume (1955). 12  Before the excavation of the khirbeh, which allowed de Vaux to connect the caves with the settlement, de Vaux gave nomenclature to the pottery of the cave, referring to Qumran: “Le plus proche des sites anciennement habités,” (“La grotte des manuscrits hébreux,” 586). He specifies in footnote 2: “Aucun indice archéologique ne met cette installation humaine [i.e. Khirbet Qumran] en relation avec la grotte où furent caches les manuscripts.” Initials such as “AF” found by Mireille Bélis on the photographic material linked to the linen textiles entrusted to Grace M. Crowfoot and on a drawer that conserved some linen textiles, interpreted as “cave near Ain Feshkha,” are to be attributed to other scholars. Cf. Mireille Bélis, “Des textiles: Catalogue et commentaires,” in Khirbet Qumrân et ʿAïn Feshkha. Vol. 2: Études d’anthropologie, de physique et de chimie (NTOA.SA 3; ed. J.-B. Humbert and J. Gunneweg; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 2003), 223–24; eadem, “The Unpublished Qumran Caves Textiles” (in this volume). Ovid R. Sellers used the definition “Aïn Feshkha Cave” in the publication of the radiocarbon dating of the textiles from cave 1Q (Ovid R. Sellers, “Radiocarbon Dating of Cloths from the ‘Aïn Feshkha Cave’,” BASOR 123 [1951]: 24–6). See also Joan E. Taylor, p. 10 in this volume. As already noted, Roland de Vaux published only the pottery in the case of cave 1Q. The cave was also identified as cave GQ14 when the caves in the rock cliffs were explored in 1952. See de Vaux, in DJD 3:6. 13  Except where the drawing of the jars takes up the whole card. 14  The typology is indicated by substituting the drawing of the object with a reference to the type. It should be noted that in some instances in the absence of a drawing there are discrepancies between the description of an object and its typological identification. So for example: Q31 is identified with Q17, but

265

identifications are really the first attempts at working out the typology that will later be developed in DJD 1. The typology here is limited to some lids and it is not applied to the other forms in the inventory, all of which are accompanied by a drawing. – on the right part:15 the photograph of the object. There are photographs of 45 out of the 57 objects that are inventoried in the HCI; 7 out of the 12 undrawn objects (but which have a typological identification) are photographed. – under the photograph: the description of the object. Here, the form, fabric, surface treatment, dimensions,16 and the state of preservation (the state of preservation only for 8 objects) are described. For the published objects, the descriptions found on the cards are nearly identical to that which is found in RB or DJD, except for some minor corrections made for publication. The state of preservation is not reported in RB or DJD, except in the case of the two lamp fragments Q56. – discovery of the objects: the objects inventoried in the HCI were found almost entirely during the excavation carried out by the Jordan Department of Antiquities in collaboration with the École Biblique et Archéologique Française and the Palestine Archaeological Museum from February 15th to March 5th 1949. On two cards— Q43 and Q44—it is indicated that the object was not found during the excavation.17 – it is sometimes noted on the card that the object was sold and to whom.18 These notes were added later, after the card had been filled in, on whatever blank space was available. The cards for the objects Q13 and Q46 in particular report the date of the sale, which was 17/02/1951 in both cases. This date sets a terminus ante quem for the completion of both cards and—we can assume—for the entire HCI. Regarding the period in which the HCI was completed we see that some objects were already presented the description of Q31 corresponds to Q19 and not Q17. See also Q25, which is identified with Q19, but the description of which matches Q18; Q29 is compared with Q12, but the description matches Q11; Q23 is compared with Q13 but the description matches Q9. 15  Except for the jars whose photographs are found on the back of the card. 16  Given in millimeters, as in RB 56 (1949). In DJD 1 the objects are measured in centimeters. 17  Q43 “Reçue de Georges, qui a pillé la grotte.” Q44 “Trouvé par le colonel Ashton, à sa première visite.” 18  A number of objects were sold to museums and institutions; for details see below the facsimile of the HCI.

266

FIDANZIO AND HUMBERT

in the preliminary report published in RB 59 (1949) (see Table 18.1), but in the same report it was also mentioned that the restoration of the materials would still take a long time.19 Non-Inventoried Objects In DJD 1 de Vaux refers to about 50 jars and almost as many lids found in cave 1Q.20 The HCI, however, describes only 12 jars and 36 lids. As with the objects of other caves,21 we also have uncatalogued objects from cave 1Q. HCI Inventory Number

Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q 10 Q 11 Q 12 Q 13 Q 14 Q 15 Q 16 Q 17 Q 18 Q 19 Q 20 Q 21 Q 22 Q 23

Shape

Drawing

Photograph

Bol Bol Bol Couvercle (assiette) Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle

X X X X

X   X

X X X X X X X X X X X X X X X X X Type Q 18 Type Q 13

X X X X X X X X X X X X X X X X X X X

19  De Vaux, “La grotte des manuscrits hébreux,” 586. 20  De Vaux, in DJD 1:8. 21  De Vaux, in DJD 3:6.

Table 18.1 The following table compares the information offered by the HCI with the data published in RB 56 (1949) and DJD 1 (unpublished objects have a grey background). RB 56 (1949) Description-Drawing Photograph

DJD 1 DescriptionDrawing

Photograph

Fig. 2.1

Fig. 2.1

PL III a.1

Fig. 2.3 Fig. 2.2

  PL II.2

Fig. 3.6 Fig. 2.4 Fig. 2.5

     

Fig. 3.8 Fig. 2.6 Fig. 2.7   Fig. 2.9         Fig. 3.7     Fig. 2.8    

PL II.5 PL II.4     PL II.1               PL III a.5    

PL XIV.1

      Fig. 2.7 Fig. 2.2 Fig. 2.5 Fig. 2.3   Fig. 2.4   Fig. 2.8             Fig. 2.6    

PL XIV.7 PL XIV.2 PL XIV.5 PL XIV.3 PL XIV.4 PL XIV.8           PL XIV.6    

a

267

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956)

HCI Inventory Number

Shape

Drawing

Q 24 Q 25 Q 26 Q 27 Q 28 Q 29 Q 30 Q 31 Q 32 Q 33 Q 34 Q 35 Q 36 Q 37 Q 38 Q 39 1Q 40 1Q 41 1Q 42 Q 43 Q 44 1Q 45 1Q 46 1Q 47 1Q 48 1Q 49 1Q 50 1Q 51 1Q 52 1Q 53 Q 54 Q 55 Q 56 e Q 57

Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Jarre Jarre Jarre Lampe Lampe Jarre Jarre Jarre Jarre Jarre Jarre Jarre Jarre Jarre Marmite Cruchette Lampes

Type Q 9 Type Q 19 Type 7 Type Q 14 Type Q 16 Type Q 12 Type Q 9 Type Q 17 Type 12 Type Q 8 X X X X X X X X X X X X X X X X X X X X X X X

Photograph

RB 56 (1949) Description-Drawing Photograph

DJD 1 DescriptionDrawing

X X     X   X X     X X   X X X X X X X X X X X c X X X       X X X

                                  Fig. 1                              

                                Fig. 2.10 Fig. 2.11 Fig. 2.12 Fig. 3.4 Fig. 3.5   Fig. 3.10b     Fig. 3.11b Fig. 3.9b       Fig. 3.2 Fig. 3.3 Fig. 3.1 (only Q 56)d

a In DJD 1 the captions to the photographs do not list the inventory number of the objects. The photographs thus ought to be understood as a general illustration of the material. By comparing the photographs published in DJD 1 with those present in the HCI it is possible to determine the inventory number of individual objects. Among the DJD 1 photographs we find the objects Q24, Q30, and Q35, but these were not presented in the chapter dedicated to the pottery, and are not recognizable without an explicit reference; in this case it is particularly evident that the photographs were used as

                                PL XV PL XVI b PL XVI b      

     

 

Photograph

PL II.3a           PL III a.6a         PL III a.4a             PL II.6 PL III b.3 PL III b.1   PL II.7               PL III a.2 PL III b.2  

a general illustration of the material. The photograph of Pl. III.3 does not match any photograph in the inventory. b The description refers to type Q41. c In the description of the object and in the drawing the base is missing but this is not the case in the photograph. d The fragments of two lamps are presented on only one card; however, since they are two different objects, they have two different inventory numbers. In DJD 1 de Vaux specifies that the two fragments are not linked and publishes only Q56.

268

FIDANZIO AND HUMBERT

Observations Inventoried but Unpublished Objects (identified in table 18.1 by the grey background) The HCI provides documentation on 32 objects which de Vaux chose not to publish: 1 bowl, 24 lids (12 of which are typologically identified with others22), 6 jars, and 1 lamp. Among these, there are objects that in the HCI functioned as representative types: Q8, Q16, Q17, Q19 (to which Q33, Q28, Q31 and Q25 refer). Furthermore two objects—Q12 and Q14—which were listed in the card index as representative types for other vessels (Q29, Q32, Q27) are published in RB 56 (1949) but not in DJD 1. From these observations we can conclude the following. Development of the Typology The preliminary report (RB 56 [1949]) offers a first account of the materials while the restoration was still in progress. The HCI represents the first attempt at developing a typology, taking into account a certain number of lids. In DJD 1 the ongoing work is resumed and critically analysed taking into account all the ceramic materials, found in the first cave possibly with the additional knowledge from the first excavation campaign at Khirbet (November– December 1951), the cliff exploration (March 1952), and

Figure 18.2

the excavation of caves 4Q, 5Q, and 6Q (September 1952).23 In DJD 1 there is a complete proposal of the typology of the pottery found in the first cave.

Typewritten Inventory 1952–1956 (see pp. 299ff.) The typewritten inventory contains the finds from the caves discovered after cave 1Q, between 1952 and 1956. It is preserved in two forms (Figs. 18.2–18.3): a card index (TCI), and a book (TB). The typewritten text of the TCI is corrected by hand on many occasions, while the TB presents the text already corrected. Thus the drafting of the TCI preceded that of the TB. The pages of the TB are numbered in sequence and uninterrupted. On pages 51–52 are found the materials from cave A and B, discovered in 1956, which are both preceded and followed by materials from caves excavated in 1952. Thus it is clear that the TB was made after the 1956 excavations. In the TCI we found an additional series of handwritten notes concerning typology not present in the TB. These notes will be discussed in the last part of this paper, where we will try to illustrate how the documentation and publication was carried out.

Typewritten card index (TCI): GQ 3–2.

22  They do not have a drawing, but refer to another lid from cave 1Q.

23   D JD 1 was published in 1955, but Harding’s foreword is dated to 1953.

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956)

Figure 18.3

269

Typewritten book (TB): p. 2.

Index The choice of preparing a TB, in which the pages are arranged in sequence, obliged the author to define the

Figure 18.4



order of presentation of the caves. The TB begins with an index:

Index of typewritten book.

Designation of the Caves

Considering that several systems to indicate the caves were applied in the inventory, with different numbering systems, it is important to note that in some cases the numbers overlap. One can distinguish the different caves

by referring carefully to the full system of designations used in the inventory, namely:24

24  For further detail see Marcello Fidanzio, “Which Cave Does This Pottery Come From? The Information Written on the Pottery

270

FIDANZIO AND HUMBERT

– GQ1–GQ40 for the natural caves in the rock cliffs explored in 1952 (among which are the caves with the manuscripts 2Q = GQ19; 3Q = GQ8; 6Q = GQ26); – X-1–X-2: two objects found on the surface during the 1952 exploration of the rock cliffs; – cave A and cave B for the two natural caves excavated in 1956 together with cave 11Q; – Gr4Q–Gr10Q for the artificial caves on the marl excavated in 1952 and 1955 (manuscripts were found in all of them25); – 11Q for the natural cave with the manuscripts excavated in 1956. For the natural caves the Q precedes the cave number, whilst for the artificial caves it follows the number. The eleventh cave discovered with manuscripts is an exception and is indicated as 11Q.26

Different Criteria

Examining the presentation order of the caves in the TB and the designation of the caves, it is apparent that different and sometimes clashing criteria were used, sometimes reflecting different ways to approach the study of the caves. First and foremost one sees the division between the natural caves in the rocky cliffs and the artificial caves in the marl terrace: this criterion is generally respected (first the natural caves are presented, then the artificial ones, and with different designation systems), except for cave 11Q. The natural caves are numbered from 1 to 40 according to a topographical order; this sequence was established in 1952, the year of the exploration of the rock cliffs (10–29 March), later enriched with cave 6Q found by the Bedouins and explored by the archaeologists in September of that year. Included in this sequence are four natural caves (GQ14 = 1Q, GQ19 = 2Q, GQ8 = 3Q, GQ26 = 6Q) in which were found the manuscripts that were numbered according to the geographical order of the cave in the cliff.27 The natural caves A and B, excavated in 1956, are presented in the pages after the natural caves GQ1–GQ40 Found in the Qumran Caves (R. de Vaux Excavations 1949–1956),” RB 122 (2015): 128–31. 25  In cave 10Q only one ostracon was found. 26  For the caves A and B see below. 27  The cave 1Q was explored again in the exploration of 1952 (1Q = GQ14), but the sherds gathered were not kept because in de Vaux’s judgement they did not add anything more regarding what was published in DJD 1 (cf. de Vaux, in DJD 3:9). For this reason this is not presented in the typewritten inventory.

and are indicated with the letters A and B “pour ne pas enterférer avec la séquence déjà établie des grottes notées en 1952.”28 Here the chronological order prevails over the topographical one. The artificial caves excavated in the marl are numbered and presented in order of their discovery. The numbering does not begin with 1 though, but with 4 (4Q): here the preponderant interest for the caves with manuscripts29 is highlighted; these caves are indicated by the numbering used to differenciate the caves with the manuscript (1Q–11Q) from the others.30 The importance of the manuscript finds inside a cave is highlighted in particular with the discovery of natural cave 11Q: this one was excavated in March 1956, as were the natural caves A and B. In the TB, however, this cave is described separately from caves A and B after the artificial caves with manuscripts, because of the manuscripts found in it.

Data in the Documentation

The typewritten inventory (in both its forms, TCI and TB) offers the following information. – the different types of artefacts and biofacts. Unlike cave 1Q, the inventory for the other caves includes many different materials: pottery, leather, iron, stone, flint, wood, bronze, copper, bone, and organic materials (natural and human-made). In the book there is a column dedicated to this information, while on the cards this is noted in the top left corner.31 – the inventory number of the object. In the TCI, the complete designation system to indicate the cave is used (see above), whilst in the TB a simplified system that omits the first part (GQ or Gr) is used. In the TB the objects of the natural caves are identified only with two numbers (the number of the cave and the number of the object) separated by a hyphen (for instance 3 – n°), whereas those of the artificial caves

28  De Vaux, in DJD 3:13. 29  Manuscripts were found in all the artificial caves listed in the inventory. In 1952, only two caves were listed, caves 4Q and 5Q. In 1955 caves 7Q to 10Q were added. 30  The numbering of the artificial caves, in addition to beginning at 4Q, is not continuous because the natural cave GQ26 corresponds to 6Q. Cave 5Q is not found in the inventory because inside of it only manuscript fragments and some animal bones were found (de Vaux, in DJD 3:26). 31  There are some problems with the flint blades, the material of which is sometimes indicated as “silex” and other times as stone.

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956)

together with 11Q have two numbers separated by the letter Q (for instance 4Qn°).32 – the date of discovery and provenance. In most cases the date of discovery is recorded. At other times a time span of some days is indicated or, in the case of cave 4Q, only the month.33 For a few pieces, instead of the date of discovery the typewritten inventory (both TCI and TB) records that they were brought by Bedouins.34 It is therefore possible to distinguish these objects from those found by the archaeologists.35 As for the objects from caves GQ1–GQ40, a pre-printed space with the abbreviation Prov. (= provenance) is filled with a correspondence between the numbering of the caves established after the exploration in March 1952 and the numbering of the caves used during the course of the exploration (for instance: cave 3 = A8). The cliff was explored by seven groups that were each composed of a member of the expedition accompanied by three or four Bedouins.36 The letter indicates the group (A–G); the number specifies the cave they explored. A second equivalence indicates those caves among the forty specified above in which the manuscripts were found (for instance: cave 19 = F5 = 2Q).37

32  The latter system overlaps with the system used to refer to the manuscripts; e.g. 7Q6 refers both to a Greek fragment and to a jar with a double inscription (we would like to thank Emile Puech for bringing this to our attention). This simplified system was already used by de Vaux in DJD 3 and recently by André Lemaire, “Inscriptions du Khirbeh, des grottes et ʿAïn Feshkha,” in Humbert and Gunneweg, Khirbet Qumrân et ʿAïn Feshkha, 373–77. It would be convenient to differentiate the references. We suggest using the prefix Gr to indicate the non-manuscript materials from caves 4Q–11Q. 33  Sometimes there is a divergence between the inventory and the dates given in DJD 3. In DJD de Vaux says that the dates are those in which the cave was discovered (DJD 3:6). For the cave GQ3 the dates given are 15–21.3.1952, but in the inventory the materials are registered with the date 14.3.1952; GQ17 15.3.1952 in DJD 3:9, but 13.3.1952 in the inventory; GQ28 23.3.1952 in DJD 3:10, but 20.3.1952 in the inventory; GQ39 20.3.1952 in DJD 3:12, but 16 or 18.3.1952 in the inventory. 34  GQ26–1; GQ26–2; GQ39–4; GQ39–5; GQ39–6. 35  GQ26–1 specifies “Acheté des Bédouins. Provenance probable: grotte 6Q = 26.” 36  De Vaux, in DJD 3:6. 37  For cave 6Q the designation is on the contrary 6Q = 26 and the inventory does not report the initial of a team because this one was discovered by the Bedouins after the exploration. For the cave GQ8 the equivalence with 3Q is not indicated, but is clearly expressed by de Vaux in DJD 3:7.

271

Sometimes the place where the object was found inside the cave is indicated, such as for the objects found in caves 7Q, 9Q, 10Q and 11Q.38 In this section of the TCI and in the same column in the TB39 a rough chronology is given for a limited number of objects: chalcolithic, iron, arabic.40 – the description, the dimensions, the drawing (or typological identification) and the photograph of the object. The typewritten inventory contains descriptions of the form, fabric, surface treatment, and state of preservation of the object. Inscriptions and graffiti are also indicated. The functional connection between a specific jar and a specific lid is rarely proposed.41 The state of preservation of objects is not published in RB or DJD.42 It offers a contribution to the discussion of the way in which de Vaux selected the objects to catalogue and the consequences of these choices.43 The inventory lists intact or complete objects, but also many incomplete objects and fragments. De Vaux took an inventory of the excavated material that seemed useful for the archaeological investigation without limiting himself to the complete vessels. The dimensions, recorded in millimeters, follow the text of the description in the TCI, while on the TB they 38  See also GQ8-4; GQ39-4; Gr4Q22. 39  Sometimes this information is found in the description; for the object GQX-2 in the TB it is noted twice. In general, there is a certain disorder, which grows in the TB compared to the TCI. There is also the case of GQ27-1 in which this information about chronology is inserted in the pre-printed space, which in the card is reserved for the date of discovery of the object: “Date: Iron Age.” 40  GQ27-1; GQ39-6; GQX-1; GQX-2; GrB-1; GrB-2; GrB-3; GrB-4; GrB5; GrB-6; Gr11Q5. To these chronological remarks, the designation “type Qumran” (as found in GQ7-1, GQ12-1, GQ12-2, GQ12-3, GQ15-1, GQ15-2, GQ29-4, GQ29-8), should probably be added if it is intended as pottery that finds its parallel in the period of occupation at Khirbet Qumran. 41  GQ15-2 with GQ15-1; GQ26-1 with GQ26-2; GQ39-5 with GQ394; Gr4Q5 (probably) with Gr4Q1; GQ12-2 with GQ12-1 (the mat GQ12-11 “Peut-être originalement au dessus de la jarre 12-1” [corrige 12-11- errata 11-1]). 42  The drawings do not always indicate that the object is incomplete; moreover there are also some problems in the definitions. For instance, GQ29-1 is described as “intact,” but the drawing and the photograph show that the object is not intact. 43  An example in this volume is Jürgen K. Zangenberg’s paper: “When reading de Vaux’s ceramic report one quickly realizes that—very much in line with contemporary practice—only complete vessels were catalogued, fragments after a short identification (“genre Qumrân”) frequently discarded. How much was missed or remained unrecorded?”. But see the context of Zangenberg’s paper.

272

FIDANZIO AND HUMBERT

have a dedicated column. Unlike the objects from cave 1Q, the dimensions of the objects from the other caves are not reported in the publications. The drawing. For the pottery a series of objects from the caves GQ1–GQ40 (and in one case from cave 7Q) are typologically identified with others. In this case there is no drawing of the object and description of the form is very limited.44 The drawing is omitted also in the case of some non-ceramic objects, or when a sherd is inventoried only because of an inscription.45 The photographs (found often in the TCI, less frequently in the TB) are taken from the Palestine Archaeological Museum (PAM) repertoire. The PAM photographs contain many objects together on one slide; the image of each object was cut from those photographs and then pasted onto single cards (TCI) or pages (TB). The cards of the TCI provide handwritten references to the PAM numbers. TCI–TB Inventory Number

Shape

Mat.

Drawingb

Photograph

GQ3-1 GQ3-2 GQ3-3 GQ3-4 GQ3-5 GQ3-6 GQ3-7 GQ3-8 GQ3-9 GQ3-10 GQ3-11 GQ3-12 GQ4-1 GQ7-1 GQ8-1 GQ8-2 GQ8-3 GQ8-4 GQ8-5

Jarre Couvercle Gobelet Marmite Couvercle Couvercle Couvercle Jarre Anneau Rondelle Lame Lame Jatte Couvercle Couvercle Couvercle Couvercle Cruche Cruche

Pot Pot Pot Pot Pot Pot Pot Pot Br Pi Sil Sil Bois Pot Pot Pot Pot Pot Pot

X X X X Type 8–6 Type 7–1 Type 3–2 (Type 2)d X X X X X X X  X X X X

  X

Non-Inventoried Objects It is notable that the phylactery cases found in cave 4Q and 5Q are not listed in the inventory even though they are known and published in DJD 3 and 6; those found in cave 8Q, however, are listed. The same holds true for the leather reinforcing tabs and thongs. The textiles are not in the inventory (see the contribution of Mireille Bélis in this volume). Regarding the pottery, in DJD 346 de Vaux makes reference to the non-catalogued sherds for the different caves.

Table 18.2

The following table compares the data of the typewritten inventory with what is published in RB 60 (1953), DJD 3, and DJD 6.47 RB 60 (1953) DescriptionDrawing

X    

  Fig. 3.6 Fig. 4.7 Fig. 4.3    

X X X X

       

X X X X X X

Fig. 3.7 Fig. 3.5 Fig. 3.10 Fig. 3.9 Fig. 4.2 Fig. 4.1

44  Typologically identified objects have limited information, which does not allow us to verify the typological link. When examining their state of preservation one can ask why the GQ7–1 lid, incomplete, was chosen as a typological reference for the intact GQ3–6 lid, for the nearly complete GQ8–24, and for GQ8–13 and for GQ8–17, each with a fragment missing. 45  The vessels GQ12–8, GQ12–9 and GQ12–10 (the latter two later defined “type uncertain;” see below) are not drawn either.

Photograph

DJD a DescriptionDrawingc

  III Fig. 2.3 PL XXII a.3 III Fig. 4.1 III Fig. 5.12 PL XXII b.5 III Fig. 5.19   Type 4 (III Fig. 4)   Type 5 (III Fig. 4) Type 1 (III Fig. 4) Type 2 (III Fig. 2)   III p. 7   III p.7   III p.7   III p.7 III p.16 PL XXII a.9 III Fig. 4.5 PL XXII a.2 III Fig. 4.18 PL XXII a.5 III Fig. 4.12 PL XXII a.6 III Fig. 4.14 PL XXII b.2 III Fig. 5.13 PL XXII b.3 III Fig. 5.15

Photograph

  III PL IV III PL V    

III PL VII III PL VII III PL VII III PL VIIe III PL IV III PL IV III PL IV III PL IV III PL V III PL V

46  De Vaux, in DJD 3:6–12. 47  For the other caves, the preliminary report is very short and without a description, drawing or photograph of the single objects.

273

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956)

TCI–TB Inventory Number

Shape

Mat.

GQ8-6 GQ8-7 GQ8-8 GQ8-9 GQ8-10 GQ8-11 GQ8-12 GQ8-13 GQ8-14 GQ8-15 GQ8-16 GQ8-17 GQ8-18 GQ8-19 GQ8-20 GQ8-21 GQ8-22 GQ8-23 GQ8-24 GQ8-25 GQ8-26 GQ8-27 GQ8-28 GQ8-29 GQ9-1 GQ12-1 GQ12-2 GQ12-3 GQ12-4 GQ12-5 GQ12-6 GQ12-7 GQ12-8 GQ12-9

Couvercle Jarre Jarre Jarre Jarre Jarre Lampe Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Lampe Jarre Couvercle Bol Couvercle Couvercle Jarre Couvercle Jarre Jarre

Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot

GQ12-10

Jarre

GQ12-11 GQ15-1 GQ15-2 GQ15-3 GQ15-4 GQ17-1 GQ17-2 GQ17-3 GQ19-1 GQ19-2

Natte Jarre Couvercle Lame Pointe Jarre Jarre Poteaux Couvercle Jarre

Drawingb

X X X X X X X Type 7-1 Type 8-6 Type 3-2 Type 8-1 Type 7-1 X X X Type 8-1 Type 8-6 Type 29-9 Type 7-1 Type 7-1 Type 8-1 Type 8-6 Type 29-9 Type 8-1 X X X X X X X X   (Type uncertain)49 Pot (Type uncertain)49 Van   Pot X Pot X Sil X Bois X Pot X Pot Bois   Pot X Pot X

Photograph

RB 60 (1953) DescriptionDrawing

X X X X X   X

Fig. 3.2 Fig. 2.7 Fig. 1.1 Fig. 2.1 Fig. 1.2    

         

         

            X X X X

            Fig. 4.5 Fig. 1.10 Fig. 1.9 Fig. 4.10

 

 

Photograph

DJD a DescriptionDrawingc

PL XXII a.1     PL XXI b.4

III Fig. 4.4 Type 1 (III Fig. 2) III Fig. 2.4 III Fig. 3.8 Type 4 (III Fig. 2)   III Fig. 2.1   III Fig. 5.6 Type 5 (III Fig. 4) Type 4 (III Fig. 4) Type 1 (III Fig. 4) Type 18 (III Fig. 4) Type 5 (III Fig. 4)   III Fig. 4.13   III Fig. 4.7   III Fig. 4.6   Type 18 (III Fig. 4)   Type 4 (III Fig. 4) Type 8 (III Fig. 4)   Type 5 (III Fig. 4)   Type 5 (III Fig. 4)   Type 18 (III Fig. 4)   Type 4 (III Fig. 4)   Type 8 (III Fig. 4)   Type 18 (III Fig. 4)   III Fig. 5.5 PL XXI b. 2 III Fig. 3.11 PL XXI b. 1 III Fig. 4.17 PL XXII b.7 III Fig. 5.7 III Fig. 4.16   III Fig. 4.2 Type 11 (III Fig. 3) III Fig. 4.22 Type 2 (III Fig. 2) Type uncertain

Photograph

III PL IV III PL VI III PL V III PL V III PL VI, VII   III PL VII

                        III PL VI III PL VI III PL V  

Type uncertain   X X X X     X    

  Fig. 2.3 Fig. 4.12     Fig. 2.4        

 

III p.8,16 III Fig. 3.10 PL XXII b.9 III Fig. 5.11   III p.9,16   III p.9,16   III Fig. 3.12   Type 10 (III Fig. 3)   III p.9   III Fig. 4.9   Type 8 (III Fig. 3)

III PL II 3 III PL VI PL V PL VII PL VII     PL VII    

274

FIDANZIO AND HUMBERT

(Cont.) TCI–TB Inventory Number

Shape

Mat.

Drawingb

Photograph

RB 60 (1953) DescriptionDrawing

Photograph

DJD a DescriptionDrawingc

Photograph

GQ19-3 GQ19-4f GQ19-5f GQ19-6abf GQ26-1 GQ26-2 GQ27-1 GQ28-1 GQ28-2 GQ28-3 GQ29-1 GQ29-2 GQ29-3 GQ29-4 GQ29-5 GQ29-6 GQ29-7 GQ29-8 GQ29-9 GQ29-10 GQ29-11 GQ29-12 GQ29-13 GQ29-14 GQ29-15 GQ29-16 GQ29-17 GQ29-18 GQ29-19 GQ29-20 GQ29-21 GQ29-22 GQ29-23 GQ29-24 GQ29-25 GQ31-1 GQ31-2 GQ31-3 GQ32-1 GQ34-1 GQ37-1 GQ37-2 GQ39-1 GQ39-2 GQ39-3

Jarre Pointe de fléche Pointe de fléche 2 feuilles de cuivre Bol Jarre Bol Jarre Couvercle Jarre Lampe Jarre Jarre Plat Couvercle Couvercle Couvercle Jarre Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Couvercle Jarre Jarre Jarre Jarre Bol Bol Bol Jarre Cruche Gobelet Bol Jarre Jarre Couvercle

Pot Fer Fer Cu Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot

X

 

 

 

Type 2 (III Fig. 2)

 

X X X X X Type 29-8 X X X  X X X X X X Type 3-2 Type 7-1 Type 7-1 X Type 3-2 Type 3-2 Type 29-9 Type 7-1 Type 7-1 Type 3-2 Type 7-1 X Type 29-2 Type 29-2 Type 29-3 Type 29-3 X X X X X X X X X X

X X

Fig. 4.9 Fig. 2.5

X X

Fig. 1.4 Fig. 1.3

X   X X X X X X X                        

Fig. 4.4 Fig. 2.2 Fig. 1.6 Fig. 4.6 Fig. 1.5 Fig. 3.1 Fig. 3.8 Fig. 2.6 Fig. 3.4                        

          X

      Fig. 4.8 Fig. 4.13 Fig. 4.11

           

          Fig. 3.3

PL XXII b.8 III Fig. 5.8 PL XXI b.3 III Fig. 3.9 III Fig. 5.1 PL XXI a2 Type 4 (III Fig. 2) PL XXI a.1 III Fig. 4.3 Type 6 (III Fig. 2) PL XXII b.1 III Fig. 5.2   Type 4 (III Fig. 2) PL XXI a.4 III Fig. 2.2 PL XXII b.4 III Fig. 5.21 PL XXI a.3 III Fig. 4.15 PL XXII a.8 III Fig. 4.20 PL XXII a.7 III Fig. 4.23   III Fig. 2.6 PL XXII a.4 III Fig. 4.8   Type 1 (III Fig. 4)   Type 5 (III Fig. 4)   Type 5 (III Fig. 4)   III Fig. 4.10   Type 1 (III Fig. 4)   Type 1 (III Fig. 4)   Type 5 (III Fig. 4)   Type 5 (III Fig. 4)   Type 5 (III Fig. 4)   Type 1 (III Fig. 4)   Type 5 (III Fig. 4)   III Fig. 4.21 Type 4 (III Fig. 2)   Type 4 (III Fig. 2)   Type 2 (III Fig. 2)   Type 2 (III Fig. 2)   III Fig. 5.9   III Fig. 5.10 PL XXII b.6 III Fig. 5.14 Type 9 (III Fig. 3)   III Fig. 5.18   III Fig. 5.17   III Fig. 5.16   Type 7 (III Fig. 3)   III Fig. 3.7   III Fig. 4.11

III PL V III PL V III PL IV III PL IV III PL V, VII   III PL IV III PL V III PL IV III PL IVg III PL IV III PL VI III PL IV                                   III PL V            

275

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956)

TCI–TB Inventory Number

GQ39-4 GQ39-5 GQ39-6 GQ39-7 GQ39-8 GQ39-9 GQ39-10 GQ39-11 GQ40-1 GQX-1 GQX-2 Gr A 1 Gr B 1 Gr B 2 Gr B 3 Gr B 4-5 Gr B 6 Gr 4Q 1 Gr 4Q 2 Gr 4Q 3 Gr 4Q 4 Gr 4Q 5 Gr 4Q 6 Gr 4Q 7 Gr 4Q 8 Gr 4Q 9 Gr 4Q 10 Gr 4Q 11 Gr 4Q 12 Gr 4Q 13 Gr 4Q 14 Gr 4Q 15 Gr 4Q 16 Gr 4Q 17 Gr 4Q 18 Gr 4Q 19 Gr 4Q 20 Gr 4Q 21 Gr 4Q 22 Gr 7Q 1 Gr 7Q 2 Gr 7Q 3

RB 60 (1953) DescriptionDrawing

Photograph

Shape

Mat.

Drawingb

Photograph

Jarre Couvercle Lampe Jarre Couvercle Couvercle Jarre Bol Jarre Cruche Jarre Tige Pointe de silex Lame de silex Pointe de flèche silex Perles Pilon Jarre Jarre Jarre Jarre Couvercle Couvercle Cruche Bol Bol Bol Assiette ou couvercle Assiette ou couvercle Marmite Cruche Couvercle Jarre Cruche Assiette Bol Bol Bol Lampe Tesson jarre (signe peint) Bol Bol

Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Br Pi Pi Pi

X X X Type 39-1 Type 39-3 Type 7-1 Type 29-3 X X X X   X X X

X X X      

Fig. 1.8 Fig. 1.7

PL XXI a.6 PL XXI a.5

     

     

X

 

 

Os Pi Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot

    X X X X X X X X  X X X 

    X X X   X X

   

   

   

   

 

 

X     X

     

     

Pot

X

Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot Pot

X  X X X X X X X X X X

Pot Pot

X X

DJD a DescriptionDrawingc

III Fig. 2.5 III Fig. 4.19 III Fig. 5.3 Type 7 (III Fig. 3) Type 11 (III Fig. 4) Type 5 (III Fig. 4) Type 2 (III Fig. 2) III Fig. 5.4 III Fig. 3.13 III Fig. 5.20 III p.13 III p.13,16 III p.13,16 III p.13,16 III p.13,16

Photograph

III PL IV III PL IV III PL VII      

 

III p.13 III p.13,16 VI 16 (2) VI 16 (1) VI 18–19 (12) VI 16 (3) VI 18–19 (15) VI 18–19 (17)

    VI PL III a b VI PL III a VI PL III a   VI PL III a VI PL III a

VI 18–19 (22)   VI 18–19 (20) VI 18–19 (6)

VI PL III a     VI PL III a

X

VI 18–19 (8)

VI PL III a

X  

VI PL III a     VI PL III a VI PL III a VI PL III a VI PL III a VI PL III a VI PL III c LEM 375 III PL VIII III PL VIII

 

 

  X X X X X X X

   

   

   

   

 

 

VI 18–19 (11) VI 18–19 (14) VI 18–19 (16) VI 18–19 (1) VI 18–19 (9) VI 18–19 (7) VI 18–19 (19) VI 18–19 (18) VI 18–19 (21) VI 18–19 (10) III p.30

X X

   

   

III Fig. 6.13 III Fig. 6.8

276

FIDANZIO AND HUMBERT

(Cont.) TCI–TB Inventory Number

Gr 7Q 4 Gr 7Q 5 Gr 7Q 6 Gr 7Q 7

Gr 7Q 8 Gr 8Q 1

Gr 8Q 2 Gr 8Q 3 Gr 8Q 4 Gr 8Q 5 Gr 8Q 6 Gr 8Q 7 Gr 8Q 8 Gr 8Q 9 Gr 8Q 10 Gr 8Q 11 Gr 8Q 12 Gr 8Q 13 Gr 8Q 14 Gr 9Q 1 Gr 9Q 2 Gr 10Q 1 Gr 10Q 2 Gr 10Q 3 11Q 1 11Q 2 11Q 3 11Q 4 11Q 5 11Q 6 11Q 7 11Q 8 11Q 9 11Q 10 11Q 11 11Q 12 11Q 13

DJD a DescriptionDrawingc

Photograph

 

III Fig. 6.11 III Fig. 6.12 III Fig. 6.5 LEM 376

III PL VIII III PL VIII III PL VIII LEM 376

   

   

Type 12 (III Fig. 6)   III p.31 III PL VIII

X     X

       

       

III p.31   III p.31 III p.31

  III PL VIII   III PL VIII

X X X X X X X X

 

 

 

 

 

 

III PL VIII III PL VIII III PL VIII III PL VIII III PL VIII III PL VIII III PL VIII III PL VIII

X

 

 

III p.31 III Fig. 6.2 III Fig. 6.6 III Fig. 6.1 III p.31 III Fig. 6.10 III Fig. 6.3 III Fig. 6.9 III Fig. 6.7 III p.31

 

 

X

 

 

III p.31 LEM 377

  III PL XXXV

Van Pot Pot Fer Cu Fer Pot Fer

  X X X X X X X

  X X X   X X

         

         

  III Fig. 6.4      

  III PL VIII      

   

   

   

   

Pot Pot Cuir Fer Fer Pi Pot

X X X X   X X

X X

   

   

   

   

 

 

 

   

   

   

RB 60 (1953) DescriptionDrawing

Photograph

   

   

 

  X

       

Cuir Pot Pot Pot Cuir Pot Pot Pot Pot  

  X X X   X X X X  

  Pot

Shape

Mat.

Drawingb

Photograph

Couvercle Jarre Jarre Tesson (décalque deux lettres grecques) Jarre Datte,noyaux de dattes, noyau d’olive, figue Lanières, pattes Peau Étoffes et ficelles Deux etuis à phylactères Etui à phylactère Couvercle Assiette Couvercle Semelle de sandale Couvercle Lampe Couvercle Jarre Dattes et noyaux de dattes corde et fichelle Tesson de jarre (inscription) Natte Lampe Couvercle Lame Boucle Ciseau ou lime Lampe Piochette-hachette Hache de sapeur Cruchette Lampe Cuir Tige Clé coudée Perle Jarre

Pot Pot Pot Pot

X X X  

X X X X

Pot  

Type 7Q5  

Cuir Cuir Cuir Cuir

X Xh X X

   

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956) a Here the roman numerals III and VI indicate the DJD volumes: VI (DJD 6) for the objects of cave 4Q, and III (DJD 3) for the other caves. Three sherds were inserted in the catalogue only because of the inscriptions: these were published in Lemaire, “Inscriptions du khirbeh,” and indicated by the initials LEM. b When a vessel is typologically identified to another one, we found the reference to that object instead of the drawing. c It was possible to list which type was attributed to each object not published by de Vaux in DJD 3 for three reasons: – the indication in the same DJD 3 of the number of objects identified with the different types in each cave. – the handwritten information on the typology on the TCI (see below). – a substantial continuity between the typology of the inventory and that of DJD 3. In this case the type of reference is indicated first, then the Figure in which it is published in the DJD, e.g. Type 4 (Fig. 4). To visually indicate the difference from the other references, the types are aligned

Observations Inventoried but Unpublished Objects The materials from cave 11Q are the most important. According to the publishing plan of the DJD series, the publication of the archaeology of cave 11Q should have come at the beginning of the first volume dedicated to the manuscripts found in it.48 The archaeology of this cave, however, was the only one left unpublished due to de Vaux’s premature death.49 The cave was discovered by the Bedouins in January/February 1956, and later excavated by the archaeologists during the fifth campaign to Khirbet Qumran (18/2–28/3 1956). The preliminary report was published in RB50 and the field notes are also known.51

48  For 11Q see de Vaux, in DJD 3:5 n. 2 “Le rapport détaillé sur la fouille que nous avons faite après cette découverte aura sa place normale en tête de la publication des textes qui proviennent de cette grotte.” 49  The archaeology of cave 4Q was posthumously published in DJD 6 (1977). Actually the chapter on the archaeology of cave 4Q is in the second DJD volume with manuscripts of that cave (already John M. Allegro, Qumrân Cave 4.I [4Q158–4Q186] [DJD 5; Oxford: Clarendon Press, 1968]). Also for cave 11Q a volume a manuscript was already published before the death of de Vaux (James A. Sanders, The Psalm Scroll of Qumrân Cave 11 [11QPs a] [DJD 4; Oxford: Clarendon Press, 1965]). 50  De Vaux, “Fouilles de Khirbet Qumrân,” 573–74. 51   Jean-Baptiste Humbert et Alain Chambon, eds., Fouilles de Khirbet Qumrân et de Aïn Feshkha. Vol. 1: Album de photographies; Répertoire du fonds photographiques; Synthèse des notes de chantier du Père Roland de Vaux (NTOA.SA 1; Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 1994), 344.

d

e f

g h

277

on the right. Since the numbering of types restarts from 1 for each group/class (jars, lids . . .), the number of the types overlap; the reference to the Figure helps disambiguate. For some non-pottery objects in DJD 3 only the indication/description is found without the drawing. In this case, instead of “Fig.”, the reference to the page is found. This definition of typology was given only later: it is handwritten in the cards and typewritten in an incoherent way in the book (see below). Misprint in DJD 3: Pl. VII where it is indicated twice 3–11 (this confirms the photograph in the inventory). This card is different from the others contained in the TCI: it is handwritten, not typewritten, and the data are not reported in the TB (see below). Misprint in DJD 3: Pl. IV where it is listed as 21–6. The photograph does not match the object description. It is rather an image of 11Q3, which has no attached photograph. We would like to thank Dennis Mizzi and Annalisa Faggi for this observation.

In the inventory there are thirteen objects from cave 11Q with descriptions, drawings and photographs. Unlike the majority of the other cases, for the objects of cave 11Q the inventory reports the place where the objects were found and in a few cases even a generic stratigraphic indication (“assez bas,” “inferieur”). The objects are presented in the order they were found, as seen by the dates (from 26–27 February for Gr11Q1 to 14 March for Gr11Q13). In the preliminary report de Vaux explains he distinguished three “periods of occupation” corresponding to three layers: in the lower layer he found pottery dated to the Chalcolithic period, amongst which there was a jar. In the inventory we find the documentation of the 11Q13 jar. Above this layer an Iron II layer, with objects dated to the VII century BCE. In the inventory we find the documentation of two lamps 11Q5 and 11Q8, and a rounded juglet 11Q7.52 Thereafter a layer “contemporary to the occupation of Khirbet Qumran” was found, of which de Vaux writes: “Elle contenait quelques objets de fer, une piochette, une ciseau (ou une lime), un couteau et peu de poterie; mais les formes sont bien caractéristiques et ont leurs parallèles au Khirbet et dans les autres grottes, en particulier une cruchette et deux couvercles en forme de bol renversé.”53 In the inventory we find the documentation of an iron pick (11Q6), an iron chisel (11Q4), an iron blade (11Q2), and a lid (11Q1). In the inventory we find also other objects not mentioned in the preliminary report (11Q3, 11Q9, 11Q10, 11Q11, 11Q12), but we only find one lid while in the above quote de Vaux mentions two lids and we do 52  The materials from the Iron Age are presented in this volume in Mariusz Burdajewicz’s contribution. 53  De Vaux, “Fouilles de Khirbet Qumrân,” 574.

278

FIDANZIO AND HUMBERT

not find a second juglet, besides that dated to the Iron Age. In the preliminary report he also mentions sherds of Iron II jars: these are kept in the Rockefeller Museum, but we don’t find them in the inventory. As a result, just as there are some materials from the other caves that were not catalogued, so too there are objects from the cave 11Q that were not registered in the inventory.54 Concerning the cave GQ19 = 2Q there are three cards in the TCI different from the others contained in the TCI, listing two arrowheads (GQ19-4 and GQ19-5) and two thin copper foils (GQ19-6 a b). These cards are handwritten and not typewritten (the handwriting is not de Vaux’s) and do not have drawings and/or photographs of the object. The data on these cards are not reported in the TB and we do not find any reference to these objects in DJD. Despite these differences, according to Humbert, these cards are part of de Vaux’s work. Comparing the typewritten inventory (both TCI and TB), the published data in RB and in DJD, it was observed that some objects were identified with a certain type only in DJD. For these objects the typewritten inventory contains the full data.55 Amongst these, GQ29-2 and GQ39-1 are representative types in the inventory with which other objects are identified (GQ29-22 and GQ29-23 for the first, GQ39-7 for the second). There are a few cases (GQ8-7, GQ8-10, GQ28-1, GQ29-2) in which an object published in RB was not published in the DJD series. The object GQ29-16 passed from one type in the inventory to another in DJD. In the typewritten inventory there are few objects of arabic pottery and few non-ceramic objects about which we only had some information. For some of these objects, we have the photograph.

Process of Documentation (Inventory and Publication)

1) The first step is the compilation of the TCI, in which a series of objects found in the caves are inventoried.56 The TCI presents a typology of the pottery. The typological classification is part of the typewritten text: it is indicated 54  For a complete list of materials found in the cave, see the table attached to Dennis Mizzi’s contribution in this volume. 55  GQ3-8, GQ8-7, GQ8-10, GQ12-6, GQ12-8, GQ12-9 (“type incertain”), GQ12-10 (“type incertain”), GQ17-2, GQ19-2, GQ19-3, GQ28-1, GQ29-1, GQ32-1, GQ39-1. 56  Other objects were collected during the excavations, but not catalogued, as de Vaux explains in DJD 3:6.

through the inventory number of the object of reference (for instance: object GQ8-13 . . . type 7-157). 2) The second step is the revision of the typology. The TCI was subsequently used during the revision of the typology. On the TCI there are handwritten notes (de Vaux’s handwriting) that use a different designation system to indicate the typology of the objects. In these notes each type is merely assigned a number (for instance: object GQ8-13 . . . type 5). The same system is used in DJD, and the typology presented in the TCI coincides perfectly with the one in DJD.58 Reading the handwritten typology notes in the TCI gives a deeper understanding of the information published in DJD 3. In DJD 3 an object from cave GQ3 is indicated as uncatalogued although the type is defined precisely;59 among the catalogued ones, however, there are two entries (GQ12-9 and GQ12-10) defined as “type incertain.”60 The reader asks himself what criteria were used to choose which objects to insert in the inventory. On the cards GQ12-9 and GQ12-10 of the TCI nothing was typewritten about typology, the phrase “Type incertain” is found only in the handwritten notes of the revision of the typology. We suggest that the two opposite situations (from one side an object with a typology precisely defined and not catalogued, on the other side two objects of the uncertain typology and catalogued) could be explained as follows: during the revision of the typology all the materials, catalogued and uncatalogued, could have been reconsidered; so it is reasonable to surmise that in the first step (compilation of the TCI) one object from cave GQ3 was not catalogued, but later (revision of typology) it was identified as a certain type. Items GQ12-9 and GQ12-10 were catalogued in TCI, but later these were not identified with a clear type. It is important to notice that the inventory was not updated or integrated with new items later on. Only the definitions of the typology were revised. DJD 3 follows the inventory with the revised typological definitions and integrates the uncatalogued object identified as a certain type. 3) The inventory is then reproduced in the form of a book (TB). As indicated, this step took place following the last excavations at Qumran, during which caves 11Q, A and B were excavated (1956). By comparing some points 57  The typology concerns a series of objects from the caves GQ1– GQ40 (in one case from cave 7Q). The reference system indicates two numbers following the word “type:” the first number is that of the cave and the second number is that of the object. 58  In the handwritten notes the objects of reference are defined as “prototype;” in the publication this definition is not found. 59  De Vaux, in DJD 3:7. 60  De Vaux, in DJD 3:8.

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956)

Figure 18.5a, b GQ 3-8 in the typewritten card index, and in the typewritten book.

279

280 of the TB with the TCI we understand that this procedure happened after the revision of the typology. In the TB the typology is usually listed following the typewritten texts on the cards of the TCI (see point 1), in which the inventory number of the object of reference is reported (for instance, type 7-1). The card for GQ3-8 in the TCI does not report any typological reference in the typewritten text, but is indicated by hand as “type 2” (see the discussion above regarding the revised designation system of the typology: only one number, the type number). In the TB “type 2” is typewritten and arranged in a column together with the reference to the typology of the preceding objects. However, this is different from the rest because it presents only one number (type 2 instead of, for instance, type 7-1).61 Whoever transcribed the text of the TCI to the TB reported the references to the typewritten typology, but in this case, since a typewritten reference was not given, s/he decided to insert the handwritten reference without realizing that it followed another system. This situation is also verified in the case of GQ12-9, GQ12-10 where “type incertain” was handwritten in the TCI and typewritten in the TB. The TB was made after the revision of the typology, but it reproduced the data typewritten in the TCI (in general, the typology coincides and is only a matter of a different reference system). In a few cases whoever prepared the TB added—in an inconsistent way—the TCI handwritten revision of the typology. 4) Finally, in the archaeological report of cave 4Q in DJD 6 (the text was finished around 196562), five uncatalogued fragments63 are presented together with the catalogued ones. De Vaux judged it useful to introduce some items excluded from the inventory to underline the similarities between the pottery of cave 4Q with the other caves. As in the previous cases, the inventory in both its forms, was not integrated with the new items at this time. The sherds received a number of references in the plate of DJD 6, but unlike the others they do not present an inventory number and are generically indicated “4Q tessons.”64 On some cards of the objects from cave 4Q in the TCI corrections are indicated in the drawings (inversion of the section drawn differently from the others, cancelling of the ribs).65 These corrections are not found 61  Furthermore in the inventory (in both its forms, TCI and TB) there are no objects from cave GQ2. 62  The DJD 6 volume was published in 1977, but in the preface (vi n. 1) Pierre Benoit clarifies that de Vaux’s text was prepared 12 years earlier. 63  These are indicated with the generic inscription found on the non-inventoried sherds conserved at the Rockefeller Museum warehouse as “4Q tesson;” de Vaux, in DJD 6:19. 64  De Vaux, in DJD 6:18–19. 65  Gr4Q1, Gr4Q3, Gr4Q4, Gr4Q5, Gr4Q6, Gr4Q14, Gr4Q15.

FIDANZIO AND HUMBERT

in the TB, but are present in DJD 6: they were made for the publication. This reconstruction helps to clarify some elements that remained obscure after studying the publications and questioning the criteria for the compilation of the inventory. In particular, it clarifies our understanding of the situation because among the non-inventoried objects from cave GQ3 the exception is found of an object of a wellestablished typology; also because on the contrary two objects (GQ12-9, GQ12-10) defined as “type incertain” are inventoried; and finally because in the presentation of the objects of cave 4Q five objects are found without an inventory number. The inventoried objects were re-examined together with those non-inventoried for the publication. The outcome of the revision is proposed in DJD 3 and 6 without modifying the inventory, which remains the basis for the publications. Conclusions The publication of de Vaux’s inventory on the materials from the caves of Qumran brings to light some unpublished items, among which are the materials from cave 11Q still entirely unpublished. There is some information in the inventory (e.g. the state of preservation of the objects) which is not reported in the publications, but is useful for research and for the debate currently taking place. Not all the items found in the excavations and subsequently preserved are present in the inventory. The inventory of de Vaux is an internal work tool, made for the study of the materials. Knowing the successive stages in which the HCI, TCI, TB and DJD 1, 3, and 6 were compiled helps to understand better how the inventory was prepared as well as to clarify some information proposed in DJD that would otherwise remain unclear. De Vaux produced good documentation according to the archaeological standards of the 1950s. His work was certainly valuable in its time, as demonstrated by the repertoire of Paul Lapp. To complete the publication of the EBAF materials from the Qumran caves the best move forward would be to publish the objects not yet catalogued but still in store at the Rockefeller Museum, and to review the objects already catalogued according to current archaeological standards. Note: in the following pages the HCI and the TCI are reproduced; however, owing to constraints of space, we cannot reproduce the TB; the upside is that the same information can be found, in an unambiguous way (see above), in the TCI, on which the TB was based.

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), HCI Q1–Q4

281

282

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), HCI Q5–Q12

283

284

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), HCI Q13–Q20

285

286

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), HCI Q21–Q28

287

288

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), HCI Q29–Q36

289

290

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), HCI Q37–Q41

291

292

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), HCI Q42–Q45

293

294

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), HCI Q46–Q49

295

296

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), HCI Q50–Q56 e 57

297

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 3-1–GQ 3-7

299

300

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 3-8–GQ 8-6

301

302

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 8-7–GQ 8-15

303

304

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 8-16–GQ 9-1

305

306

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 12-1–GQ 15-1

307

308

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 15-2–GQ 19-5

309

310

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 19-6–GQ 29-3

311

312

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 29-4–GQ 29-16

313

314

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 29-17–GQ 37-1

315

316

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 37-2–GQ 39-10

317

318

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GQ 39-11–Gr 4Q 1

319

320

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GR 4Q 1bis–Gr 4Q 9

321

322

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GR 4Q 10–Gr 4Q 19

323

324

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GR 4Q 20–Gr 7Q 4

325

326

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GR 7Q 5–Gr 8Q 7

327

328

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GR 8Q 8–Gr 9Q 2

329

330

FIDANZIO AND HUMBERT

APPENDIX: DE VAUX ’ S INVENTORY OF THE EXCAVATIONS (1949–1956), TCI GR 10Q 1–11Q 10

331

332

FIDANZIO AND HUMBERT

Bibliography Abel, Félix-Marie. “Notes complémentaires sur le mer Morte.” RB 38 (1929): 237–60. Adan-Bayewitz, David. Common Pottery in Roman Galilee: A Study of Local Trade. BISNELC. Ramat-Gan: Bar-Ilan University Press, 1993. ———, et al. “Preferential Distribution of Lamps from the Jerusalem Area in the Late Second Temple Period (Late First Century B.C.E.–70 C.E.).” BASOR 350 (2008): 37–85. Adler, Yonatan. Review of Tangled Up in Text: Tefillin and the Ancient World, by Yehudah B. Cohn. Zion 66 (2001): 383–87 [Hebrew]. ———. “Identifying Sectarian Characteristics in the Phylacteries from Qumran.” RevQ 89 (2007): 79–92. ———. “The Content and Order of the Scriptural Passages in Tefillin: A Reexamination of the Early Rabbinic Sources in Light of the Evidence from the Judean Desert.” In Albert I. Baumgarten, et al., eds. Halakhah in Light of Epigraphy. JAJSup 3. Göttingen: Vandenhoeck & Ruprecht, 2011, 205–29. ———. “Two Types of Multi-Compartment Tefillin Cases from the Judean Desert Caves.” Forthcoming. Adler, Yonatan and Ada Yardeni. “Remains of Tefillin from Naḥal Ṣeʾelim (Wadi Seiyal): A Leather Case and Two Inscribed Fragments (34Se 1 A–B).” Forthcoming. Aharoni, Miriam, and Yohanan Aharoni. “The Stratification of Judahite Sites in the 8th and 7th Centuries B.C.E.” BASOR 224 (1976): 73–90. Aharoni, Yohanan. “The Caves of Naḥal Ḥever.” ʿAtiqot 3 (1961): 148–62. ———. “Expedition B.” IEJ 11 (1961): 11–24. ———. Excavations at Ramat Rahel: Seasons 1959 and 1960. Serie Archeologica 2. Rome: Centro di Studi Semitici, Università degli Studi, 1962. ———. “Expedition B—The Cave of Horror.” IEJ 12 (1962): 186–99. ———. Excavations at Ramat Rahel: Seasons 1961 and 1962. Serie Archeologica 6. Rome: Centro di Studi Semitici, Università degli Studi, 1964. ———, ed. Beer-Sheba I: Excavations at Tel Beer-Sheba, 1969– 1971 Seasons. PIA 2. Tel Aviv: Institute of Archaeology, Tel Aviv University, 1973. Albright, William F. The Excavation of Tell Beit Mirsim in Palestine. Vol. 1, The Pottery of the First Three Campaigns. AASOR 12. New Haven: American Schools of Oriental Research, 1932. ———. The Excavation of Tell Beit Mirsim. Vol. 3, The Iron Age. AASOR 21–22. New Haven: Yale University Press, 1943. ———. “Are the ʿAin Feshkha Scrolls a Hoax?” JQR 40 (1949): 41–49.

———. “The Dead Sea Scrolls of St. Mark’s Monastery.” BASOR 118 (1950): 5–6. Alexander, Philip S., and Geza Vermes, eds. Qumran Cave 4.XIX: Serekh Ha-Yaḥad and Two Related Texts. DJD 26. Oxford: Clarendon Press, 1998. Allegro, John M. The Treasure of the Copper Scroll. London: Routledge & Kegan Paul, 1960. ———. Search in the Desert. Garden City, NY: Doubleday, 1964. ———. Qumrân Cave 4.I (4Q158–4Q186). DJD 5. Oxford: Clarendon Press, 1968. Amar, Zohar. “The Balsam Wood in the Temple Incense.” Tehumin 17 (1997): 473–79 [Hebrew]. ———. “The Ash and the Red Material from Qumran.” DSD 5 (1998): 1–15. Amit, David, and Hanan Eshel. “A Tetradrachm of Bar Kokhba from a Cave in Naḥal Ḥever.” INJ 11 (1993): 33–35. ———, and Hanan Eshel. “The Bar Kokhba Period Finds from the Tetradrachm Cave.” In Eshel, and Amit. Refuge Caves of the Bar Kokhba Revolt, 189–204 [Hebrew]. Ariel, Donald T., ed. “The Coins from the Surveys and Excavations of Caves in the Northern Judean Desert.” ʿAtiqot 41/2 (2002): 281–304. ———. “The Numismatic Finds.” In Itah, Kam, and Ben Haim. “Region X,” 172. ———. “The Numismatic Finds.” In Wolff. “The Excavation of Cave IV/12,” 80. ———. “Coins from Tel Mikhal (Tel Michal).” ʿAtiqot.English Series 52 (2006): 71–88. ———. “Coins of En-Gedi.” In Stern. En-Gedi Excavations I, 423–27. ———. “Coins.” In Hillel Geva, ed. Jewish Quarter Excavations in the Old City of Jerusalem Conducted by Nahman Avigad, 1969– 1982. Vol. 4, The Burnt House of Area B and Other Studies; Final Report. Jerusalem: Israel Exploration Society, 2010, 236–47. ———. “Coins from Excavations at Qumran.” Unpublished manuscript. ———, and Jean-Philippe Fontanille. The Coins of Herod: A Modern Analysis and Die Classification. AJEC 79. Leiden: Brill, 2012. Arif, Aida S. A Treasury of Classical and Islamic Coins: The Collection of Amman Museum. London: A. Probsthain, 1986. Arubas, Benny, and Haim Goldfus. “The Kilnworks of the Tenth Legion Fretensis.” In John H. Humphrey, ed. The Roman and Byzantine Near East: Some Recent Archaeological Research. JRASS 14. Portsmouth, RI: Journal of Roman Archaeology, 1995, 95–107.

334 ———, and Haim Goldfus, eds. Excavations on the Site of the Jerusalem International Convention Center (Binyanei Ha⁠ʾuma): A Settlement of the Late First to Second Temple Period, the Tenth Legion’s Kilnworks, and a Byzantine Monastic Complex; The Pottery and Other Small Finds. JRASS 60. Portsmouth, RI: Journal of Roman Archaeology, 2005. Atkinson, Kenneth. “Representations of History in 4Q331 (4QpapHistorical Text C), 4Q332 (4QHistorical Text D), 4Q333 (4QHistorical Text E), and 4Q468e (4QHistorical Text F): An Annalistic Calendar Documenting Portentous Events?” DSD 14 (2007): 125–51. Avigad, Nahman. “The Palaeography of the Dead Sea Scrolls and Related Documents.” In Chaim Rabin, and Yigael Yadin, eds. Aspects of the Dead Sea Scrolls. ScrHier 4. Jerusalem: Magnes Press, 1958, 56–87. ———. “Expedition A.” IEJ 11 (1961): 6–10. ———. “Expedition A—Naḥal David.” IEJ 12 (1962): 169–83. Bagatti, Bellarmino. “P. Augusto Spijkerman (1920–1973).” LA 23 (1973): 387–90. Bailey, Donald M. Catalogue of the Lamps in the British Museum. Vol. 1, Greek, Hellenistic and Early Roman Pottery Lamps. London: British Museum Publications, 1975. ———. Catalogue of the Lamps in the British Museum. Vol. 2, Roman Lamps Made in Italy. London: British Museum Publications, 1980. ———. Catalogue of the Lamps in the British Museum. Vol. 3, Roman Provincial Lamps. London: British Museum Publications, 1988. Baillet, Maurice. “Textes des grottes 2Q, 3Q, 6Q, 7Q, à 10Q.” In Baillet, Milik, and de Vaux. Les ‘petites grottes’ de Qumrân, 45–168. ———. “Nouveaux phylactères de Qumrân (XQ Phyl 1–4): à propos d’une édition récente.” RevQ 27 (1970): 403–15. ———. “504. Paroles des luminaires (premier exemplaire: DibHama).” In Baillet. Qumrân Grotte 4.III, 105–36. ———. Qumrân Grotte 4.III (4Q482–4Q520). DJD 7. Oxford: Clarendon Press, 1982. ———, Józef T. Milik, and Roland de Vaux, eds. Les ‘petites grottes’ de Qumrân: Exploration de la falaise; Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q; Le rouleau de cuivre. DJD 3. 2 vols. Oxford: Clarendon Press, 1962. Balla, Marta. “The Search for Interregional Contacts between Different Groups of Jews Living in the Dead Sea Area during the Early Roman Period on the Basis of Pottery.” In Gunneweg, and Greenblatt. Outdoor Qumran and the Dead Sea, http://micro5.mscc.huji.ac.il/~msjan/balla.pdf. Bar Adon, Pesach. “Expedition C.” IEJ 11 (1961): 25–35. ———. “Expedition C—The Cave of the Treasure.” IEJ 12 (1962): 215–26. ———. “The Judaean Desert and Plain of Jericho.” In Kochavi. Judaea, Samaria and the Golan, 91–152 [Hebrew].

BIBLIOGRAPHY ———. “Another Settlement of the Judean Desert Sect at ʿEn el-Ghuweir on the Shores of the Dead Sea.” BASOR 227 (1977): 1–25. ———. The Cave of the Treasure: The Finds from the Caves in Naḥal Mishmar. Jerusalem: Israel Exploration Society, 1980. ———. “The Hasmonean Fortresses and the Status of Khirbet Qumran.” EI 15 (1981): 349–52 [Hebrew]. ———. “Excavations in the Judean Desert.” ʿAtiqot 9 (1989) [Hebrew]. Bar Nathan, Rachel. “Pottery and Stone Vessels of the Herodian Period (1st century B.C.–1st century A.D.).” In Ehud Netzer, ed. Greater Herodium. Qedem 13. Jerusalem: Hebrew University, 1981, 54–70. ———. Hasmonean and Herodian Palaces at Jericho: Final Reports of the 1973–1987 Excavations. Vol. 3, The Pottery. Jerusalem: Israel Exploration Society/Hebrew University, 2002. ———. Masada VII: The Yigael Yadin Excavations 1963–1965; Final Reports; The Pottery of Masada. Jerusalem: Israel Exploration Society/Hebrew University, 2006. ———. “Qumran and the Hasmonean and Herodian Winter Palaces of Jericho: The Implication of the Pottery Finds on the Interpretation of the Settlement at Qumran.” In Galor, Humbert, and Zangenberg. Qumran, the Site of the Dead Sea Scrolls, 263–77. ———, and Irina Eisenstadt. “The Ceramic Corpus from the Roman Estate at Jericho: Late 1st-Early 2nd Centuries C.E.” In Bar Nathan, and Gärtner. Hasmonean and Herodian Palaces at Jericho, Vol. 5: 3–84. ———, and Judit Gärtner. Hasmonean and Herodian Palaces at Jericho: Final Reports of the 1973–1987 Excavations. Vol. 5, The Finds from Jericho and Cypros. Jerusalem: Israel Exploration Society/Hebrew University, 2013. ———, and Ronit Gitler-Kamil. “Typology of the Herodian 3 Pottery.” In Bar Nathan. Hasmonean and Herodian Palaces at Jericho, Vol. 3: 145–92. Barag, Dan, ed. “Studies in Memory of Leo Mildenberg.” INJ 14 (2000–2002). ———, and Malka Hershkovitz. “Lamps from Masada.ˮ In Joseph Aviram, Gideon Foerster, and Ehud Netzer, eds. Masada IV: The Yigael Yadin Excavations 1963–1965; Final Reports. Jerusalem: Israel Exploration Society/Hebrew University, 1994, 1–147. Barshad, Dror, and Idan Shaked. “Region III: Survey and Excavations of Caves in Lower Wadi el-Makkuk.” ʿAtiqot 41/1 (2002): 33–41 [Hebrew]; ʿAtiqot 41/2 (2002): 35–41. Barthélemy, Dominique. “Textes Bibliques.” In Barthélemy, and Milik. Qumran Cave 1, 49–76. ———, and Józef T. Milik. “Les Textes.” In Barthélemy, and Milik. Qumran Cave 1, 43–155. ———, and Józef T. Milik, eds. Qumran Cave 1. DJD 1. Oxford: Clarendon Press, 1955.

BIBLIOGRAPHY Baruch, Yuval, Gabriel Mazor, and Debora Sandhaus. “Region XI: Survey and Excavations of Caves along the Fault Escarpment above Ḥorbat Qumran.” ʿAtiqot 41/1 (2002): 189–98 [Hebrew]; ʿAtiqot 41/2 (2002): 177–83. Bélis, Mireille. “Des textiles: Catalogues et commentaires.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol. 2: 207–76. ———. “Révision commentée des différents systèmes de numérotation.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol. 2: 409–15. Ben Dov, Jonathan. Head of All Years: Astronomy and Calendars at Qumran in their Ancient Context. STDJ 78. Leiden: Brill, 2008. ———, and Daniel Stökl Ben Ezra. “4Q249 Midrash Moshe: A New Reading and Some Implications.” DSD 21 (2014): 131–49. Ben Hayyim, Zeev. The Literary and Oral Tradition of Hebrew and Aramaic amongst the Samaritans. Vol. 2. Jerusalem: Bialik Institute/Academy of the Hebrew Language, 1957 [Hebrew]. Ben Tor, Amnon. “Excavations at Horbat ʿUsa.” ʿAtiqot 3 (1966): 1–24 [Hebrew]. Benoit, Pierre, Józef T. Milik, and Roland de Vaux, eds. Les grottes de Murabbaʿât. DJD 2. 2 vols. Oxford: Clarendon Press, 1961. Berlin, Andrea M. “Pottery and Pottery Production in the Second Temple Period.” In Arubas, and Goldfus. Excavations on the Site of the Jerusalem International Convention Center, 29–60. Bernick-Greenberg, Hannah. “The Ceramic Assemblages and the Wheel-Made Pottery Typology.” In Rudolph Cohen, and Hannah Bernick-Greenberg, eds. Excavations at Kadesh Barnea (Tell el-Qudeirat) 1976–1982. IAA Reports 34. Jerusalem: Israel Antiquities Authority, 2007, 131–86. Bijovsky, Gabriela. “The Coins.” In Hirschfeld. En-Gedi Excavations II, 157–232. Birnbaum, Solomon. “The Dates of the Cave Scrolls.” BASOR 115 (1949): 20–22. Bishop, Ronald L., and M. James Blackman. “Instrumental Neutron Activation Analysis of Archaeological Ceramics: Scale and Interpretation.” Accounts of Chemical Research 35 (2002): 603–10. Blau, Lajos. “Wie lange stand die althebräische Schrift bei den Juden im Gebrauch?” In Marcus Brann, and Ferdinand Rosenthal. Gedenkbuch zur Erinnerung an David Kaufmann. Breslau: Schottländer, 1900, 44–57. Bonani, Georges, et al. “Radiocarbon Dating of the Dead Sea Scrolls.ˮ ‛Atiqot 20 (1991): 27–32. ———, et al. “Radiocarbon Dating of Fourteen Dead Sea Scrolls.” Radiocarbon 34 (1992): 843–49. Bottini, Giovanni Claudio. “Michele Piccirillo (1944–2008) francescano di Terra Santa e archeologo.” LA 58 (2008): 479–500. ———. “Ricordo di Michele Piccirillo.” Notiziario Studium Biblicum Franciscanum; Anno Accademico 2008–2009 (2009): 8–24.

335 ———, and Massimo Luca, eds. Michele Piccirillo francescano archeologo tra scienza e provvidenza. SBF Museum 15. Milano: Edizioni Terra Santa, 2010. Brooke, George J. “Deuteronomy 5–6 in the Phylacteries from Qumran Cave 4.” In Paul, et al. Emanuel, 57–70. ———. Qumran and the Jewish Jesus: Reading the New Testament in the Light of the Scrolls. Grove Biblical Series 35. Cambridge: Grove Books, 2005. ———. “Florilegium (4Q174).” EDEJ, 646–47. ———. “Between Scroll and Codex?: Reconsidering the Qumran Opisthographs.” In James K. Aitken, Katharine J. Dell, and Brian A. Mastin, eds. On Stone and Scroll: Essays in Honour of Graham Ivor Davies. BZAW 420. Berlin: W. de Gruyter, 2011, 123–38. ———. “From Jesus to the Early Christian Communities: Modes of Sectarianism in the Light of the Dead Sea Scrolls.” In Adolfo D. Roitman, Lawrence H. Schiffman, and Shani Tzoref, eds. The Dead Sea Scrolls and Contemporary Culture: Proceedings of the International Conference held at the Israel Museum, Jerusalem ( July 6–8, 2008). STDJ 93. Leiden: Brill, 2011, 411–34. Broshi, Magen. “The Archaeology of Qumran—A Reconsideration.” In Devorah Dimant, and Uriel Rappaport, eds. The Dead Sea Scrolls: Forty Years of Research. STDJ 10. Leiden: Brill, 1992, 103–15. ———. “468e. 4QHistorical Text F.ˮ In Pfann. Qumran Cave 4.XXVI: Cryptic Texts; Alexander, et al. Miscellanea, Part 1, 406–22. ———, and Hanan Eshel. “How and Where did the Qumranites Live?” In Parry, and Ulrich. The Provo International Conference on the Dead Sea Scrolls, 267–73. ———, and Hanan Eshel. “Residential Caves at Qumran,” DSD 6 (1999): 328–48. Brown, Judith A. John Marco Allegro: Maverick of the Dead Sea Scrolls. Grand Rapids: Eerdmans, 2005. Brownlee, William H. “Muhammad ed-Deeb’s Own Story of His Scroll Discovery.” JNES 16 (1957): 236–39. ———. “Edh-Dheeb’s Story of His Scroll Discovery.” RevQ 12 (1962): 483–94. Bruce, Frederick F., and Stephen J. Pfann. “Qumran.” EncJud Vol. 16: 771–72. Busa, Anna. Die Phylakterien von Qumran (4Q128.129.135.137) aus der Heidelberger Papyrussammlung. PHeid Neue Folge 15. Heidelberg: C. Winter, 2015. Butcher, Kevin. Coinage in Roman Syria: Northern Syria, 64 BC– AD 253. London: RNS, 2004. Buxeda i Garrigós, Jaume, Miguel A. Cau Ontiveros, and Vassilis Kilikoglou. “Chemical Variability in Clays and Pottery from a Traditional Cooking Pot Production Village: Testing Assumptions in Pereruela.” Archaeometry 45/1 (2003): 1–17. Callegher, Bruno. “Note su Augustus Spijkerman numismatico (1920–1973),” LA 64 (2014): 615–47.

336 ———. “Henri Seyrig, Auguste Spijkerman et les institutions de recherches à Jérusalem.” Syria forthcoming. Campbell, Jonathan. The Exegetical Texts. CQS 4. London: T&T Clark, 2004. Cansdale, Lena. Qumran and the Essenes: A Re-evaluation of the Evidence. TSAJ 60. Tübingen: Mohr Siebeck, 1997. Carmignac, Jean. “Vestiges d’un Pesher de Malachie (?).” RevQ 13 (1964): 97–100. Carswell, John. “Fastenings on the Qumrân Manuscripts.” In de Vaux, and Milik. Qumrân Grotte 4.II, 23–28. Casey, Patrik J. Understanding Ancient Coins: An Introduction for Archaeologists and Historians. London: Batsford, 1986. Chambon, Alain. Tell el-Farʾah. Vol. 1, L’âge du Fer. Mémoire 31. Paris: Éditions Recherche sur les Civilisations, 1984. Charlesworth, James, et al. “Biblical Scrolls from Naḥal Ḥever and ‘Wadi Seiyal’: Introduction.” In Charlesworth, et al. Miscellaneous Texts from the Judean Desert, 133–36. ———, et al., eds. Miscellaneous Texts from the Judaean Desert. DJD 38. Oxford: Clarendon Press, 2000. Cheynet, Jean-Claude, Cécile Morrisson, and Werner Seibt, eds. Sceaux byzantins de la collection Henri Seyrig: Catalogue raisonné. Paris: Bibliothèque Nationale, 1991. Ciecieląg, Jerzy. “Coins from the So-Called Essene Settlements on the Dead Sea Shores.” In Kapera. Mogilany 1995, 105–15. Clamer, Christa. Fouilles archéologiques de ʿAïn ez-Zâra/ Callirrhoé: Villégiature hérodienne. BAH 147. Beirut: Institut français d’archéologie du Proche-Orient, 1997. ———. “Jewellery Finds from the Cemetery.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol. 2: 171–83. Cohen, Rudolph, and Yigal Yisraeli. “The Excavations of Rock Shelter XII/50 and in Caves XII/52–53.” ʿAtiqot 41/1 (2002): 207–13 [Hebrew]; ʿAtiqot 41/2 (2002): 185–86. Cohn, Yehudah B. Tangled Up in Text: Tefillin and the Ancient World. BJS 351. Providence, RI: Brown University, 2008. Collins, John J. Beyond the Qumran Community: The Sectarian Movement of the Dead Sea Scrolls. Grand Rapids, MI: Eerdmans, 2010. Corbo, Virgilio. “Prefazione.” In Augustus Spijkerman. Cafarnao. Vol. 3, Catalogo delle monete della città. SBF.CMa 19. Jerusalem: Franciscan Printing Press, 1975, 10. Cotton, Hannah M., and Erik W. Larson. “4Q460/4Q350 and Tampering with Qumran Texts in Antiquity?” In Paul, et al. Emanuel, 113–25. ———, and Ada Yardeni. “General Introduction.” In Cotton, and Yardeni. Aramaic, Hebrew and Greek, 1–6. ———, and Ada Yardeni, eds. Aramaic, Hebrew and Greek Documentary Texts from Naḥal Ḥever and Other Sites: With an Appendix Containing Alleged Qumran Texts (The Seiyâl Collection II). DJD 27. Oxford: Clarendon Press, 1997. Cox, Ben B. “Other Finds: Beads.” In Lapp, and Lapp. Discoveries in the Wâdī ed-Dâliyeh, 105–6.

BIBLIOGRAPHY Crawford, Michael H. Roman Republican Coinage. 2 vols. Cambridge: Cambridge University Press, 1974. Croom, Alexandra T. Roman Clothing and Fashion. Charleston, SC: Tempus, 2000. Cross, Frank M. “The Oldest Manuscripts from Qumran.” JBL 74 (1955): 147–72. ———. “The development of the Jewish Scripts.” In George E. Wright, ed. The Bible and the Ancient Near East: Essays in Honor of William Foxwell Albright. Garden City, NY: Doubleday, 1961, 133–202. A slightly revised version in Cross. Leaves from an Epigrapher’s Notebook, 3–43. ———. “Écriture et chiffres.” In Baillet, Milik, and de Vaux. Les ‘petites grottes’ de Qumrân, 215–21. ———. “Epigraphic Notes on Hebrew Documents of the Eighth-Sixth Centuries B.C.: II. The Murabba⁠ʾât Papyrus and the Letter Found near Yabneh-Yam.” BASOR 165 (1962): 34–42. ———. “El-Buqeiʾa.” NEAEHL Vol. 1: 267–69. ———. The Ancient Library of Qumran and Modern Biblical Studies. Garden City, NY: Doubleday, 1958; 3rd edition Sheffield: Sheffield Academic Press, 1995. ———. “The Scripts of the Dâliyeh (Samaria) Papyri.” In Cross. Leaves from an Epigrapher’s Notebook, 44–46. ———. Leaves from an Epigrapher’s Notebook: Collected Papers in Hebrew and West Semitic Palaeography and Epigraphy. HSS 51. Winona Lake, IN: Eisenbrauns, 2003. ———, and Esther Eshel. “Ostraca from Khirbet Qumran.” In Pfann. Qumran Cave 4.XXVI: Cryptic Texts; Alexander, et al. Miscellanea, Part 1, 497–514. ———, and Józef T. Milik. “Explorations in the Judaean Buqêʾah.” BASOR 142 (1956): 5–17. Crowfoot, Elisabeth. “Other Finds: Textiles.” In Lapp, and Lapp. Discoveries in the Wâdī ed-Dâliyeh, 60–78. Crowfoot, Grace M. “The Linen Textiles.” In Barthélemy, and Milik. Qumran Cave 1, 18–38. ———, and Elisabeth Crowfoot. “The Textiles and Basketry.” In Benoit, Milik, and de Vaux. Les grottes de Murabbaʿât, 51–63. Crown, Alan. “An Alternative View of Qumran” In Moshe Bar Asher, and Moshe Florentin, eds. Samaritan, Hebrew and Aramaic Studies Presented to Professor Abraham Tal. Jerusalem: Bialik Institute, 2005, 1–24. Dahari, Uzi. Monastic Settlements in South Sinai in the Byzantine Period: The Archaeological Remains. IAA Reports 9. Jerusalem: Israel Antiquities Authority, 2000. ———. “Region XIV: Survey and Excavations of Caves along the Fault Escarpment from Naḥal Kidron to Naḥal Deragot.” ʿAtiqot 41/2 (2002): 209–19. Dar, Shimon. “The Search for Scrolls in the Judaean Desert Caves in the Years 1950–1960—An Archaeological Memoir.” BAIAS 28 (2010): 141–57. Davies, Philip R. “The ‘Damascus’ Sect and Judaism.” In John C. Reeves, and John Kampen, eds. Pursuing the Text: Studies in

BIBLIOGRAPHY Honor of Ben Zion Wacholder on the Occasion of his Seventieth Birthday. JSOTSup 184. Sheffield: Sheffield Academic Press, 1994, 70–84. Reprinted in Davies. Sects and Scrolls: Essays on Qumran and Related Topics. SFSHJ 134. Atlanta, GA: Scholars Press, 1996, 163–77. De Groot, Alon, and Donald T. Ariel. “Ceramic Report.” In Ariel, ed. Excavations at the City of David 1978–1985 Directed by Yigal Shiloh. Vol. 5, Extramural Areas. Qedem 40. Jerusalem: Institute of Archaeology, Hebrew University, 2000, 91–154. Deichgräber, Reinhard. “Fragmente einer Jubiläen-Handschrift aus Höhle 3 von Qumran.” RevQ 19 (1965): 415–22. Diez Fernandez, Florentino. Cerámica común romana de la Galilea: Aproximaciones y diferencias con la cerámica del resto de Palestina y regiones circundantes. Jerusalem/Madrid: Escuela Biblica, 1983. Dimant, Devorah. “The Qumran Manuscripts: Contents and Significance.” In Dimant, and Lawrence H. Schiffman, eds. Time to Prepare the Way in the Wilderness: Papers on the Qumran Scrolls by Fellows of the Institute for Advanced Studies of the Hebrew University, Jerusalem, 1989–1990. STDJ 16. Leiden: Brill, 1995, 23–58. ———. “The Composite Character of the Qumran Sectarian Literature as an Indication of Its Date and Provenance.” RevQ 88 (2006): 615–30. ———. “The Vocabulary of the Qumran Sectarian Texts.” In Frey, Claussen, and Kessler. Qumran und die Archäologie, 347–95. Diringer, David. “Early Hebrew Script versus Square Script.” In David W. Thomas, ed. Essays and Studies Presented to Stanley Arthur Cook. London: Taylor’s Foreign Press, 1950, 35–49. Donceel, Robert. “Reprise des travaux de publication des fouilles au Khirbet Qumrân.” RB 99 (1992): 557–73. ———. “Poursuite des travaux de publication du matériel archéologique de Khirbet Qumrân: Les lampes en terrecuite.ˮ In Kapera. Mogilany 1995, 99–104. ———, and Pauline Donceel-Voûte. “The Archaeology of Khirbet Qumran.” In Wise, et al. Methods of Investigation of the Dead Sea Scrolls, 1–38. Dothan, Moshe. “The Excavations at Nahariya: Preliminary Report (Seasons 1954/55).” IEJ 6 (1956): 14–25. Doudna, Gregory L. “Dating the Scrolls on the Basis of Radiocarbon Analysis.ˮ In Flint, and VanderKam. The Dead Sea Scrolls After Fifty Years, Vol. 1: 430–71. ———. “Redating the Dead Sea Scrolls Found at Qumran: The Case for 63 BCE.” QC 8/4 (1999). ———. 4Q Pesher Nahum: A Critical Edition. JSPSup 35. Sheffield: Sheffield Academic Press, 2001. ———. “The Legacy of an Error in Archaeological Interpretation: The Dating of the Qumran Cave Scroll Deposits.” In Galor, Humbert, and Zangenberg. Qumran, the Site of the Dead Sea Scrolls, 147–57.

337 ———. “The Sect of the Qumran Texts and its Leading Role in the Temple in Jerusalem During Much of the First Century BCE: Toward a New Framework for Understanding.” In Stacey, and Doudna. Qumran Revisited, 75–124. Doyen, Jean-Marc. “Cliométrie et numismatique contextuelle: compter et quantifier le passé? Petite histoire de la méthode (1960–2011).” JAN 1 (2011): 9–46. Driver, Godfrey R. The Judaean Scrolls: The Problem and a Solution. New York: Schocken Books, 1965. Drori, Amir, et al. “Operation Scroll.” In Twentieth Archaeological Conference in Israel: Abstracts. Jerusalem: Israel Exploration Society, 1994, 12–17 [Hebrew]. Dušek, Jan. Aramaic and Hebrew Inscriptions from Mt. Gerizim and Samaria between Antiochus III and Antiochus IV Epiphanes. CHANE 54. Leiden: Brill, 2012. Edge, Raymond L. “The Use of Palaeo-Hebrew in the Dead Sea Scrolls: Paleography and Historiography.” Ph.D. diss., University of Texas at Austin, 1995. Edwards, Douglas R., ed. Religion and Society in Roman Palestine: Old Questions, New Approaches. London: Routledge, 2004. Eisenman, Robert. “The 1988–1922 California State University Dead Sea Walking Cave Survey and Radar Groundscan of the Qumran Cliffs.” QC 9 (2000): 123–30. Enzel, Yehouda, et al. “Late Holocene Climates of the Near East deduced from Dead Sea Level Variations and Modern Regional Winter Rainfall.” Quarternary Research 60 (2003): 263–73. Eshel, Esther. “4QDeutn—A Text that has Undergone Harmonistic Editing.” HUCA 62 (1991): 117–54. ———. “477. 4QRebukes Reported by the Overseer.” In Pfann. Qumran Cave 4.XXVI: Cryptic Texts; Alexander, et al. Miscellanea, Part 1, 474–84. ———, Hanan Eshel, and Ada Yardeni. “448. 4QApocryphal Psalm and Prayer.” In Esther Eshel, et al. eds. Qumran Cave 4.VI: Poetical and Liturgical Texts, Part 1. DJD 11. Oxford: Clarendon Press, 1998, 403–25. ———, and Amos Kloner. “An Aramaic Ostracon of an Edomite Marriage Contract from Maresha, Dated 176 B.C.E.” IEJ 46 (1996): 1–22. Eshel, Hanan. “A Coin of Bar Kokhba from a Cave in Wadi el-Māckūk.” INJ 9 (1986–1987): 51–52. ———. “Nailed Sandals in the Jewish Sources in Light of the Finds of the Refuge Caves.” In Eshel, and Amit. Refuge Caves of the Bar Kokhba Revolt, 225–32 [Hebrew]. ———. “Ḥever, Naḥal: Archaeology.” EDSS Vol. 1: 357–59. ———. “Murabbaʿat, Wadi: Written Material.” EDSS Vol. 1: 583–86. ———. “Documents of the First Jewish Revolt from the Judean Desert.” In Andrea M. Berlin, and Andrew J. Overman. The First Jewish Revolt: Archaeology, History and Ideology. London: Routledge, 2002, 157–63.

338 ———. “6Q30, a Cursive Sin, and Proverbs 11.” JBL 122 (2003): 544–46. ———. “Gleaning of Scrolls from the Judean Desert.” In Hempel. The Dead Sea Scrolls, 49–87. ———. “Excavations in the Judean Desert and at Qumran under Israeli Jurisdiction.” In Devorah Dimant. The Dead Sea Scrolls in Scholarly Perspective: A History of Research. STDJ 99. Leiden: Brill, 2012, 381–99. ———, and David Amit, eds. Refuge Caves of the Bar Kokhba Revolt. Vol. 1. Tel Aviv: Israel Exploration Society, 1998 [Hebrew]. ———, Yosi Baruchi, and Roi Porat. “Fragments of a Leviticus Scroll (ArugLev) Found in the Judean Desert in 2004.” DSD 13 (2006): 55–60. ———, and Magen Broshi. “The Archaeological Remains on the Marl Terrace around Qumran.” Qadmoniot 30 (1997): 129–33 [Hebrew]. ———, and Magen Broshi. “Excavations at Qumran, Summer of 2001.” IEJ 53 (2003): 61–73. ———, and Magen Broshi. “Three Seasons of Excavations at Qumran.” JRA 17 (2004): 321–32. ———, and Esther Eshel. “Fragments of Two Aramaic Documents which were Brought to the Abiʿor Cave during the Bar Kokhba Revolt.” EI 23 (1992): 276–85 [Hebrew]. ———, and Haggai Misgav. “A Document from the Fourth Century B.C.E. from a Cave in Ketef Jericho.” Tarbiz 56 (1987): 461–477 [Hebrew]. ———, and Haggai Misgav. “A Fourth Century B.C.E. Document from Ketef Yeriḥo.” IEJ 38 (1988): 158–76. ———, and Boaz Zissu. “Roman Coins from the ‘Cave of the Sandal’ West Jericho.” INJ 13 (1994–1999): 70–77. ———, and Boaz Zissu. “Excavations and Surveys: Ketef Jericho, 1993.” IEJ 45 (1995): 292–98. ———, and Boaz Zissu. “Finds from the Bar Kokhba Period in the Caves at Ketef Jericho.” In Eshel, and Amit. Refuge Caves of the Bar Kokhba Revolt, Vol. 1: 113–51 [Hebrew]. ———, and Boaz Zissu. “Naḥal Mikhmash, Magharat el-Jāi.” ESI 110 (1999): 75–76 [Hebrew]; 56*–57*. ———, and Boaz Zissu. “Jericho: Archaeological Introduction.” In Charlesworth, et al. Miscellaneous Texts from the Judean Desert, 3–20. ———, and Boaz Zissu. “Coins from the el-Jai Cave in Naḥal Mikhmash (Wadi Suweinit).” INJ 14 (2000–2002): 169–75. ———, and Boaz Zissu. “Region VIII: Survey and Excavations of Caves along the Cliff Slopes of Triangulation Point 86, on the Fringes of Jebel Maʿar el-Bas.” ʿAtiqot 41/1 (2002): 141–50 [Hebrew]; ʿAtiqot 41/2 (2002): 117–23. ———, Boaz Zissu, and Amos Frumkin. “Two Refuge Caves in Wadi Suweinit.” In Eshel, and Amit. Refuge Caves of the Bar Kokhba Revolt, Vol. 1: 93–109 [Hebrew]. Eshel, Itzhak. “Two Pottery Groups from Kenyon’s Excavations on the Eastern Slope of Ancient Jerusalem.” In Eshel, and Kay

BIBLIOGRAPHY Prag, eds. Excavations by K. M. Kenyon in Jerusalem 1961–67. Vol. 4, The Iron Age Cave Deposits on the South-east Hill and Isolated Burials and Cemeteries Elsewhere. BAMA 6. Oxford: Oxford University Press, 1995, 1–157. Farhi, Yoav, and Randall Price. “The Numismatic Finds from the Qumran Plateau Excavations 2004–2006, and 2008 Seasons.” DSD 17 (2010): 210–25. Fidanzio, Marcello. “Which Cave Does This Pottery Come From? The Information Written on the Pottery Found in the Qumran Caves (R. de Vaux Excavations 1949–1956).” RB 122 (2015): 128–31. Fields, Weston W. The Dead Sea Scrolls: A Full History. Vol. 1, 1947–1960. Leiden: Brill, 2009. Findlater, George, et al. “The Wadi Faynan Project: The South Cemetery Excavations, Jordan 1996; A Preliminary Report.” Levant 30 (1998): 69–83. Fischer, Moshe, and Oren Tal. “Pottery.” In Fischer, Gichon, and Tal, eds. ‘En Boqeq: Excavations in an Oasis on the Dead Sea, Vol. 2: 29–67. ———, Mordechai Gichon, and Oren Tal, eds. ‘En Boqeq: Excavations in an Oasis on the Dead Sea. Vol. 2, The Officina, An Early Roman Building on the Dead Sea Shore. Mainz: Philipp von Zabern, 2000. Fitzmyer, Joseph A. “331. 4QpapHistorical Text C.ˮ In Pfann. Qumran Cave 4.XXVI: Cryptic Texts; Alexander, et al. Miscellanea, Part 1, 275–80. ———. “332. 4QHistorical Text D.ˮ In Pfann. Qumran Cave 4.XXVI: Cryptic Texts; Alexander, et al. Miscellanea, Part 1, 281–86. ———. “333. 4QHistorical Text E.ˮ In Pfann. Qumran Cave 4.XXVI: Cryptic Texts; Alexander, et al. Miscellanea, Part 1, 287–89. ———. A Guide to the Dead Sea Scrolls and Related Literature. Grand Rapids, MI: Eerdmans, 2008. Flint, Peter W., and James C. VanderKam, eds. The Dead Sea Scrolls After Fifty Years: A Comprehensive Assessment. 2 vols. Leiden: Brill, 1998–1999. Fraade, Steven D. Legal Fictions: Studies of Law and Narrative in the Discursive Worlds of Ancient Jewish Sectarians and Sages. JSJSup 147. Leiden: Brill, 2011. Franken, Hendricus J. “The Twelve Pottery Classes.” In Franken, and Margreet L. Steiner. Excavations in Jerusalem 1961–1967. Vol. 2, The Iron Age Extramural Quarter on the South-East Hill. BAMA 2. Oxford: Oxford University Press, 1990, 99–121. Freedman, David N., and Kenneth A. Mathews. The PaleoHebrew Leviticus Scroll (11QpaleoLev). Winona Lake, IN: Eisenbrauns, 1985. Freud, Liora. “Pottery: Iron Age.” In Itzhak Beit Arieh, ed. Tel ʿIra: A Stronghold in the Biblical Negev. MSSMNIA 15. Tel Aviv: Institute of Archaeology, Tel Aviv University, 1999, 189–209. ———. “Iron Age Pottery.” In Itzhak Beit Arieh, ed. Horvat ʿUza and Horvat Radum: Two Fortresses in the Biblical Negev.

BIBLIOGRAPHY MSSMNIA 25. Tel Aviv: Institute of Archaeology, Tel Aviv University, 2007, 77–121. ———, and Itzhak Beit Arieh. “Pottery.” In Itzhak Beit Arieh, ed. Horvat Qitmit: An Edomite Shrine in the Biblical Negev. MSSMNIA 11. Tel Aviv: Institute of Archaeology, Tel Aviv University, 1995, 209–57. Frey, Jörg, Carsten Claussen, and Nadine Kessler, eds. Qumran und die Archäologie: Texte und Kontexte. WUNT 278. Tübingen: Mohr Siebeck, 2011. Galor, Katharina, Jean-Baptiste Humbert, and Jürgen Zangenberg, eds. Qumran, the Site of the Dead Sea Scrolls: Archaeological Interpretations and Debates; Proceedings of a Conference Held at Brown University, November 17–19, 2002. STDJ 57. Leiden: Brill, 2006. Gamble, Harry Y. Books and Readers in the Early Church: A History of Early Christian Texts. New Haven, CT/London: Yale University Press, 1995. García Martínez, Florentino. “Greek Loanwords in the Copper Scroll.” In García Martínez, and Gerard P. Luttikhuizen, eds. Jerusalem, Alexandria, Rome: Studies in Ancient Cultural Interaction in Honour of A. Hilhorst. JSJSup 82. Leiden: Brill, 2004, 119–45. ———. “Qumrán en el siglo XXI: Cambios y perspectivas después de 50 años de estudios.” MEAH 55 (2006): 309–34. ———. “¿Sectario, no-sectario, o qué? Problemas de una taxonomía correcta de los textos qumránicos.” RevQ 91 (2008): 383–94. ———. “Cave 11 in Context.” In Hempel. The Dead Sea Scrolls, 199–209. ———. “Reconsidering the Cave 1 Texts Sixty Years After Their Discovery: An Overview.” In Daniel Falk, et al., eds. Qumran Cave 1 Revisited: Texts from Cave 1 Sixty Years after Their Discovery; Proceedings of the Sixth Meeting of the IOQS in Ljubljana. STDJ 91. Leiden: Brill, 2010, 1–13. ———, and Eibert J. C. Tigchelaar. The Dead Sea Scrolls Study Edition. Leiden/Grand Rapids, MI: Brill/Eerdmans, 2000. ———, Eibert J. C. Tigchelaar, and Adam S. Van der Woude. “11QUnidentified Wads.” In García Martínez, Tigchelaar, and Van der Woude. Qumran Cave 11.II, 445–46. ———, Eibert J. C. Tigchelaar, and Adam S. van der Woude, eds. Qumran Cave 11.II: 11Q2–18, 11Q20–31. DJD 23. Oxford: Clarendon Press, 1997. Gebhart, Hans, et al. “Bemerkungen zur kritischen Neuaufnahme der Fundmünzen der römischen Zeit in Deutschland.” JNG 7 (1956): 9–76. Getzov, Nimrod, et al. Horbat ʿUza: The 1991 Excavations. Vol. 2, The Late Periods. IAA Reports 42. Jerusalem: Israel Antiquities Authority, 2009. Geva, Hillel. “Hellenistic Pottery from Areas W and X-2.” In Geva, ed. Jewish Quarter Excavations in the Old City of Jerusalem Conducted by Nahman Avigad, 1969–1982. Vol. 2, The Finds

339 from Areas A, W and X-2; Final Report. Jerusalem: Israel Exploration Society, 2003, 113–75. ———, and Malka Hershkovitz. “Local Pottery of the Hellenistic and Early Roman Periods.” In Hillel Geva, ed. Jewish Quarter Excavations in the Old City of Jerusalem Conducted by Nahman Avigad, 1969–1982. Vol. 3, Area E and Other Studies; Final Report. Jerusalem: Israel Exploration Society, 2006, 94–192. Gibson, Shimon, and Joan E. Taylor. “Roads and Passes Round Qumran.” PEQ 140 (2008): 225–27. Gitin, Seymour. Gezer III: A Ceramic Typology of the Late Iron II, Persian and Hellenistic Periods at Tell Gezer. Annual of the Nelson Glueck School of Biblical Archaeology 3. Jerusalem: Hebrew Union College, 1990. Gitler, Haim. “A Comparative Study of Numismatic Evidence from Excavations in Jerusalem.” LA 46 (1996): 317–62. ———, and Oren Tal. The Coinage of Philistia of the Fifth and Fourth Centuries BC: A Study of the Earliest Coins of Palestine. Collezioni Numismatiche: Materiali Pubblici e Privati 6. Milan: Ennerre, 2006. Glessmer, Uwe. “Calendars in the Qumran Scrolls.” In Flint, and VanderKam. The Dead Sea Scrolls After Fifty Years, Vol. 2: 213–78. Golb, Norman. Who Wrote the Dead Sea Scrolls? The Search for the Secret of Qumran. New York: Scribner, 1995. Goranson, Stephen. “Sectarianism, Geography, and the Copper Scroll.” JJS 43 (1992): 282–87. Gorin-Rosen, Yael. “Glass.” In Wolff, “The Excavation of Cave IV/12,” 81. ———. “The Glass Vessels from Cave VIII/28.” ʿAtiqot 41/2 (2002): 143–45. Grant, Elihu, and George E. Wright. Ain Shems Excavations V. Biblical and Kindered Studies 8. Haverford, PN: Haverford College, 1939. Greenfeld, Jonas C. “The Texts from Naḥal Ṣeʾelim (Wadi Seiyal).” In Julio Trebolle Barrera, and Luis Vegas Montaner, eds. The Madrid Qumran Congress: Proceedings of the International Congress on the Dead Sea Scrolls, Madrid 18–21 March, 1991. Vol. 2. STDJ 11. Leiden: Brill, 1992, 661–65. Gropp, Douglas M. Wadi Daliyeh II: The Samaria Papyri from Wadi Daliyeh; Moshe Bernstein, et al., eds. Qumran Cave 4.XXVIII: Miscellanea, Part 2. DJD 28. Oxford: Clarendon Press, 2001. Gunneweg, Jan. “Archaeology and Archaeometry at Qumran.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol. 2: xv–xx. ———. “Engedi-Qumran-Schoeyen and Allegro Jars: Four Similar Looking Dead Sea Jars that All Have a Different Provenience: Appearances Provide Sometimes the Wrong Information.” In Gunneweg, and Greenblatt. Outdoor Qumran and the Dead Sea, http://micro5.mscc.huji.ac.il/ ~msjan/aeqsj.pdf.

340 ———, Annemie Adriaens, and Joris Dik, eds. Holistic Qumran: Trans-Disciplinary Research of Qumran and the Dead Sea Scrolls; Proceedings of the NIAS-Lorentz Center Qumran Workshop 21–25 April 2008. STDJ 87. Leiden: Brill, 2010. ———, and Marta Balla. “Neutron Activation Analysis: Scroll Jars and Common Ware.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol. 2: 3–53. ———, and Marta Balla. “Possible Connection Between the Inscriptions on Pottery, the Ostraca and Scrolls Found in the Caves.ˮ In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol 2: 389–96. ———, and Marta Balla. “Was the Qumran Settlement a Mere Pottery Production Center? What Instrumental Neutron Activation Revealed.” In Gunneweg, Adriaens, and Dik. Holistic Qumran, 39–61. ———, and Charles Greenblatt, eds. Outdoor Qumran and the Dead Sea: Its Impact on the Indoor Bio- and Material Cultures at Qumran and the Judean Desert Manuscripts; Proceedings of the Joint Hebrew University and COST Action D-42 Cultural Heritage Workshop held at the Hebrew University of Jerusalem in May 25–26, 2010. Jerusalem: Hebrew University, 2012. ———, Charles Greenblatt, and Annemie Adriaens, eds. Bioand Material Cultures at Qumran: Papers from a COST Action G8 Working Group Meeting held in Jerusalem, Israel on 22–23 May 2005. Stuttgart: Fraunhofer IRB, 2006. Hadad, Shulamit. Excavations at Bet Shean. Vol. 1, The Oil Lamps from the Hebrew University Excavations at Bet Shean. QR 4. Jerusalem: Institute of Archaeology, Hebrew University/ Israel Exploration Society, 2002. Hadas, Gideon. “Nine Tombs of the Second Temple Period at ‘En Gedi.” ʿAtiqot 24 (1994): 3–65 [Hebrew]; 1*–8*. Hanson, Richard S. “Paleo-Hebrew Scripts in the Hasmonean Age.” BASOR 175 (1964): 26–42. Harding, Gerald L. “The Dead Sea Scrolls.” PEQ 81 (1949): 112–16. ———. “Khirbet Qumran and Wady Muraba‛at: Fresh Light on the Dead Sea Scrolls and New Manuscript Discoveries in Jordan.” PEQ 84 (1952): 104–9. ———. “Introductory, the Discovery, the Excavation, Minor Finds.” In Barthélemy, and Milik. Qumran Cave 1, 3–7. Hayes, John W. Ancient Lamps in the Royal Ontario Museum. Toronto: Royal Ontario Museum, 1980. Hempel, Charlotte. “Qumran Communities: Beyond the Fringes of Second Temple Society.” In Stanley E. Porter, and Craig Evans, eds. The Scrolls and the Scriptures: Qumran Fifty Years After. JSPSup 26. Sheffield: Sheffield Academic Press, 1997, 43–53. ———. “Interpretative Authority in the Community Rule Tradition.” DSD 10 (2003): 59–80. ———. “Emerging Communal Life and Ideology in the S Tradition.” In Florentino García Martínez and Mladen

BIBLIOGRAPHY Popović, eds. Defining Identities: ‘We’, ‘You’ and ‘the Other’ in the Dead Sea Scrolls; Proceedings of the Fifth Meeting of the IOQS in Groningen. STDJ 70. Leiden: Brill, 2007, 43–61. ———, ed. The Dead Sea Scrolls: Texts and Context. STDJ 90. Leiden: Brill, 2010. ———. “ ‘Haskalah’ at Qumran: The Eclectic Character of Qumran Cave 4.” In Hempel. The Qumran Rule Texts in Context, 303–37. ———. The Qumran Rule Texts in Context: Collected Studies. TSAJ 154. Tübingen: Mohr Siebeck, 2013. ———. “The Long Text of the Serekh as Crisis Literature.” RevQ 105 (2015): 3–24. Hendin, David. “The Metrology of Judean Small Bronze Coins.” AJN 21 (2009): 105–21. Herzog, Zeʾev, et al. “The Israelite Fortress at Arad.” BASOR 254 (1984): 1–34. Hirschfeld, Yizhar. The Judean Desert Monasteries in the Byzantine Period. New Haven: Yale University Press, 1992. ———. Qumran in Context: Reassessing the Archaeological Evidence. Peabody, MA: Hendrickson, 2004. ———. En-Gedi Excavations II. Final Report (1996–2002). Jerusalem: Israel Exploration Society/Institute of Archaeology, Hebrew University, 2007. ———. “A Settlement of Hermits above ‘En Gedi’.” Tel Aviv 27 (2008): 103–55. ———, and Donald T. Ariel. “A Coin Assemblage from the Reign of Alexander Jannaeus Found on the Shore of the Dead Sea.” IEJ 55 (2005): 66–89. ———, and Shimon Riklin. “Region II: Survey and Excavations in the Upper Wadi el-Makkuk Caves.” ʿAtiqot 41/1 (2002): 9–30 [Hebrew]; ʿAtiqot 41/2 (2002): 5–19. Holmgren, Richard, Anders Kaliff, and Jan Svennson. “The Hermit Life on al-Lisân Peninsula—Results of the Swedish Dead Sea Expedition: a Preliminary Report.” ADAJ 49 (2005): 167–88. Howgego, Christopher. “The Relationship of the Issar to de Denar in Rabbinic Literature.” INJ 8 (1984–1985): 59–64. Humbert, Jean-Baptiste. “L’espace sacré à Qumrân: Propositions pour l’archéologie.” RB 101 (1994): 161–214. ———. “Arguments en faveur d’une résidence pré-essénienne.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol. 2: 467–82. ———. “L’agglomérat métallique KhQ 960 et son contexte.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol. 2: 397–405. ———. “Reconsideration of the Archaeological Interpretation.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol. 2: 419–24. ———. “The Chronology during the First Century B.C., de Vaux and his Method: a Debate.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol. 2: 425–44.

BIBLIOGRAPHY ———. “Some Remarks on the Archaeology of Qumran.” In Galor, Humbert, and Zangenberg. Qumran, the Site of the Dead Sea Scrolls, 19–39. ———. “Activités de l’École Archéologique Française de Jérusalem. L’archéologie de Qumrân. Le dossier numismatique.” In École Biblique et Archéologique Française de Jérusalem. Rapport d’Activités 2007–2008. Jerusalem: EBAF archives, 48–57. ———, and Alain Chambon, eds. Fouilles de Khirbet Qumrân et de Aïn Feshkha. Vol. 1, Album de photographies; Répertoire du fonds photographique; Synthèse des notes de chantier du Père Roland de Vaux. NTOA.SA 1. Fribourg/Göttingen: Éditions Universitaires/Vandenhoeck & Ruprecht, 1994. ———, and Alain Chambon, eds. The Excavations of Khirbet Qumran and Ain Feshkha: Synthesis of Roland de Vaux’s Field Notes. Translated and revised by Stephen J. Pfann. NTOA.SA 1B. Fribourg/Göttingen: Éditions Universitaires/ Vandenhoeck & Ruprecht, 2003. ———, and Jan Gunneweg, eds. Khirbet Qumrân et ʿAïn Feshkha. Vol. 2, Études d’anthropologie, de physique et de chimie. NTOA.SA 3. Fribourg/Göttingen: Éditions Universitaires/ Vandenhoeck & Ruprecht, 2003. Hutchesson, Ian. “63 BCE: A Revised Dating for the Depositation of the Dead Sea Scrolls.” QC 8 (1999): 177–94. Ibrahim, Fawzi. “Region XII: The Excavation of Cave XII/49.” ʿAtiqot 41/1 (2002): 202–5 [Hebrew]; ʿAtiqot 41/2 (2002): 185. ———. “The Excavation of Cave XII/56.” ʿAtiqot 41/1 (2002): 215–16 [Hebrew]; ʿAtiqot 41/2 (2002): 186. ———, and Nadav Hameiri. “The Excavation of Cave XII/61.” ʿAtiqot 41/1 (2002): 217–18 [Hebrew]; ʿAtiqot 41/2 (2002): 186. Itah, Michel, Yoni Kam, and Ronny Ben Haim. “Region X: Survey and Excavations of Caves along the Fault Escarpment South of Almog Junction and West of Qalya.” ʿAtiqot 41/1 (2002): 175–86 [Hebrew]; ʿAtiqot 41/2 (2002): 169–76. Jackson-Tal, Ruth E. “The Glass Finds from Cave VIII/9 (‘The Large Caves Complex’).” ʿAtiqot 41/1 (2002): 167–68 [Hebrew]; ʿAtiqot 41/2 (2002): 127. ———. “The Glass Vessels.” In Sion. “Regions IV and VI,” ʿAtiqot 41/1 (2002): 78 [Hebrew]; ʿAtiqot 41/2 (2002): 63. ———. “Glass Vessels from En-Gedi.” In Hirschfeld, En-Gedi Excavations II, 474–506. ———. “The Glass Finds from the Hasmonean and Herodian Palaces at Jericho.” In Bar Nathan, and Gärtner. Hasmonean and Herodian Palaces at Jericho, Vol. 5: 100–29. James, Frances W. The Iron Age at Beth Shan: A Study of Levels VI–IV. Philadelphia: University Museum, University of Pennsylvania, 1966. Jiménez Bedman, Francisco. El misterio del Rollo de Cobre de Qumrán: Análisis Lingüístico. Biblioteca Midrásica 25. Estella: Verbo Divino, 2002.

341 Jokiranta, Jutta, and Hanna Vanonen. “Multiple Copies of Rule Texts or Multiple Rule Texts: Boundaries of the S and M Documents.” In Mika S. Pajunen, and Hanna Tervanotko, eds. Crossing Imaginary Boundaries: The Dead Sea Scrolls in the Context of Second Temple Judaism. Helsinki: Finnish Exegetical Society, 2015, 11–60. Jull, A. J. Timothy, et al. “Radiocarbon Dating of Scrolls and Linen Fragments from the Judaean Desert.” Radiocarbon 37 (1995): 11–19. Reprinted in ʿAtiqot 28 (1996): 85–91. Kapera, Zdzisław J., ed. Mogilany 1995: Papers on the Dead Sea Scrolls Offered in Memory of Aleksy Klawek. Qumranica Mogilanensia 15. Cracow: Enigma Press, 1998. ———, and Robert Feather. Doyen of the Dead Sea Scrolls: An In-Depth Biography of Józef Tadeusz Milik (1922–2006). Qumranica Mogilanensia 17. Cracow/Mogilany: Enigma Press, 2011. Karageorghis, Vassos. Two Cypriotes Sanctuaries of the End of the Cypro-Archaic Period. Rome: Consiglio Nazionale delle Ricerche, 1977. Kelso, James L., and Dimitri C. Baramki. Excavations at New Testament Jericho and Khirbet en-Nitla. AASOR 29–30. New Haven: American Schools of Oriental Research, 1955. Kennedy, Charles A. “The Development of the Lamp in Palestine.” Berytus 14 (1963): 67–115. Killebrew, Ann E. “Cosmetics.” EDSS Vol. 1: 148–49. King, Philip J. and Lawrence E. Stager. Life in Biblical Israel. Library of Ancient Israel. Louisville, Kentucky: Westminster John Knox Press, 2001. Klijn, Albertus F. J. “A Library of Scriptures in Jerusalem?” In Kurt Treu, ed. Studia Codicologica. TU 124. Berlin: Akademie Verlag, 1977, 265–72. Klimowsky, Ernest W. “Leo Kadman: In Memoriam.” INJ 2 (1964): 3–6. Kloner, Amos, and Yigael Tepper. The Hiding Complexes in the Judean Shephelah. Tel Aviv: Hakibbutz Hameuchad Publishing House, 1987 [Hebrew]. ———, and Boaz Zissu. “Hiding Complexes in Judaea: An Archaeological and Geographical Update on the Area of the Bar Kokhba Revolt.” In Schäfer. The Bar Kokhba War Reconsidered, 181–216. Knibb, Michael A. “The Community of the Dead Sea Scrolls: Introduction.” DSD 16 (2009): 297–308. Kochavi, Moshe, ed. Judaea, Samaria and the Golan: Archaeological Survey 1967–1968. Jerusalem: Carta, 1972. Kuhn, Karl G., ed. Phylakterien aus Höhle 4 von Qumran. Abhandlungen der Heidelberger Akademie der Wissenschaften, Philosophisch-Historische Klasse 1. Heidelberg: C. Winter, 1957. Lacerenza, Giancarlo. “Un nouveau fragment en écriture paléohébraïque.” RevQ 75 (2000): 441–47.

342 Lacoudre, Noël, et al. “L’amas métallique KhQ 960 du site de Khirbet Qumrân.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol 2: 397–404. Ladizhinskaya, Victoria. “Jewelry and Metal Artefacts of Middle Bronze Age II and the Early Roman Period from Cave VIII/28.” ʿAtiqot 41/2 (2002): 147–55. Lamon, Robert S., and Geoffrey M. Shipton. Megiddo I: Seasons of 1925–34; Strata I–V. Oriental Institute Publication 42. Chicago: University of Chicago Press, 1939. Lange, Armin. “The Qumran Dead Sea Scrolls—Library or Manuscript Corpus?” In Florentino García Martínez, Annette Steudel, and Eibert J. C. Tigchelaar, eds. From 4QMMT to Resurrection: Mélanges qumraniens en hommage à Émile Puech. STDJ 61. Leiden: Brill, 2006, 177–93. ———, and Ulrike Mittmann-Richert. “Annotated List of the Texts from the Judaean Desert Classified by Content and Genre.” In Tov. The Texts from the Judaean Desert, 115–64. Laperrousaz, Ernest-Marie. Qoumrân: L’établissement essénien des bords de la Mer Morte; Histoire et archéologie du site. Paris: A. & J. Picard, 1976. ———. “La grotte 3 de Qoumrân et le ‘Rouleau de cuivre’: L’origine et la nature du ‘Rouleau de cuivre’.” In Laperrousaz, ed. Qoumrân et les manuscrits de la mer Morte: Un cinquantenaire. Paris: Éditions du Cerf, 1997, 207–13. Lapp, Nancy L., and George W. E. Nickelsburg Jr. “The Roman Occupation and Pottery of ʿArâq en-Naʿsaneh (Cave II).” In Lapp, and Lapp. Discoveries in the Wâdī ed-Dâliyeh, 49–54. Lapp, Paul W. Palestinian Ceramic Chronology 200 B.C.-A.D. 70. New Haven: American Schools of Oriental Research, 1961. ———. “Wâdī ed-Dâliyeh.” RB 72 (1965): 405–9. ———, and Nancy L. Lapp, eds. Discoveries in the Wâdī ed-Dâliyeh. AASOR 41. Cambridge, MA: American Schools of Oriental Research, 1974. Lawrenz, John C. “The Jewish Coins.” In Arlotte D. Tushingham. Excavations in Jerusalem 1961–1967. Vol. 1. Toronto: Royal Ontario Museum, 1985, 156–66. Le Moyne, Jean. Les Sadducéens. Paris: Librairie Lecoffre, 1972. Le Rider, Georges. “Henri Seyrig 1895–1973.” RSN 52 (1973): 167–71. Lefkovits, Judah K. The Copper Scroll—3Q15: A Reevaluation. STDJ 25. Leiden: Brill, 1999. Leith, Mary J. W. Wadi Daliyeh I: The Wadi Daliyeh Seal Impressions. DJD 24. Oxford: Clarendon Press, 1997. Lemaire, André. “Writing and Writing Materials.” ABD Vol. 6: 999–1008. ———. “Inscriptions du khirbeh, des grottes et de ʿAïn Feshkha.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol. 2: 341–88. Leonard, Robert D. “Numismatic Evidence for the Dating of Qumran: A Review Article.” QC 7 (1997): 225–34. Levy, Brooks. “Later Tyrian Shekels: Dating the Crude Issues; Reading the Controls.” In Carmen Alfaro, Carmen Marcos, and Paloma Otero, eds. XIII Congreso Internacional de

BIBLIOGRAPHY Numismatica, Madrid 2003: Actas-Proceedings-Actes. Vol. 1. Madrid: Ministerio de Cultura, 2005, 885–90. Libby, Willard F. “Radiocarbon Dates, II.” Science 114 (1951): 291–296. Lieberman, Saul. Tosefeth Rishonim: A Commentary. Vol. 3. Jerusalem: Bamberger & Wahrmann, 1939. Liphschitz, Nili. “Timber Analysis of Household Objects in Israel: A Comparative Study.” IEJ 48 (1998): 77–90. Loffreda, Stanislao. La Ceramica di Macheronte e dell’Herodion (90 a.C.–135 d.C.). SBF.CMa 39. Jerusalem: Franciscan Printing Press, 1996. London, Gloria A. “On Fig Leaves, Itinerant Potters, and Pottery Production Locations in Cyprus.” In Patrick E. McGovern, Michael D. Notis, and W. David Kingery, eds. Cross-craft and Cross-cultural Interactions in Ceramics. Westerville, OH: American Ceramic Society, 1989, 65–80. Lönnqvist, Kenneth. The Report of the Amman Lots of the Qumran Silver Coin Hoards: New Chronological Aspects of the Silver Coin Hoard Evidence from Khirbet Qumran at the Dead Sea. Amman: National Press, 2007. ———, and Minna Lönnqvist. “The Numismatic Chronology of Qumran: Fact and Fiction.” NC 166 (2006): 121–65. Luria, Ben Zion. The Copper Scroll from the Judaean Desert. PIBRS 14. Jerusalem: Kiryat Sepher, 1963 [Hebrew]. Ma⁠ʾayeh, Farah S. “Recent Archaeological Discoveries in Jordan.” ADAJ 4–5 (1960): 114–16. Macdonald, Michael. “A Bibliography of Gerald Lankester Harding.” ADAJ 24 (1980): 8–12. Magen, Yitzhak. The Stone Vessel Industry in the Second Temple Period: Excavations at Hizma and the Jerusalem Temple Mount. JSP 1. Jerusalem: Israel Exploration Society/Israel Antiquities Authority/Staff Officer of Archaeology—Civil Administration of Judea and Samaria, 2002. ———. “Qalandiya—A Second Temple-Period Viticulture and Wine-Manufacturing Agricultural Settlement.” In Magen, et al. The Land of Benjamin. JSP 3. Jerusalem: Staff Officer of Archaeology, Civil Administration of Judea and Samaria/ Israel Antiquities Authority, 2004, 29–144. ———, and Yuval Peleg. “Back to Qumran: Ten Years of Excavation and Research, 1993–2004.” In Galor, Humbert, and Zangenberg. Qumran, the Site of the Dead Sea Scroll, 55–113. ———, and Yuval Peleg. The Qumran Excavations 1993–2004: Preliminary Report. JSP 6. Jerusalem: Staff Officer of Archaeology, Civil Administration of Judea and Samaria/ Israel Antiquities Authority, 2007. Maggen, Michael. “The Conservation of Papyri Documents dated to the 4th Century BCE.” Restaurator 18/3 (1997): 153–61. Magness, Jodi. “The Community at Qumran in Light of Its Pottery.” In Wise, et al. Methods of Investigation of the Dead Sea Scrolls, 39–50. ———. “The Chronology of the Settlement at Qumran in the Herodian Period.” DSD 2 (1995): 58–65.

BIBLIOGRAPHY ———. “The Chronology of Qumran, Ein Feshkha, and Ein El-Ghuweir.” In Kapera. Mogilany 1995, 55–76. ———. The Archaeology of Qumran and the Dead Sea Scrolls. Grand Rapids, MI: Eerdmans, 2002. ———. “Women at Qumran?” In Leonard V. Rutgers, ed. What Athens Has to Do with Jerusalem: Essays on Classical, Jewish, and Early Christian Art and Archaeology in honor of Gideon Foerster. ISACR 1. Leuven: Peeters, 2002, 89–123. Reprinted in Magness. Debating Qumran, 113–49. ———. Debating Qumran: Collected Essays on Its Archaeology. ISACR 4. Leuven: Peeters, 2004. ———. “Why Scroll Jars?” In Edwards. Religion and Society in Roman Palestine, 146–61. ———. “The Pottery from the 1995 Excavations in Camp F at Masada.” BASOR 353 (2009): 75–107. ———. Stone and Dung, Oil and Spit: Jewish Daily Life in the Time of Jesus. Grand Rapids, MI: Eerdmans, 2011. Maier, Johann. “Tanḥumin and Apocryphal Lamentations.” EDSS Vol. 2: 915. Mason, Eric F., et al., eds. A Teacher for All Generations: Essays in Honor of James C. VanderKam. JSJSup 153. Vol. 1. Leiden: Brill, 2011. Mathews, Kenneth A. “The Background of the Paleo-Hebrew Texts at Qumran.” In Carol L. Meyers, and Michael O’Connor, eds. The Word of the Lord Shall Go Forth: Essays in Honor of David Noel Freedman in Celebration of His Sixtieth Birthday. Winona Lake, IN: Eisenbrauns, 1983, 549–68. Mazar, Amihai, and Nava Panitz-Cohen. Timnah (Tel Batash) II: The Finds from the First Millenium BCE. Qedem 42. Jerusalem: Institute of Archaeology, Hebrew University, 2001. Mazar, Benjamin, Moshe Dothan, and Immanuel Dunayevsky. “En-Gedi: The First and Second Seasons of Excavations, 1961– 1962.” ʿAtiqot 5 (1966). Mazar, Eilat, and Benjamin Mazar. Excavations in the South of the Temple Mount: The Ophel of Biblical Jerusalem. Qedem 29. Jerusalem: Institute of Archaeology, Hebrew University, 1989. McLaren, James S. “Going to War against Rome: The Motivation of the Jewish Rebels.” In Mladen Popović, ed. The Jewish Revolt against Rome: Interdisciplinary Perspectives. JSJSup 153. Leiden: Brill, 2011, 129–53. McLean, Mark D. The Use and Development of Palaeo-Hebrew in the Hellenistic and Roman Periods. Unpubl. Ph.D. diss., Harvard University, 1982. McNamee, Kathleen. Sigla and Select Marginalia in Greek Literary Papyri. PapyBrux 26. Brussels: Fondation Égyptologique Reine Élisabeth, 1992. Meshorer, Ya⁠ʾakov. “The En-Gedi Hoard.” In Herbert A. Cahn, and Georges Le Rider, eds. Actes du 8ème Congrès International de Numismatique. Vol. 1. Paris: Association Internationale des Numismates Professionnels, 1976, 111–12.

343 ———. Sylloge Nummorum Graecorum: The Collection of the American Numismatic Society; Part 6; Palestine-South Arabia. New York: American Numismatic Society, 1981. ———. “The Coins of Masada.” In Yigael Yadin, and Joseph Naveh, eds. Masada I: The Yigael Yadin Excavations 1963– 1965; Final Report; The Aramaic and Hebrew Ostraca and Jar Inscriptions. Jerusalem: Israel Exploration Society/Hebrew University, 1989, 69–132. ———. “The Coins from Qumran.” INJ 15 (2006): 19–23. ———, Gabriela Bijovsky, and Wolfgang Fischer-Bossert, eds. Coins of the Holy Land: The Abraham and Marian Sofaer Collection at the American Numismatic Society and the Israel Museum. Ancient Coins in North American Collections 8. 2 vols. New York: American Numismatic Society, 2012. Metso, Sarianna. “The Primary Results of the Reconstruction of 4QSe.” JJS 44 (1993): 303–8. ———. The Textual Development of the Qumran Community Rule. STDJ 21. Leiden: Brill, 1997. ———. “In Search of the Sitz im Leben of the Community Rule.” In Parry, and Ulrich. The Provo International Conference on the Dead Sea Scrolls, 306–15. ———. “Whom does the Term Yaḥad identify?” In Charlotte Hempel, and Judith M. Lieu, eds. Biblical Traditions in Transmission: Essays in Honour of Michael A. Knibb. JSJSup 111. Leiden: Brill, 2006, 213–35. Michniewicz, Jacek. Qumran and Jericho Pottery: A Petrographic and Chemical Provenance Study. Seria Geologia 20. Poznan: Wydawnictwo Naukowe UAM, 2009. ———, and Miroslaw Krzyśko. “The Provenance of Scroll Jars in the Light of Archaeometric Investigations.” In Humbert, and Gunneweg. Khirbet Qumrân et ʿAïn Feshkha, Vol 2: 59–99. Mildenberg, Leo, Ulrich Hübner, and Ernst Axel Knauf, eds. Vestigia Leonis: Studien zur antiken Numismatik Israels, Palästinas und der östlichen Mittelmeerwelt. NTOA 36. Fribourg/ Göttingen: Universitätsverlag/Vandehoeck & Ruprecht, 1998. Milik, Józef T. “Le travail d’édition des fragments manuscripts de Qumrân.” RB 63 (1956): 60–62. ———. Ten Years of Discovery in the Wilderness of Judaea. SBT 26. London: SCM Press, 1959. ———. “Textes hébreux et araméens.” In Benoit, Milik, and de Vaux. Les grottes de Murabbaʿât, 65–205. ———. “Deux jarres inscrites provenant d’une grotte de Qumrân.” In Baillet, Milik, and de Vaux. Les ‘petites grottes’ de Qumrân, 37–41. ———. “Le rouleau de cuivre provenant de la grotte 3Q (3Q15).” In Baillet, Milik, and de Vaux. Les ‘petites grottes’ de Qumrân, 211–302. ———. “Texte de la Grotte 5Q.” In Baillet, Milik, and de Vaux. Les ‘petites grottes’ de Qumrân, 167–97. ———. The Books of Enoch: Aramaic Fragments of Qumrân Cave 4. Oxford: Clarendon Press, 1976.

344 ———. “Tefillin, Mezuzot et Targums (4Q128–4Q157).” In de Vaux, and Milik. Qumrân Grotte 4.II, 33–92. ———, and Henri Seyrig. “Trésor monétaire de Murabbaʿât.” RN 6/1 (1958): 11–26. Millard, Alan. Reading and Writing in the Time of Jesus. New York: New York University Press, 2000. Misgav, Haggai. “The Ostraca.” In Győző Vörös. Machaerus. Vol. 1, History, Archaeology and Architecture of the Fortified Herodian Royal Palace and City Overlooking the Dead Sea in Transjordan; Final Report of the Excavations and Surveys, 1807–2012. SBF.CMa 53. Milano: Edizioni Terra Santa, 2013, 259–72. Mitchell-Innes, Alfred. “What is Money?” Banking Law Journal 30/5 (1913): 377–408. ———. “The Credit Theory of Money.” Banking Law Journal 31/1(1914): 151–68. Mizzi, Dennis. The Archaeology of Khirbet Qumran: A Comparative Approach. Ph.D. diss., University of Oxford, 2009. ———. “The Glass from Khirbet Qumran: What does it tell us about the Qumran Community?” In Hempel. The Dead Sea Scrolls, 99–198. Młynarczyk, Jolanta. “Terracotta Oil Lamps from Qumran: The Typology.” RB 120 (2013): 99–133. Mommsen, Hans. “Short Note: Provenancing of Pottery—The Need for an Integrated Approach?” Archaeometry 46/2 (2004): 267–71. Morgenstern, Matthew, and Michael Segal. “6. XḤev/ SePhylactery.” In Charlesworth, et al. Miscellaneous Texts from the Judean Desert, 183–91. Mumcuoglu, Kosta Y. “Human Parasites from Qumran and the Surrounding Regions in Israel.” In Gunneweg, Greenblatt, and Adriaens. Bio- and Material Cultures at Qumran, 57–61. Murphy, Catherine M. Wealth in the Dead Sea Scrolls and in the Qumran Community. STDJ 40. Leiden: Brill, 2001. Nahmias, Keren, and Tzahi Gal. “Modiʿin.” ESI 111 (2000): 63–64 [Hebrew]; 49*–50*. Nakman, David. “The Contents and Order of the Biblical Sections in the Tefillin from Qumran and Rabbinic Halakhah: Similarity, Difference, and Some Historical Conclusions.” Cathedra 112 (2004): 19–44 [Hebrew]. Naveh, Joseph. Early History of the Alphabet: An Introduction to West Semitic Epigraphy and Palaeography. 2nd ed. Jerusalem: Magnes Press, 1987. ———. “341. 4QExercitium Calami C.” In Pfann. Qumran Cave 4.XXVI: Cryptic Texts; Alexander, et al. Miscellanea, Part 1, 291–93. Netzer, Ehud, Masada III: The Yigael Yadin Excavations 1963–1965; Final Reports; The Buildings, Stratigraphy and Architecture. Jerusalem: Israel Exploration Society/Hebrew University, 1991.

BIBLIOGRAPHY Newell, Edward T. The Coinage of the Eastern Seleucid Mints: From Seleucus I to Antiochus III. Numismatic Studies 1. New York: American Numismatic Society, 1938. ———. The Coinage of the Western Seleucid Mints: From Seleucus I to Antiochus III. Numismatic Studies 4. New York: American Numismatic Society, 1941. Newman, Judith H. “Speech and Spirit: Paul and the Maskil as Inspired Interpreters of Scripture.” In Jörg Frey, and John R. Levison, eds. The Holy Spirit, Inspiration, and the Cultures of Antiquity: Multidisciplinary Perspectives. Ekstasis 5. Berlin: W. de Gruyter, 2014, 243–66. Newsom, Carol. “Apocryphon of Joshua.” In George J. Brooke, et al., eds. Qumran Cave 4.XVII: Parabiblical Texts, Part 3. DJD 22. Oxford: Clarendon Press, 1996, 237–88. Nickelsburg, George W. E., Jr. “Other Finds: Miscellaneous Small Finds.” In Lapp, and Lapp. Discoveries in the Wâdī ed-Dâliyeh, 101–2. North, Robert. “The Damascus of Qumran Geography.” PEQ 87 (1954): 34–48. Nys, Karin. “Terre cuite.” In Sabine Fourrier, and Antoine Hermary, eds. Amathonte VI: Le sanctuaire d’Aphrodite des origines au début de l’époque imperial. Études chypriotes 17. Athènes/Paris: École française/De Boccard, 2006, 93. Oren, Eliezer D. The Northern Cemetery of Beth Shan. Leiden: Brill, 1973. ———, and Uriel Rappaport. “The Necropolis of Maresha— Beth Govrin.” IEJ 34 (1984): 114–53. Oziol, Therese. Les lampes du musée de Chypre: Salamine de Chypre VII. Paris: De Boccard, 1977. Parry, Donald W., and Eugene Ulrich, eds. The Provo Inter­ national Conference on the Dead Sea Scrolls: Technological Innovations, New Texts, and Reformulated Issues. STDJ 30. Leiden: Brill, 1998. Patrich, Joseph. “Caves of Refuge and Jewish Inscriptions on the Cliffs of Naḥal Michmas.” EI 18 (1985): 153–66 [Hebrew]. ———. “Hideouts in the Judean Wilderness: Jewish Revolutionaries and Christian Ascetics Sought Shelter and Protection in Cliffside Caves.” BAR 15/5 (1989): 32–42. ———. “Khirbet Qumran in Light of New Archaeological Explorations in the Qumran Caves.” In Wise, et al. Methods of Investigation of the Dead Sea Scrolls, 73–95. ———. “Qumran Caves.” ESI 13 (1995), 64. ———. “Did Extra-Mural Dwelling Quarters Exist at Qumran?” In Schiffman, Tov, and VanderKam. The Dead Sea Scrolls Fifty Years After their Discovery, 720–27. ———, and Benny Arubas. “A Juglet containing Balsam Oil (?) from a Cave near Qumran.” IEJ 39 (1989): 43–59. ———, Benny Arubas, and Benny Agur. “Monastic Cells in the Desert of Gerasimus near the Jordan.” In Frédéric Manns, and Eugenio Alliata, eds. Early Christianity in Context: Monuments

BIBLIOGRAPHY and Documents. SBF.CMa 38. Jerusalem: Franciscan Printing Press, 1993, 277–96. ———, Benny Arubas, and Eyal Naor. “Jewish Caves of Refuge in the Cliffs of Naḥal Mikhmas.” Qadmoniot 19 (1986): 45–50 [Hebrew]. Paul, Shalom M., et al., eds. Emanuel: Studies in Hebrew Bible, Septuagint and Dead Sea Scrolls in Honor of Emanuel Tov. VTSup 94. 2 vols. Leiden: Brill, 2003. Peacock, David P.S., and David F. Williams. Amphorae and the Roman Economy: An Introductory Guide. London: Longman, 1982. Peskowitz, Miriam B. Spinning Fantasies: Rabbis, Gender and History. Berkeley: University of California Press, 1997. Pfann, Stephen J. “4Q298: The Maskîl’s Address to All Sons of Dawn.” JQR 85 (1994): 203–35. ———. “Sites in the Judaean Desert Where Texts Have Been Found.” In Emanuel Tov, and Stephen J. Pfann, eds. Companion Volume to the Dead Sea Scrolls Microfiche edition. 2nd revised edition. Leiden: Brill, 1995, 109–19. ———, ed. Qumran Cave 4.XXVI: Cryptic Texts; Philip S. Alexander, et al., eds. Miscellanea, Part 1. DJD 36. Oxford: Clarendon Press, 2000. ———. “The Writings in Esoteric Script from Qumran.” In Schiffman, Tov, and VanderKam. The Dead Sea Scrolls Fifty Years After their Discovery, 177–90. ———. “Reassessing the Judean Desert Caves: Libraries, Archives, Genizas and Hiding Places.” BAIAS 25 (2007): 147–70. ———. “The Mount Zion Inscribed Stone Cup: Preliminary Observations.” In David Amit, Orit Peleg-Barkat, and Guy D. Stiebel, eds. New Studies in the Archaeology of Jerusalem and Its Region: Collected Papers. Vol. 4. Jerusalem: Israel Antiquities Authority, 2010, 44*–53*. Plenderleith, Harold J. “Technical Note on unwrapping of Dead Sea Scroll Fragments.” In Barthélemy, and Milik. Qumran Cave 1, 39–40. Politis, Konstantinos D. “The Discovery and Excavations of the Khirbet Qazone Cemetery and its Significance Relative to Qumran.” In Galor, Humbert, and Zangenberg. Qumran, the Site of the Dead Sea Scrolls, 213–21. Poole, John B. The Nature, Origins and Techniques of Manufacture of those of the Dead Sea Scrolls which are Made from Animal Skins. Ph.D. diss., University of Leeds, 1959. Popović, Mladen. “Roman Book Destruction in Qumran Cave 4 and the Roman Destruction of Khirbet Qumran Revisited.” In Frey, Claussen, and Kessler. Qumran und die Archäologie, 239–91. ———. “Qumran as Scroll Storehouse in Times of Crisis? A Comparative Perspective on Judaean Desert Manuscript Collections.” JSJ 43 (2012): 551–94.

345 ———. “The Ancient ‘Library’ of Qumran between Urban and Rural Culture.” In Sidnie White Crawford, and Cecilia Wassén, eds. The Dead Scrolls at Qumran and the Concept of a Library. STDJ 116. Leiden: Brill, 2015, 155–167. ———. “The Manuscript Collections: An Overview.” In George Brooke, and Charlotte Hempel, eds. T&T Clark Companion to the Dead Sea Scrolls. London: T&T Clark, forthcoming. Porat, Roi, Hanan Eshel, and Amos Frumkin. “Finds from the Bar Kokhba Revolt from Two Caves at En Gedi.” PEQ 139 (2007): 35–53. ———, Hanan Eshel, and Amos Frumkin. “The ‘Caves of the Spear’: Refuge Caves from the Bar-Kokhba Revolt North of ‘En-Gedi.” IEJ 59 (2009): 21–46. ———, Hanan Eshel, and Amos Frumkin. “The Christmas Cave in the Lower Part of Naḥal Kidron.” In Hanan Eshel, and Roi Porat, eds. Refuge Caves of the Bar Kokhba Revolt. Vol. 2. Jerusalem: Israel Exploration Society, 2009, 31–52 [Hebrew]. Price, Martin J., ed. Coin Hoards. Vol. 1. London: Royal Numismatic Society, 1975. Puech, Emile. “Abécédaire et liste alfabétique de noms hébreux du début du IIe s. A.D.ˮ RB 87 (1980): 118–26. ———. “Inscriptions funéraires palestiniennes: tombeau de Jason et ossuaires.” RB 90 (1983): 481–533. ———. “In memoriam. L’Abbé Jean Starcky (1909–1988).” RevQ 53 (1989): 3–6. ———. “Notes en marge de 11QPaléoLévitique: Le fragment L, des fragments inédits et une jarre de la grotte 11.” RB 96 (1989): 161–83. ———. “Sept fragments grecs de la lettre d’Hénoch (1 Hén 100, 103, 105) dans la grotte 7 de Qumrân (= 7QHén gr).” RevQ 70 (1997): 313–23. ———. “521. 4QApocalypse messianique.” In Puech. Qumrân Grotte 4.XVIII, 1–38. ———. “524. 4QRouleau du Temple.” In Puech. Qumrân Grotte 4.XVIII, 85–114. ———. Qumrân Grotte 4.XVIII: Textes hébreux (4Q521–4Q528, 4Q576–4Q579). DJD 25. Oxford: Clarendon Press, 1997. ———. “L’alphabet cryptique A en 4QSe (4Q259).” RevQ 71 (1998): 429–35. ———. “Les langues et les écritures dans les manuscrits de la mer Morte.” In Rina Viers, ed. Des signes pictographiques à l’alphabet: La communication écrite en Méditerranée; Actes du colloque, 14–15 mai 1996, Villa grecque Kérylos, Fondation Théodore Reinach (Beaulieu-sur-Mer). Paris/Nice: Éditions Karthala/Association Alphabets, 2000, 175–211. ———. “Note additionnelle sur le fragment en paléo-hébreu.” RevQ 75 (2000): 449–51. ———. Qumrân Grotte 4.XXII: Textes araméens; Première partie 4Q529–549. DJD 31. Oxford: Clarendon Press, 2001.

346 ———. “Apports des manuscrits de Qumrân à la croyance à la résurrection dans le Judaïsme ancien.ˮ In André Lemaire, and Simon C. Mimouni, eds. Qoumrân et le Judaïsme du tournant de notre ère: Actes de la Table Ronde, Collège de France, 16 novembre 2004. CREJ 40. Leuven: Peeters, 2006, 81–110. ———. “L’ostracon de Khirbet Qumrân (KhQ 1996/1) et une vente de terrain à Jéricho, témoin de l’occupation essénienne à Qumrân.” In Anthony Hilhorst, Emile Puech, and Eibert J. C. Tigchelaar, eds. Flores Florentino: Dead Sea Scrolls and Other Early Jewish Studies in Honour of Florentino García Martínez. JSJSup 122. Leiden: Brill, 2007, 1–29. ———. “L’épigraphie de Qumrân: Son apport à l’identification du site.” RevQ 95 (2010): 433–40. ———. “L’épilogue de 4QMMT revisité.” In Mason, et al. A Teacher for All Generations, Vol. 1: 309–39. ———. “La constitution de groupements de livres normatifs à Qumrân.” RB 119 (2012): 45–57. ———. “4Q225 revisité: Un midrash essénien?ˮ RevQ 102 (2013): 169–209. ———. “La Lettre essénienne MMT dans le manuscrit 4Q397 et les parallèles.ˮ RevQ 105 (2015): 99–135. ———. The Copper Scroll Revisited. STDJ 112. Leiden: Brill, 2015. ———. “Les copies du livre de Josué dans les manuscrits de la mer Morte: 4Q47, 4Q48, 4Q123 et XJosué.ˮ RB 122 (2015): 480–506. Qimron, Elisha. The Dead Sea Scrolls: The Hebrew Writings. Vol. 1. Jerusalem: Yad Ben Zvi, 2010. Rabin, Ira. “Archaeometry of the Dead Sea Scrolls.” DSD 20 (2013): 124–42. ———, et al. “On the Origin of the Ink of the Thanksgiving Scroll (1QHodayota).” DSD 16 (2009): 97–106. Rahmani, Leir Y. “The Coins from the Cave of Horror.” IEJ 12 (1962): 200. ———. “The Coins.” ʿAtiqot 5 (1966): 51–52. Rasmussen, Kaare L., et al. “Cleaning and Radiocarbon Dating of Material from Khirbet Qumran.” In Gunneweg, Greenblatt, and Adriaens. Bio- and Material Cultures at Qumran, 139–64. ———, et al. “The Effects of Possible Contamination on the Radiocarbon Dating of the Dead Sea Scrolls I: Castor Oil.” Radiocarbon 43 (2001): 127–32. ———, et al. “Reply to Israel Carmi (2002): ‘Are the 14C Dates of the Dead Sea Scrolls Affected by Castor Oil Contamination?’ ” Radiocarbon 45 (2003): 497–99. ———, et al. “The Effects of Possible Contamination on the Radiocarbon Dating of the Dead Sea Scrolls II: Empirical Methods to Remove Castor Oil and Suggestions for Redating.” Radiocarbon 51 (2009): 1005–22. Reece, Richard, et al. “Jerusalem: the Coins.” In Kay Prag, ed. Excavations by K.M. Kenyon in Jerusalem 1961–1967. Vol. 5,

BIBLIOGRAPHY Discoveries in Hellenistic to Ottoman Jerusalem; Centenary Kathleen M. Kenyon 1906–1978. LevantSup 7. Oxford: Oxbow Books, 2008, 411–31. Reed, Ronald. Ancient Skins, Parchments and Leathers. London: Seminar Press, 1972. ———, and John B. Poole. “A Study of Some Dead Sea Scroll and Leather Fragments from Cave 4 at Qumran: Part I—Physical Examination.” Proceedings of the Leeds Philosophical and Literary Society (Scientific Section) 9/1 (1962): 1–13. Reed, Stephen A. “Find-Sites of the Dead Sea Scrolls.” DSD 14 (2007): 199–221. Reed, William L. “The Qumran Caves Expedition of March 1952.” BASOR 135 (1954): 8–13. Regev, Eyal. “Between Two Sects: Differentiating the Yaḥad and the Damascus Covenant.” In Hempel. The Dead Sea Scrolls, 431–49. Reich, Ronny. “A Note on the Function of Room 30 (the ‘Scriptorium’) at Khirbet Qumran.” JJS 46 (1995): 157–60. ———. “Women and Men at Masada: Some Anthropological Observations based on the Small Finds (Coins, Spindles).” ZDPV 117 (2001): 149–63. Reifenberg, Adolf. Ancient Jewish Coins. 2nd revised edition. Jerusalem: Rubin Mass, 1947. Rofé, Alexander. “Further Manuscript Fragments of the Jubilees in the Third Cave of Qumran.” Tarbiz 34 (1965): 333–36 [Hebrew]. Rohrhirsch, Ferdinand. Wissenschaftstheorie und Qumran: Die Geltungsbegründungen von Aussagen in der biblischen Archäologie am Beispiel von Chirbet Qumran und En Feschcha. NTOA 32. Fribourg/Göttingen: Universitätsverlag/ Vandenhoeck & Ruprecht, 1996. ———. “Die Geltungsbegründungen der Industrie-RollenTheorie zu Chirbet Qumran und En Feschcha auf dem methodologischen Prüfstand: Relativierung und Widerlegung.” DSD 6 (1999): 267–81. Rosenthal-Heginbottom, Renate. Römische Bildlampen aus Östlichen Werkstätten. Wiesbaden: Harrasowitz, 1981. ———. “Imported Hellenistic and Roman Pottery.” In Ephraim Stern, ed. Excavations at Dor: Final Report. Vol. 1B, Areas A and C; The Finds. QR 2. Jerusalem: Institute of Archaeology, Hebrew University, 1995, 183–288. Rothstein, David. From Bible to Murabbaʿat: Studies in the Literary, Textual and Scribal Features of Phylacteries and Mezuzot in Ancient Israel and Early Judaism. Ph.D. diss., University of California, Los Angeles, 1992. Russell, Kenneth W. “The Earthquake Chronology of Palestine and Northwest Arabia from the 2nd through the Mid-8th Century A. D.” BASOR 260 (1985): 37–59. Safrai, Shmuel. “The Synagogue.” In Safrai, and Menaḥem Stern, eds. The Jewish People in the First Century: Historical Geography, Political History, Social, Cultural and Religious Life

BIBLIOGRAPHY and Institutions. CRINT 1. Vol. 2. Assen: Van Gorcum, 1976, 908–44. Sagiv, Nahum. “Tel Goded.” ESI 14 (1995): 112–14. Sanders, James A. The Psalms Scroll of Qumrân Cave 11 (11QPsa). DJD 4. Oxford: Clarendon Press, 1965. Schäfer, Peter, ed. The Bar Kokhba War Reconsidered: New Perspectives on the Second Jewish Revolt against Rome. TSAJ 100. Tübingen: Mohr Siebeck, 2003. Schick, Tamar. The Cave of the Warrior: A Fourth Millennium Burial in the Judean Desert. IAA Report 5. Jerusalem: Israel Antiquities Authority, 1998. ———. “The Early Basketry and Textiles from Caves in the Northern Judaean Desert.” ʿAtiqot 41/2 (2002): 223–39. Schiffman, Lawrence H. Review of Qumrân Grotte 4.II, by Roland de Vaux and Józef T. Milik. JAOS 100 (1980): 170–72. ———. Reclaiming the Dead Sea Scrolls: The History of Judaism, the Background to Christianity, the Lost Library of Qumran. Philadelphia, PA: Jewish Publication Society, 1994. ———, Emanuel Tov, and James C. VanderKam, eds. The Dead Sea Scrolls Fifty Years After their Discovery: Proceedings of the Jerusalem Congress, July 20–25, 1997. Jerusalem: Israel Exploration Society/Shrine of the Book, 2000. Schofield, Alison. “Between Center and Periphery: The Yaḥad in Context.” DSD 16 (2009): 330–50. ———. From Qumran to the Yaḥad: A New Paradigm of Textual Development for ‘The Community Rule’. STDJ 77. Leiden: Brill, 2008. Schuller, Eileen. “433a. 4QpapHodayot-like Text B.” In Esther G. Chazon, et al., eds. Qumran Cave 4.XX: Poetical and Liturgical Texts, Part 2. DJD 29. Oxford: Clarendon Press, 1999, 237–45. ———, and Moshe Bernstein. “372. 4QNarrative and Poetic Compositionb.” In Gropp. Wady Daliyeh II; Bernstein et al., Qumran Cave 4.XXVIII: Miscellanea, Part 2, 165–97. ———, and Moshe Bernstein. “373. 4QNarrative and Poetic Compositionc.” In Gropp. Wady Daliyeh II; Bernstein, et al. Qumran Cave 4.XXVIII: Miscellanea, Part 2, 199–204. Sellers, Ovid R. “Radiocarbon Dating of Cloths from the ʿAïn Feshkha Cave.” BASOR 123 (1951): 24–6. Shahar, Yuval. “The Underground Hideouts in Galilee and Their Historical Meaning.” In Schäfer. The Bar Kokhba War Reconsidered, 217–40. Shamir, Orit. “Textiles and Garments from Qumran: Chalcolithic and Roman Period.” In Gunneweg, Greenblatt, and Adriaens. Bio- and Material Cultures at Qumran, 285–96. ———, and Alisa Baginski. “The Later Textiles, Basketry and Cordage from Caves in the Judaean Desert.” ʿAtiqot 41/2 (2002): 241–55. ———, and Naama Sukenik. “Qumran Textiles and the Garments of Qumran’s Inhabitants.” DSD 18 (2011): 206–25. ———, and Naama Sukenik. “Qumran Textiles and the Clothing of the Inhabitants of Qumran.” In Gunneweg and Greenblatt.

347 Outdoor Qumran, http://micro5.mscc.huji.ac.il/~msjan/orit .pdf. Shanks, Hershel, ed. Frank Moore Cross: Conversations with a Bible Scholar. Washington DC: Biblical Archaeology Society, 1994. Sharabani, Marcia. “Monnaies de Qumrân au Musée Rockefeller de Jérusalem.” RB 87 (1980): 274–84. Sillar, Bill. Shaping Culture: Making Pots and Constructing Households; An Ethnoarchaeological Study of Pottery Production, Trade and Use in the Andes. BAR.IS 883. Oxford: Archaeopress, 2000. Singer-Avitz, Lily. “Arad: The Iron Age Pottery Assemblages.” Tel Aviv 29 (2002): 110–214. Sion, Ofer. “Regions IV and VI: Survey and Excavations of Caves along the Jebel Abu Saraj Cliff.” ʿAtiqot 41/1 (2002): 45–89 [Hebrew], ʿAtiqot 41/2 (2002): 43–70. Sivan, Renee. Ancient Lamps in the Schloessinger Collection. Qedem 8. Jerusalem: Hebrew University, 1978. Smith, Robert H. “The ‘Herodian’ Lamp of Palestine: Types and Dates.ˮ Berytus 14 (1963): 53–65. ———. “The Household Lamps of Palestine in Intertestamental Times.” BA 27 (1964): 101–24. ———. “The Household Lamps of Palestine in New Testament Times.” BA 29 (1966): 1–27. Snyder, Gregory H. “Naughts and Crosses: Pesher Manuscripts and Their Significance for Reading Practices at Qumran.” DSD 7 (2000): 26–48. Augustus Spijkerman and Jean Starcky, “Un nouveau lot de monnaies palestiniennes,” RB 65 (1958): 568–84. Stacey, David. “Seasonal Industries at Qumran.” BAIAS 26 (2008): 7–29. ———. “A Reassessment of the Stratigraphy of Qumran.” In Stacey, and Doudna. Qumran Revisited, 7–74. ———, and Gregory L. Doudna. Qumran Revisited: A Reassessment of the Archaeology of the Site and its Texts. BAR. IS 2520. Oxford: Archaeopress, 2013. Stager, Lawrence E. “Ancient Agriculture in the Judean Desert: A Case Study of the Buqêʾah Valley in the Iron Age.” Ph.D. diss., Harvard University, 1975. ———. “Farming in the Judean Desert during the Iron Age.” BASOR 221 (1976): 145–58. Stegemann, Hartmut. Die Essener, Qumran, Johannes der Täufer und Jesus: Ein Sachbuch. 4th edition. Freibourg/Basel/Wien: Herder, 1994; ———. The Library of Qumran: On the Essenes, Qumran, John the Baptist, and Jesus. Grand Rapids, MI: Eerdmans, 1998. ———, and Jürgen Becker. “Zum Text von Fragment 5 aus Wadi Murabbaʿat.” RevQ 11 (1962): 443–48. ———. “The Wadi Murabbaʿat Caves.” NEAEHL Vol. 3:833–35. Stern, Ephraim. “Murabbaʿat, Wadi: Archaeology.” EDSS Vol. 1: 581–83.

348 ———, ed. En-Gedi Excavations I: Final Report (1961–1965). Jerusalem: Israel Exploration Society/Institute of Archaeology, Hebrew University, 2007. Stern, Sacha. “Qumran Calendars and Sectarianism.” In Timothy H. Lim, and John J. Collins, eds. The Oxford Handbook of the Dead Sea Scrolls. Oxford: Oxford University Press, 2010, 232–53. Steudel, Annette. Der Midrasch zur Eschatologie aus der Qumrangemeinde (4QMidrEschata.b): Materielle Rekonstruktion, Textbestand, Gattung, traditionsgeschichtliche Einordnung des durch 4Q174 (“Florilegium”) und 4Q177 (“Catena A”) repräsentierten Werkes aus den Qumranfunden. STDJ 13. Leiden: Brill, 1994. Stiebel, Guy. “Armis et Litteris: The Military Equipment of Early Roman Palestine in Light of the Archaeological and Historical Sources.” Ph.D. diss., University of London, 2007. Stökl Ben Ezra, Daniel. “Old Caves and Young Caves: A Statistical Reevaluation of a Qumran Consensus.” DSD 14 (2007): 313–33. ———. “Further Reflections on Cave 1 and 11: A Response to Florentino García Martínez.” In Hempel. The Dead Sea Scrolls, 211–23. ———. “Wie viele Bibliotheken gab es in Qumran?” In Frey, Claussen, and Kessler. Qumran und die Archäologie, 327–46. ———. Lehrbuch Qumran. Forthcoming. Strange, James F. “The 1996 Excavations at Qumran and the Context of the New Hebrew Ostracon.ˮ In Galor, Humbert, and Zangenberg. Qumran, the Site of the Dead Sea Scrolls, 41–54. Strugnell, John. “The Historical Background to 4Q468g [= 4Qhistorical B].ˮ RevQ 73 (1999): 137–38. Sukenik, Eliezer L. Megilot genuzot mitokh genizah keduma shenimtseʾah bemidbar Yehudah [Hidden Scrolls from the Genizah Found in the Judaean Desert]. 2 vols. Jerusalem: Bialik Institute, 1948/1949 [Hebrew]. Sussman, Varda. “The Oil Lamps.” In Joseph Patrich. Archaeological Excavations at Caesarea Maritima: Areas CC, KK and NN; Final Report. Vol. 1, The Objects. Jerusalem: Eisenbrauns, 2008, 208–300. Sydenham, Edward A. The Coinage of the Roman Republic. London: Spink & Son, 1952. Tal, Nir, and Gershon Oron. “Region XIII: Survey and Excavations of Caves along the Fault Escarpment South of Qumran.” ʿAtiqot 41/1 (2002): 221–27 [Hebrew], 187–93. Talmon, Shemaryahu. “Hebrew Fragments from Masada.” In Talmon, and Yadin. Masada VI, 1–149. ———. “Calendrical Documents and Mishmarot.” In Talmon, Jonathan Ben Dov, and Uwe Glessmer, eds. Qumran Cave 4.XVI: Calendrical Texts. DJD 21. Oxford: Clarendon Press, 2001, 1–166. ———, and Yigael Yadin, eds. Masada VI: The Yigael Yadin Excavations 1963–1965; Final Reports; Hebrew Fragments

BIBLIOGRAPHY from Masada. Jerusalem: Israel Exploration Society/Hebrew University, 1999. Taylor, Joan E. “The Cemeteries of Khirbet Qumran and Women’s Presence at the Site.” DSD 6 (1999): 285–323. ———. “Khirbet Qumran in the Nineteenth Century and the Name of the Site,” PEQ 134 (2002): 144–64. ———. “Khirbet Qumran in Period III.” In Galor, Humbert, and Zangenberg. Qumran, the Site of the Dead Sea Scrolls, 133–46. ———. “ ‘Roots, Remedies and Properties of Stones’: The Essenes Qumran and Dead Sea Pharmacology.” JJS 60 (2009): 226–44. ———. “Buried Manuscripts and Empty Tombs: The Qumran Genizah Theory Revisited.” In Aren M. Maeir, Jodi Magness, and Lawrence H. Schiffman, eds. ‘Go Out and Study the Land’ ( Judges 18:2): Archaeological, Historical and Textual Studies in Honor of Hanan Eshel. JSJSup 148. Leiden: Brill, 2011, 269–315. ———. The Essenes, the Scrolls, and the Dead Sea. Oxford: Oxford University Press, 2012. ———, et al. “Qumran Textiles in the Palestine Exploration Fund, London: Radiocarbon Dating Results.” PEQ 137 (2005): 159–67. ———, and Shimon Gibson. “Qumran Connected: The Qumran Pass and Paths of the North-western Dead Sea.” In Frey, Claussen, and Kessler. Qumran und die Archäologie, 163–209. Tebes, Juan M. “The Pottery Assemblage of Jerusalem’s NeoBabylonian Destruction Level: A Review and Discussion.” Antiguo Oriente 9 (2011): 277–341. Testa, Emmanuele. Herodion IV: I graffiti e gli ostraka. SBF.CMa 20. Jerusalem: Franciscan Printing Press, 1972. Thiering, Barbara. “The Date and the Order of Scrolls, 40 BCE to 70 CE.” In Schiffman, Tov, and VanderKam. The Dead Sea Scrolls Fifty Years After their Discovery, 191–98. Thomas, Samuel I. The “Mysteries” of Qumran: Mystery, Secrecy, and Esotericism in the Dead Sea Scrolls. EJL 25. Atlanta, GA: SBL, 2009. Tigchelaar, Eibert J.C. “In Search of the Scribe of 1QS.” In Paul, et al. Emanuel, 439–52. ———. “On the Unidentified Fragments of DJD XXXIII and PAM 43.680: A New Manuscript of 4QNarrative and Poetic Composition, and Fragments of 4Q13, 4Q269, 4Q525 and 4QSb (?).” RevQ 83 (2004): 477–85. ———. “Assessing Emanuel Tov’s ‘Qumran Scribal Practice’.” In Sarianna Metso, Hindy Najman, and Eileen Schuller, eds. The Dead Sea Scrolls: Transmission of Traditions and Production of Texts. STDJ 92. Leiden: Brill, 2010, 173–207. ———. “The Dead Sea Scrolls.” EDEJ: 163–80. Tournay, Raymond J. “Les anciens manuscrits hébreux récemment découverts.” RB 56 (1949): 204–33. Tov, Emanuel. “The Socio-Religious Background of the PaleoHebrew Biblical Texts Found at Qumran.” In Peter Schäfer,

BIBLIOGRAPHY Hubert Cancik, and Hermann Lichtenberger, eds. Geschichte– Tradition–Reflexion: Festschrift für Martin Hengel zum 70. Geburtstag. Vol. 1. Tübingen: Mohr Siebeck, 1996, 353–74. ———. “Tefillin of Different Origin from Qumran?” In Yair Hoffman, and Frank S. Pollak, eds. A Light for Jacob: Studies in the Bible and the Dead Sea Scrolls in Memory of Jacob Shalom Licht. Jerusalem: Bialik Institute, 1997, 44*–54*. ———. “A Qumran Origin for the Masada Non-biblical Texts?” DSD 7 (2000): 57–73. ———. “Further Evidence for the Existence of a Qumran Scribal School.” In Schiffman, Tov, and VanderKam. The Dead Sea Scrolls Fifty Years After their Discovery, 199–216. ———, ed. The Texts from the Judaean Desert: Indices and an Introduction to the Discoveries in the Judaean Desert Series. DJD 39. Oxford: Clarendon Press, 2002. ———. “The Biblical Texts from the Judaean Desert: An Overview and Analysis of the Published Texts.” In Edward D. Herbert, and Emanuel Tov, eds. The Bible as Book: The Hebrew Bible and the Judaean Desert Discoveries. New Castle, DE: Oak Knoll Press, 2002, 139–66. ———. Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert. STDJ 54. Leiden: Brill, 2004. ———. “The Special Character of the Texts Found in Qumran Cave 11.” In Esther G. Chazon, David Satran, and Ruth A. Clements, eds. Things Revealed: Studies in Early Jewish and Christian Literature in Honor of Michael E. Stone. JSJSup 89. Leiden: Brill, 2004, 187–96. Reprinted in Emanuel Tov. Hebrew Bible, Greek Bible and Qumran: Collected Essays. TSAJ 121. Tübingen: Mohr Siebeck, 2008, 421–27. ———. “The Sciences and the Reconstruction of the Ancient Scrolls: Possibilities and Impossibilities.” In Armin Lange, et al., eds. The Dead Sea Scrolls in Context: Integrating the Dead Sea Scrolls in the Study of Ancient Texts, Languages, and Cultures. VTSup 140/1. Vol. 1. Leiden: Brill, 2011, 3–25. ———. “The Dead Sea Scrolls and the Textual History of the Masoretic Text.” In Nóra Dávid, et al., eds. The Hebrew Bible in Light of the Dead Sea Scrolls. FRLANT 239. Göttingen: Vandenhoeck & Ruprecht, 2012, 41–53. Tsafrir, Yoram, and Boaz Zissu. “A Hiding Complex of the Second Temple Period and the Time of the Bar Kokhba Revolt at ʿAin-ʿArrub in the Hebron Hills.” In John H. Humphrey, ed. The Roman and Byzantine Near East. Vol. 3. JRASS 49. Portsmouth, RI: Journal of Roman Archaeology, 2002, 7–36. Ussishkin, David. “Excavations at Tel Lachish 1973–1978.” Tel Aviv 5 (1978): 1–97. Van der Plicht, Johannes, and Kaare L. Rasmussen. “Radiocarbon Dating and Qumran.” In Gunneweg, Adriaens, and Dik. Holistic Qumran, 99–121. Van Lessen, Michael. “The Coinage Pattern of Palestine in the First and Second Centuries A.D.” NCirc 79/6 (1971): 240–41.

349 VanderKam, James C. Calendars in the Dead Sea Scrolls: Measuring Time. Literature of the Dead Sea Scrolls. London: Routledge, 1998. ———, and Peter Flint. The Meaning of the Dead Sea Scrolls: Their Significance for Understanding the Bible, Judaism, Jesus, and Christianity. San Francisco: HarperSanFrancisco, 2002. Vaughn, Kevin J., and Hector Neff. “Tracing the Clay Source of Nasca Polychrome Pottery: Results from a Preliminary Raw Material Survey.” JARS 31 (2004): 1577–86. Vaux, Roland de. “La cachette des manuscrits hébreux.” RB 56 (1949): 234–37. ———. “La grotte des manuscrits hébreux.” RB 56 (1949): 586–609. ———. “Exploration de la région de Qumrân: Rapport préliminaire.” RB 60 (1953): 540–61. ———. “Fouilles au Khirbet Qumrân: Rapport préliminaire.” RB 60 (1953): 83–106. ———. “Les grottes de Murabba⁠ʾat et leurs documents.” RB 60 (1953): 245–67. ———. “Fouilles au Khirbet Qumrân: Rapport préliminaire sur la deuxième campagne.” RB 61 (1954): 206–36. ———. “La poterie.” In Barthélemy, and Milik. Qumran Cave 1, 8–17. ———. “Fouilles de Khirbet Qumrân: Rapport préliminaire sur les 3e, 4e et 5e campagnes.” RB 63 (1956): 533–77. ———. “Une hachette essénienne?” VT 9 (1959): 399–407. ———. L’archéologie et les Manuscrits de la Mer Morte. The Schweich Lectures 1959. London: Oxford University Press for the British Academy, 1961; Revised edition. Translated by David Bourke. Archaeology and the Dead Sea Scrolls. 1973. ———. “Fouilles de Feshkha.” RB 66 (1959): 225–55. ———. “Archéologie.” In Benoit, Milik, and de Vaux. Les grottes de Murabbaʿât, 3–50. ———. “Archéologie.” In Baillet, Milik, and de Vaux. Les ‘petites grottes’ de Qumrân, 3–36. ———. “Esséniens ou Zélotes? A propos d’un livre récent.ˮ RB 73 (1966): 212–35. ———. “Archéologie.” In de Vaux, and Milik. Qumrân Grotte 4.II, 3–22. ———. Die Ausgrabungen von Qumran und En Feschcha: Die Grabungstagebücher. German translation and notes by Ferdinand Rohrhirsch, and Bettina Hofmeir. NTOA.SA 1A. Fribourg/Göttingen: Universitätsverlag/Vandenhoeck & Ruprecht, 1996. ———, and Józef T. Milik, eds. Qumrân Grotte 4.II: I. Archéologie; II. Tefillin, Mezuzot et Targums (4Q128–4Q157). DJD 6. Oxford: Clarendon Press, 1977. Vermes, Geza. “Historiographical Elements in the Qumran Writings: A Synopsis of the Textual Evidence.” JJS 58 (2007): 121–39.

350 Voegtli, Hans. “Bibliographie Herbert A. Cahn, 1975–1984.” In Beatrice Schärli, and Hans Voegtli, eds. Festschrift Herbert A. Cahn: Zum 70. Geburtstag gewidmet und herausgegeben vom Circulus Numismaticus Basiliensis. Basel: Circulus Numismaticus Basiliensis, 1985, xiii–xv. Wagemakers, Bart, and Joan E. Taylor. “New Photographs of the Qumran Excavations from 1954 and Interpretations of L.77 and L.86.” PEQ 143 (2011): 134–56. Walker, Dennis. “Notes on Qumran Archaeology: The Geographical Context of the Caves and Tracks.” QC 3 (1993): 93–100. ———. “Qumran Cave 4 and the 1952 Survey.” QC 3 (1993). ———, and Robert H. Eisenman. “The 1990 Survey of Qumran Caves.” QC 2 (1992): 45–49. Wampler, Joseph C. Tell en-Nasbeh: Excavated under the Direction of the late William Frederic Badè. Vol. 2, The Pottery. Berkeley/ New Haven: Palestine Institute of Pacific School of Religion/ American Schools of Oriental Research, 1947. Wartenberg, Ute. “Herbert A. Cahn, 1915–2002.” American Numismatic Society Magazine 1/2 (2002): 33. ———. “Herbert A. Cahn, 1915–2002.” The Celator 16/9 (2002): 32. Webster, Brian. “Chronological Index of the Texts from the Judaean Desert.” In Tov. The Texts from the Judaean Desert, 351–446. Weinberg, Gladys D., and Dan Barag. “Other Finds: Glass Vessels.” In Lapp, and Lapp. Discoveries in the Wâdī ed-Dâliyeh, 103–5. Weise, Manfred. Kultzeiten und Kultischer Bundesschluss in der “Ordensregel” vom Toten Meer. SPB 3. Leiden: Brill, 1961. Wexler, Lior, ed. “Preface.” ʿAtiqot 41/2 (2002): v–xviii. ———, ed. “Surveys and Excavations of Caves in the Northern Judean Desert (CNJD)—1993.” ʿAtiqot 41 (2002). White Crawford, Sidnie. “Qumran: Caves, Scrolls, and Buildings.” In Mason, et al. A Teacher for all Generations, Vol. 1: 253–73. Wise, Michael O. “Accidents or Accidence: A Scribal View of Linguistic Dating of the Aramaic Scrolls from Qumran.” In Wise. Thunder in Gemini and Other Essays on the History, Language and Literature of Second Temple Palestine. JSPSup 15. Sheffield: Sheffield Academic Press, 1994, 103–51. ———. “Dating the Teacher of Righteousness and the Floruit of his Movement.” JBL 122 (2003): 53–87. ———. “Murabbaʿāt and the First Jewish Revolt.” In Joel L. Kraemer, and Michael G. Wechsler, eds. Pesher Naḥum: Texts and Studies in Jewish History and Literature from Antiquity through the Middle Ages Presented to Norman (Naḥum) Golb. SAOC 66. Chicago: Oriental Institute, University of Chicago, 2012, 293–315. ———, et al., eds. Methods of Investigation of the Dead Sea Scrolls and the Khirbet Qumran Site: Present Realities and Future Prospects. ANYAS 722. New York: New York Academy of Science, 1994.

BIBLIOGRAPHY Wolff, Samuel. “The Excavation of Cave IV/12.” ʿAtiqot 41/2 (2002): 73–83. Wolters, Albert M. “The Copper Scroll and the Vocabulary of Mishnaic Hebrew.” RevQ 55 (1990): 483–95. Wood, Bryant G. “To Dip or Sprinkle? The Qumran Cisterns in Perspective.” BASOR 256 (1984): 45–60. Wouters, Helena. “Archaeological Glass from Khirbet Qumran: An Analytical Approach.” In Gunneweg, Greenblatt, and Adriaens. Bio- and Material Cultures at Qumran, 171–90. ———, et al. “Antique Glass from Khirbet Qumrân: Archaeological Context and Chemical Determination.” BIRPA 28 (1999/2000): 9–40. Yadin, Yigael. “Expedition D.” IEJ 11 (1961): 36–52. ———. “Expedition D—The Cave of the Letters.” IEJ 12 (1962): 227–57. ———. The Finds from the Bar Kokhba Period in the Cave of Letters. Jerusalem: Israel Exploration Society, 1963. ———. Masada: Herod’s Fortress and the Zealots’ Last Stand. New York: Random House, 1966. ———. Tefillin from Qumran (XQ Phyl 1–4). Jerusalem: Israel Exploration Society/Shrine of the Book, 1969. ———. Bar Kokhba: The Rediscovery of the Legendary Hero of the Last Jewish Revolt against Imperial Rome. London: Weidenfeld & Nicolson, 1971. ———. “The Ben Sira Scroll from Masada.” In Talmon, and Yadin. Masada VI, 151–252. ———, et al. Hazor. Vol. 1, An Account of the First Season of Excavations, 1955. James A. Rothschild Expedition at Hazor. Jerusalem: Magnes Press, 1958. ———, et al. The Documents from the Bar Kokhba Period in the Cave of Letters: Hebrew, Aramaic and Nabatean-Aramaic Papyri. Judean Desert Studies 3. Jerusalem: Israel Exploration Society, 2002. Yardeni, Ada. “A Draft of a Deed on an Ostracon from Khirbet Qumrân.” IEJ 47 (1997): 233–37. ———. “Documentary Texts Alleged to be from Qumran Cave 4.” In Cotton, and Yardeni. Aramaic, Hebrew and Greek, 283–318. ———. “234. 4QExercitium Calami A.” In Pfann. Qumran Cave 4.XXVI: Cryptic Texts; Alexander, et al. Miscellanea, Part 1, 185–86. ———. “360. 4QExercitium Calami B.” In Pfann. Qumran Cave 4.XXVI: Cryptic Texts; Alexander, et al. Miscellanea, Part 1, 297. ———. Textbook of Aramaic, Hebrew and Nabataean Documentary Texts from the Judaean Desert and Related Material. 2 vols. Jerusalem: Hebrew University, Ben-Zion Dinur Center, 2000. ———. “A Note on a Qumran Scribe.” In Meir Lubetski, ed. New Seals and Inscriptions, Hebrew, Idumean and Cuneiform. HBM 8. Sheffield: Phoenix Press, 2007, 287–98.

BIBLIOGRAPHY Yellin, Joseph, Magen Broshi, and Hanan Eshel. “Pottery of Qumran and Ein Ghuweir: The First Chemical Exploration of Provenience.” BASOR 321 (2001): 65–78. Yezerski, Irit. “Pottery of Stratum V.” In Stern. En-Gedi Excavations I, 86–129. Young, Ian. “The Stabilization of the Biblical Text in the Light of Qumran and Masada: A Challenge for Conventional Qumran Chronology?” DSD 9 (2002): 364–90. ———. “The Biblical Scrolls from Qumran and the Masoretic Text: A Statistical Approach.” In Marianne Dacy, Jennifer Dowling, and Suzanne Faigan, eds. Feasts and Fasts: A Festschrift in Honour of Alan David Crown. Mandelbaum Studies in Judaica 11. Sydney: Mandelbaum Publishing, 2005, 81–139. ———. “The Contrast Between the Qumran and Masada Biblical Scrolls in the Light of New Data: A Note in Light of the Alan Crown Festschrift.” In Shani Tzoref, and Ian Young, eds. Keter Shem Tov: Collected Essays on the Dead Sea Scrolls in Memory of Alan Crown. Piscataway, NJ: Gorgias Press, 2013, 113–19. Zangenberg, Jürgen K. “Opening Up Our View: Khirbet Qumran in a Regional Perspective.” In Edwards, Religion and Society in Roman Palestine, 170–87. ———. “Zwischen Zufall und Einzigartigkeit: Bemerkungen zur jüngsten Diskussion über die Funktion von Khirbet Qumran und die Rolle einiger ausgewählter archäologischer Befunde.” In Frey, Claussen, and Kessler. Qumran und die Archäologie, 121–46.

351 Zayadine, Fawzi. “Hellenistic Pottery from the Estate of the Tobiads.” In Dina Frangié, and Jean-François Salles, eds. Lampes antiques du Bilad es Sham: Jordanie, Syrie, Liban, Palestine; Ancient lamps of Bilad es Sham; Actes du Colloque de Pétra-Amman, 6–13 novembre 2005. AH. Paris: De Boccard, 2011, 167–81. Zias, Joseph E. “The Cemeteries of Qumran and Celibacy: Confusion Laid to Rest?” DSD 7 (2000): 220–53. Zimhoni, Orna. “The Pottery.” In Volkmar Fritz, and Aharon Kempinski, eds. Ergebnisse der Ausgrabungen auf der Hirbet el-Msas (Tel Masos) 1972–1975. ADPV 6. Vol. 1. Wiesbaden: Otto Harrassowitz, 1983, 127–30. ———. “The Pottery of Levels III and II.” In David Ussishkin, ed. The Renewed Archaeological Excavations at Lachish (1973– 1994). MSSMNIA 22. Vol. 4. Tel Aviv: Emery and Claire Yass Publications in Archaeology, 2004, 1789–899. Zissu, Boaz, et al. “Coins from the Bar Kokhba Revolt Hidden in Meḥarat Ha-Teḥomim (Mŭghâret Umm et Tûeimîn), Western Jerusalem Hills.” INJ 17 (2009–2010): 113–47. ———, and David Hendin. “Further Remarks on Coins in Circulation during the Bar Kokhba War: TeʾOmim Cave and Horvat ʿEthri Coin Hoards.” In David M. Jacobson, and Nikos Kokkinos, eds. Judaea and Rome in Coins 65 BCE–135 CE: Papers Presented at the International Conference Hosted by Spink, 13th–14th September 2010. London: Spink & Son, 2012, 215–28.

Index of Modern Authors Abel, Félix-Marie 60n44 Adan-Bayewitz, David 113n50, n53, 185–6, 185n10, n13, 186n15 Adler, Yonatan 4, 161n1, 163n5, n8, 164n14, 165n21, 166n26, 167nn29–30, 168n35, n37, n39, 169n41 Adriaens, Annemie 28n133, 134n25, 137n2, 148n75, 154n108, 201n39, 235n108, 244n28 Agur, Benny 151n84 Aharoni, Miriam 249n24, 252n97 Aharoni, Yohanan 14, 16, 16n42, 115n71, 142n26–27, 163n5, n8, 167n29, 168n39, 249n24, n30, n37, 250n44, n59, 251n72, 252n91, n97–98, n104 Aitken, James K. 85n52 Albright, William F. 239–40, 239n8, 240n9, 246, 251n77, 252n99, n112, 253n118 Alexander, Philip S. 71n49, 83n30, 84nn45–7, 85n51, 88n9, 100n24, 214n11 Alfaro, Carmen 232n100 Allegro, John 16–7, 16n45, 17n48, 23–4, 36, 73n73, 201, 277n49 Alliata, Eugenio 151n84, 222n17, 223n21 Amar, Zohar 21, 21n83 Amit, David 16n41, 18n59, 23n93, 29nn146–7, 88n7, 141n19, 142n28 Ariel, Donald T. 38n6, 140n16, 153n105, 160nm, np, 182n25, 221n7, 224n29, 228n64, 230n74, 232n100, 233n102, n104, 234n107, 235n109, 250n50, 252n93 Arif, Aida F. 224n29, 228n50–1, n58, n61, 229n65 Arubas, Benny 19nn63–4, 21n77, 109n12, 151n84, 152n97, 160nm, 185n14, 186nn16–7, 202n51, 248n12, 255 Atkinson, Kenneth 239n6 Augé, Cristian 227–8 Avigad, Nahman 14, 99n15, 141n23, 248n23, 255 Aviram, Joseph 14, 111n30, 158nd, 160no Avi-Yonah, Michael 186 Bagatti, Bellarmino 221n6 Baginsky, Alisa 136n30, 143n48, 206n80 Bailey, Donald M. 114n63, 116n79, n86 Baillet, Maurice 13n22, n29, 36n2, 71nn50–4, 72, 72nn56–8, n62, nn64–5, n68, 73nn69–71, 74n92, 75n97, 76n109, 77, 77n120, nn123–4, 80n1, 83, 83n28, 100n20, 105n44, 109n3, 115, 127n11, 138n4, 163n4, n8, 168n39, 169n40, 177n3, 186n22, 188n43, 196n4, 202n45, 216n23, 241n15, 247n4, 263n4 Balla, Marta 111, 111n34, 113n51, 149n77, 187n39, 188, 188nn51–2, 189–94, 189n54, nn58–60, 189n62, 190nn63–4, n66–71, 191nn74–6, nn78–82, 193n106, nn109–10, 200, 200n35, 201, 201n39, n42, 213n4, 214nn9–10, 215, 215n13, n15, 234n108, 243n24 Bar Adon, Pesach 16–7, 16n43, 17nn49–54, 27n118, 36, 39n12, n14, 141n25, 158nd, 203, 241, 241n16, 251n63, n75, n83 Bar Asher, Moshe 238n2 Bar Nathan, Rachel 19n65, 110, 110n25, 111, 111n29, n35, 112n40, n43, 113nn53–4, 149n77, 152n91, 158nd, 185n9, 187, 187nn29–31, n37, 191n78, 200, 200n36, 238, 238n3, 242n21, 243, 243n26, 244n27 Barag, Dan 111n30, 113n53, 115nn72–3, 140n15, 222n10 Baramki, Dimitri C. 149n77 Barrois, Augustin-George 60 Barshad, Dror 143n48 Barthélemy, Dominique 10n3, 37n2, 70n37, 109n1, 112, 123n2, 125n7, 138n4, 162n4, 163n8, 169n40, 186n21, 196n3, 240n10, 263n4 Baruch, Yuval 138n6, 139–40n13, 143n41, 160nq, 205n79, 248n20

Baruchi, Yosi 23n101 Baumgarten, Albert I. 167n30 Becker, Jürgen 164n11 Beit Arieh, Itzhak 249n29, n36, 250n55 Bélis, Mireille 3, 6, 41n21, 42n22, n24, 123nn5–6, 137n2, 158nd, 159ng, ni, 160nm, 197n6, 201, 265n12, 272 Ben Dov, Jonathan 82n26, 83, 83n29, 89n12 Ben Haim, Ronny 22n86, 110n14, 116n75, 138n6, 139n13, 140nn13–14, 143n41, 144n51, 160nm, 160nno–p, 205n79, 248n12, 255 Ben Hayyim, Zeev 92n32 Ben Tor, Amnon 185n14 Benoit, Pierre 10n13, 59n41, 95n44, 97n7, 113n55, 141n20, 142n3, 179n11, 208n88, 223n19, 248n21, 275n63 Berlin, Andrea 181n22, 186, 186n17 Bernick-Greenberg, Hannah 249n32, n35, 250n48, 251n80, 252n109 Bernstein, Moshe 16n44, 71n55 Bijovsky, Gabriela 230n83, 231n85, 232nn98–100, 234n107 Birnbaum, Solomon 239, 239n7 Bishop, Ronald L. 193, 194n113 Blackman, James M. 193, 194n113 Blau, Lajos 89n14 Bonani, Georges 95n42, 102n32, 105n46, 178n6, 179n10 Bottini, Giovanni Claudio 221, 221n2, 222n17 Bourke, David 10n8 Bowden, Andrew 83n35 Brann, Marcus 89n14 Brooke, George J. 68, 68n13, 75n105, 80n1, 81n11, 82n20, 84n39, 85n52, 167n27, n31, 177n1 Broshi, Magen 22, 22nn89–91, 23, 23n94, n96, 36, 67, 67n2, n4, 100n26, 113n47, 138, 138n8, 139n13, 140n13, 142n37, 143n41, 144, 144n54, 145, 145nn56–8, 147nn68–9, 148nn70–4, 159ng, 170n45, 171n54, 172n55, 181n18, nn20–1, 188, 188nn40–1, nn43–6, nn48–50, 189n60–2, 191n77, 192n94, 199nn24–5, 202, 204, 204nn67–71, 206, 224n29, 235n110 Brown, Judith A. 16n46 Brownlee, William H. 125n8 Bruce, Frederick F. 13n23, 82n20, 242, 242n18 Burdajewicz, Mariusz 5, 247n2, 249n25, 250n40, 251n61, 253n123, Busa, Anna 163n8 Butcher, Kevin 229n66 Buxeda i Garrigós, Jaume 194nn119–121 Cahn, Herbert A. 222, 222n11, 234n107 Callegher, Bruno 5, 224n27, 230n76 Campbell, Jonathan 82n22, 84n38 Cancik, Hubert 90n21 Cansdale, Lena 145, 145n60, 197n6, 201n44, 202, 202n46, 203n59 Carmi, Israel 178n7 Carmignac, Jean 75, 75n106 Carswell, John 31n159, 171n54 Casey, Patrich J. 232n96, 233 Cau Ontiveros, Miguel A. 194nn119–21 Chambon, Alain 10n9, 13n28, 14n33, 28n132, 46n29, 109nn1–2, n4, 112n40, nn43–4, 114n61–2, 115n66–8, 116n77, 148n75, 158nd, nf, 159nnf–g, 184n3, 186n26, n28, 190n67, 196n3, 197n6, nn8–9, n12, 198n16, 199n23, 214nn7–8, n10, 224n29, 225n31, n33, 226n34, 241n13, 253n114, 277n51 Charlesworth, James 22n87, 49n35, 141n17, 162n3, 163n8

Index of Modern Authors Chazon, Esther 78n132, 82n20, 85n52 Chernov, Elena 158nd Cheynet, Jean-Claude 221n8 Ciecieląg, Jerzy 224n29 Clamer, Christa 28, 28n134, 187n32 Claussen, Carsten 26n112, 67n6, 72n61, 80n9, 81n10, n17, 142n36, 182n25, 200n31, 207n82, 208n85 Clements, Ruth A. 78n132, 82n20 Cohen, Rudolf 138n6, 140n14, 143n41, 160nr, 249n32 Cohn, Yehudah B. 161n1, 164, 164nn10–1, n15, 165, 165n18, 167n27, n29, n31, 168n35, n38 Collins, John J. 69n32, 81, 81n11, n13, 83n31, 150n81 Contenson, Henri de 73 Corbo, Virgilio 221n6 Cotton, Hannah M. 29, 29n150, 30, 30n151, n153, 72, 72n60, 93n34, 142n36, 162n3, 163n8, 217n32 Coüasnon, Charles 55 Cox, Ben B. 140n15 Crawford, Michael H. 228n62 Croom, Alexandra T. 154n109 Cross, Frank M. 26, 26n114, 69n31, 73, 73nn80–1, 80, 80n8, 95, 95n44, 97n6, 98n13, 99nn14–7, 100n20, n22, 101nn27–9, 214n11, 216n21, 216n25–6, 217n31, 240, 240n11–2, 246, 254 Crowfoot, Elisabeth 140n15, 141n22 Crowfoot, Grace M. 123, 123n2, 125–6, 125n8, 128–9, 129n17, 130–1, 130n18, 135, 138n4, 139n13, 141n22, 159ng, 264, 265n12 Crown, Alan D. 238, 238n2 Dacy, Marianne 238n2 Dahari, Uzi 151n84, 248n20, 255 Dar, Shimon 14n38 Dávid, Nóra 244n29 Davies, Philip R. 91n11 De Groot, Alon 250n50, 252n93 Deichgräber, Reinhard 72, 72n67 Dell, Katharine J. 85n52 Diez Fernandez, Florentino 114n61, 115n68 Dik, Joris 201n39, 235n108, 244n28 Dimant, Devorah 16n41, 80nn1–2, 81–2, 81nn15–7, 82nn20–1, n24, 84n41, 145n58, 169n43 Diringer, David 90, 90n19 Donceel, Robert 110–1, 110n24, 111n36, 112n37, n40, n43, nn45–6, 113, 113n48, nn51–2, 148n75, 223nn18–20, 224n29, 225n30, 227n46, n49, 228nn52–3, 230n73, n78, 231, 231nn88–9, n93 Donceel-Voûte, Pauline 103n40, 148n75, 224n29, 225n30, 228nn52–3, 230n73, 231, 231n89, n93 Dothan, Moshe 250n57, 252n111 Doudna, Gregory L. 5, 93n33, 95n42, 99n15, 102n32, 238nn1–2, 239n6, 243nn23–4, n27, 244n28 Dowling, Jennifer 238n2 Doyen, Jean-Marc 232n96 Driver, Godfrey R. 143n38 Drori, Amir 22, 22n85, 36 Dunayevsky, Immanuel 250n57 Dušek, Jan 97n5 Edge, Raymond L. 89n13, 90n21 Edwards, Douglas R. 25n106, 205n78 Eisenman, Robert 19, 19n71, 20n74, 21nn78–81, 197n7, 202–4, 203nn60–2, 204nn63–6, 248n11 Eisenstadt, Irinia 149n77 Enzel, Yehouda 28n139

353 Eshel, Esther 18n57, 97n4, 100n23, 101n27, 167n27, 214n11, 216n21, nn25–6, 218nn35–6 Eshel, Hanan 9, 9n1, 16n41, n47, 17n55, 18, 18nn56–9, 22–3, 22n84, n87, nn89–91, 23nn93–4, nn96–102, 29nn145–7, 36, 67–8, 67n2, n4, 76, 76n112, 98n14, 100n23, 113n47, 138, 138n8, 139n13, 140n13, 141nn17–20, n24, 142nn27–8, n37, 143n41, nn43–4, nn46–7, 144, 144n54, 145nn56–8, 146n66, 147nn68–70, 148nn71–4, 158nf, 159ng, 170n45, 171n54, 172n55, 181n18, nn20–2, 188, 188nn40–1, nn43–6, nn48–50, 188nn60–2, 191n77, 192n94, 199nn24–5, 201–2, 204, 204nn67–71, 206, 224n29, 225nn110–3, 235n115 Eshel, Itzhak 250n42, 251n65, 252n89 Evans, Craig 81n11 Evans, Jane DeRose 233n105, 234n107 Faggi, Annalisa 153n102, 158nf Faigan, Suzanne 238n2 Falk, Daniel K. 67n8, 82n18, 137n1 Farhi, Yoav 28n131, 224n29 Feather, Robert 197nn11–2, 198n14 Fidanzio, Marcello 5, 188n43, 269n24 Fields, Weston W. 9n2, 21n82, 135n29, 196nn3–4, 197n6, nn8–10, nn12–13, 198n16–9, nn21–3 Findlater, George 39n16 Fischer, Moshe 149n77, 187n32 Fischer-Bossert, Wolfgang 230n83, 231n85, 232nn98–100 Fitzmyer, Joseph A. 10n16, 11n17, 26n109, 27n117, 100n24 Flint, Peter W. 83n29, 95nn41–3, 102n32, 238n1 Florentin, Moshe 238n2 Foerster, Gideon 111n30 Fontanille, Jean-Philippe 228n64, 232n100, 233n102, n104 Fourrier, S. 253n119 Fraade, Steven D. 83n32 Frangié, Dina 111n31 Franken, Hendricus J. 250n60, 251, 251n64, n76 Freedman, David N. 89nn13–4 Freud, Liora 249n29, n36, 250n55, 251nn70–1, nn87–8 Frey, Jörg 26n112, 67n6, 72n61, 80n9, 81n10, n17, 83n35, 109n31, 142n36, 182n25, 199n31, 207n82, 208n85 Fritz, V. 249n31 Frumkin, Amos 16n47, 18, 23, 23nn97–100, n102, 29n145, 141n19, n24, 201, 235n115 Gal, Tzahi 142n30 Galor, Katharina 23n92, 27n126, 39n15, 39n17, 111n28, 112n40, 143n37, 148n75, 149n77, 184n5, 200n36, 214n6, 215n12, 238n1 Gamble, Harry Y. 91n28 García Martínez, Florentino 2, 6, 67–8n8, 70nn35–6, n39, 71n45, n48, 74n88, 78n131, n137, 79n138, 80n2, 81n11, 81–82n18, 82nn20–1, n26, 94, 94nn36–7, 137n1, 145n55, 150n81, 153n104, 163n9, 170nn46–47, 171n51, 181n18 Gärtner, Judit 149n77, 151n91, 158nd Gebhart, Hans 228n56 Getzov, Nimrod 185n14 Geva, Hillel 113n57, 149n77, 158nd, 234n107 Gibson, Shimon 26n112, 27, 27n115, 199–200, 199n31, 200n32, 202n48, 205n75 Gichon, Mordechai 149n77, 187n32 Gitin, Seymour 249n34, n38, 250n53, n56, 251n66, 252n96, n100 Gitler, Haim 232n98, 234n107 Gitler-Kamil, Ronit 149n77 Glessmer, Uwe 82n26, 83n29 Golb, Norman 87, 88n4, 172n62

354 Goldfus, Haim 185n14, 186nn16–7 Goranson, Stephen 73n74 Gorin-Rosen, Yael 140n16, 141n17 Grant, Elihu 253n118 Greenblatt, Charles 23, 23n103, 28n133, 134n25, 137n2, 148n75, 154n109, 200n35, 201nn41–3 Greenfeld, Jonas C. 162n3 Gropp, Douglas 16n44, 71n55 Gunneweg, Jan 23, 23n103, 28nn133–4, 38n7, 41n21, 101n30, 110n17, 111, 111n34, 113n51, 123n6, 131n23, 134n25, 137n2, 148n75, 149n77, 154n108, 187n39, 188–94, 188nn51–3, 189nn54–6, nn58–60, n62, 190nn63–4, nn66–71, 191nn74–6, nn78–82, 193n106, nn109–10, 197n6, 200, 200n35, n37, 201n39, nn41–3, 213n1, n4, nn9–10, 215, 215nn13–5, 234–5n108, 243n24, 244n28, 247n3, 264n12, 271n32 Hadad, Shulamit 116n83 Hadas, Gideon 158ne Hadidi, Adnan 222n12 Hameiri, Nadav 138n6, 139–40n13 Hanson, Richard S. 89n14 Harding, Gerald L. 9–10, 10n4, n12, 36n2, 109n1, 126, 138n4, 139n13, 143n41, 152n100, 159ng, 162n4, nn13–4, 165n16, 184, 184nn1–2, 196n3, 226, 226n34, 240, 240n10, 260, 264n9 Hayes, John W. 116, 116 n78, n86 Hempel, Charlotte 2, 9, 9n1, 67, 68n8, 74, 74nn89–90, n92, n94, 75n98, 78n131, 80n1, 81n11, n14, 82n18, n26, 83n34, 84nn36–7, n42, 85nn49–50, n53, 86n54, 94, 94nn36–7, 137n1, 145n55, n63, 150n81, 170n46, 171n51, 173, 173n61, 177n1, 181n18 Hendin, David 232n101, 233n102, 235n116 Herbert, Edward D. 173n59 Hermary, A. 253n119 Hershkovitz, Malka 111n30, 113n53, n57, 115n72–3 Herzog, Ze’ev 249n28, 250n46, 251n69, n79, 252n92, n107, n110 Hilhorst, Anthony 98n13, 216n21 Hirschfeld, Yizhar 28, 28n138, 38n6, 39n13, 109n1, 114–5, 114n59, 145, 145n60, 146nn66–7, 151n84, n91, 153n108, 154n109, 158nd, 196n3, 198n20, 204, 204n71, n73, 205n78, 234nn107–8, 241, 242n16 Hoffman, Yair 166n26 Hofmeir, Bettina 27n123, 148n75, 224n29 Holmgren, Richard 59nn42–3 Howgego, Christopher 229n66 Hübner, Ulrich 222n10 Humbert, Jean-Baptiste 2, 5–6, 10n9, 13n28, 14n33, 23n92, 26–7, 27n126, 28n132, n134, 31, 38n3, n7, n9, 39n15, n17, 41n21, 46n29, 51n36, 80, 80n5, 101n30, 105n44, 109nn1–2, n4, 110–2, 110n17, 111n26, n28, n34, 112n38, n40, nn43–4, 114nn61–2, 115nn66–8, 116, 116n77, 123, 123n6, 131, 131nn22–3, 133, 137n2, 139n10, 143n37, 148n75, 149n77, 158nd, nf, 159nnf–g, 184n3, n5, 186n26, n28, 187n39, 188n53, 190n67, 196n3, 197n6, nn8–9, n12, 198n16, 199n23, 200nn35–6, 213n1, n4, nn6–8, 214n10, 215nn12–4, 223, 223n19, n21, 224n29, 225n31, n33, 226n34, 227n48, 229n67, 230n74, 238n1, 241, 241n13, n16, 243, 247n3, 253n122, 265n12, 271n32, 277n51, 278 Humphrey, John H. 142n31, 182n24, 185n14 Hutchesson, Ian 238, 238n2 Ibrahim, Fawzi 138n6, 139n13, 140n14, 160ns, 248n15, nn17–8, 250n57, 255 Itah, Michel 22n86, 110n14, 116n75, 138n6, 139n13, 140n14, 143n41, 144n51, 160nm, no–p, 205n79, 248n12, 255 Jackson-Tal, Ruth E. 143n46, n49, 151n91, 152n91, 159nj Jacobson, David M. 235n116

Index of Modern Authors James, Frances W. 253n117 Jiménez Bedman, Francisco 73n84 Jokiranta, Jutta 84n44 Jones, Vendyl 9n2, 17–8, 19n66, 21 Jull, Timothy A.J. 100n19, 102n32, 178n6 Kadman, Leo 221 Kaliff, Anders 59nn42–3 Kam, Yoni 22n86, 110n14, 116n75, 138n6, 139n13, 140n14, 143n41, 144n51, 160nm, nno–p, 205n79, 248n12, 255 Kampen, John 81n11 Kapera, Zdzisłav 110n24, 197nn11–2, 198n14, 203n60, 224n29, 238n4 Karageorghis, Vassos 253n120 Kelso, James L. 149n77 Kempinski, A. 249n31 Kennedy, Charles A. 116, 116n81 Kessler, Nadine 26n112, 67n6, 72n61, 80n9, 81n10, n17, 142n36, 182n25, 200n31, 207n82, 208n85 Kilikoglou, Vassilis 194nn119–121 Killebrew, Ann E. 153n107, 154n108 King, Philip J. 158nd Kingery, David W. 194n117 Klijn, Albertus F.J. 91n29 Klimowsky, Ernest W. 222n9 Kloner, Amos 58n39, 97n4, 182n24, Knauf, Ernst A. 222n10 Kochavi, Moshe 17n49 Kokkinos, Nicos 235n116 Kool, Robert 221n7 Kraemer, Joel L. 169n41, 182n22 Krzyśko, Miroslaw 188–9, 188n53, 189n57, 191–2, 191nn83–9, 192, 192nn90–93, 194, 194n116, 215, 215n14, nn16–7, n19 Kuhn, Karl G. 163n8 Lacerenza, Giancarlo 97n9 Lacoudre, Noël 148n75 Ladizhinskaya, Victoria 141n17 Lamon, Robert S. 253n115 Lange, Armin 70, 70n41, 74n91, 75, 80n2, 82n20, 95n41, 170n47, 217, 217n33 Laperrousaz, Ernest-Marie 73n76, 145n58, 224n29, 238 Lapp, Nancy L. 16n44, 140n15 Lapp, Paul W. 16, 16n44, 111n31, 115, 115n69, 140n15, 263n6, 280 Larson, Erik W. 29, 29n150, 30, 30n151, 72, 72n60, 142n36 Lawrenz, John C. 234n107 Le Moyne, Jean 90n20 Le Rider, George 221n8, 234n107 Lefkovits, Judah K. 73n75 Leith, Mary J.W. 16n44 Lemaire, André 92n29, 101n30, 103n39, 213, 213n1, 216, 216n24, n27, 271n32, 277na Leonard, Robert D. 224n29 Levana, Tsfania 146n64 Levi, Thomas 18n57 Levison, John R. 83n35 Levy, Brooks 232n100 Libby, Willard F. 134, 134n25 Lichtenberger, Hermann 90n21 Lieberman, Saul 165n17 Lieu, Judith M. 81n11, 150n81 Lim, Timothy H. 83n31 Liphschitz, Nili 158ne

355

Index of Modern Authors Loffreda, Stanislao 113, 113n54, 114n58, 149n77 London, Gloria A. 194n117 Lönnqvist, Kennet 28n141, 222n15, 223nn23–4, 224n29, 225n30, 226n39, 227n45, 228n52, nn54–5, n57, n59–60, 229n67, n70, 230n73, 231nn94–5, 233n105, 234n106, n108 Lönnqvist, Minna 222n15, 223nn23–4, 224n29, 225n30, 226n39, 227n45, 228n54, 229n70, 230n73, 233n105, 234nn94–5, n106, n108 Lubetski, Meir 71n46, 82n23, 104n42 Luca, Massimo 221n2 Luria, Ben Zion 73n79 Luttikhuizen, Gerard P. 74n88 Ma⁡ʾayeh, Farah S. 187n33 Macdonald, Michael 226n34 Maeir, Aren M. 26n111, 80n6, 137n1, 169n42, 207n82 Magen, Yitzhak 4, 21–2, 22n84, 27, 27n124, 29, 29n143, 36, 39n17, 111, 111n28, 135, 135n27, 146n64, 148n75, 149n77, 184–7, 184nn5–6, 185nn7–9, nn11–12, 187nn34–6, 193, 241, 242n16, 247n3 Maggen, Michael 18n56 Magness, Jodi 4, 24n106, 26n111, 68n11, 80n6, 82, 82n19, 110, 110n22, n25, 112, 112nn39–40, nn43–44, 113n53, n57, 137n1, 144n53, 147n68, 150n80, nn82–3, 151n84, 153nn105–7, 154n108, 160no, 169n42, 170n45, 172n55, 186n20, 190n66, 195, 200–1, 204, 204n72, 207n82, 208n87, 213nn2–3, 215n13, 234n108, 238, 238n4 Maier, Johann 84n38 Manns, Frédéric 151n84 Marcos, Carmen 232n100 Mason, Eric F. 68n22, 80n1, 102n34, 145n55, 169n44, 190n65, 196n2, 213n2 Massabeh, Arwa; 130 Massonnat, Gérard 2 Mastin, Brian A. 85n52 Mathews, Kenneth A. 89nn13–5, 90n18, 97n9 Mazar, Amihai 250n52, 251n67, 252n106 Mazar, Benjamin 249n27, 250n43, nn57–8, 252n103 Mazar, Eilat 249n27, 250n43, n58, 252n103 Mazor, Gabriel 138n6, 139–40n13, 143n41, 160nq, 205n79, 248n20, 255 McGovern, Patrick E. 194n117 McLaren, James S. 182n22 McLean, Mark D. 89nn13–4 McNamee, Kathleen 92n31 Meshorer, Ya⁡ʾakov 224n29, 230n83, 231n85, 232nn98–100, 234n107 Metso, Sarianna 67n8, 81n11, 82n18, 84nn45–6, 150n81, 166n26 Meyers, Carol L. 89n15, 97n9 Michniewicz, Jacek 111, 111n33, 113n51, 187n39, 188–9, 188n53, 189n57, 191–4, 191nn83–9, 192nn90–3, nn95–8, 193nn99–108, nn110–2, 194n116, 215, 215n14, nn16–7, n19 Mildenberg, Leo 222, 222n10 Milik, Józef T. 10n3, n13–5, 11n18, 13n22, n27, 14, 14n36, 21, 26–7, 26n114, 27n116, n119, 30–1, 31n154, 36n2, 37n2, 59n41, 69n31, 70n37, 71n50, 72n59, n63, 73, 73n72, n78, nn82–3, n86, 74n93, 75, 75n100, 75nn102–4, 75n107, 76, 76n110, 80n1, n7, 82, 82n25, 84, 84nn45–6, 85n52, 88, 95, 95n44, 97n7, 98n13, 99n18, 100n22, 101n28, 104n44, 105n44, 109n1, n3, 112, 113n55, 115, 123n2, 125n7, 126n10, 127n11, 135, 138n4, n7, 141n20, 145n58, 162nn3–4, 163n4, nn6–8, 164nn13–5, 165n15–6, n20, n22, 166n25, 167n27, 168, 168n33, n35–6, n39, 169nn39–40, 170n46, 171, 171nn53–4, 172n57, 173n58, 177n3, 179nn11–2, 182n26, 186nn21–3, 188n43, 196n3, n4, 197, 197n12, 198, 198n14, n16, 202n45, 208n88, 216n23, 229n66, 235n114, 240n10, 241n15, 247n4, 248n21, 263n4 Millard, Alan 91n28 Mimouni, Simon C. 103n39

Misgav, Haggai 18n56, 61n46, 98n14 Mitchell-Innes, Alfred 233, 233n103 Mittmann-Richert, Ulrike 70n41, 74n91, 217, 217n33 Mizzi, Dennis 1, 3, 6, 9, 24nn105–6, 145n63, 148n75, 153n102, 158nf, 177n2, 180n16, 181n19, 182n23, 247n1, 277nh, 278n55 Młynarczyk, Jolanta 3, 27, 27n125, 29, 29n144, 98n13, 110n18–9, n22, n25, 112n46, 114nn61–2, 115n66, 116n74, 116n77, 242, 242n20, 243n22 Mommsen, Hans 189n58, 194n114, n118 Montaner, Vegas L. 162n3 Morgenstern, Matthew 163n8, 169n39 Morrisson, Cécile 221n8 Moulhérat, Christophe 126, 131–2 Mumcuoglu, Kosta Y. 154n109 Murphy, Catherine M. 224n29 Nahmias, Keren 142n30 Nakman, David 164–5n15, 167nn27–8, 168n35, n38, 172n57 Naor, Eyal 21n77 Naveh, Joseph 90, 90n20, 217n30, 234n107 Neff, Hector 194n115 Nenner-Soriano, Ravit 158nd Netzer, Ehud 19n65, 88n5, 111n30, 113n54, 238, 243, 244n27 Newell, Edward T. 226, 226n37 Newman, Judith H. 83n35 Newsom, Carol 85n105 Nickelsburg, George W.E. Jr. 140n15 North, Robert 38, 38n8 Notis, Michael D. 194n117 Nys, Karin 253n119 O’Connor, Michael 89n15, 97n9 Oren, Eliezer D. 116n84, 253n116 Oron, Gershon 138n6, 139n13, 140n13, 160nt, 205n79, 248n11, n19, 255 Otero, Paloma 232n100 Overman, Andrew J. 181n22 Oziol, Therese 253n121 Pajunen, Mika S. 84n44 Palleschi, Vincenzo 153n102, 158nf Panitz-Cohen, Nava 250n52, 251n67, 252n106 Parry, Donald W. 22n89, 137n1, 138n8, 150n81, 170n45, 199n24 Patrich, Joseph 18–22, 18nn60–1, 19nn62–4, nn67–8, nn70–71, n73, 20nn74–5, 21n77, 22, 29, 29n148, 36, 109, 109nn12–3, 110n15, 113n53, 117, 138, 138n5, 139n9, n13, 140nn13–4, 144n51, 145, 145n59, 151, 151n84, n86, n90, 152n92, nn94–9, 159nj, 160nj, nnm–n, 170n45, 181n18, nn20–1, 202–4, 202n45, nn47–52, 203nn54–9, 204n71, n74, 205, 205n77, 206, 207n83, 208nn85–6, 247, 247n11, 248n12, 255 Peacock, David P.S. 185n10 Peleg, Yuval 4, 21, 22n84, 27, 27n124, 29, 29n143, 39n17, 111, 111n28, 135, 135n27, 148n75, 149n77, 184–5, 184nn5–6, 185nn7–9, nn11–2, 186, 193, 241, 242n16, 247n3 Peskowitz, Miriam B. 154n108 Pfann, Stephen J. 5n1, 13n23, 18n61, 33, 68–9, 68nn17–21, 71n49, 74, 76, 76nn115–6, 78, 78n129, 81n10, 82n20, 83n30, 88, 88nn7–11, 100n24, 109n1, n12, 110n15, 114nn61–2, 115n66–8, 116n77, 137n1, 145, 145n61, 148n75, 164n12, 196nn2–4, 197n6, nn8–9, n12, 198nn21–3, 214n8, n11, 224n29, 241n13, 242, 242n18, 243n24 Piccirillo, Michele 221, 221n2, 223, 223n19, 231n92 Plenderleith, Harold J. 126n9 Politis, Konstantinos D. 39n15

356 Pollak, Frank S. 166n26 Poole, John B. 162n4, 164n14 Popovic, Mladen 4, 9, 9n1, 11n19, 25, 25n107, 26n111, 30, 30n152, 67n1, 68n9, 69, 69nn27–9, 72, 72n61, 77–8, 77n122, 78n128, 80, 80n1, n9, 81n10, 82n20, 105n44, 137n1, 142n36, 143n43, n45, 150n81, 153n103, 169n44, 171n49, 172n55, 177, 177n1, n3, 178nn4–5, 180nn13–5, 181n17, 182nn22–3, n25, 183nn27–9, 198n15, 205n76, 207n82, 208n86, n88, nn90–1, 209nn93–4 Porat, Roi 16n47, 23, 23nn97–102, 29n145, 141n24, 201, 235n115 Porter, Stanley E. 81n11 Prag, Kay 234n107, 250n42 Price, Randall 28, 28n131, 224n29 Puech, Emile 3, 6, 73n74, n77, 76–7, 76n111, 77n118, 84–5, 84nn45–47, 85n48, 90n20, 96n3, 97nn8–9, 98n13, 100n19, n21, n25, 102nn33–4, 103nn37–9, 105n46, 179n9, 216n21, n25, 229n71, 271n32 Qimron, Elisha 84n45 Rabin, Chaim 99n15 Rabin, Ira 71n47 Rahmani, Leir Y. 142n27, 234n107 Rappaport, Uriel 116n84, 145n58 Rasmussen, Kaare L. 28n133, 134n25, 178nn7–8, 244n28 Ravetz, Alison 233 Reece, Richard 234n107 Reed, Ronald 162n4, 164n14 Reed, Stephen A. 162n2 Reed, William L. 11n21, 197n6, n8 Reeves, John 81n11 Regev, Eyal 81n11 Reich, Ronny 87n1, 103n40, 154n108, 158nd Reifenberg, Adolf 223n23, 229–30, 230n79 Riklin, Shimon 146nn66–7 Roberts, Charles 77 Rofé, Alexander 72, 72n66 Rohrhirsch, Ferdinand 27n123, 87n3, 148n75, 224n29, 231n84, nn86–7 Roitman, Adolfo D. 81n11 Rosenthal, Ferdinand 89n14 Rosenthal-Heginbottom, Renate 115–6, 115n73, 116n78, n82, n85 Rothschild, James A. 253n113 Rothstein, David 165n22 Russell, Kenneth W. 27n128 Rutgers, Leonard 153n106 Safrai, Shmuel 92n29 Sagiv, Nahum 142n29 Salles, Jean-François 111n31 Sanders, James A. 277n49 Sandhaus, Debora 138n6, 139n13, 140n13, 143n41, 160nq, 205n79, 248n20, 255 Satran, David 78n132, 82n20 Schäfer, Peter 58n39, 90n21, 182n24 Schärli, Beatrice 222n11 Schick, Tamar 22n88, 158nd, 206n80 Schiffman, Lawrence H. 26n111, 68n11, 80, 80nn1–2, n4, n6, 81n11, 88n11, 95n42, 137n1, 145n59, 166n26, 169n42–3, 170n45, 172n57, 181n18, 204n71, 207n82 Schofield, Alison 69, 69n32, 81, 81n11, n13, 150n81 Schuller, Eileen 71n55, 85n52, 166n26 Segal, Michael 163n8, 169n39

Index of Modern Authors Seibt, Werner 221n8 Sellers, Ovid R. 265n12 Seyrig, Henri 5, 11n18, 221, 221n8, 223–4, 223n25, 224n27, 226–8, 227n45, 228n49, 228n55, 229n66, 229n70, 235n14 Shahar, Yuval 58n39, 182n24 Shaked, Idan 143n48 Shamir, Orit 5n1, 24n104, 105n45, 129, 131, 134, 134n26, 137n2, 143n48, 163n5, 201n43, 206n80 Shanks, Hershel 254n125 Sharabani, Marcia 223, 223nn19–20, n25, 224n29, 226–7, 226nn40–1, 227n42, n45, 228n55 Shiloh, Yigal 251n84 Shipton, Geoffrey M. 253n115 Sillar, Bill 194, 194n115, n117 Singer-Avitz, Lily 250n54, 252n95 Sion, Ofer 143n49, 146nn66–7 Sivan, Renee 115, 115n70, n73 Smith, Robert H. 110–1, 110n23, n25, 111n27, n32, nn40–1, 113–6, 113n49, n53, n56, 114n64, 115n65, n69, 116n80 Snyder, Gregory H. 93n33 Spijkerman, Augustus 5, 221–3, 221n4, n6, 222n13, n15, 223n23, 224n27, n29, 227n42, 229n70, 230, 230n72, nn76–8, 231nn93–4, 236–7 Stacey, David 205n78, 238, 238nn1–2, 242–3, 242n19, 243n27, 244n27 Stager, Lawrence E. 26n113, 158nd Starcky, Jean 229, 229n71, 230n72 Stegemann, Hartmut 70, 80n3, 82n20, 87, 87n2, 143n39, 164n11, 206n81 Steiner, Margreet L. 250n60 Stern, Ephraim 18n57, 116n82, 141n22, 234n107, 235n114, 249n26 Stern, Menaḥem 92n29 Stern, Sacha 83n31 Steudel, Annette 80n2, 84, 84n40, 170n47 Stiebel, Guy D. 20, 20n75, 160no Stökl Ben Ezra, Daniel 3, 67–9, 67nn5–8, 68n10, 73, 74n87, 76–8, 76n114, 77n117, n119, n121, 78n130, n133, 80n3, 81, 81n10, n18, 82n18, n21, 89n12, 94–5, 94n38, n40, 104n44, 137n1, 145, 145n55, n61, 147n68, 153n105, 169n43, 170, 170n48, 171n51, 181n18, 208, 208n85, n90, 217n31 Strange, James F. 214, 214n6, 216 Strugnell, John 14, 69n31, 80n1, 100n26, 145n58 Stutchbury, Howard 16 Sukenik, Eliezer L. 68, 68n12, 169n42 Sukenik, Naama 5n1, 24n104, 105n45, 131, 134, 134n26, 137n2, 201n43 Sussman, Varda 113n53 Svennson, Jan 59nn42–43 Sydenham, Edward A. 226, 226n38, 228–9 Tal, Nir 138n6, 139n13, 140n13, 160nt, 205n79, 248n11, 248n19, 255 Tal, Oren 149n77, 187n32 Talmon, Shemaryahu 82n26, 83nn29–30, 90nn22–3, 96n2, 104n41 Taylor, Joan E. 1, 10n6, 23n92, n95, 25n108, 26nn111–2, 27n115, n121, n127, 29n149, 68, 68n11, nn14–6, 69, 78, 78n127, 80, 80nn6–7, 82n20, 90n19, 109n11, 123n4, 134n25, 137n1, 142n34, n37, 143n37, nn39–41, 145, 145n55, n61, 148n76, 149n77, 150n82, 151n85, 152n99, 152n101, 153nn106–8, 154n108, 158nf, 169n42, 172n55, 199–200, 199n31, 200n32, 202n48, 205n75, 207n82, 209n92, 215n12, 247n10, 265n12 Tebes, Juan M. 252n101 Tepper, Yigael 58n39 Tervanotko, Hanna 84n44 Testa, Emmanuele 97n8

Index of Modern Authors Thiering, Barbara 95n42 Thomas, David W. 90n19 Thomas, Samuel I. 86n55 Tigchelaar, Eibert J.C. 67n8, 69–70, 69n34, 70n40, n42, 71n45, n48, 72n55, 74, 75nn95–6, 79, 79n141, 80n2, 81n11, 82n18, n20, n27, 98n13, 163n9, 166n26, 170n47, 216n21 Tournay, Raymond J. 10n4, 223nn18–9 Tov, Emanuel 2, 3, 69n33, 78, 78n132, 82n20, n22, 84n45, 85n52, 88n11, 90n21, 91n26, 92n30, 93n35, 95n42, 96n1, 145n59, 166n26, 168, 168n34, 170n45, n47, 173nn58–60, 181n18, 187n38, 196n3, 204n71, 217n33, 244–5, 244n29 Trebolle Barrera, Julio C. 162n3 Treu, Kurt 92n29 Tsafrir, Yoram 142n31, 182n24 Tushingham, Arlotte D. 234n107, 250n58, 252n103 Tzoref, Shani 81n11, 238n2 Ulrich, Eugene C. 22n89, 138n8, 150n81, 170n45, 199n24 Ussishkin, David 250, 250n39, n45 Van Beek, Gus W. 227 Van der Plicht, Johannes 244n28 Van der Woude, Adam S. 71n45, 82n27, 163n9 Van Lessen, Michael 229n66 VanderKam, James C. 83n29, 88n11, 95nn41–3, 102n32, 145n59, 166n26, 170n45, 181n18, 204n71, 238n1 Vanonen, Hanna 84n44 Vaughn, Kevin J. 194n115 Vaux, Roland de 1, 5, 9–14, 10nn4–5, nn7–8, nn10–1, n13, n20, 13n22, nn24–8, nn30–1, 14n32, nn34–5, n37, 19, 19n72, 21, 26–31, 27n120, nn122–3, 28n129–30, n132, n136–7, n141, 29n142, n144, 31nn155–7, n159–60, 36–7, 36–7n2, 40–1, 41n21, 42n23, 43, 43n25, 46, 46nn26–8, 48–9, 48nn30–2, 49nn33–4, 51–2, 55, 57–8, 58nn37–8, 59n41, 61, 61n47, 67, 67n3, 69, 71n50, 72–3, 72n59, 73n85, 75–8, 75n101, 76n108, 77n125, 78n126, nn134–5, 80n1, n7, 82, 82n25, 87n1, 95n44, 96n1, 97n7, 98n13, 101, 101n28, 104–5nn43–4, 109–12, 109n1, nn3–10, 110nn16–7, n21, 111n36, 112n40, nn43–4, 113n55, 114–6, 114nn59–62, 115nn66–8, 116n77, n88, 123, 126, 126n10, 127nn11–3, 133–6, 133n24, 138–9, 138n4, n7, 139n9, n13, 140n14, 141nn20–2, 142nn32–3, n35, 143n41, n45, 144, 144n53, 145n55, 147nn68–9, 148n75, 149n77, 150n80, nn82–3, 151, 151n87, n89, 152n100, n102, 153n105, 158nnd–f, 159nnf–j, 160nnk–m, 162nn3–4, 163nn4–5, nn8–9, 164nn13–4, 169n44, 170nn45–6, 171, 171n51, n53, 172n55, n57, 177n3, 179n11, 181nn20–1, 184, 184nn3–4, 186, 186nn18–9, nn21–5, n27, 187n32, 188nn42–3, 190n63, 193n105, nn107–8, n110, 195–206, 196nn3–4, 197nn5–6, nn8–12, 198n14, n16, n18, nn20–3, 199n23, nn26–7, n29, 200n34, 201n40, n44, 202n53, 203n59, 205n76, 208, 208nn85–8, 213n3, 214, 214n7, n10, 215n12, 216n23, 221, 222n14, 223n23, 224n29, 229nn69–70, 230n75, 231n84, nn86–7, n94, 238, 240–4, 241nn14–5, 242n17, 243nn24–5, 247–9, 247nn3–9, 248n16, 248nn21–2, 249n33, 253, 255, 263–4, 263nn3–4, 263n6, 264n12, 266, 266nn19–21, 267nd, 268, 270n24,

357 nn27–8, n30, 271–2, 271nn32–3, nn36–7, n43, 272n46, 277–80, 277nc, nn48–50, n53, 278n56, nn59–60, 280nn62–4 Vermes, Geza 84nn45–7, 85n51, 239n6 Viers, Rina 97n9 Voegtli, Hans 222n11 Vörös, Győző 61n46 Wagemakers, Bart 150n82 Walker, Dennis 197n7, 198n16, 199, 199nn30–1, 202–4, 202n48, 203n60–2, 204nn63–6, 208n89 Wampler, Joseph C. 252n90, 253n118 Wartenberg, Ute 222n11 Wassén, Cecilia 183n29 Webster, Brian 94, 94n39, 96n1, 100n19, 102n31, 104n44 Wechsler, Michael G. 169n41, 182n22 Weinberg, Gladys D. 140n15 Weise, Manfred 83n33 Wexler, Lior 248n13–4, n20, 255 White Crawford, Sidnie 5, 62, 68, 68n22, 69, 69n23–6, 80n1, n8, 82n20, n23, 145n55, 169n44, 172n55, 183n29, 190n65, nn72–3, 196n2, 199n28, 200n34, 207n82, 208n91, 213n2 Williams, David F. 185n10 Wise, Michael O. 18n60, 76, 76n113, 109n12, 110n22, 138n5, 148n75, 169n41, 170n45, 181n18, 182n22, 202n45, 224n29, 239, 239nn5–6 Wolff, Samuel 140n16, 143n42 Wolters, Albert 73n84 Wood, Bryant G. 145n58 Wouters, Helena 148n75 Wright, George E. 99n15, 240n11, 253n118 Yadin, Yigael 14–5, 15n39–40, 18, 59n40, 90n22, 96n2, 99n15, 113n55, 116n87, 127n14, 142n26, 162n3, 163n4, nn6–8, 164n14, 165n16, 169n40, 178, 187, 187n38, 234n107, 243, 253n113 Yardeni, Ada 3, 30, 30n153, 71, 71n46, 82, 82n23, 93n34, 99n15, 100n23, n26, 101n27, 104, 104n42, 162n3, 163n5, n8, 164n14, 167n29, 168n39, 169n41, 216n21, n25, 217–8, 217nn28–9, n32, 218n34 Yellin, Joseph 187–92, 188nn40–1, nn43–6, nn48–50, 189nn60–2, 191n77, 192n94, 194 Yezerski, Irit 249n26, 250n41, n49, 251n62, n74, n82, 252n102, 253n123 Yisraeli, Yigal 138n6, 140n14, 143n41, 160nr Young, Ian 238, 238n2, 245 Zangenberg, Jürgen K. 1, 4–5, 23n92, 27n126, 39n15, n17, 109n11, 111n28, 112n40, 143n37, 148n75, 149n77, 184n5, 200n36, 205n78, 207n82, n84, 209n93, 214n6, 215n12, 238n1, 271n43 Zayadine, Fawzi 111n31 Zias, Joseph E. 28, 28n135, 123 Ziegler, J. 11–2, 14, 21, 33 Zimhoni, Orna 249n31, 250n45, n47, n51, 251n68, n73, n78, n81, nn85–86, 252n94, n105, n108, n110 Zissu, Boaz 18, 18n58, 22, 22n87, 29n146, 141nn17–9, 142n31, 143nn43–4, nn46–7, 146n66, 182n24, 235nn111–2, n116

Index of Sites and Place Names Abila 187 Abu Saraj, Jebel 146n66 Cave IV/12 140, 143, 143n42 Cave VI/52 143 Abu Tabaq, Khirbet 26–7 Akko 252 Alexandria 91, 133 Almog 19, 138 Amathus 253 Amman (Philadelphia) 3, 111, 123, 126, 131, 134, 226–8, 226n41, 227nn42–3, n46, n48 Andes 194 Antioch 19, 152, 160nj Arad 249–52 ʿArugot, Naḥal 14, 23, 191 Ashkelon 11, 27, 222, 230 Asia 136 Bab edh-Dhra⁠ʾ 38n4, 39n15 Bahariyah 41n21 Batash, tel (Timnah) 250–2 Beersheba 249, 251–2 Beirut 228 Beni Naim 40 Bethlehem 27, 125n8 Beit Mirsim, Tell 249, 251–3 Beit ʿUmmar 190 Beth Shean 114, 253 Beth Shemesh 253 Beth Zafafa 183 Beth Zur 111 Bosra 38 Buqei⁠ʾa, el- 24, 26–27, 29 Burj el-Bahar 39 Caesarea 27, 91, 234n107 Christmas Cave 16–8, 23–4, 59, 201, 201n43 Cilicia 116 Cnidus 114 Coin Cave 17 Cypros 188 Cyprus 116, 194n117, 253 Dayr al-Qattar 59 Daliyeh, Wadi ed- 16, 97–8 Cave 2 (Araq el-Na‛asaneh) 16, 18, 140 Damascus 38, 41n20 Dana, Wadi 39 Darajeh, Wadi 18, 27 David, Naḥal (Wadi Sdeir) 14, 90–1, 178 Cave of the Pool 14, 141, 141n23, 248–50, 252, 254–5, 259 Har Yishai Cave 23, 141 Dead Sea (Lac Asphaltite) 1–2, 4–5, 9–10, 13–7, 24–8, 31–2, 37–41, 41n 21, 47, 59, 61, 95, 110, 125, 136, 138, 148, 148–9n77, 177, 182, 185, 187–8, 191, 194, 196, 200, 205, 215n19, 235, 247, 249, 251, 253–4 Dura Europos 216n23 Edom 191, 215–6 Egypt 87, 91, 100–1, 111, 114

ʿEin Arrub 142 ʿEin Boqeq (Umm al-Baghaq) 38, 148n77, 185, 187, 187n32 ʿEin el-Ghuweir 17, 24, 38–9, 38n5, 91, 183, 188, 203–4, 251, 251n63 ʿEin et-Turaba 17, 24, 38–9, 205, 251, 251n63 ʿEin ez-Zara (Callirhoe) 38–40, 38n3, 39n11, 185, 187–8, 187n32, 192–3 ʿEin Feshkha 10–2, 14, 16–17, 19, 21, 24, 26–29, 38, 38n3, n5, 39n10, 41n21, 42, 87, 101, 109n10, 110, 112, 123, 127, 133–4, 138, 148n77, 185, 187–8, 187n32, 190, 203–5, 221, 223n19, 230, 243–4 En Gedi (Tel Goren) 1–2, 5, 14, 16, 18, 23–4, 26–7, 32, 38–40, 38n5, 41n20, 148–9n77, 158nnd–e, 182–3, 185, 191, 234n107, 235, 249–54 En Raḥel 158ne Ephesus 91, 114 Europe 231 Farʾah, Wadi 21 Far⁠ʾah, Tell el- 253 Farafra 41n21 Feinan 39 Galilee 58, 116, 185, 185n14, 203, 232 Gamla 19, 234n107 Gerizim 97 Gezer 249–52 Ghor es-Safi 38, 40 Goded, Tel 142 Golan 186 Habibi, Wadi el- 21 Ḥammam, Wadi 235 Hananya, Kfar 185–6 Hasa, Wadi al- 38 Ḥazor 252 Hebron 40, 58, 149n77, 188, 190–2, 215, 229n66 Herodium 19, 27, 39, 113, 149n77 Ḥever, Naḥal (Wadi Habra) 14, 16, 18–9, 21, 64, 88, 91, 93n34, 95, 142, 162n3, 163–4, 163n8, 168, 169n39, 172, 178–83, 209, 240 Cave of the Letters (Cave 5/6) 14, 59, 116, 141, 160no, 178, 180 Cave of the Horrors (Cave 8) 14, 16, 59, 142, 178 Cave of the Tetradrachm 18, 29, 142 Hyrcania 26, 38n3, 39 Idumaea 116 ʿIra, Tell 249, 251 Israel 14, 85, 87–8, 90, 92, 113, 154n109, 158nd, 163, 185, 213 Jericho 1–2, 16, 18, 21–2, 24, 26–7, 28n140, 32, 37–8, 40, 42, 47, 91, 98, 110–1, 113, 117, 143, 149n77, 154n109, 158nd, 187–8, 190–4, 198, 200, 205, 215, 235, 238, 242–4 Jericho, Ketef 22, 24, 146n66, 177 Cave of the Sandal (Cave VIII/28) 18, 29, 141, 235 Abi’or Cave 18, 18n58, 22, 141, 143, 177, 178n4 Large Cave Complex (Cave VIII/9) 143 Jerusalem 21, 27, 38, 39n10, 40, 68, 73–4, 91, 96, 100n25, 101, 105, 110, 113, 113n51, 115–7, 149n77, 158nnd–e, 183, 185, 187–92, 194, 218, 222–3, 225–7, 225n30, n36, 227n42, n44, 229–30, 235, 245, 249–52 Binyanei Ha‌ʾuma (Givʿat Ram) 185–6 City of David 250–2 Kidron Valley 27 Ophel 249

359

Index of Sites and Place Names Jib, el- 191, 194 Jordan River 10, 111 Jordan Valley 123 Jordan 187, 221, 227 Judea 3–4, 16, 25, 35, 37–40, 47, 59, 112, 116, 137, 140, 147n68, 150, 183, 185, 188, 192, 194, 213, 215, 215n14, 232–3, 239, 243, 245, 249. Judean Desert 2–4, 9, 20, 24, 32, 37–8, 40, 59, 68–70, 87–8, 90–2, 95–7, 138, 139n9, 140, 143, 146, 147n68, 150, 150n84, 151n91, 153n106, 154, 158nnd–e, 160nnj–k, nm, 161, 163–6, 168, 177, 179–180, 182, 205, 244–8, 251, 253 Judean Hills 249, 251, 253 Kadesh Barnea 249–52 Kerak, Wadi 38 Khabra, Wadi 97, 105n44 Kidron (Wadi en-Nar) 10, 16, 21, 23, 39n10, 40, 91, 201 Kufrayn, al- 111 Lachish 249–251 Limassol-Kommissariato 253 Lisan 38, 48, 59–61, 191 Lugano 136 Machaerus 38, 38n3, 39n10, 40, 60, 113, 149n77 Mafjar, Wadi el- 18, 177 Makkuk, Wadi el- 18, 146n66, 235 Cave III/8 143 Cave of the Warrior 22 Maoza 182 Maqari, Khirbet 26 Mareshah (Tell Sandahannah) 97, 114n41, 116 Marrazah, Wadi Caves of the Spear 23, 29 Mazraa 38–9 Mar Saba 10 Masada 19, 27, 39, 41n20, 59–61, 88–91, 95–6, 101, 104–5, 111–2, 115, 149n77, 158nnd–e, 181–2, 185, 185n9, 187–8, 191, 200, 209, 234n107, 242–4, 243n27 Masos, Tel 249–52 Matsia, Wadi 18 Mazen, Khirbet 17, 38–9, 39n10, 192–3, 199, 205 Megiddo 249, 253 Meẓad Ḥasidim 27 Michmash, Naḥal (Wadi Suweinit) 21 Magharat el-Jai Cave 141, 143, 235 el-ʿAleiliyat Caves 21 Migdal 234n107 Mishmar, Naḥal 91 Cave of the Treasure 16, 141 Miẓpe Shalem, Kibbutz 27, 205 Modiʿin 142 Mo’a 158ne Motza 39, 149n77, 186, 188, 190–2, 215n14 Meḥarat Ha-Teḥomim (Mughâret Umm et Tûeimîn) 235 Murabbaʿat, Wadi (Naḥal Darga) 2, 9–11, 13–4, 16–17, 21, 23–7, 27n118, 32, 59, 91, 95–6, 105n44, 123, 141, 141n22, 143, 143n45, 162n3, 163, 163n8, 164n11, 168, 168n39, 169n41, 171–2, 177–182, 203–4, 208–9, 208n88, 235, 240, 248–50, 252–4, 259 Cave 1 10, 113n55, 116, 141, 141n20, 163–4, 164n14, 166, 172, 248, 255, 258 Cave 2 10, 25, 30, 141, 141n21, 248–50, 253, 255, 257–8 Cave 3 141n22, 248, 255, 257–8 Cave 5 10, 141n22

Nabatea 39, 191, 215 Naṣbeh, Tell en- 111, 252–3 Near East 225 Negev 249–51 Nessana 158ne Neuchâtel 221n8 Nicosia 253 Niran 205 Oboda 158ne Palestine 27n118, 58, 92n30, 128, 130, 146n63, 186, 221, 222n10, 225, 232, 252–3 Paris 221n8, 226 Pergamon 91 Petra 27 Pheretae 21 Phoenicia 116 Qalandiya 149n77, 187 Qalia, Kibbutz 19 Qarantal Monastery 18, 22 Qaṣr al-Abiadh 59–60 Qaṣr al-Tuba 59–60 Qaṣr el-Yahud 251, 251n63 Qazone, Khirbet 39, 39n15, 59, 183 Qelt, Wadi 16 Qitmit, Horvat 249 Qôm, Khirbet el- (Maqqedah) 97n4 Qumran 1–6, 10–1, 13, 16–9, 21–30, 32, 34–40, 38n3, 39n10, 42, 47–9, 55, 58–62, 73–7, 80, 87, 89, 90n18, 91n25, 92n30, 93–9, 97n9, 101, 100n23, 103n40, 105, 107–10, 115–8, 122–3, 133–4, 136, 141, 143–4, 145n58, 146, 147n68, 148–54, 149n77, 153n107, 159nf, 160nm, 162n4, 163–4, 164n11, 169, 169n43, 172–3, 172n57, 173nn59–60, 178–80, 183–5, 187–91, 193–7, 197nn6–7, 200–9, 200n32, 201n39, 213, 213n4, 215, 216n20, 217–8, 223n19, 238–46, 248, 263n6, 265n12 Khirbeh, Settlement, Site 3–5, 9–11, 13, 22, 25–6, 37, 48, 61, 63, 67n3, 76, 80, 82, 89, 92n30, 96, 96n1, n3, 97n8, 98, 101, 104–5, 105n44, 109n10, 110–4, 117, 134, 137–8, 144–5, 147–8, 147n68, 149n79, 150–3, 152n99, 158nd, 160nk, 169, 169n44, 170n47, 177, 183, 186, 189–91, 193–7, 199–200, 202–6, 208, 213, 215, 218–9, 221–5, 230, 232, 235, 240, 247–51, 253–4, 265n12, 271n40, 277 Caves 1–5, 9–10, 18, 22, 24–5, 29, 34, 42, 61–3, 73, 79, 81, 94, 96, 99, 109–10, 109n10, 112, 117, 122–3, 126, 128, 131, 137–40, 137n2, 142–51, 144n50–2, 147n68, 149n77, 151n84, 152–5, 153n106, 158nd, 161–2, 164, 164n14, 166, 169–70, 169n40, 170n45, 173n60, 177, 179–83, 195, 200, 238–40, 242, 244–7, 244n28, 249, 254, 263 Qumran, Region/Area 2, 13, 122, 138, 142, 144–5, 144n50, n54, 145n58, 147n68, 148, 151–3, 155, 181, 199, 204–5, 209n92, 224 Qumran, Wadi 11, 13, 18, 28, 30, 47–51, 58, 61, 138, 142, 193–4, 215n19 1Q (GQ14, Ain Feshkha cave) 2–4, 9–11, 13, 22, 24–6, 42–4, 42n22, 46, 61–6, 74–9, 81–4, 84n43, 89, 92–5, 104, 104–5n44, 109–12, 109n1, 114, 117, 125–31, 134–6, 139nn12–3, 140nn13–14, 151–2, 152n101, 154–5, 159ng, 162–6, 162nn3–4, 163n8, 164nn13–4, 169, 169n40, nn42–3, 171–2, 171n51, 173n58, 181–2, 184, 186, 188, 190–1, 196–7, 196n3, 197n6, 199, 205, 207n81, 208–9, 213, 219, 239–43, 246, 263–4, 263n5, 265n12, 266, 268, 270, 270n27, 272 2Q (GQ19, F5) 3, 11, 13, 42–3, 61–2, 71–2, 74–5, 77, 81, 89, 94, 104, 140n14, 164n12, 190, 196, 199, 208–9, 213, 264n8, 270–1, 278 3Q (GQ8, G8) 3, 11, 13–4, 19, 22, 24–5, 42–3, 45, 67–8, 70, 72–5, 77, 81, 94, 104, 109, 109n6, 124, 133–4, 139n13, 140n14, 160nm, 164n12, 186, 188, 190–1, 193, 196–7, 199, 202–3, 203n58, 205, 208–9, 208n85, 213, 247–8, 270, 271n38

360 Qumran (cont.) 4Q (4Qa, 4Qb) 2–5, 9, 13, 15, 22, 26, 29–32, 31n156, n159, 47–9, 51–62, 67–9, 71–2, 71n45, n49, 74–86, 74n92, 84n43, 88–9, 92–5, 94n40, 104–5, 104–5n44, 109–10, 109n2, 117, 136, 139n13, 142, 147n68, 161–6, 162nn3–4, 163n4, n6, n8, 164nn13–15, 165n20, 168–72, 168n34, n39, 169n40, n43, 170nn45–6, n48, 171n53, 173n58, 181, 186, 188, 190, 196, 198, 208–9, 209n92, 213, 215n19, 217–8, 217n31, n33, 220, 242, 246, 264n8, 268, 270–2, 270nn29–30, 271n33, 277n50, 280 5Q 4, 13, 22, 30, 48–9, 51–3, 55, 57, 67–9, 75, 77–8, 81, 84n43, 94, 94n40, 139n13, 162–4, 163n6, 164n14, 166, 169, 171–2, 181, 186, 196, 198, 205, 209, 209n92, 213, 264n8, 268, 270, 270nn29–30, 271n33, 272 6Q (GQ26) 3, 13, 42, 67–8, 74–7, 81, 89, 94, 104, 186, 188, 188n43, 190, 196, 198, 208–9, 213, 220, 264n8, 268, 270, 270n30, 271n33, n36, n38 7Q 3, 13, 22, 29–30, 48–51, 55, 58, 67–9, 77–8, 94, 109, 109n3, 139–40n13, 142, 148, 169, 172n55, 181, 186, 186n18, 188–90, 196, 198, 205, 209, 209n92, 213, 220, 264n8, 270–2, 270n29, 271n33, 278n58 8Q 3–4, 13, 22, 29–30, 31n159, 48–51, 55, 58, 67–9, 75, 77–8, 94, 104, 109, 109n4, 131–4, 139–40n13, 142, 147n68, 148, 162–3, 163n4, n8, 164–66, 164nn13–4, 168–9, 168n39, 169n40, 171–2, 172n55, 173n58, 181, 186, 186n18, 188, 190, 196, 198, 204–5, 209, 209n92, 213, 220, 246, 264n8, 270, 270n29, 271n33, 272 9Q 13, 22, 29–30, 48–9, 51, 55, 58–9, 67–9, 77–8, 94, 139–40n13, 142, 147n68, 148, 169, 172n55, 177, 181, 186, 186n18, 190, 196, 198, 205, 209, 209n92, 213, 220, 247, 255, 259, 264n8, 270–1, 270n29, 271n33 10Q 13, 22, 30, 48–9, 51–2, 55, 67–9, 78, 109, 109n5, 139–40n13, 142, 181, 186, 186n18, 190, 196, 198, 205, 209, 209n92, 213, 220, 264n8, 270–1, 270n29, 271n33 11Q 1–4, 14, 16–9, 22, 24–6, 29, 32, 34–5, 42, 46, 61–2, 67nn3–4, 71–2, 74–5, 77–9, 83, 84n43, 89, 94, 94n40, 104, 110, 117, 124, 131–4, 139, 139–40nn12–3, 144n51, 151–5, 152n102, 153n105, 158–9nf, 163–4, 163n9, 164n12, 170nn45–6, 171, 171n51, 177, 181, 181n18, 186, 190, 196–7, 199, 202–3, 203n58, 208–9, 208n85, 213, 220, 247, 252, 255, 257–9, 263, 263n4, 265n12, 270–1, 271n33, 277–8, 277n50, 280 GQ1 42, 139n13 GQ2 (P13?) 19, 42, 109, 160nm, 248, 255, 280n61 GQ3 (A8) 13, 42, 139n12, 154–5, 158nd, 271, 271n33, 278–9 GQ4 13, 139n12, 155 GQ6 13, 139n13, 247, 255 GQ7 42 GQ8 see 3Q GQ9 109, 109n7 GQ12 42, 125, 134, 139n13, 147n68, 197–8, 205 GQ13 13, 247, 255 GQ14 see 1Q GQ15 13, 42, 46–7, 155, 159nh GQ17 13, 42, 109, 109n8, n10, 139n12, 155, 159ni, 202, 271n33 GQ19 see 2Q GQ20 (XI/20) 134, 139–40n13 GQ21 42 GQ22 42 GQ23 13, 28, 139n13 GQ24 139n13 GQ26 see 6Q GQ27 13, 247, 255 GQ28 42, 271n33 GQ29 (Timothy’s Cave) 42, 46, 109, 109n9, 128, 134–136, 140n14, 181, 197–8, 197n12, 198n14, 205 GQ30 42

Index of Sites and Place Names GQ31 42, 139n13 GQ32 12, 42 GQ33 12, 13 GQ34 13 GQ36 13 GQ37 (P37, FQ37) 12–4, 19–21, 24, 30, 42, 139, 139nn12–3, 140n14, 144n51, 151–2, 156, 159–60nj, 181n21, 202–3, 205, 247, 255 GQ38 13 GQ39 13, 42, 127, 247, 252, 255, 271n34 GQ1-GQ40 270–2, 278n58 Cave A (de Vaux) 13, 139nn12–3, 154, 156, 160nk, 247, 255, 264n8, 268, 270 Cave B (de Vaux) 13, 154, 156, 247, 255, 264n8, 268, 270 X (de Vaux) 270 Cave A (Broshi and Eshel) 22 Cave C (Broshi and Eshel) 22, 204 Cave D (Broshi and Eshel) 148 Cave E (Broshi and Eshel) 22 Cave F (Broshi and Eshel) 22, 204 Point 1X (Broshi and Eshel) 22 Point 2X (Broshi and Eshel) 22, 148 Cave X/35 (see P13)  Cave X/42 139n13 Cave X/47 255 Cave X/51 (Cave of the Leather) 139n12–3, 157, 159ni, 160no Cave X/57 (Cave of the String) 139n13, 157, 160np Cave X/60 139–40n13 Cave XI/6 139–40n13, 206 Cave XI/7 139n13 Cave XI/10 206 Cave XI/14 139n13 Cave XI/15 248, 255 Cave XI/16 139–40nn12–3, 157, 160nq Cave XI/18 139n13 Cave XI/20 see GQ20  Cave XII/49 139n13, 140n14, 248–9, 255, 259 Cave XII/53 140n14, 157 Cave XII/56 157, 248, 250, 255 Cave XII/61 139n13 Cave XIII/2 139–40n13, 145n13, 157, 255 Cave XIII/3 (see GQ37) Cave XIII/13 248, 251, 254, 255 Cave XIV/55 248, 254, 255 Cave XIV/56 248, 254, 255 P13 (Cave X/35, Cave of the Balsam Oil Juglet, GQ2?) 3, 19, 42, 109–10, 112, 115n66, 116–7, 139nn12–3, 140nn13–4, 144n51, 151–2, 156, 160nm, 202, 235, 248, 255, 259, 280n62 P24 3, 18, 19n66, 22, 29, 110, 117, 139–4nn12–3, 140n14, 144n51, 151–2, 154, 156, 160n n, 202 P37 see GQ37 Cave of the Column (Twin Cave) 9n2, 17–8, 21–2, 25 Ibrahim’s Caves 16, 24 Radum, Ḥorvat 251, 251n88 Ramat Raḥel 250, 252 Ras Feshkha 199 Red Sea 61 Rujm el-Bahr 251, 251n63 Sabar Cave 23 Sagha, Hiam el- 183 Samara, Wadi el- 138

361

Index of Sites and Place Names Samaria 232 Samrah, Khirbet es- 26, 39 Sdeir, Wadi see Naḥal David Ṣeʾelim, Naḥal (Wady Seiyal) 14, 16, 85, 89, 91, 95, 162n3, 163–4, 163n5, 168, 169n41, 171 Cave 34 163–4, 163n5, 164n14, 166, 168, 168n39, 172 Seiyal, Wadi see Naḥal Ṣeʾelim Shephelah 2, 58, 249, 251 Sinai 42n22, 150n84 Sorek, Naḥal 186 Wadi Suweinit, see Naḥal Michmash Syria 96, 100 Syro-Palestine 116, 228

Taʿamireh, Wadi 27 Tekoa, Naḥal 27 Tiberias 11 Timnah see Batash, tel Transjordan 38 Umm Hadar 111 ʿUza, Ḥorvat 185n14, 250–1, 251n88 Washington 227 Zion, Mount 88 Zoora (Zoara) 38–9, 59