Sei sulla pagina 1di 815




T h e F i r s t T h o u s a n d Ye a r s


YA L E U N I V E R S I T Y P R E S S /

N E W H AV E N & L O N D O N

Published with the assistance of the Ronald and Betty Miller Turner Publication Fund and the Lucius N. Littauer Foundation. Copyright 2005 by Yale University. All rights reserved. This book may not be reproduced, in whole or in part, including illustrations, in any form (beyond that copying permitted by Sections 107 and 108 of the U.S. Copyright Law and except by reviewers for the public press), without written permission from the publishers. Set in Janson type by Tseng Information Systems, Inc. Printed in the United States of America by Edwards Brothers, Ann Arbor, Michigan. The Library of Congress has catalogued the rst edition as follows: Levine, Lee I. The ancient synagogue : the rst thousand years / Lee I. Levine. p. cm. Includes bibliographical references and index. isbn 0-300-07475-1 (alk. paper) 1. SynagoguesHistoryTo 1500. 2. JudaismHistoryPost-exilic period, 586 b.c.210 a.d. 3. JudaismHistory Talmudic period, 10425. I. Title. bm653.l38 1999 296.6'5'0901dc21 98-52667 cip

A catalogue record for this book is available from the British Library. The paper in this book meets the guidelines for permanence and durability of the Committee on Production Guidelines for Book Longevity of the Council on Library Resources. 10 9 8 7 6 5 4 3 2 1

isbn 0-300-10628-9 (pbk.)


Preface to the Second Edition Preface to the First Edition Chronolog y one Introduction Sources and Methodology History of Research

ix xi xv 1



two Origins State of Research The City-Gate as Synagogue Forerunner The Emergence of the Synagogue in the Second Temple Period Between City-Gate and Synagogue Concluding Remarks three Pre- 70 Judaea Galilee: some methodological considerations / nazareth / capernaum / tiberias / gamla Jerusalem Judaean Desert: masada /




herodium / qumran The Coastal and Shephelah Regions: dor / caesarea / qiryat sefer / modiin (khirbet umm el-umdan) Other Proposed Sites The Judaean Synagogue in Perspective
four The Pre- 70 Diaspora Egypt: epigraphical and papyrological evidence / philo / a rabbinic tradition Berenice (Cyrene) Italy: ostia / rome Delos Asia Minor and Greece: josephus / new testament / the acmonia inscription The Kingdom of Bosphorus Syria The Diaspora Synagogue in Perspective f i v e The S e c o n d Te m p l e S y n a g o g u e Its Role and Functions The Synagogue as a Community Center The Synagogue as a Religious Institution: torah reading / reading from the prophets (haftarah) / study and instruction / sermons / targumim / communal prayer The First-Century Synagogue in Historical Perspective



si x L ate Roman Palestine ( 7 0 Fo u r t h C e n t u r y C . E . ) Sources and Methodology: rabbinic materialabundance and scarcity / post-70 synagogues: the archaeological evidence? / synagogue building in the mid third century Continuity and Change Synagogues in Roman PalestineFurther Observations seve n Byzantine Palestine Diversity within Commonality Figural Art in Historical Perspective Jewish Motifs in the Byzantine-Christian Context The Synagogues Enhanced Religious Dimension The Impact of Christianity Concluding Remarks eight Diaspora S ynagogues Archaeological Sites: dura europos / gerasa / apamea / sardis / priene / aegina / plovdiv / stobi / ostia / bova marina / naro ( ammam-lif) / elche Synagogues of Rome Babylonian Synagogues The Centrality of the Diaspora Synagogue Between Unity and Diversity Religious Leadership Synagogues in the Diaspora and PalestineSimilarities and Dierences Some Concluding Thoughts







nine The Building Location The Synagogue Complex Architecture Orientation Atriums and Water Installations Entrances Benches and Columns Partitions and Balconies Bimot, Tables, and Platforms Cathedra of Moses The Torah Shrine The Eternal Light and the Menorah Art Iconoclasm A Rabbinic Source on the Synagogue Interior Inscriptions Liturgy and Architecture: The Worship Setting ten The Communal Dimension Community Control Institutional Functions: meeting place / court / charity / place of study / library / place of residence / place for individual needs eleven Leadership Archisynagogue Other Ocials: archon / pater synagoges / mater synagoges / presbyter (elder) / grammateus / phrontistes / azzan / teacher / minor officials Synagogue Ocials in Jewish Palestine and the Diaspora Conclusions t welve The Pat riarch (Nasi ) and the S y nagogue Rabbinic Literature Archaeological Evidence Epiphanius The Theodosian Code Patriarchal Involvement in the Synagogue thirteen The Sages and the Synagogue Non-Rabbinic Sources Rabbinic Sources: dissonance between the sages and the synagogue / synagogue versus bet midrash / the sages and figural art Rabbinic Involvement in the Synagogue Rabbis and Liturgy Concluding Remarks f o u r t e e n Wo m e n i n t h e S y n a g o g u e Attendance Seating Liturgical Roles Benefactors Ocials Women in Synagogue Life: Historical Perspectives fif tee n P riests Benefactors Ocials Priests in Synagogue Ritual 313







sixteen Liturg y 530 Methodological Considerations The Second Century: torah reading / the amidah / the shema liturgy / other liturgical developments /


rabbinic liturgy and the second-century synagogue / early christian and rabbinic liturgies Late Antiquity: the third century / differences between palestine and babylonia / prayer / qedushah / archaeological evidence for prayer / the torah reading and its accompanying activities / piyyut Beyond Late Antiquity
seventeen Iconog raph y: The L imits of Interpretation 593 Methodological Considerations Comparisons with Christian Art The Use and Abuse of Literary Sources Limitations in the Interpretation of Jewish Art: menorah / helios and the zodiac signs / programmatic explanations Concluding Perspectives eighteen Diachronic and Synchronic Dimensions The Synagogue in Context Architectural Evidence Art Communal Center Inscriptions Liturgy Sanctity Degrees of Hellenization Unique Jewish Components Conclusions nineteen Epilogue Glossary List of Abbreviations Bibliog raph y Illust ration Credits Source Index Subject Index


637 641 645 649 731 735 763


ale University Press has graciously agreed to republish The Ancient Synagogue in a revised, paperback edition. Such a revision has become a desideratum owing to the deluge of synagogue-related material that has been published since the submission of my original manuscript to the Press in 1998. Over the past six years, studies addressing every conceivable aspect of the ancient synagogue have appeared, ranging from excavation reports and monographs to articles in edited volumes and a plethora of journals. Thus, updating the original volume is appropriate, as is the decision to publish a paperback edition that will be accessible to a wider audience. This has also aorded the opportunity to reformulate and rene some of my analyses as well as to correct mistakes that inadvertently appeared in the rst edition. I welcome this opportunity to acknowledge my appreciation to Yale University Press, and especially to its editorial director, Jonathan Brent, for agreeing to undertake this project.


his volume is the fruit of years of teaching and research connected with the ancient synagogue. It was only after living in Israel for several years and seeing rsthand the steady stream of archeological discoveries associated with this institutionthe buildings as well as their artistic and epigraphical remainsthat this subject rst engaged my attention. The opportunity to share this fascinating material with students further stimulated my interest and curiosity. This led to the editing of several volumes on the subject in the 1980s, Ancient Synagogues Revealed and The Synagogue in Late Antiquitythe former a series of articles presenting the latest ndings of archeological excavations of synagogues, the latter a collection of the papers delivered at an international conference sponsored by the Jewish Theological Seminary in New York. Since then, I have published a number of articles on a range of topics dealing with various aspects of the ancient synagogue. The subject matter in this volume has been organized both diachronically and synchronically. The primary division is chronological, with the destruction of the Temple in the year 70 serving as a watershed. The reason for this division is twofold. First, the presence of the Temple and its subsequent destruction were powerful factors in shaping the religious role of the synagogue. As long as the Temple existed, no institution could compete with its prominence, sanctity, or religious authenticity. The Temple was the Jewish religious institution par excellence, and its demise created a vacuum in Jewish life, which was lled in large part by the synagogue.


preface to the first edition

A second factor in choosing the year 70 as a watershed is related to the disparate nature of the sources at our disposal before and after that year. The literary sources change dramatically, and the quality and quantity of the archeological and epigraphical material for late antiquity far exceeds that which was available earlier. Following the Introduction and a chapter on the origins of the synagogue, we shall focus on the synagogues of Judaea and the Diaspora, and then on the role of these synagogues in Jewish society in the rst century c.e. Such an arrangement allows for the initial presentation and analysis of the relevant data, followed by a synthesis of the material and a discussion of a variety of issues relating to the functioning of the synagogue in the late Second Temple period. The remainder of the book is devoted to the post-70 era. After a series of chapters on the synagogues development in late antique Palestine and the Diaspora, our attention focuses on the synagogue as an institutionthe physical dimension of the building, its communal aspects, and its leadership, as well as a number of specic groups within Jewish society that played a signicant role in this institution: the Patriarchs, the rabbis, women, and priests. A chapter is devoted to the liturgical developments within the synagogue, an aspect richly addressed by rabbinic and other material. There I describe how the Jewish worship context evolved in late antiquity until it reached a form quite similar to that which exists in most liturgical contexts today. Another chapter deals with the interpretation of Jewish art and examines what can and cannot be ascertained given the evidence available, and a nal chapter discusses the internal and external (i.e., diachronic and synchronic) forces that shaped the synagogue of late antiquity. A few words regarding the use of this book are in order. Owing to its size, an attempt has been made to keep footnotes as unencumbered as possible. Therefore, shortened references have been used throughout, with full bibliographical details appearing at the end of the volume. Translations of verses from the Old Testament are taken from The Holy Scriptures According to the Masoretic Text of the Jewish Publication Society of America (Philadelphia, 1917); a number of emendations have been made in the translations at the discretion of the author. Translations of verses from the New Testament follow the Revised Standard Version (RSV). Page references to critical editions appear in a designated section of the bibliography. On rare occasions, several editions of a single work are cited. For example, Liebermans edition is used when citing the rst four sedarim of the Tosefta, and Zuckermandels edition for the last two. However, when no pagination appears, the standard uncritical edition is being cited. When relevant information appears in the printed, and not the critical, edition, the former is cited without a page reference. At times I have cited an older critical edition rather than a newer one, in which case the name of the edition is cited as well (e.g., for Tanuma or Pesiqta de Rav Kahana). Translations of Greek and Latin sources have been taken from the Loeb Classical Library unless otherwise indicated. In citing collections of epigraphical material, reference is made to the inscription number (no.) in a given corpus; whenever a number appears alone (not preceded by p.), the reference is to a page in that edition.

preface to the first edition


Several of the chapters in this book have appeared as articles, although each has undergone extensive revision and expansion. Chapter 9, on the synagogues physical dimension, originally appeared in the Hebrew journal Cathedra in 1990, while Chapter 13, a study of the relationship between the sages and the synagogue, was originally published in The Galilee in Late Antiquity in 1992. Chapter 2, on the origins of the synagogue, is a revision of an article that appeared in the Journal of Biblical Literature in 1996. All appear here with the permission of the publishers. Since I began working on this volume several years ago, I have enjoyed the support of several institutions and academic frameworks. The Rockefeller Foundation made it possible for me to spend six weeks at its magnicent academic center in Bellagio, Italy. My sabbatical stay at Yale University and the Jewish Theological Seminary, along with grants from the Hebrew University of Jerusalem and the Memorial Foundation for Jewish Culture, aorded me further opportunities to make substantial progress in my research. I also have beneted immensely from the comments and suggestions of many colleagues who were kind enough to read parts of the manuscript: Professors G. Blidstein, S. Fine, G. Foerster, I. Gafni, R. Jacoby, R. Kalmin, S. Reif, A. Shinan, P. van der Horst, B. Visotzky, and Z. Weiss. My thanks are also due to a number of graduate students who helped in the nal stages of the research: Jill Borodin, Joshua Kulp, and Jennifer Tobenstein. I am indebted to ani Davis for her superb editing skills and meticulous reading of the manuscript in its many versions. Her insights and suggestions are found throughout this book. Finally, I thank Charles Grench, Mary Pasti, and the sta of Yale University Press for the professional and supportive attention given to every phase of this books production.


1000586 b.c.e. ca. 950 621 586 586536 53870 (c.e.) 538332 536 516 458 444 33263 332 301 ca. 200 175 167 164 14063 63

First Temple period Building of the First Temple by Solomon King Josiahs reforms Destruction of the First Temple Exilic period Second Temple period Restoration period First wave of returnees from Babylonian exile Completion of the Second Temple Ezra arrives in Jerusalem Nehemiah arrives in Jerusalem; public reading of the Torah Hellenistic period Alexander the Great conquers Judaea Judaea under Ptolemaic rule Judaea under Seleucid rule Hellenistic reforms in Jerusalem Antiochus persecutions in Judaea Maccabean purication of the Temple Hasmonean era Pompey conquers Judaea


374 6 c.e. 2636 40 6674 6996 70 74 ca. 70225 70132 ca. 7080 ca. 90120 132135 ca. 140180 ca. 180225 193235 ca. 200220 ca. 220400 ca. 220500 279 306337 324 ca. 326 351 361363 379395 438 614 638 Herods reign Judaea incorporated into Roman provincial system Pontius Pilate in Judaea Caligulas attempt to place his statue in the Jerusalem Temple First revolt against Rome Flavian dynasty Destruction of the Second Temple Conquest of Masada Tannaitic age Yavnean era Era of Yoanan ben Zakkai in Yavneh Era of Rabban Gamaliel II in Yavneh Second revolt against Rome (Bar-Kokhba revolt) Ushan era Era of R. Judah I Severan era Codication of the Mishnah Amoraic period (Palestine) Amoraic period (Babylonia) Death of R. Yoanan bar Napa, the leading Palestinian amora Reign of Constantine I Christianity recognized as ocial religion of the Roman Empire Discovery of the cross and Golgotha Gallus revolt Reign of Julian Reign of Theodosius I Publication of the Theodosian (II) Code Persian conquest of Palestine Arab conquest of Jerusalem




he synagogue, one of the unique and innovative institutions of antiquity, was central to Judaism and left indelible marks on Christianity and Islam as well.1 As the Jewish public space par excellence, the synagogue building was always the largest and most monumental in any given Jewish community and was often located in the center of the town or village. In the Hellenistic and early Roman periods, the term synagogue () was used to refer to the community, its central building, or both. Luke uses the term to denote both meanings in the same chapter (Acts 13:14, 43), as do the Jews of Berenice in one of their inscriptions. In Asia Minor, Rome, and Judaea, synagogue referred to a building, but in a number of inscriptions from Bosphorus, the community was clearly intended. In Bosphorus, Egypt, and Delos, the word proseuche [, house of worship] referred to the building. The term synagogue will be used in this volume to refer to the communal framework that evolved sometime in the Second Temple period and constituted the focus of Jewish life in Late Antiquity. It is entirely possible that some communities initially met on premises other than a synagogue building or called their central institution by another name. By the second century c.e., however, synagogue had become a universal term for the building in which communal activities were held.2
1. See, for example, Krinsky, Synagogues of Europe, 1520. 2. Scholars have long discussed the dierences between these two terms, synagoge and proseuche. While


In comparison to the Jerusalem Temple, which it came to replace as the central religious institution in Jewish life, the synagogue was revolutionary in four major areas.3 Location. The synagogue was universal in nature. Not conned to any one site, as was the ocial sacricial ritual of the post-Josianic era, the synagogue enabled Jews to organize their communal life and worship anywhere. Leadership. The functionaries of the synagogue were not restricted to a single caste or socioreligious group. In principle, anyone could head the institution. Priests may have played a central role in its religious aairs as well, owing to their knowledge and experience in liturgical matters and not necessarily because of their priestly lineage per se. Synagogue leadership wasin theory, at leastopen and democratic (in certain functions and places, regarding women as wellsee Chap. 14). Participation. In addition to the communal dimension, the congregation was directly involved in all aspects of synagogue ritual, be it scriptural readings or prayer service. This stands in sharp contrast to the Jerusalem Temple setting, where people entering the sacred precincts remained passive and might never have even witnessed the sacricial proceedings personally unless they themselves were oering a sacrice. In many cases, visitors to the Temple remained in the Womens Court without being able to view what was transpiring in the inner Israelite or Priestly Courts.4 Moreover, non-Jews were explicitly banned from the Temple precincts under penalty of death (warning inscriptions were set up around the sacred precincts), whereas the synagogue was open to all; in many places, particularly in the Diaspora, non-Jews attended the synagogue regularly and in signicant numbers. Worship. Perhaps the most distinct aspect of the synagogue was that it provided a context in which a dierent form of worship other than that of the Jerusalem Temple developed. Over the course of Late Antiquity, the synagogue came to embrace a wide range of religious activities, including scriptural readings, communal prayers, hymns, targum, sermons, and piyyut. Instead of the silence that characterized the Temples sacricial cult, the synagogue placed a premium on public recitationcommunal prayer, as well as the reading, translation, and exposition of sacred texts.
some have assumed that they refer to two very dierent institutions (e.g., Gutmann, Synagogue Origins, 3; Runesson, Origins, 42976), most assume, correctly in my opinion, that these terms refer to one and the same institution, each highlighting a dierent dimension (e.g., Hengel, Proseuche und Synagoge, 2754; Httenmeister, Synagoge und Proseuche, 16381). Neither of these terms was uniquely Jewish, as both were borrowed from pagan culture. In the course of time, however, and certainly by the rst century c.e., they had become largely associated with the Jewish community. 3. On the uniqueness of the synagogue as a religious institution, see Elbogen, Jewish Liturgy, 188; Heinemann, Prayer, 1319; S. Safrai, Synagogue, 9089; Fleischer, On the Beginnings of Obligatory Jewish Prayer, 400401; L. I. Levine, Second Temple Synagogue, 7. 4. Pagan temples sometimes barred people from entering; at times they were open to all; see Stambaugh, Functions of Roman Temples, 571; MacMullen, Paganism, 44.


The centrality of the text in the synagogues liturgical agenda was indeed revolutionary; the communal reading and study of the Bible made this institution, from its inception, radically dierent from other Jewish religious frameworks of antiquity. The Jerusalem Temple, the temples of Elephantine and Leontopolis, and the public liturgy of Qumran all had entirely dierent foci. In fact, the synagogue was likewise unique vis--vis contemporary pagan religious contexts, wherein hymns, prayers, and recitations formed the primary nonsacricial liturgy. However, the primary importance of the synagogue, as a whole, throughout antiquity lay in its role as a community center. By the rst century c.e., the synagogue had become the dominant institution on the local Jewish scene in both the Diaspora and Judaea, with, of course, the sole exception of pre-70 Jerusalem. No other communal institution that might conceivably have competed with the synagogue for communal prominence is ever mentioned in our sources. Within the connes of the synagogue the Jewish community not only worshipped, but also studied, held court, administered punishment, organized sacred meals, collected charitable donations, housed the communal archives and library, and assembled for political and social purposes. As a communal institution, the synagogue was fundamentally controlled and operated by the local community. Running such an institution may have been the concern either of the community as a whole, as was most likely the case in villages and towns, or primarily of the local urban aristocracy, which often assumed responsibility for the building and maintenance of such structures.5 In contrast to pagan temples and Christian churches, for which architectural and organizational models used throughout the Roman and Byzantine Empires were often the norm, synagogues were generally locally based and autonomous. As a result, we see a broad range of styles and practices associated with this institution throughout antiquity. This varietyfrom architectural patterns, artistic expressions, and inscriptions to prayer, Torah reading, sermons, targum, and piyyutcharacterized the synagogue of antiquity, constituting what Peter Brown has called in another context an exuberant diversity. 6 The extent of this diversity has become abundantly clear over the past generation or two with the dramatic increase in archaeological material and greater sophistication in the analysis and evaluation of our literary sources. As a result, we are aware of striking regional dierences even within Roman-Byzantine Palestine, not to speak of the far-ung Diaspora. In several cases we have become aware of very dierent types of synagogues even within a given city. Nevertheless, despite this diversity, the institution exhibited a remarkable uniformity. Its basic role as a community center and the range of activities and religious functions conducted therein, as well as its orientation, ornamentation, symbolism, and sanctity,
5. Baron, Jewish Community, I, 5354, 13441. See below, Chap. 10. 6. Brown, Art and Society, 18.


were, in varying degrees, common to synagogues throughout antiquity. These shared characteristics are evident in both archaeological and literary sources. The synagogue evolved signicantly throughout the course of antiquity. Its communal dimension continued to remain central between the rst and seventh centuries c.e. yet, as noted, the religious component of the institution changed dramatically in scope and prominence. The synagogue did not emerge at rst as a quintessentially religious institution, although some dimension of religious activity was undoubtedly present from the outset. Only in Late Antiquity (from the secondthird centuries c.e. onward) did the religious component develop and expand to become the decisive feature of the synagogue. The synagogue was thus transformed from a community center with a religious component into a house of worship that included an array of communal activities. This transformation is most strikingly attested by the synagogues of ancient Palestine, and, despite the relative paucity of evidence, such developments can be detected in the Diaspora as well. The synagogue had become a miqdash meat ( ,) a lesser or diminished sanctuary.7 In some respects the synagogue came to replace the Temple. Whereas the latter had served as the main focus of Jewish religious life throughout the Second Temple period, after the destruction in 70 c.e. many of its customs and prerogatives were gradually assumed by the synagogue. The impetus for these changes came from several quarters. Certainly, internal Jewish developments, rst and foremost among which was the destruction of the Temple, played a signicant role. No less important, however, were the evolving Empire-wide social and religious contexts in which the synagogue operated. Greco-Roman inuences were clearly in evidence in many physical aspects of the synagogue, as were Christian models by Late Antiquity. With regard to the church, an ironic reversal took place between the rst and seventh centuries c.e. Whereas nascent Christianity drew heavily on religious and liturgical elements derived from contemporary Second Temple Jewish life, this trend was largely reversed after the ascendancy and dominance of the church in the Byzantine period, as Jewish life generally, and the synagogue in particular, began absorbing elements of contemporary Christian practice. More than any other Jewish institution of antiquity, the synagogue demonstrates a fascinating synthesis of Jewish and non-Jewish elements within a single framework. While some features of the synagogue reect earlier Jewish customs and beliefs, others, just noted, derive from the surrounding pagan and later Christian worlds. The integration of these elements in every aspect of the institutionfrom the physical dimension of art and architecture to the spiritual dimension of liturgyoers a glimpse into the diverse and dynamic nature of Jewish life at the time, socially, religiously, and culturally. As we shall see, the Jewish response to these stimuli was far from monolithic; while many ele7. B Megillah 29a; on this source, see below, Chap. 6.


ments were adopted or adapted, others were ignored. Furthermore, whatever reactions there were might change markedly from one community to the next. The various ways these external models were combined with practices identied at the time as Jewish are intriguing. Often they coexisted with no apparent tension, and we can only speculate as to how a community might have understood such a synthesis. For example, the zodiac motif, depicting the four seasons and the sun god Helios, riding in his chariot, is invariably found on the mosaic oors of synagogues alongside panels depicting distinctively Jewish symbols, such as the Torah shrine, menorah, lulav, and shofar. In the recently discovered synagogue at Sepphoris, biblical scenes and Tabernacle/Templerelated items are featured together with the zodiac pattern. Clearly, many Jewish communities integrated non-Jewish models into their synagogue framework without feeling threatened or compromised in any way. Because of its centrality and importance in the community, the synagogue played an integrative role in ancient Jewish society. The inclusiveness of its activities, ranging from social to religious and from political to educational, underscores this fact. An impressive array of religious forms found expression within its walls, some of older Second Temple period vintage (scriptural readings, sermons, and targumim), some of post-70 origin (communal prayer, piyyut, and religious art). Moreover, all segments of the community came within the purview of the synagogue in one way or anotherthe common folk of both genders and all ages, village and town leaders, the wealthy urban aristocracy, various economic and social associations, the Patriarchate and those associated with that oce, the rabbis, and other religious gures within the community. The study of the synagogue has far-reaching implications in addition to tracing the important role of this institution in Jewish society. Given its centrality, there is much to learn about the communities per se via this institution: How did the communities dene themselves? What was the nature of their leadership? What were their religious and cultural agendas? What was their relationship to the pagan and Christian surroundings, as well as to the Roman and Byzantine authorities (both secular and religious)? In light of the growing wealth of information regarding the ancient synagogue, many conceptions regarding Jewish history of Late Antiquity have undergone serious revision. The location of synagogue remains, for example, has aorded a much fuller picture of Jewish settlement in Byzantine Palestine than was heretofore known. At times, these remains have conrmed much of what we know from other sources, i.e., that the post-70 Jewish settlement was concentrated in the Galilee in particular, as well as in the large coastal cities of the country. In other cases, however, archaeological nds have supplemented extant literary sources by indicating that Jewish settlement also ourished in other areas, e.g., the eastern, southern, and western peripheries of Judaea and in the Golan, areas that have been largely ignored in literary sources. Moreover, our assessment of the sociological, political, and cultural dimensions of


Jewish life in Late Antiquity has been totally transformed by the cumulative nds relating to the ancient synagogue. Until recently it was almost universally assumed, in the historiographical tradition reaching back to the nineteenth century, that the Jewish communities of Late Antiquity suered ever-increasing persecution and discrimination and that, as a result, these communities, particularly those in Byzantine Palestine, were severely reduced in status and diminished in size. On the basis of the data now available, this picture must be seriously revised. Synagogues, in fact, were to be found the length and breadth of Byzantine Palestine; some were built anew; others underwent periodic renovation. Jewish cultural activityfar from being stiedcontinued to ourish throughout these centuries: artistic expression was extensive, synagogue prayer and poetry were rened and expanded, sermonic and targumic materials were compiled and edited, new halakhic and liturgical forms were created, new types of apocalyptic literature were written, and new forms of synagogue poetry, magic, and mystical experiences crystallized. A similar reevaluation has taken place with regard to the Diaspora. The picture of these far-ung communities as suering legal discrimination, church hostility, and occasional persecution now has to be balanced by evidence of toleration, stability, prosperity, and even of Judaisms continued appeal to non-Jews. Both literary and archaeological data oer evidence of this more positive dimension. The attraction of Judaism for Antiochan Christians, the participation of non-Jews in the communal activities of the Aphrodisian Jewish community, and the centrality and prominence of many synagogue buildings in their respective urban settings (rst and foremost Sardis, but not exclusively) make it crystal clear that Diaspora Jewry, at least in part, continued to ourish throughout Late Antiquity. Synagogue studies have also opened up new vistas regarding our understanding of the nature of Judaism throughout Late Antiquity. It was once assumed (and, as it turns out, quite gratuitously) that the synagogue and Jewish religious life generally followed rabbinic dictates: what the rabbis legislated, the community then adopted. Reality, however, appears to have been far more complex. Synagogue remains oer a variety of cultural, artistic, and religious expressions, some of which appear far from compatible with rabbinic dicta. It is indisputable that the rabbis were a signicant factor in Jewish society by the end of Late Antiquity; whether they wielded denitive authority or a normative inuence in communal aairs, even in religious matters, does not appear to have been the case very often. Only now are we beginning to recognize the many dierent cultural and religious currents at play in Jewish society at the time that helped to create the rich mosaic of beliefs and practices we know today.


Given its centrality in Jewish life, it is not surprising that the synagogue is mentioned frequently in a wide variety of sources. Nevertheless, while literary and archaeo-


logical materials abound, both in absolute terms and in comparison to other Jewish institutions in antiquity, our understanding of the synagogue is hampered by the diverse types of sources available, by their varying foci, and by the resultant discontinuity of information from one locale to another and from one period to the next. For example, our knowledge of Diaspora synagogues in the pre-70 era rests in large part on literary remains (the New Testament, Philo, and Josephus), but for Late Antiquity (Babylonia excepted) the evidence is drawn almost exclusively from archaeological material. Thus, the discrepancy in the sources available for the two periods is sharp and almost irreconcilable. The issue of continuity is only slightly improved with respect to Roman Palestine. The limited amount of sources for the pre-70 period is compensated by an abundance of material from the late Roman and Byzantine eras. Both rabbinic sources and archaeological data oer a relatively detailed picture of the synagogue in its Late Antique Palestinian setting. The former is a particularly rich source in this regard, although utilizing these sources presents a plethora of methodological challenges. In addition to the usual issues relating to the reliability of textual traditions and attributions, the late editing of many rabbinic works, and the dating of unattributed statements, we must constantly question how reective this corpus of information is regarding synagogues generally. Do rabbinic traditions preserve unique cases that may have been of interest to the sages but were not necessarily reective of the institution as a whole? If so, how limited are they? Perhaps they are more representative of rabbinic synagogues, i.e., those in which the rabbis tended to congregate, than of the majority of synagogues serving the community at large. Or are these sources merely an indication of what the rabbis wished to see rather than what was usually the case? Complicating this matter further is the issue of the nature and extent of rabbinic inuence and involvement in this institution. As we can no longer assume that the rabbis ipso facto wielded authority over synagogue aairs in Late Antiquity, their comments about the institution and its leadership must be treated with a measure of circumspection. With the exception of Babylonia, rabbinic material does not usually relate to the Diaspora. As a result, we possess only vague notions about these Jewish communities from the rabbinic perspective, and little idea as to the ties between them and the rabbis. Generally speaking, each corpus of evidence has its own particular raison dtre, with the synagogue often playing only a marginal role. Each type of source tends to concentrate on certain dimensions of the institution without making any attempt to oer a more comprehensive picture. Rabbinic literature, for example, focuses on the synagogues liturgical and, at times, social aspects; Josephus on its political role; and the New Testament on the synagogue within the context of Jesus and Pauls missions. On a larger scale, archaeology tends to focus on the communal dimension (in its broadest terms) while literary material concentrates on the religious component.8 Nevertheless, even if
8. See my First Century c.e. Synagogue, 124.


only partial, the information oered by each source is invaluable. Taken together, a fairly comprehensive picture of the synagogue does emerge, and as the range of data is impressive both geographically and chronologically, so, too, are the subjects addressed both directly and indirectly: the physical aspects of the synagogue; artistic expressions; cultural proclivities; communal activities; leadership roles; and liturgical matters. Let us review the main sources to which we will have recourse throughout our study, beginning with the pre-70 era. Josephus notes the existence of synagogues in Judaea within the context of his political narrative of the rst century. In both Dor and Caesarea, the synagogue became a center of controversy during the political struggles between Jews and pagans; in Tiberias, the synagogue (here called proseuche) served the Jewish community as a meeting place for its political deliberations at the outset of the revolt in 6667 c.e. Regarding the Diaspora, Josephus cites a series of Roman documents from the latter half of the rst century b.c.e. aecting Jewish communities in a number of locales, particularly Asia Minor. The rights accorded the Jews under Roman rule are clearly articulated in these privilegia, and on several occasions the latter make explicit reference to the synagogue; more often, they mention activities and functions that we can safely assume took place in this institution. Philo makes only a few passing references to the synagogue or proseuche. In the course of describing the pogroms of 38 c.e., he notes the existence of Alexandrian synagogues, and particularly one monumental and lavishly ornamented building. On several occasions, he mentions certain aspects of the Sabbath morning Torah-reading ritual, as well as other worship practices of the Therapeutae and Essenes. A number of books in the New Testament speak of synagogue-related matters. All the gospels recount Jesus activity in Galilean synagogues, and Luke is especially expansive in his opening account of Jesus preaching in Nazareth. The information in Acts for both Jerusalem and the regions of Asia Minor and Greece is also of primary importance to our understanding of the pre-70 institution. Rabbinic material has preserved a number of traditions that we may assume describe pre-70 synagogues. Of particular importance is the Toseftas relatively elaborate description of a rst-century Alexandrian synagogue. However, many of these rabbinic sources, including the Bavli and most midrashic compilations, are quite late and thus of questionable historical value. Although the pre-70 archaeological material is scanty, it is still of cardinal importance. While remains of at least six synagogue buildings are attestedve in Judaea and one (perhaps two) in the Diaspora (Delos, and possibly Ostia)inscriptions provide the bulk of archaeological evidence from this period. The Theodotos inscription from Jerusalem, that of Julia Severa from Acmonia in Asia Minor, a number of catacomb inscriptions from Rome, three synagogue inscriptions from Berenice (Cyrene), ve from Delos, six from the Bosphorus, and sixteen (or parts thereof ) from Egypt (including papyrologi-


cal evidence) oer us a varied and far-ranging picture of the institution in the rst century c.e. This situation changes dramatically in the post-70 era. As noted, archaeological material for the Diaspora becomes far more abundant than previously. Thirteen buildings have been identied as synagogues of Late Antiquity; nine additional sites are less certain. Hundreds of inscriptions have been found, most dedicatory in nature and deriving from synagogue buildings, others stemming from a funerary context and mentioning someone associated in his or her lifetime with a synagogue. An even greater abundance of archaeological evidence is to be found with regard to Roman-Byzantine Palestine.9 To date, well over one hundred synagogues have been identied and close to two hundred inscriptions retrieved, almost two-thirds of which are in Hebrew or Aramaic, the remainder in Greek. The score of amulets found in synagogue contexts are likewise of great value, not to speak of the dozens of mosaic oors, some of which are remarkably well preserved. Literary remains relating to the synagogue in this period are scattered among the writings of Byzantine authors (e.g., Scriptores Historiae Augustae), church fathers (Epiphanius, John Chrysostom), the Theodosian Code, and Justinians Novellae. However, the overwhelming preponderance of literary material derives from rabbinic sources. More than four hundred pericopes explicitly mention the synagogue from either an historical, communal, or liturgical perspective; many hundreds, if not thousands, more deal with liturgical activities, most of which undoubtedly took place within the walls of this institution. Other Jewish texts relating to synagogue matters include the mystical Hekhalot texts, the various targumim, Byzantine piyyutim, and Byzantine halakhic material, such as the list of variant practices in Palestine and Babylonia. The methodological problems mentioned above with respect to rabbinic literature hold truemutatis mutandiswith regard to other sources as well. Many of the statements preserved are problematic because of their selectivity, tendentiousness, or fragmentary nature. Even if a report appears to be accurate in and of itself, we are rarely certain whether the instance described was a local phenomenon or one reective of other times and places. Nevertheless, with the required modicum of caution and awareness of these issues, a not insignicant amount of information about the synagogue can be retrieved.

To attempt to reconstruct the history of research on the ancient synagogue is, in essence, to relate the histories of research in each of the many areas pertaining to this institution. One may trace the history of research on the synagogue building by focusing on
9. On the criteria for identifying structures as synagogues, see below, Chap. 9.



archaeological reports, on its worship dimension by following publications treating liturgy, and on its artistic and sociocultural dimensions by reviewing those works that concentrate on Jewish art and epigraphical remains. Only on rare occasion has there been an attempt to integrate the various strands into one historical account of the institution. Despite the enormous diversity in the development of the various elds of research, there are many common characteristics among them. Until the mid-twentieth century, only a handful of scholars mastered each of these disciplines. The assumption shared by most of them is that the development of the synagogue underwent a steady linear progression, and thus much of their work focused on tracing this evolution. Let me oer two examples of this phenomenon. The rst concerns the synagogue building. In its classical expression, primarily within the realm of architecture, the once regnant theory states that the physical appearance of the synagogue evolved in three distinct phases in Late Antiquity. This particular theory crystallized over the rst half of the twentieth century, beginning with the rst serious study of the ancient synagogue, Kohl and Watzingers Antike Synagogen in Galilaea (in 1916). Based upon an overall survey and limited excavation probes carried out between the years 1905 and 1907 in eleven Galilean and Golan buildings (including one on the Carmel), these scholars established criteria that became axiomatic in synagogue studies for decades, i.e., that the Galilean synagogues were built in the late second and early third centuries, constituted a recognizable architectural group, and were modeled architecturally and artistically after buildings in Roman Syria, especially the Roman basilica. Several decades later, in 1934, Sukenik published his Schweich lectures under the title Ancient Synagogues in Palestine and Greece, in which he added a second tier to this theory by identifying a series of Christian basilica-like synagogue structures (e.g., Naaran, Bet Alpha, Gerasa) that date to the sixth and seventh centuries. These, he claimed, belonged to another, later, synagogue type. In 1949, Sukenik further developed this distinction, adding ammat Gader, useifa, and Jericho to the list. Finally, in the 1950s, a third type was proposed by Goodenough and Avi-Yonah, who posited a broadhouse, or transitional, stage, with the synagogue of Eshtemoa serving as the primary example of this type (see below, Chap. 9). The basis of the above approach was the assumption that synagogue typology is linked to its chronology: dierent types of synagogues were built at dierent times. Thus, at any given moment one particular type of building predominated, although there may have been some overlap in the transition from stage to stage. A similar approach characterized the study of Jewish prayer, particularly with regard to the development of the Amidah, or Shemoneh Esreh, which, owing to its centrality and importance, was the primary focus of many early studies. Already in the nineteenth century, a number of scholars began setting the parameters for the subject when they posited a linear development for this prayer, i.e., that it evolved from a shorter, simpler version into a longer, more complex one. Zunz addressed the subject, and was followed in



the rst half of the twentieth century by Kohler, Finkelstein, Elbogen, and others. The core of the Amidah was generally traced back to the Persian or early Hellenistic periods and was described as having developed incrementally in response to internal and external factors (e.g., competing Jewish ideologies, external political and cultural inuences, persecutions, the destruction of the Temple) until it was fully standardized under Rabban Gamaliel of Yavneh ca. 100 c.e. The study of Jewish prayer was revolutionized by the ndings from the Cairo Genizah, not the least of which was an awareness of the liturgical diversity in the rst millennium c.e. Not only did this material aect our understanding of the development of prayer, it also related to the cycles of scriptural readings for the Sabbath and festivals. Inspired and facilitated by the texts from the Cairo Genizah, scholars such as Mann and Bchler wrote extensively on the triennial Torah and prophetic readingsand their variations as reected in these early documents. The rst half of the twentieth century also produced several formidable works intended to serve as systematic compilations of the known data relating to the synagogue. S. Krauss Synagogale Altertmer (1922) focused primarily on the material culture of the institution as reected in rabbinic literature and the limited archaeological evidence then available. Elbogens still denitive Der jdische Gottesdienst in seiner geschichtlichen Entwicklung (1913; revised Hebrew version 1972; English version 1993) focused on the synagogues liturgical and, secondarily, physical and communal dimensions. Since the mid-twentieth century, a veritable explosion has taken place in synagogue studies. The reasons for this ourish of activity are many: (1) The sheer number of synagogues unearthed has risen dramatically. Buildings were discovered in urban centers such as Caesarea, Tiberias, Bet Shean, and Sepphoris, as well as in rural settings such as Maon (Nirim). Following the 1967 Arab-Israeli war, synagogues were found in Gaza, Susiya, and En Gedi, as were a host of structures in the Golan and the southern Judaean hills. At the same time, the number of synagogues discovered in the Diaspora has almost doubled; the newly identied synagogues include the building in Ostia and the monumental edice at Sardis. All of these nds have generated a renewed interest in the synagogue building per se, as well as in related areas. (2) New elds of inquiry have been created, particularly as a result of a series of seminal studies, foremost among which is the thirteen-volume work of Goodenough, Jewish Symbols in the Greco-Roman Period, published between 1953 and 1968. In fact, his eorts at collecting a hitherto unimaginable array of Jewish artistic works from Late Antiquity spurred the creation of the eld of ancient Jewish art. Goodenough, however, had aimed to do much more, hoping that his collection of material would serve a much grander design of reconceptualizing the nature and form of Judaism in Late Antiquity. This goal, however, was never achieved. The study of Jewish art was given a signicant boost with the publication of the nal excavation report in 1956 of the Dura Europos synagogue. Discovered in 1932, this build-



ing is by far the most sensational and revolutionary ancient synagogue ever unearthed. Kraelings ocial publication on the site was soon followed by Goodenoughs detailed and controversial analysis in volumes 911 of his Jewish Symbols (1964). These two monumental works ushered in a plethora of studies in a host of elds, the ramications of which continue to our own day. Similar groundbreaking eorts were taking place in epigraphical studies. The rst major corpus of Jewish epigraphical evidence was published by Frey. While volume 1 of his Corpus Inscriptionum Judaicarum had already appeared in 1936, the publication of volume 2, including inscriptions from the eastern Mediterranean (Asia Minor, Syria, and Palestine) and North Africa, appeared only posthumously in 1952. For all its inadequacies in no small part probably due to the exigencies surrounding its publicationthis singular achievement provided the springboard for many other collections that appeared in the coming decades, each of which has focused on specic geographical areas or types of inscriptions. Such studies include the corpus of catacomb inscriptions that provided the basis for Leons The Jews of Ancient Rome (1960), Rutgers The Jews in Late Ancient Rome (1995); the corpus of Egyptian inscriptions appended to volume 3 of Corpus Papyrorum Judaicarum, edited by Tcherikover, Fuks, and Stern (1964), Lifshitz Donateurs et fondateurs dans les synagogues juives (1967), the Hebrew volumes of Naveh, On Stone and Mosaic (1978), and Roth-Gerson, The Greek Inscriptions from the Synagogues in Eretz-Israel (1987) and The Jews of Syria as Reected in the Greek Inscriptions (2001); and the three volumes of the Jewish Inscriptions Project of the Faculty of Divinity at the University of Cambridge: Horbury and Noys Jewish Inscriptions of Graeco-Roman Egypt (1992) and Noys two-volume Jewish Inscriptions of Western Europe (199395). (3) This era also witnessed methodological and conceptual breakthroughs that breathed new life and insights into elds that had lain dormant for decades. Heinemanns Prayer in the Talmud: Forms and Patterns (Hebrew, 1964; English trans., 1977) introduced into the study of Jewish prayer a form-critical approach widely used in biblical studies, and he single-handedly revitalized the historical study of Jewish prayer over these past forty years. Heinemann espoused the theory that various prayer forms developed in different contexts and by dierent circles within Jewish society, and these were transmitted orally from generation to generation and from place to place; thus, a plethora of versions were in circulation simultaneously. As a result of his inuence, the linear approach to the development of Jewish prayer has generally been abandoned, as well as the assumption that it was created by a Jewish leadership. A similar development subsequently took place in the archaeological realm. With the discovery of the remains of numerous additional structures, the earlier-mentioned concept of a linear development of architectural types of synagogues has generally become pass and has been replaced by the assumption that dierent architectural types were in use at one and the same time (see below, Chap. 9). (4) An abundance of newly published material has revolutionized a number of areas



in the study of Jewish liturgy. We have already taken note of the extraordinary documentation provided by the Cairo Genizah, which aected practically every area of synagogue studies. Renewed interest in targumic texts has taken place through the discovery of new sources. On the one hand, the Qumran nds have revealed fragments of Aramaic targumim in use in the pre-70 era; on the other, the sensational discovery of the Neoti manuscript of the Torah in the Vatican library in 1949 has brought to light the earliest, most complete targum text known to date. This latter discovery was particularly instrumental in generating a renaissance in targumic studies. The increased interest in the study of midrash over the past generation has led to renewed scholarly attention in the homiletical material found in rabbinic literature. Heinemanns form-critical approach to Jewish prayer has revitalized the examination of other rabbinic genres as well; he himselfuntil his death in 1978devoted numerous studies to the rabbinic homily, particularly the proem. Moreover, interest in early Jewish liturgy has spilled over to the eld of scriptural lectionary, an area that had not been seriously addressed since earlier in the century. Attention was directed toward uncovering traces of the triennial lectionary cycle in extant literary works, and studies such as Guildings The Fourth Gospel and Jewish Worship (1960), though controversial, have been illuminating in this regard. The area of piyyut has received a great deal of scholarly attention since the 1960s. Building on the earlier work of Schirmann and Zulay, major contributions have been made in this eld by Mirsky, Z. M. Rabinowitz, Yahalom, and especially Fleischer. The study of this religious and literary genre was begun by Zunz over a century ago; however, the increased availability of Genizah material has revolutionized this eld. Instead of assuming a medieval origin and a Muslim context for the piyyut, as had once been the case, it is universally conceded today that this genre crystallized in Byzantine Palestine, when literally thousands of such poems were written. To date, we know the names of no less than twenty paytanim who ourished in Late Antiquity, as well as other anonymous ones. Almost unknown several generations ago, Yannai is now identied as a sixth-century poet who penned at least two thousand poems, as probably did the greatest of the ancient paytanim a generation later, Elazar Ha-Qallir (or Qillir). (5) The urry of activity in synagogue studies is also an expression of the dramatic worldwide growth in Jewish studies within institutions of higher learning in the last part of the twentieth century. This has been noticeable in Israel since 1967, with the dramatic increase in the number and size of the countrys universities; in North America, where Jewish studies began growing at a feverish pace in the mid-1960s; and, to a lesser extent, in England, Germany, and elsewhere in Europe. As a result, the sheer quantity of scholarly material being produced on synagogue-related subjects is far greater now than it was a generation or two ago. (6) Finally, it is important to note that the marked interest in the ancient synagogue is in no small measure due to the attention Christian scholars have accorded the Jew-



ish background of the New Testament and early Christianity. Although such activity was already apparent at the outset of the century with Schweitzers The Quest of the Historical Jesus (German 1906; English 1926), a much greater openness and receptivity has been evident from the mid-twentieth century onward, such as Davies Paul and Rabbinic Judaism (1948). Among the reasons for this change are the following: the post-Holocaust Christian reassessment of attitudes toward Jews and Judaism; the impact of the creation of the State of Israel; the exploration of Christian roots within Judaism stimulated by the Qumran evidence; and, nally, the eects of the prevailing atmosphere of tolerance and understanding generated by the changes in Catholic policy in the decisions of Vatican II. The resultant contributions of Christian scholars to Jewish studies over the last generation have been a major catalyst in bringing synagogue-related studies to the fore. The heightened interest in synagogue studies has spurred a number of attempts to summarize the institutions history in broad outline, including I. Levys The Synagogue: Its History and Function (1963), Hrubys Die Synagoge: Geschichtliche Entwicklung einer Institution (1971), and Schrages extensive entry on the synagogue () in volume 7 of the Theological Dictionary of the New Testament (English trans. 1971).10 A plethora of studies devoted to the Second Temple synagogue has appeared in recent years. Three major monographs address various aspects of this phenomenon: Binder, Into the Temple Courts (1999); Runesson, The Origins of the Synagogue (2001); and Claussen, Versammlung (2002). A fourth monograph, Gibson, Jewish Manumission Inscriptions of the Bosporus Kingdom (1999), oers a detailed analysis of these most unusual inscriptions from the rst centuries c.e. In addition, a slew of recent articles deal with a wide range of synagogue-related issues, such as the synagogues of Rome (Richardson); the Jerusalem Temple and the synagogue (Binder, Flesher); collegia and synagogue (Richardson); the Diaspora synagogue generally (Rajak, Fitzpatrick-McKinley); and the debate regarding the date of the Galilean synagogue (Magness, E. M. Meyers, Strange, and Aviam). This explosion in synagogue-related research persists to this very day. Not only does material continue to issue forthbe it new primary data from excavations, Genizah material, or secondary studiesbut more rened methodological approaches and new assumptions underpin most scholarly works. We have already noted that the neat linear approach has largely been replaced by a more complex, multifaceted one. Moreover, it is now generally assumed that institutional developments such as that of the synagogue were not determined by authorities from above, nor did they occur at the same time or in a monolithic fashion; rather, a more diversied and dynamic approach is to be preferred, one which appears to be far more reective of the actual historical processes. A greater sensitivity to methodological issues includes a more sophisticated use of literary sources, especially the New Testament and rabbinic literature. In the case of the
10. See also Hoppe, Synagogues and Churches; Bloedhorn and Httenmeister, Synagogue, 26797; Schrage, , 798847.



latter, this involves a more systematic distinction between earlier (tannaitic) and later (amoraic) material as well as between sources of Palestinian and Babylonian provenance. Not only are the tendentiousness and selectivity of these literary sources more widely acknowledged, so, too, are the literary and theological agendas of each of these writings. At the same time, a growing number of scholars are more inclined to explore the social and cultural implications of archaeological nds, and interdisciplinary studies, although still in their infancy in this area, are more in evidence of late. Moreover, scholarly treatments are now generally more attuned to external inuences on Jewish society than had hitherto been the case. Rather than questioning whether such factors existed, eorts are now directed at determining the degree of outside inuence in what areas of Jewish life, in which geographical regions, among which classes of society, and under what historical circumstances. The mid-twentieth century was pivotal in redening this eld of inquiry. Liebermans works (1942, 1950) and Hengels magnum opus (German, 1969; English trans., 1974) framed an era in which a series of pathbreaking studies, including those of Goodenough, M. Smith, Scholem, Bickerman, Schalit, and Tcherikover, were written.11 In recent decades, new aspects of the ancient synagogue have been addressed in addition to the older, more traditional ones. The role of women in the synagogue (Brooten), the place and authority of the rabbis (L. Levine, Goodman, and S. J. D. Cohen, Hezser, and S. Schwartz), the iconographic interpretation of synagogue art (Fine, Khnel, Z. Weiss, and H. Kessler), the borrowing of pagan and Christian motifs and symbols and their meaning in a Jewish context (Foerster, Ovadiah, Habas, Talgam, and Hachlili) these subjects and more have engaged scholarly attention of late. Also noteworthy is the burgeoning number of studies focusing on the Diaspora synagogue. While Sukenik, Goodenough, and Kraeling addressed this subject earlier on,12 it was the discovery of the Sardis building, following the nal publication of the Dura Europos excavations, that ushered in a remarkably fruitful era of research and inquiry about this Diaspora institution. The main catalyst has been Kraabel, who himself participated in the Sardis dig and has since written a plethora of studies on the Sardis building as well as various aspects of the Diaspora synagogue.13 Other notable contributors to this eld in11. See my Judaism and Hellenism. 12. Sukenik, Ancient Synagogues; Goodenough, Jewish Symbols, II, 70100; Kraeling, Excavations at Dura: Synagogue. 13. See the following studies by Kraabel: Judaism in Western Asia Minor; Hypsistos and the Synagogue at Sardis, 8193; Melito the Bishop and the Synagogue at Sardis, 7785; Paganism and Judaism, 1333; Diaspora Synagogue, 477510; Jews in Imperial Rome, 4158; Social Systems of Six Diaspora Synagogues, 7991; Excavated Synagogues, 22736; Roman Diaspora, 44564; Impact of the Discovery, 17890; Synagoga Caeca, 21946; Unity and Diversity, 4960; (and A. Seager), Synagogue and the Jewish Community, 17890. Many of these studies have been collected in a festschrift in his honor; see Overman and MacLennan, Diaspora Jews and Judaism.



clude Hengel, Seager, Foerster, Ovadiah, White, Levinskaya, Feldman, Rutgers, Richardson, Runesson, and Trmper.14 The two Diaspora synagogue sites that functioned in the rst two centuries c.e. (Delos and Ostia) have merited a number of illuminating studies (Runesson, White, and Trmper). The study of the ancient synagogue has been facilitated by an ever-growing corpus of scholarly aids. Recent decades have witnessed the appearance of convenient handbooks on ancient synagogues, such as those of Saller, Chiat, and Z. Ilan.15 Up-to-date summaries of archaeological nds have been made available in the New Encyclopedia of Archaeological Excavations in the Holy Land,16 as well as in a collection of rabbinic material relating to synagogue sites by Httenmeister and Reeg.17 Hachlilis volumes, Ancient Jewish Art and Archaeology in the Land of Israel (1988) and Ancient Jewish Art and Archaeology in the Diaspora (1998), provide a broad overview of synagogue material, particularly the artistic dimension, and, in eect, update Goodenoughs major work in many ways. Several computer data bases have made much of rabbinic literature available;18 in recent years at least nineteen volumes of essays on the ancient synagogue have appeared, not to speak of several dozen monographs and hundreds of articles oering overviews of the subject or detailed studies of the synagogue in one or more of its manifestations.19 The appearance of basic compendia is nowhere more evident than in the eld of epig14. Hengel: Proseuche und Synagoge, 2754; Die Synagogeninschrift von Stobi, 11048. Seager: Building History, 42535; Architecture of the Dura and Sardis Synagogues, 14993; Synagogue at Sardis, 17884; Recent Historiography, 3947; (and A. T. Kraabel), Synagogue and the Jewish Community, 16878. Foerster: Survey of Ancient Diaspora Synagogues, 16471; Remains of a Synagogue at Corinth, 185; Fifth Century Synagogue in Leptis Magna, 5358. Ovadiah: Ancient Synagogues in Asia Minor, 85766; Ancient Synagogues from Magna Grecia, 920. White: Delos Synagogue Revisited, 13360; Building Gods House, 60101; Social Origins; Feldman: Diaspora Synagogues, 4866. Rutgers: Diaspora Synagogues, 6795; Hidden Heritage of Diaspora Judaism; Richardson: Early Synagogues as Collegia, 90109; Architectural Case, 90117; Runesson: Synagogue, 2999; Trmper: Oldest Original Synagogue, 51398. Mention should also be made of Trebilcos Jewish Communities, which, while not dealing with the synagogue per se, still contains an enormous trove of material regarding Asia Minor. 15. Saller, Second Revised Catalogue of Ancient Synagogues; Chiat, Handbook of Synagogue Architecture; Z. Ilan, Ancient Synagogues. 16. E. Stern, NEAEHL. 17. Httenmeister and Reeg, Antiken Synagogen. 18. Bar-Ilan University Responsa Project; Davka CD-ROM Judaic Classics Library. 19. These collections of essays include: L. Levine, Ancient Synagogues Revealed; idem, Synagogue in Late Antiquity; Gutmann, Dura-Europos Synagogue; idem, Synagogue; idem, Ancient Synagogues; Z. Safrai, Ancient Synagogue; Oppenheimer et al., Synagogues in Antiquity; Overman and MacLennan, Diaspora Jews and Judaism; Hachlili, Ancient Synagogues in Israel; Klil-Hahoresh, Synagogues; S. Elizur et al., Knesset Ezra; Urman and Flesher, Ancient Synagogues; Fine, Sacred Realm; idem, Jews Christians, and Polytheists; Kee and Cohick, Evolution of the Synagogue; Avery-Peck and Neusner, Judaism in Late Antiquity; Olsson and Zetterholm, Ancient Synagogue; Olsson et al., Synagogue of Ancient Ostia; Tabory, Kenishta.



raphy. As noted, a number of important corpora have appeared over the past twenty years: Navehs collection of Hebrew and Aramaic inscriptions, Roth-Gersons collections of Greek inscriptions from synagogues in Israel (1987) and Syria (2001),20 the volumes of Horbury and Noy on Egypt and Western Europe, Kroll and Cross publications of the Sardis inscriptions,21 as well as Kants discussion of ancient Jewish inscriptions in Greek and Latin.22 Most recently, three volumes of Inscriptiones Judaicae Orientis have been published, covering the provinces of Eastern Europe, Asia Minor, Syria, and Cyprus.23 Finally, new synagogue-related elds have emerged. Of cardinal importance are the studies on ancient Jewish mysticism (Hekhalot literature). Some of these compositions contain prayers and incantations identical (or almost identical) to prayers known from synagogue liturgy. What the precise relationship was between these mystical circles, the rabbis, and the actual synagogue prayers at the time is far from certain, but some sort of connection seems to have existed. Another important eld is Jewish magic. Beginning with the publication of Sepher HaRazim in 1966 by Margalioth and continuing with the volumes by Naveh and Shaked (Amulets and Magical Bowls: Aramaic Incantations of Late Antiquity, 1985; Magic Spells and Formulae: Aramaic Incantations of Late Antiquity, 1993), the scholarly world has been presented with a convenient collection of Jewish magical material from Late Antiquity. To these must be added the texts on magic from the Genizah of Late Antiquity edited by Schiman and Swartz (Hebrew and Aramaic Incantation Texts from the Cairo Genizah, 1992). Despite this prodigious display of research, almost all these studies have been directed to one or another specic eld of research: archaeology, liturgy, epigraphy, art history, etc. Most are highly specialized, concentrating on a particular dimension of the synagogue or on a particular type of evidence.24 The aim of the present study is to integrate the data from these diverse elds into a comprehensive account of this pivotal Jewish institution over a thousand-year period. It aims to trace the synagogues growth and development from a variety of perspectives in light of the forces at play from within the Jewish community and from without.
20. Roth-Gerson, Jews of Syria. 21. Kroll, Greek Inscriptions of the Sardis Synagogue, 5127; Cross, Hebrew Inscriptions from Sardis, 319. 22. Kant, Jewish Inscriptions in Greek and Latin, 671713. 23. Noy, Panayotov, and Bloedhorn, I: Eastern Europe; Ameling, II: Kleinasien; Noy and Bloedhorn, III: Syria and Cyprus. I have not yet had the opportunity to consult these volumes. 24. An exception to the above are the integrative studies regarding sacred space that have recently appeared; see Branham, Sacred Space under Erasure, 37594; idem, Vicarious Sacrality, 31945; Fine, This Holy Place.





etermining the origin and early development of the synagogue has presented modern scholarship with a seemingly insurmountable challenge. As often happens with institutions, movements, and ideas of revolutionary proportions, the forces that shape new initiatives, especially in their embryonic stages, remain shrouded in mystery. For a period of time these initiatives crystallize outside the limelight of history, only to appear later in our sources in a relatively developed form. Such, indeed, was the case with the ancient synagogue. Despite our understandable interest in knowing when such an institution rst appeared, what factors were decisive in its development, who was responsible for it, and where exactly this creation took place, the sources at our disposal are oblivious to these issueseither because they are not as historically oriented as we would have wished or perhaps because these early formative stages simply were not worthy of comment at the time. The earliest hard evidence we have for the existence of a synagogue appears in a number of inscriptions from third-century b.c.e. Ptolemaic Egypt that mention a proseuche. To date, about twelve such inscriptions and papyri have been discovered from the Hellenistic period, and the earliest archaeological remains of a synagogue building on the island of Delos in the Aegean indicate a second- or possibly mid-rst-century b.c.e. date of construction.1
1. For a detailed presentation of the evidence, see below, Chap. 4. The relationship between the synagoge and proseuche has been discussed on a number of occasions. While several scholars clearly dieren-


historical development of the synagogue

It is not until the rst century c.e. that the synagogue emerges into the full light of history as the central communal institution of Jewish communities throughout Judaea and the Diaspora. Major cities, such as Jerusalem, Alexandria, Rome, and probably Antioch, each boasted a number of such institutions, and villages, towns, and cities throughout the Roman Empire probably had at least one. The number and range of both literary and archaeological sources referring to the synagogue at this time are indeed impressive.2 The existence of a proseuche in third-century b.c.e. Egypt indicates that the synagogues roots in the Diaspora go back at least to this time. What are the implications of this date as regards the institutions origins in the rest of the Diaspora and in Judaea? Are we to follow such evidence alone and therefore posit a Ptolemaic Egyptian origin? Or can we assume that the synagogues beginnings may date back to the Restoration, Exilic, or perhaps even First Temple times?

As might be expected, the theories propounded to explain the emergence of the synagogue range far and wide. Some scholars date its origins to as early as the ninth or eighth century b.c.e., others look to the late seventh century. Most have opted for either a sixth-century date, at the time of the Babylonian exile, or a fth-century Judaean setting, thus assuming that the synagogue emerged as a result of the initiatives of Ezra and Nehemiah. Some scholars have posited dates ranging from the fourth down to the rst centuries b.c.e. This wide spectrum of opinion is both chronological and geographical. While some have placed the rst appearance of the synagogue in a Babylonian or Egyptian setting, others assume that Judaea provided the original social and religious context. The reasons for this variety of opinions abound. First and foremost is the sheer absence of data. With no clear-cut references at hand, the only recourse has been to speculate on the time, place, and circumstances surrounding the synagogues emergence. In reality, such eorts have been inuenced by one or more of the following considerations: (1) determining a major event in Jewish history that may have given rise to this new and revolutionary institution; (2) locating a reference in the Bible that may point to the existence of a setting that could be called a proto-synagogue or a synagogue-in-the-making; (3) construing the absence of references to synagogues, particularly in Judaean literature of the Persian and Hellenistic periods, as an indication that the synagogue did not exist then; (4) following only the existing hard evidence, i.e., that of third-century b.c.e. Egypt or rst-century c.e. Palestine, as the decisive factor in determining origins. Secondly, there is a lack of clarity as to the meaning of the term synagogue (tiate between the two, regarding them as very distinct institutions (see below, note 4), most assume that they were basically the same, at most with a somewhat dierent emphasis. See my Second Temple Synagogue, 2021, and the literature cited therein. 2. See below, Chaps. 35.



) and, consequently, what characteristics and developments should determine its time and place of origin. As noted, the term may have referred to a congregation, any group of people (qahal or edah [ ], in the Septuagint, Ben Sira, and targumim), or a building.3 And even if congregation is meant, the synagogue of asidim ( I Macc. 2, 42) and the synagogue of scribes ( ibid., 7, 12) were probably no more than groups of like-minded pious individuals, thus rendering the term meaningless for our purposes. On the other hand, to look for a specic type of building might not be a fruitful exercise, for many dierent settings could have been used by a synagogue-community: a public building constructed expressly for religious purposes, a multipurpose communal building, a private residence or part of a home converted for public use. Alternatively, the synagogue might be dened solely on the basis of its functions and status within a community, that is, its institutional role (e.g., organization, leadership, activities, etc.). In addition to synagoge (place of assembly), a number of other terms were used (almost always in the Diaspora) for this institution, each apparently emphasizing a dierent dimension or characteristic; the choice of a name depended on local needs and self-perceptions (proseuche [house of prayer]; to hieron [sanctuary]; to hagios topos [a holy place]; eucheion [a place of prayer]; sabbateion [a Sabbath meeting place]; didaskeleion [a place of instruction]; bet am [house of people]; amphitheater; and temple).4 Thirdly, what was the nature of this institution at its very beginning? Was it primarily a social-communal institution or was it, rst and foremost, a religious one? And if the latter, which particular religious activity was focallistening to a prophet? praying? fasting? reading the Torah? Fourthly, how might one determine the emergence of the synagogue, or of any other institution, for that matter? Is it the rst appearance of any of the aforementioned activities? If so, then what about the issue of continuity? Perhaps a particular phenomenon surfaced in an early context but then disappeared for a long while, leaving no trace in Jewish religious practice other than an isolated literary reference? For example, should the Torah-reading ceremony described in Neh. 810 be considered the beginning of the synagogue, since it may have been the origin of the practice of Torah reading? Or was the ceremony of Ezra and Nehemiah only a onetime occurrence conducted under very special circumstances? If so, the Torah-reading ceremony of the later synagogue would then have been the outcome of other factors, with the earlier setting serving, at best, as some sort of historical precedent for a later development. Finally, one is faced with the issue of whether to place credence in later rabbinic tradi3. See above, Chap. 1; and TDNT, VII, 80210. 4. On these terms, see L. Levine, Second Temple Synagogue, 1314; D. Binder, Into the Temple Courts, 91154; and below, Chap. 4, notes 237240, and Chap. 5. Scholars assuming dierent institutions on the basis of dierent names include: Gutmann, Synagogue Origins, 3; Runesson, Origins, 42976; and Rajak, Synagogue and Community, 2238.


historical development of the synagogue

tions. What should one make of the rabbinic claims that the miqdash meat of Ezek. 11:16 and the maon ( ,dwelling place) of Ps. 90:1 are indeed evidence for the synagogues existence? 5 Such traditions are highly questionable and are even more compromised by the contradictory claims made elsewhere in rabbinic literature regarding the origins of several fundamental synagogue practices. For example, various sources maintain that the weekly scriptural readings originated with either Moses, a body of prophets and elders, the Men of the Great Assembly, a group of former prophets, or early asidim.6 With regard to the Amidah, dierent sources declare that this prayer originated with the biblical Patriarchs, Moses, a group of 120 elders and prophets, or the Yavnean Simeon Hapaquli.7 In view of these disparate notions, it seems certain that such traditions are little more than anachronistic musings, perhaps for homiletical purposes, and are of little historical worth. Given this formidable array of issues, it should come as no surprise that the theories regarding synagogue origins are diverse and, at times, far aeld. In the absence of explicit references, scholars have clutched at any and all hints and allusions, utilizing them as evidence for the institutions emergence. The more important of these theories may be briey summarized.8 A number of scholars have focused on the First Temple period as the context for the synagogues origins. Several have pointed to Solomons references to the Temple as a place of prayer, others to the custom of regularly visiting a prophet on the Sabbath and New Moon.9 Of greater attraction has been the presumed impact of Josiahs reforms in 622/621 b.c.e., which prohibited sacrices outside Jerusalem. This is considered by many to have been the crucial factor that led to the creation of a new religious framework based on non-sacricial worship.10 Finally, terms such as bet haam ( , Jer. 39:8) and
5. B Megillah 29a. See also the reference in Pesiqta Rabbati 26 (p. 129b) to the prophet Zephaniah preaching in the synagogue, as did Jeremiah in the marketplace and Huldah to the women. So, too, with regard to the targumim, which frequently associate biblical gures with the synagogue; see, for example, Targum Jonathan to Exod. 18:12; Judg. 5:9, as well as the targumim to I Chron. 16:31 and Isa. 1:13. 6. Moses: Y Megillah 4, 1, 75a. Prophets and elders: Mekhilta of R. Ishmael, Beshala, 1 (p. 154); Mekhilta de R. imon b. Jochai, Beshala 15, 22 (p. 103). Great Assembly: B Berakhot 33a. Former prophets: SifreDeuteronomy 343 (p. 395). asidim: Midrash on Psalms 17, 4 (p. 64a). 7. Biblical patriarchs: B Berakhot 26b; Midrash Hagadol, Genesis 19:27; 28:11 (pp. 324, 498). Moses: Y Berakhot 7, 11c; Y Megillah 3, 8, 74c; Midrash on Psalms 19, 22 (p. 82b). 120 elders, prophets, and Simeon Hapaquli: B Megillah 17b18a. 8. For useful surveys regarding the origins of the synagogue, see Weingreen, Origin of the Synagogue, 6884; Rowley, Worship, 21345; Hruby, Die Synagoge, 1930; Gutmann, Origin of the Synagogue, 3640; idem, Synagogue Origins, 16; L. Levine, Second Temple Synagogue, 810; and the recent Hachlili, Origin of the Synagogue: A Reassessment, 3447; and Runesson, Origins, 67167. 9. I Kgs. 8; II Kgs. 4:23; Finkelstein, Origin of the Synagogue, 4959; idem, Pharisees, 563; Weingreen, Origin of the Synagogue, 6884; I. Levy, Synagogue, 12. See also Levenson, From Temple to Synagogue, 14366. 10. Morgenstern, Origin of the Synagogue, 192.; Weingreen, Origin of the Synagogue, 6884;



moadei el ( , Ps. 74:8) have also been invoked to substantiate a pre-Exilic date of origin.11 A second approach, preferred by most scholars over the years, places the origin of the synagogue in a sixth-century Babylonian Exilic setting.12 The destruction of the First Temple and the subsequent exile have been viewed as traumatic and decisive factors leading to the creation of an alternative form of religious worship. The gathering of elders to hear the words of a prophet (Ezek. 8:1; 14:1) and, to a lesser extent, the later rabbinic interpretation of Ezekiels lesser or diminished sanctuary (B Megillah 29a) as referring to the synagogue have been invoked to substantiate this dating.13 Another view regards the Restoration period, particularly the activities of Ezra and Nehemiah in fth-century Jerusalem, as the time of the synagogues beginnings.14 Given the centrality of the Torah reading and its concomitant exposition in a later synagogue context, some have viewed the public reading and explanation of the Torah by Ezra in the
Weinfeld, Deuteronomy, 44. Some have argued that the synagogue was a presupposition of Josiahs reforms (von Waldow, Origin of the Synagogue Reconsidered, 26984); others have opined that it was a consequence of them (Wellhausen, Geschichte, 184). Some claim to have discerned early sermons and catechetical instruction (presumably in a synagogue setting) in the books of Jeremiah and Deuteronomy. Both Judaean and Babylonian settings have also been suggested (Nicholson, Preaching to the Exiles). 11. So, for example, Lw, Gesammelte Schriften, IV, 5.; idem, Der Synagogale Rituus, 97. See also Donner, Argumente zur Datierung des 74. Psalms, 4150; Gelston, Note on Psalm lxxiv 8, 8286; Rowley, Worship, 21821 and literature cited there. For vigorous reservations, see Haran, Priest, Temple and Worship, 181 n. 11. Another event sometimes invoked as a catalyst for the creation of the synagogue is the desecration of the Temple by King Manasseh, who reigned from 687 to 642 b.c.e. (II Kgs. 21:47). It is assumed that under these circumstances alternative secretive religious frameworks were sought (Finkelstein, Origin of the Synagogue, 5253). See also Janssen, who, in Juda in der Exilszeit, 156., opines that the beginnings of the synagogue may have taken form in the meetings of prophetic disciples in the pre-Exilic period. 12. A view already propounded by tenth-century Sherira Gaon of Pumbeditha (Iggeret Harav Sherira Gaon [pp. 72.]; see also Gutmann, Sherira Gaon, 20912) and by sixteenth-century Sigonio, in his De republica Hebraeorum (63.; reprinted in Ugolino, Thesaurus antiquitatum sacrarum, IV, ccxci). See also Schrer, History, II, 426; Wellhausen, Geschichte, 196.; Bacher, Synagogue, 619; Elbogen, Jewish Liturgy, 189; Herford, Pharisees, 8991; S. Krauss, Synagogale Altertmer, 5266; G. F. Moore, Judaism, I, 283; Baron, Jewish Community, I, 5963, 73; idem, Social and Religious History of the Jews, I, 126; Landsberger, House of the People, 15354; Janssen, Juda in der Exilszeit; Rowley, Worship, 22425; Sandmel, Judaism and Christian Beginnings, 143; Shanks, Judaism in Stone, 21.; Levenson, From Temple to Synagogue, 14366. 13. Another line of argument dates the synagogue to sixth-century Judaea, associating it with either the fast days then being commemorated (Zech. 8) or the institution of the maamadot, assumed by some to have crystallized around that time (Landman, Origin of the Synagogue, 32225). On the maamadot, see below. 14. Mentioned as a possibility by both Schrer (History, II, 426) and Bacher (Synagogue, 619), but advocated by S. Safrai (Synagogue, 90913) and M. Smith (Jewish Religious Life, 25860). See also Weinfeld, Biblical Roots, 54763.


historical development of the synagogue

month of Tishri, 444 b.c.e., as the catalyst for the emergence of the synagogue. Proponents of this view either seek support by invoking later rabbinic traditions that view the age of Ezra as the critical period in the development of synagogue liturgy, in particular the Torah-reading ceremony,15 or refer to Persian imperial policy as having fostered this practice. Most recently, Runesson has suggested that the Persian context played a crucial role in the emergence of the synagogue. Basing himself on theories that posit the centrality and importance of imperial policy in Persian Jerusalem,16 he claims that the Torahreading ceremony itself was a Jewish implementation of an empire-wide Persian initiative in the sixth and fth centuries b.c.e. 17 Over the past century, scholarly opinion has generally been divided over these last three options: a seventh-century b.c.e. date focusing on the Josianic reforms, a sixthcentury Babylonian Exilic venue, or a fth-century b.c.e. Persian Jerusalem venue with the Torah-reading ceremony at its center. All three relate to the synagogue as a primarily religious institution, the rst two as a place of worship (for prayer and prophetic discourse) in lieu of sacrices, the last as a liturgical-scriptural context in which the Torah reading was its focus. Whatever their popularity, these options far from exhaust the scope and variety of views on the subject, particularly in recent decades. In 1931, Zeitlin proposed a fourthcentury setting, viewing the synagogue as a primarily communal institution, as indicated by its name, synagoge. He posits that the inhabitants of the towns and villages in PersianHellenistic Judaea would congregate to discuss communal aairs; these town meetings, at rst convened primarily for social and political purposes, gradually assumed a religious stamp under Pharisaic inuence. A similar idea had already been put forth by Lw, although he himself dated the synagogues emergence to the First Temple period.18 More recently, Hengel has argued for a Hellenistic Egyptian setting.19 Following a line of argument articulated earlier by Bousset, Gressmann, Friedlnder, and others,20 Hengel bases his claim on several considerations. First of all, he follows the existing evidence, namely, that the rst clear-cut indication of the existence of synagogues appears
15. Y Megillah 4, 1, 75a; B Bava Qama 82a; B Megillah 3a; Elbogen, Jewish Liturgy, 19192. 16. Blenkinsopp, Mission, 40921; Hoglund, Achaemenid Imperial Administration. 17. Runesson, Origins, 259303, 39598; and more succinctly, see idem, Persian Imperial Politics, 6389. For reactions (some sympathetic, most reserved, if not opposed) to this approach of attributing a signicant degree of Persian inuence on the composition and promulgation of the Torah, see the collection of articles in Watts, Persia and Torah. 18. Zeitlin, Studies, 113 (= Origin of the Synagogue, 6981). Landman (Origin of the Synagogue, 31725) oers a variation of Zeitlins approach, which dates the book of Esther to this period and associates the word ,which appears only here in biblical literature, with the synagogue (bet knesset) (Studies, 7). On Lw, see above, note 11. 19. Hengel, Proseuche und Synagoge, 16180 (= Gutmann, Synagogue, 3150). 20. According to Bousset and Gressmann (Die Religion des Judentums, 172) and Friedlnder (Synagoge und Kirche, 56.), the synagogue rst appeared in pre-Hasmonean Egypt.



in inscriptions from third-century b.c.e. Hellenistic Egypt. Secondly, he argues that the synagogue community was, in fact, a Jewish form of the Hellenistic religious association prevalent in Ptolemaic Egypt. The emergence of the synagogue, then, was not only an outcome of developments within the Jewish community, but also the product of forces from without. Griths carried Hengels argument one step further, claiming an Egyptian origin for the synagogue on the basis of architectural and functional parallels between the Egyptian Jewish proseuche and Egyptian pagan temples.21 Beyond citing various parallels meticulously enumerated by Dion some years earlier,22 e.g., the existence of a pylon (i.e., gateway) at the entrance to the building, a double colonnade, an adjacent grove, and an asylum, Griths points to the association of a place of worship with a place of learning (Per Ankh, the House of Life, a library, a place of certain religious rites)a feature that he claims links pagan and Jewish institutions in Hellenistic Egypt.23 Returning to a Judaean setting, Rivkin has suggested that the synagogue was a Pharisaic creation dating to the Hasmonean era.24 Given the far-ranging upheavals within Jewish society, resulting from a successful revolt and the attainment of political sovereignty several decades later, Rivkin argued that the synagogue reects many new trends and developments in Hasmonean society: Pharisaic ascendancy, emerging individualism, and democratization. Furthermore, he tried to turn the absence of any prior concrete data regarding the synagogue in Judaea to his advantage, asserting that Ben Siras failure to mention a synagogue at the beginning of the second century b.c.e. is a telling indication that such an institution did not exist at that time. Rivkin assumes that Ben Siras description of contemporary Jewish religious leadership and institutions is so detailed as to justify his argumentum ex silentio. Also relying heavily on the silence of Hellenistic Jewish sources, Grabbe revived a suggestion made earlier by, inter alia, Jost, Zunz, and Friedlnder, namely, that the Ju21. Griths, Egypt and the Rise of the Synagogue, 115; idem, Legacy, 102836. 22. Dion, Synagogues et temples, 4575. 23. The weakness of Griths argument is due not only to the very limited evidence available for such Egyptian temples. Even if one were to suppose that the Per Ankh served as a place of instruction, it appears to have been conned to priestly circles and focused heavily on funerary rites, including some sort of medical and cultic education as well; see Gardiner, House of Life, 15779; Grabbe, Judaism, I, 95. This is indeed a far cry from the practice of weekly scriptural readings for the entire Jewish community in the proseuche that included a gamut of topics found in the Torah and Prophets. Moreover, such a suggestion ignores any earlier traditions that the Jews might have brought with them from Judaea regarding the reading of Scriptures, e.g., Deuteronomy or Nehemiah 8. See in this regard the interesting remark made by Hecataeus, that the high priest Ezekias read to his friends from Scriptures that he brought from Judaea to Egypt in the late fourth century ( Josephus, Against Apion 1, 18789, and comments in A. Kasher, Josephus, Against Apion, I, 17981). For a more extensive critique of this theory, see my First-Century Synagogue. 24. Rivkin, Ben Sira and the Non-Existence of the Synagogue, 32048. This approach has also been adopted by Gutmann, Synagogue Origins, 34.


historical development of the synagogue

daean synagogue is a late phenomenon. Since the earliest Judaean synagogue is attested only in the late rst century b.c.e., Grabbe concludes that the institution itself is postMaccabean.25 Indeed, both Jewish and pagan sources of the pre-Roman period fail to mention the existence of a Judaean synagogueneither Hecataeus of Abdera, Manetho, nor Mnaseas of Patara.26 Agatharchides (second century b.c.e.), in his description of religious practices specically in the Jerusalem Temple, observes the following:
The people known as Jews, who inhabit the most strongly fortied of cities, called by the natives Jerusalem, have a custom of abstaining from work every seventh day; on those occasions they neither bear arms nor take any agricultural operations in hand, nor engage in any other form of public service, but pray with outstretched hands in the temples until the evening [emphasis mineLL].27

The above reference to temples ( ) has been interpreted by some as referring to synagogues.28 However, it is more likely that the reference was indeed to the Jerusalem Temple and that the use of the plural here was inadvertent; pagan writers would have considered the use of the plural, temples, most natural. In any case, a well-known and oft-frequented religious setting in Jerusalem at that time could only have been the Temple, and praying all day with outstretched hands is rather unusual and would make some sense, if at all, in the Temple and not a synagogue setting. The only other possibility is that Agatharchides was referring to certain Diaspora synagogues, where Sabbath worship (though not necessarily praying with outstretched hands) might have lasted much or most of the day.29 However, this was most unlikely, and thus a Temple setting was undoubtedly intended.


Having thus reviewed the more salient suggestions put forth by scholars regarding the origins of the ancient synagogue, I wish to propose a dierent approach.30 In addressing this issue in the past, scholars have almost invariably tried to pinpoint a historical context or event that led to the emergence of this institution. Given the state of our sources

25. Jost, Geschichte des Judenthums, III, 136; Zunz, Die Gottesdienstlichen Vortrge der Juden, 3; Friedlnder, Synagoge und Kirche, 53. (noting the synagogues rst appearance in Judaea); Grabbe, Synagogues in Pre-70 Palestine, 40110; idem, Judaism, II, 529. 26. See M. Stern, GLAJJ, passim. 27. Josephus, Against Apion 1, 209 (GLAJJ, I, 106109). 28. S. J. D. Cohen, Pagan and Christian Evidence, 16162; Feldman, Jew and Gentile, 159; A. Kasher, Josephus, Against Apion, I, 21415; Binder, Into the Temple Courts, 412; Runesson, Origins, 34647. 29. See, for example, Philos Hypothetica 7, 13; and below, Chap. 4. 30. See also L. Levine, Nature and Origin, 42548, here modied; see below.



or, more precisely, the lack of any solid evidence, such eorts have clearly become exercises in studied guesswork; as a result, prevailing theories on this subject range over a period of more than eight hundred years. With but few exceptions, these theories share several assumptions: that the religious component of the ancient synagogue was primary, and that dramatically new religious circumstances gave rise to this institution. Implicit in most of these theories is the view that some kind of liturgical activity, be it listening to a Divine prophecy, the recital of public prayer, or the introduction of scriptural readings, played a crucial and denitive role in the formation of the early synagogue. However, as noted at the outset of this chapter, the formative stages of a new phenomenon are very often not marked by any one moment of dramatic innovation. The synagogue may not have resulted from a crisis or a specic decision by one person or community to initiate something boldly new. We may well be dealing with a much more subtle and nuanced process that took place over decades, if not centuries, and operated at a dierent pace in various locales. Only at a later stage, when the synagogue had reached a sucient level of crystallization, could one look back and say that a novel institution had indeed been created. Thus, it may be helpful to revisit the question of origins from a dierent perspective. Instead of combing the earlier sources for clues regarding the time and place of the synagogues origins, I suggest a dierent starting point, namely, a period from which we have some solid evidence about the nature of the synagogue and how it functioned. Armed with what we know about the synagogue when it appears in the full light of history, we may then look to an earlier period and ask ourselves where such activities had once taken place. We may then have some clue as to how, why, and from where the institution referred to in the rst century as a synagogue (or as a proseuche somewhat earlier in Egypt) rst developed. A broad sociological and institutional approach may thus be warranted in trying to understand the synagogues origins rather than searching for a moment of national crisis or religious innovation that might have sparked its creation. Let us turn for a moment to the rst-century c.e. synagogue. In the following chapters we will discuss the characteristic features of the Second Temple synagogue in detail. As background for our present discussion, suce it to say that both the synagogue and the proseuche of this period were rst and foremost communal institutions where the entire gamut of activities connected with any Jewish community found expression. As documented in rst-century sources, the building might have been used as a courtroom, school, hostel, a place for political meetings, social gatherings, housing charity funds, a setting for manumissions, meals (sacred or otherwise), and, of course, a number of religious-liturgical functions.31 On the assumption, then, that the rst-century synagogue served as a center for a variety of communal as well as religious functions and activities, we now are in a position
31. See below, Chap. 5; and R. A. Horsley, Galilee, 22233.


historical development of the synagogue

to look for the framework that served the same (or similar) purposes in earlier centuries. When seen in this light, it becomes clear that the location for most, if not all, of these activities in earlier times was the city-gate, the main communal setting in cities and towns in the biblical period.32 The city-gate as the focal point of communal activity is well attested in biblical and non-biblical literature.33 It served as a marketplace (II Kgs. 7:1), as well as a setting where a ruler would hold court and where prophets would speak (I Kgs. 22:10; Jer. 38:7).34 As the gate was a popular meeting-place for public gatherings, a variety of communal activities were conducted there. So, for example, Hezekiah appointed battle ocers over the people; then, gathering them to him in the square of the city-gate [ ,] he rallied them (II Chron. 32:6).35 Those who came regularly to the gate were the populace at large (see Ruth 3:11) and the town elders and leaders.36 The transaction between Abraham and Ephron the Hittite took place at the city-gate (Gen. 23:10, 18), and a number of legal documents from Nuzi conclude with the formula the tablet was written after the proclamation at the gate. 37 Announcement of the settlement of a negotiation at the gate aorded maximum pub32. This suggestion was proposed by Silber (following Lw, Gesammelte Schriften, IV, 5.) in his doctoral thesis at the University of Denver and was summarized by the same author in a pamphlet entitled The Origin of the Synagogue (1915). Silber, however, following Lw, dates this transition to the Solomonic period and further concludes that the synagogue was secular in origin, only later acquiring a distinct religious character. Decades later, the city-gate thesis was revisited by Hoenig (Ancient City-Square, 44876), who, however, confused the town square of Greco-Roman times with the city-gate of a Near Eastern setting (on this, see Ward-Perkins, Cities of Ancient Greece and Italy, 12) while freely utilizing later rabbinic material to supplement what little is known on the subject from pre-70 times. Moreover, much of his argument rests on the very problematic reading of reov ha-ir ( ) in place of the not uncommon reference to ever ha-ir ( ) in rabbinic literature. On the city-gate in biblical times, see Khler, Hebrische Mensch, 14371; de Vaux, Ancient Israel, 152 53; Encyclopaedia Biblica, VIII, 231; Frick, City in Ancient Israel, 7891; Kenyon, Royal Cities, 5761; and esp. Fritz, City in Ancient Israel, 3537, 13840. 33. CAD, A/1, 8288, s.v. abullu; ibid., B, 1920, s.v. babu; Frick, City in Ancient Israel, 8384, 114 27; Encyclopaedia Biblica, VIII, 23236; Otto, Zivile Funktionen, 18897. 34. Other references to rebuking the people at the gates include Isa. 29:21 and Amos 5:10. On the communal signicance of the city-gate, see Ephal and Naveh, Jar of the Gate, 5965. 35. On the discovery of such a square between the main and outer gates at the biblical site of Tel Dan, see Biran, Biblical Dan, 23549; idem, Tel Dan: Five Years Later, 177; idem, Dan, 32731; and below, notes 44 and 45. 36. See Prov. 31:23: Her husband is renowned at the city-gate [lit., gates], sitting among the elders of the land. See also ibid., 31:31. At times, people might come to the gate to simply spend time, socialize, and keep abreast of the latest news; see Esth. 4:6; II Kgs 7: 34. 37. See, for example, Pfeier and Speiser, One Hundred Nuzi Texts, 115. See also de Vaux, Patriarches hbreux, 25 and n. 1. The gates in Deut. 6:9 and 11:20 may refer to city-gates; see Tigay, Deuteronomy, 79.



licity and signied the assent of the entire community, and meals marking the fulllment of a commandment (e.g., eating the tithe or food for the needy) were eaten there.38 Moreover, prophetic activity often took place there so as to reach the greatest number of people (Isa. 29:21; Amos 5:10). One of the primary activities at the city-gate was judiciary.39 City elders would assemble there to dispense justice: His father and mother shall take hold of him and bring him out to the elders of his town at the public place [i.e., the gate] of his community (Deut. 21:19; 17:5; 22:24; see also Ps. 69:13); and the prophet Amos advised: Hate evil and love good, and establish justice in the gate (5:15; see also Zech. 8:16). The importance of the city-gate as a place for settling personal aairs in the presence of the community is vividly reected in Ruth 4:12: Meanwhile, Boaz had gone to the gate and sat down there. And now the redeemer whom Boaz had mentioned passed by. He called, Come over and sit down here, So-and-so! And he came over and sat down. Then [Boaz] took ten elders of the town and said, Be seated here; and they sat down. 40 These ten elders (referred to by Speiser as city-fathers) undoubtedly convened there regularly to ociate as judges, arbitrators, and witnesses to business transactions.41 They were the civil judiciaryas opposed to the priestly judiciary of the Jerusalem Temple and the local sanctuaries.42 The gate as the heart of a city is also reected in the fact that a conqueror might place his throne there as a sign of his rule. Nebuchadnezzars ocers did so (Jer. 39:3) and thus fullled the prophets dire prediction (ibid., 1:1516): For I am summoning all the peoples of the kingdoms of the north, declares the Lord. They shall come, and shall each set up a throne before the gates of Jerusalem, against its walls roundabout, and against all the towns of Judah. And I will argue My case against them for all their wickedness. They have forsaken Me and sacriced to other gods and worshiped the works of their hands. 43 A king might sit at the city-gate to hear the peoples grievances. So, for example, following Absaloms death, Joab urged David to terminate his mourning and to sit at the
38. According to Deut. 26:12, the tithes given to the Levite, stranger, orphan, and widow on the third year of a sabbatical cycle were to be eaten at the city-gate. 39. See McKenzie, Judicial Procedure at the Town Gate, 100104. For rabbinic interpretations of biblical references to judicial proceedings at the city-gate, see Halivni (Weiss), Location of the Bet Din, 18191. 40. The concept of ten as a quorum appears in Josephus description of the Essenes (War 2, 146), at Qumran (CD 13, 1; 1QS 6:3; 6, 7; 10:14), and, of course, in later rabbinic literature (e.g., M Megillah 4, 3). 41. As in the story of Abraham and Ephron in Genesis 23. See Speiser, Coming and Going, 2023; Reviv, Early Elements and Late Terminology, 19091; and, for a dissenting opinion, G. Evans, Coming and Going, 2833. 42. Weinfeld, Deuteronomy, 235. One of the gates of Jerusalem under Josiah was named after Joshua, governor of the city (II Kgs. 23:8). 43. On the meting out of punishment against an entire city, see Deut. 13:17.


historical development of the synagogue

gate so that all the troops could come before the king (II Sam. 19:89). On another occasion, Ahab, king of Israel, and Jehoshaphat, king of Judah, sat at the gate of Samaria prior to a battle in Gilead, summoning the prophets to support their venture (I Kgs. 22:10). Similarly, according to II Chron. 32:6, Hezekiah assembled the people at the city-gate to strengthen their resolve in the face of Sennacheribs imminent attack; a century or so later, we are told, Zedekiah also sat at a city-gate called Benjamin (Jer. 38:7). Lastly, the city-gate was also a place for performing religious functions. In the ancient Near East, people often gathered at the city-gate to worship gods, as is evidenced by the cultic objects found near the gates of Megiddo Va, Beersheba IV, Tel Dan, and Bethsaida.44 Regarding the city-gate at Tel Dan, Biran, following Barnett, notes that it may have been the site of religious ceremonies in antiquity:
We may consider this to be also a ceremonial route. This could depend to a certain extent on the interpretation of the unique structure found in the square between the outer and main gates. This structure is rectangular with an open space where a throne or pedestal was set. Two decorated column-bases were found in situ, a third in the debris and of the fourth only an imprint was left. . . . Our suggested reconstruction shows a canopied structure which could have served the king when he sat at the gate (e.g., I Kgs. 22:10) or it could have served as a pedestal for the statue of a god.45

In terms of explicit biblical evidence, II Kgs. 23:8 has the following to say about Josiahs reforms of 622/621 b.c.e.:
He brought all the priests from the towns of Judah [to Jerusalem] and deled the shrines where the priests had been making oeringsfrom Geba to Beer-sheba. He also demolished the shrines of the gates, which were at the entrance of the gate of Joshua, the city prefect which was on a persons left [as he entered] the city gate.

In the post-Exilic period, a gate area was utilized by Ezra and Nehemiah: The entire people assembled as one man in the square before the Water Gate, and they asked Ezra the scribe to bring the scroll of the Teaching of Moses with which the Lord had charged Israel (Neh. 8:1). Archaeological data conrm the fact that the biblical (i.e., Iron Age) gate was the site of many communal functions. In contrast to the early Middle Bronze Age II gate, which appears to have fullled much more of a defensive role, the twenty or so known Iron

44. See NEAEHL, I, 172 (Beersheba), 32729 (Dan). On Megiddo, see Herzog, Das Stadttor in Israel, 164; on Bethsaida, see below. On this subject generally, see Dever, Recent Archaeological Discoveries, 12862; and esp. Bernett and Keel, Mond, Stier und Kult am Stadttor, passim; Blomquist, Gates and Gods, 70131. 45. Biran, To the God Who Is in Dan, 143; see also Barnett, Bringing the God into the Temple, 1020; and Biran, Biblical Dan, 23845. The role of the city-gate in religious processions and ritual may well be reected in the words of Ps. 24:79: O gates, lift up your heads! . . . so the king of glory may come in!



Age II gate complexes (ca. 1000580 b.c.e.) diered signicantly.46 Whereas gates in the second millennium had rooms usually separated by partition walls to serve as independent units, the Iron Age II gates had rooms that opened onto the main passagewayeither two, four, or six chambers, some large (in one instance, reaching 9 m. long) and often containing benches and stone water basins.47 These areas and, even more importantly, the adjacent open spaces, usually inside but at times outside the gate (in which case there was often another circumvallating wall), provided the setting for the many civil functions noted above.48 A striking example of a complex gate system is found at Tel Dan, built some time in the ninth century b.c.e., presumably by Ahab, where three gates (outer, main, and upper) were preceded by a paved square, a courtyard, and a royal processional way, respectively (g. 1). The latter two gates had four sentry rooms each.49 Perhaps an even more dramatic instance of city-gate cults has come to light at Bethsaida, on the north shore of the Sea of Galilee. At the northern tower, a stepped structure or podium, an iconic stele (a bulls head), a basin, and several vessels were found, and at the southern one, a niche. If the excavators identication of the site as a bama is correct, then Bethsaida would constitute a remarkable example of a gate cult. Four aniconic stelae found elsewhere in the gate complex only serve to reinforce this identication.50 G. Evans, who has collected much valuable material in this regard, recapitulates the importance of the city-gate as follows: A study of the texts in which the term appears shows clearly that the gate, together with the street which lay behind it, just within the walls, was a centre of political and legal activity, as well as of trade . . . the gate was the scene of many activities which, in a western city, were carried on in a central square. 51 McCown has summarized this phenomenon thusly: For the ancient Hebrew, the city
46. The subject has been addressed over the past several decades by Herzog in a series of studies, rst in Hebrew (The City-Gate in Eretz-Israel), then in German translation (see above, note 44), and more recently in English (Settlement and Fortication Planning, 26574; Archaeology of the City). See also E. Stern, Hazor, Dor and Megiddo, 1230. 47. Six-chambered gates have been discovered at Megiddo, azor, Gezer, Lachish, and Ashdod. Fourchambered gateways were found at Ashdod, Dan, Beersheba, Megiddo, and Dor, while two-chambered gateways were excavated at Dor, Mt. Gerizim, and Megiddo. See NEAEHL, passim; Fritz, City in Ancient Israel, 7778 (Dan), 81 (azor), 8896 (Megiddo), 104 (Lachish), 111 (Beersheba). See also E. Stern and Sharon, Tel Dor, 1986, 2037; and Magen, Mount Gerizim, 7376. On the controversial question, whether there are any chronological implications to the dierent-numbered chambers, see above, note 46. On the earlier Mesopotamian gate and its development from the third to rst millennia b.c.e., see Damerji, Development of the Architecture of Doors and Gates, 18198. 48. See also Encyclopaedia Biblica, VIII, 23743; and A. Mazar, Archaeology of the Land of the Bible, 467 70. Esth. 4:6 speaks of a city square or street next to the Kings Gate. 49. Biran, Biblical Dan, 23553; Blomquist, Gates and Gods, 5767. 50. Arav and Freund, Bethsaida, II, 2542; Bernett and Keel, Mond, Stier und Kult am Stadttor, 17, 22 32; Blomquist, Gates and Gods, 4957. 51. G. Evans, Gates and Streets, 112 (cited from p. 1).


historical development of the synagogue

[To view this image, refer to the print version of this title.]

1. Model of the city gate at Tel Dan (Tel Dan Museum).

gate was much more than a means of ingress and egress, much more than an important part of the citys defenses. It was also the center (even though at one side) of the citys social, economic and judicial aairs. 52


Excavations indicate that a major change took place in the gate area in the Hellenistic period. Architectural traditions that had held sway for centuries were now being revised, in part owing to sustained and intensive contact with the Hellenistic world.53 From the available dataand one must admit that the evidence here is fragmentary, at bestit is clear that Hellenistic and, later, Roman gates had no accompanying rooms and usually no adjacent open area or square.54 Instead, they were usually anked by round or square
52. McCown, City, 634. 53. E. Stern, Excavations at Tell Mevorakh, 1727; idem, Walls of Dor, 1114; idem, Dor, 2068; Sharon, Fortication of Dor, 10513; see also idem, Local Traditions, 5. See also van Beek and van Beek, Canaanite-Phoenician Architecture, 70*77*. 54. Arav, Settlement Patterns, 159.



towers. The gate itself was no more than a passageway between these towers. This change may have been inspired by developments in Greek fortication systems in general, which themselves appear to have resulted from technological advances in the art of warfare, particularly the introduction of ballistae, catapults, and siege engines.55 While the gate area of third-century b.c.e. Marisa, on the southern coastal plain of Judaea, still boasted a square surrounded by buildings (an area evidently used for assorted religious, administrative, and military activities),56 gate areas from other Hellenistic sites in Israel reect this new pattern. So, for example, the gate at Dor changed dramatically, from the casemate-type wall, which had been in vogue for ve hundred years, to a simpler Hellenistic style, having no adjacent open areas, rooms, or separate buildings.57 Hellenistic building techniques are also evidenced in the Hellenistic town of Samaria-Sebaste,58 and by the rst century b.c.e., Herod was using this Hellenistic gate model at both Sebaste and Caesarea (g. 2).59 Although meagerly preserved, the Ginat gate in Jerusalem, discovered by Avigad, may also reect a similar Hellenistic model.60 On the basis of the above discussion, it becomes evident that most of the activities carried out within the context of the rst-century synagogue are already documented for the city-gate area in the First Temple period. By the Hellenistic period, when the biblical city-gate complex was transformed from a center of urban activity into a simple, functional gate for entrance and exit, the activities previously associated with the city-gate and the adjacent square were relocated to a building that eventually came to be known as the synagogue. An important caveat is in order here. A change in the function of the city-gate is documented for the Hellenistic period at certain sites located primarily in the coastal area of Judaea and relate almost exclusively to several pagan cities in this region (e.g., Dor). Whether other parts of the country were similarly aected at this time is unknown. We lack any kind of signicant archaeological evidence for Jewish settlements in Judaea and
55. See Shatzman, Ballistra Stones from Tel Dor, 94104; Sharon, Fortication of Dor, 10513; and, more generally, Winter, Greek Fortications, 22233, 32432; McNicoll, Hellenistic Fortications, 68, 4648, and esp. the concluding chapter written by Milner (pp. 20723); Shatzman, Res Militares, 184 96. See also idem, Artillery in Judaea, 46182. 56. Avi-Yonah, Maresha, 94851. The transitional character of Hellenistic Marisawith a heavy emphasis on its pre-Hellenistic oriental elementsis argued by G. Horowitz, Town Planning, 93111. 57. E. Stern et al., Excavations at Dor, I, 27778. See also idem, Walls of Dor, 1113; Sharon, Phoenician and Greek Ashlar Construction Techniques, 2142. 58. Crowfoot, Kenyon, and Sukenik, Buildings in Samaria, 2427. 59. Samaria-Sebaste: ibid., 31, 3941; Avigad, Samaria (City), 1307. On a round Hellenistic tower at Akko, similar to the one found at Samaria, see Dothan, Fortications of Ptolemais, 7174. Caesarea: Frova, Scavi di Caesarea Maritima, 24971. 60. Avigad, Discovering Jerusalem, 50, 69; see also Arav, Settlement Patterns, 159. At times, a freestanding gate was built with no accompanying walls, as in rst-century c.e. Tiberias; see NEAEHL, IV, 1470 71.


historical development of the synagogue

2. Plan of the northern city-gate of Herodian Caesarea.

their fortications during the Persian and Hellenistic periods. The fact that Jerusalem had a wall at this time is amply documented (e.g., Neh. 23; Josephus, War 5, 14255), and we have had occasion to note that Ezra and Nehemiah gathered the people at the citys Water Gate for the public reading of Scriptures. Nevertheless, we have no inkling what this Jerusalem gate area might have looked like or if, in fact, it was a city-gate rather than an internal one serving the Temple precincts. Besides Jerusalem, we have no information regarding other Judaean towns or villages.61 It is quite possible that many, if not most, of these places had no fortications whatsoever at this time and that communal activities had already been taking place earlier in an outdoor framework or, in the course of time, inside a public building such as the synagogue.62 The implications of the above discussion point to the fact that, in contrast to the communis opinio, the emergence of the Judaean synagogue was not the outcome of any specic event or crisis but rather a gradual development during the Persian and Hellenistic
61. On the paucity of archaeological material for this period, especially in the hinterland, and on Persias general policy of limiting urban fortications, see Ahlstrm, History of Ancient Palestine, 82829; E. Stern, Archaeology of the Land of the Bible, 46168. 62. Here I am modifying the suggestion for a specically Hellenistic dating of the synagogues origin argued in my article Nature and Origin. In some places, it might have crystallized earlier. The import of the lack of physical evidence for specically Jewish centers was not fully realized then, and therefore a Persian context is also possible.



3. Plan of the Gamla synagogue integrating the city wall.

periods. It is impossible to oer a specic date for the process that transpired at various paces some time between the fth and rst centuries b.c.e. 63 What may be no more than a fortuitous circumstance, the structure at Gamla, which has been generally regarded as one of the earliest known synagogue buildings in Judaea dating to the late rst century b.c.e., seems to reect the earlier city-gate tradition (g. 3). Located neither in the middle of the settlement nor on a particularly high spot, as were many village synagogues in later centuries, this synagogue was situated at the eastern edge of the town, between two main streets and near what was undoubtedly the entrance to the city.64 Had Gamla existed centuries earlier, a city-gate area might well have been

63. The assumption that the synagogue evolved from a communal setting such as the biblical citygate may also explain the association of the azzan with this institutionat least as attested in later rabbinic sources. In the ancient Near East, the azzan functioned as a city administrator, chief magistrate, or mayor (as was Arad-Khipa of Jerusalem and other local Canaanite rulers as mentioned in the el-Amarna letters); see Neeman, Contribution of the Amarna Letters, 2021; he is also mentioned as a religious functionary in early antiquity, e.g., the Babylonian hazzanu of a temple, although this position appears to have been derivative. It is possible that the azzan, who became a permanent xture in the later synagogue, was formerly associated with a village or town (i.e., city-gate) context; however, no source makes this connection explicit. On the hazzanu in the ancient Near East, see Weisberg, Guild Structure, 93; Kaufman, Akkadian Inuences, 55; Menzel, Assyrische Tempel, I, 289; II, 232 nn. 386465; CAD, II, 163 65. For an example of the religious dimension of this term, see Waterman, Royal Correspondence, I, 25455 (Letter 366). 64. Gutman, Synagogue at Gamla, 3034; idem, Gamala, 45963. As can be seen at the site, the synagogue is located adjacent to the city fortications. It should be noted, however, that Gamlas wall was added only in the mid-rst century c.e. in preparation for the war against Rome.


historical development of the synagogue

located there. The structure at Gamla, then, would have preserved vestiges of the earlier tradition of communal public space, located at or near the entrance to the city.


The Torah-reading ceremony, the central religious component of Second Temple synagogue ritual (see below, Chap. 5), may have taken place initially at the city-gate or some other designated communal space. When precisely this custom emerged as a common practice within the Jewish community is unknown. No Greek author of the Hellenistic period, nor any author of the apocryphal and pseudepigraphal books of the last centuries b.c.e., refers to such a practice. However, to assume that the practice did not exist is gratuitous; such an argumentum ex silentio is of limited signicance. Indeed, references to any kind of institutional practiceeven regarding the Jerusalem Templeare not very numerous for these centuries, and it is quite possible that the practice of reading the Torah existed for a long period of time, but may not have merited comment.65 There are certainly enough biblical passages that might have inuenced the development of the Torah-reading custom (e.g., Deut. 6:7; 31:913) or at least legitimized it ex post facto. Earlier theories have considered the original Torah-reading ceremony as either a polemic against undesirable foreign inuences (Samaritans or Hellenism) or, following Deuteronomy, a means of teaching the ordinary Jew about his tradition and religion.66 It has also been suggested that exposure to the Greek world and Greek models may have played a role in creating or, alternatively, strengthening an already existing Torah-reading ceremony (this may well have been an important factor in accounting for Diaspora practice).67 Most scholars once considered the Torah-reading ceremony a substitute for Temple worship deriving from the Babylonian Diaspora.68 In fact, however, it may well have developed parallel to, and not in opposition to or as a substitute for, the sacricial cult; rather, it constituted an additional form of worship and signied increased communal participation in a religious setting. Indeed, the earliest reference we have to a communal Torah-reading ceremony is that of the maamadot, a local ceremony held when a regions
65. See, for example, I Chron. 17: 79, ostensibly describing Jehosaphats day but more likely the time of the Chronicler. On the development of the Mosaic canon, see P. R. Davies, Scribes and Schools, 89106. 66. Samaritans: Bchler, Reading of the Law and Prophets (1) 424. Similarly, Bchler regards the Hellenistic decrees referred to by Josephus as reective of the Jewish-Samaritan controversy of the late Second Temple period; see idem, Die Tobiaden und Oniaden, 14371. Hellenism: Leszynsky, Die Sadduzer, 133. Jewish tradition: J. Mann, Bible as Read, I, 4: Rather the positive aim and need of familiarizing the ordinary Jew on the leisure days of the Jewish calendar with a knowledge of his religion should be regarded as the main raison dtre of this institution. See also Elbogen, Jewish Liturgy, 13031. 67. Kugel and Greer, Early Biblical Interpretation, 56. 68. Fishbane, Biblical Interpretation, 113.



priests (and other lay representatives) ociated in the Temple during an assigned week.69 According to the Mishnaha late source for our purposes but, in this instance, probably reecting Second Temple practicea portion of the Scriptures was read publicly in the town or village on each day of that week: The ordinary Jews [non-priests and nonLevites] associated with each priestly division would gather in their towns and read [the scriptural section dealing with] the creation story [Gen. 12:4]. 70 This mishnah is valuable to our discussion on three counts: (1) it attests to the early practice of Torah reading in a local setting;71 (2) it tells us that the reading was performed by the community at large, in this case by those unable to participate in the Temple worship in Jerusalem; (3) a synagogue per se is not mentioned, but a general reference is made to gathering in towns. Although later renditions of this tradition, followed by most commentators, freely introduce the notion that these scriptural readings took place in the local synagogue,72 such interpretations are wholly gratuitous. The mishnah mentions only towns, and the specic Sitz im Leben may have been the gate area or some other public space.73
69. On the division into twenty-four courses of priests, Levites, and Israelites, see Schrer, History, II, 24550; S. Safrai, Pilgrimage, 21720; Sperber, Mishmarot and Maamadot, 8993; see also Elbogen, Jewish Liturgy, 19092. On the formative period of this institution, see also Liver, Chapters in the History of the Priests and Levites, 3352, esp. pp. 4952. On one of the Temple roles of the head of a maamad, see M Tamid 5, 6 (cf. also M Sotah 1, 5). 70. M Taanit 4, 2; see comments by Albeck, Mishnah, II, 49596. At rst glance, there seems to be little connection between the reading of the creation story and the Temple service. Two possible explanations come to mind: 1. The association may derive from the apparent connection between the building of the Tabernacle (Exod. 2540) and the creation story, on the one hand, i.e., the subdivision of the former into seven parts (six deal with the actual process of building, i.e., with creation, and the last with the Sabbath law), and, on the other hand, the completion of the Tabernacle on New Years Day (Exod. 40:17), traditionally associated with the Sabbath after the Creation. In other verses as well the Sabbath is associated with the Temple (Lev. 19:30; 26:2). Such a connectionalready made in rabbinic sourcesmay also explain why the period of service for each priestly course extended from Sabbath to Sabbath, reminiscent of the week of creation. See Cassuto, Commentary on Exodus, 33435, 447.; Weinfeld, Sabbath Temple and the Enthronement of the Lord, 50112; Sarna, Exploring Exodus, 196220. 2. In rabbinic literature, there is a tradition that the creation of the world began on the Temple Mount, on its foundation stone ( .) A related tradition speaks of Adam being created from the dust of the Temple Mount, where he was later buried. The Christians knew of this tradition and subsequently transferred Adams burial spot to Golgotha, in reaction to which later rabbinic sources claim he was buried in the Cave of Machpelah in Hebron. See Ginzberg, Legends of the Jews, I, 12; V, 1416, 12526; Gafni, Pre-Histories of Jerusalem, 1016. See also Terrien, Omphalos Myth, 31538. 71. The division into maamadot was already known to the author of I Chronicles (24:118) in the fourth or third century b.c.e., and the customs described in the mishnah may well have crystallized as early as this time. 72. See, for example, B Taanit 27b. See also Albeck, Mishnah, II, 341; Schrer, History, II, 293. 73. A number of scholars have viewed the maamadot ceremony as the setting from which the syna-


historical development of the synagogue

Of particular interest in the above-cited mishnah is the fact that the local Torahreading ceremony of the maamad was clearly parallel to the Temple ritual, i.e., it was meant to serve as a substitute for those unable to be in Jerusalem. As noted, the emergence and evolution of the synagogue have been viewed in the past as a competitive development vis--vis the Jerusalem Temple, and many have even characterized the synagogue as a Pharisaic institution that emerged in response to the Sadducean-run Temple.74 Howgogue as we know it emerged. See, for example, Hruby, Die Synagoge, 1617; Bowker, Targums and Rabbinic Literature, 910; Zeitlin, Rise and Fall of the Judaean State, I, 179; Petuchowski, Liturgy of the Synagogue, 46. On the textual and historical issues of this post-70 ceremony, see D. Levine, Communal Fasts and Rabbinic Sermons, 6696, 10319. Another early rabbinic tradition that may have a bearing on the role of a public space in Second Temple towns and villages is found in M Bikkurim 3, 2, where the following ritual of bringing rstfruits to the Temple is described: How do they bring up [to Jerusalem] the rstfruits? All the villages in the [area of the] maamad gather together in the maamads city and lodge in the city square [plaza or gate area, ] and they would not enter the houses and early in the morning the person in charge would say, Let us arise and go up to Zion, to [the house of ] the Lord our God ( Jer. 31:5). Perhaps of signicance is the fact that the Tosefta (Bikkurim 2, 8 [p. 292]) speaks of the azzanim of the synagogue accompanying the processions to Jerusalem. Finally, the mishnaic tradition of public fast-day ceremonies conducted in the public square may also have a bearing on the question at hand (M Taanit 2; T Taanit 1, 813 [pp. 32528]; and later parallels). However, the specic literary tradition that has been preserved appears to be a second-century c.e. version, to wit, the mention of a nasi and av bet din, the recitation of an elaborate Amidah prayer (M Taanit 2, 24), and the reference to a particular incident and specic sages of this time (ibid., 2, 5; T Taanit 1, 13 [pp. 32728]). 74. The list of scholars making this assumption is long and impressive. For a sampling, see Herford, Pharisees, 88109; R. M. Grant, Historical Introduction to the New Testament, 27475; Tcherikover, Hellenistic Civilization, 12425 (referring to scribes instead of Pharisees); Hengel, Judaism and Hellenism, I, 82; idem, Pre-Christian Paul, 57; Hengel and Deines, E. P. Sanders Common Judaism, 3233; Gutmann, Synagogue Origins, 4; Hanson, People Called, 353. Flesher (Palestinian Synagogues, 28) has similarly posited a sharp contrast between the priestly Temple cult and the synagogue, although without Pharisaic-Sadducean overtones. Cf., however, Saldarini, Pharisees, Scribes and Sadducees, 5253; Grabbe, Synagogues in Pre-70 Palestine, 4089; G. F. Moore, Judaism, I, 28687; Finkelstein, Pharisees, 568 69; Perrot, Reading of the Bible, 150; and Schaper, Pharisees, 421. Given the assumed spiritual prominence of the Pharisees in Jewish life of the Second Temple period and the universality of the synagogue as a religious institution, it was more than natural in the past to posit Pharisaic dominance and inuence over the synagogue. In no small part, the polemics of the New Testament were responsible for this misconception, pitting Jesus against the Pharisees, often in a synagogal context (Matt. 12:14; 23:16). Primarily, however, these views reect the ongoing belief that the Pharisees determined the normative religious tradition in the Second Temple period. This point of view has in large measure been rejected today; see, for example, M. Smith, Palestinian Judaism, 7381; Aune, Orthodoxy in First Century Judaism? 110; S. J. D. Cohen, From the Maccabees to the Mishnah, 16064; idem, Were Pharisees and Rabbis? 89105; Grabbe, Synagogues in Pre-70 Palestine, 4089; E. P. Sanders, Judaism, 4772, 380412. If, then, Pharisaic Judaism was not normative, and if many literary compositions of the Second Temple period often ascribed to them by modern scholars are, in fact, not theirs (e.g.,



ever, the truth of the matter is, the Pharisees had little or nothing to do with the early synagogue, and there is not one shred of evidence pointing to a connection between the two. No references associate the early Pharisees (the Pairs and others) with the synagogue, nor is there anything in early synagogue liturgy that is particularly Pharisaic. A possible connection between the judicial process once conducted at the city-gate and the synagogue may be found in the book of Susannah. The focus of this story is a meeting attended by elders and townspeople for the purpose of adjudication. According to the Septuagint version, the trial took place in the synagogue (); the revised version of Theodotion merely states that the people assembled. 75 Scholars generally regard the Septuagint tradition as older and assume that it had a Semitic Vorlage of Palestinian provenance, and that it was probably composed in the Persian or, more likely, Hellenistic period.76 Thus, we may have here an early attestation of the term synagogue, referring either to a separate building or an outdoor area that served as the setting for a major communal activity.77 An advanced stage in the transition to a synagogue building can justiably be dated to the late Hellenistic or early Roman period, i.e., the rst century b.c.e., from when the earliest archaeological remains of synagogues appear (see below, Chap. 3). Our argument hereunfortunately, but of necessityrests heavily on the lack of evidence. Had the synagogue been a known and recognized institution in Judaean society, one might well have expected it to be mentioned at least once in the many literary works of the third to rst centuries b.c.e. 78 Yet, despite his numerous references to Jewish elites and religious institutions of the day, Ben Sira takes no note of it. We also might have expected I and II Maccabees to comment on the impact of Antiochus persecutions on the functioning of synagogues, had they existedyet, they do not. Purity laws, circumcision, the Sabbath, festivals, kashrut, and, of course, the Temple are all mentioned, but not a word is said about the synagogue.79 Compare this silence to descriptions of rst-century c.e.

on the Psalms of Solomon; see ODell, Religious Background of Psalms, 24157; see also E. P. Sanders, Paul, 4024), then there is little justication in speaking of their control over the local synagogues. 75. Charles, Apocrypha and Pseudepigrapha, I, 649; C. A. Moore, Daniel, Esther and Jeremiah: Additions, 1014. 76. C. A. Moore, Daniel, Esther and Jeremiah: Additions, 9192. 77. The reference in Matt. 6:5 to those who pray in the synagogues and on the street corners is intriguing and has been the basis of much speculation. However, there are too many unknowns in the above statement for it to be of much use. In the rst place, the passage in Matthew may be referring to private, and not communal, prayer. Secondly, the statement may be referring to the exception rather than the rule. It might be interpreted that such behavior is typical of the arrogant and insincere and that ordinary people do not and should not act thusly. 78. For the literature dating to this period, see Nickelsburg, Jewish Literature, 71160; Hengel, Judaism and Hellenism, I, 110.; M. Stone, Apocalyptic Literature. 79. I Macc. 1, 4164; II Macc. 6.


historical development of the synagogue

pogroms and anti-Jewish incidents in both Judaea and the Diaspora, where the synagogue featured prominently in the writings of both Josephus and Philo.80 The same silence permeates the books of Tobit and Jubilees, the Letter of Aristeas, the Testaments of the Twelve Patriarchs, the book of Enoch, and elsewhere. The synagogue as a place of worship or central communal institution in Jewish life is nowhere to be detected. Similarly, third- and second-century b.c.e. non-Jewish writers who describe the Jewish sceneparticularly its religious componentomit all reference to this institution.81 When the word synagogue does, in fact, appear in the late second-century b.c.e. book of I Maccabees, it refers to a group of people: an assembly of asidim (2, 42), an assembly of scribes (7, 12), or the large national assembly ( ) that acclaimed Simon the Maccabee as ruler in 141 b.c.e. (14, 28).82 In the Septuagint, the term invariably refers to a congregation as a whole and not a meeting-place.83

From our discussion above, we may safely conclude that the synagogue as a distinct and recognizable institution began to crystallize owing to a shift in urban planning. The earlier setting for the synagogues many functions was the biblical city-gate, with its manifold activities. This setting served as such throughout Canaan and the Near East for centuries, if not for a millennium or more, and thus biblical Israel was adopting a well-known model. The later move into a building in place of the open-air city-gate setting was likewise a not unfamiliar phenomenon in the Hellenistic age. Within Jewish society, this move constituted the beginnings of the institution that became known as the synagogue. The shift from the city-gate as a center of activity to a setting within the city may have aected Jerusalem and its Temple as well. One of the striking aspects of the rst-century
80. See below, Chaps. 3 and 4. This line of argument must, however, be tempered for the pre-rstcentury c.e. Egyptian Jewish literary sources likewise make no mention of a proseuche in the Hellenistic period, although that institution is well attested in epigraphical and papyrological sources. 81. This point has already been noted on numerous occasions in the previous two centuries, at the very least from the time of Bauer (Beschreibung, II, pp. 125., down to Rivkin, Ben Sira and the NonExistence of the Synagogue, 32054). The interpretation of Ps. 74:8 (they burned all Gods tabernacles in the land) as a reference to the destruction of synagogues during the Antiochan persecutions is hardly accepted today (see Goldstein, I Maccabees, 138 n. 208). 82. See my Jerusalem, 99101. 83. See TDNT, VIII, 805. It may be precisely for this reason, i.e., the association in Second Temple literature of the word synagogue with congregation, that the word also came to be applied to a building that housed communal activities. That the synagogue building belonged to the community as a whole is clearly attested in early rabbinic tradition; see, for example, M Nedarim 5, 5; T Bava Metzia 11, 23 (p. 125); Lieberman, TK, IX, 320; Y Megillah 3, 2, 74a; Y Yevamot 12, 13a; Genesis Rabbah 81, 1 (pp. 969 72); and below, Chap. 10.



Jerusalem Temple is the fact that it had become the hub of various activities not only for those living within the city but also for Jews throughout Judaea and beyond. More precisely, it was on the Temple Mount that a wide variety of social, political, religious, judicial, and economic functions took place.84 Never before had the Temple area functioned in such a comprehensive capacity. When did this plethora of functions move to the Temple Mount area? It, too, may have occurred with the disappearance of the city-gate as a place of assembly and the subsequent establishment of a vibrant Temple Mount area. There can be little doubt that the Hasmonean renovations on the Temple Mount, and most certainly Herods enlargement of the area to twice its former size, were done not only to accommodate the increased numbers of pilgrims or to satisfy personal ambitions, but also to house the new functions envisioned for the Temple Mount. When Herod planned this area architecturally as a Hellenistic temenos on the one hand and as the equivalent of a Greek agora or Roman forum on the other, he was, in fact, centralizing in this public area activities once conducted in other parts of the city, primarily the city-gate area. This development thus coincided with the wider relocation of communal functions from city-gates to other settings, as was the trend elsewhere in the Hellenistic East. If, indeed, this assumption is to be granted, it might then explain a number of interesting and hitherto enigmatic phenomena, such as the ceremonial reading of the Torah within the Temple precincts. This was the only formal liturgical activity not specically associated with sacrices (in contrast to the levitical psalms, which were meant to accompany the sacricial ritual) that took place there. Annually on Yom Kippur, as well as on the Sukkot festival at the end of each sabbatical year ( ,Haqhel, in Deut. 31), the high priest would read from the Torah in public.85 Such proceedings were clearly ancillary to the Temples main agenda and, as such, were conducted in the Womens Court and not in the Priests Court, thus opening this activity to a wider public gathered in the Temple precincts. It is quite possible that the Temple Haqhel ceremony, at least, might have been a carryover from the earlier city-gate setting and paralleled the Torah-reading ceremony that was developing in the synagogue at this time. Given the possible shift of venue from Jerusalems city-gate to the Temple Mount, the puzzling mention of two functionaries, the rosh knesset ( ) and azzan knesset ( ,) with regard to these Torah-reading ceremonies may also gain a measure of clarity. Among the references to the synagogue in the rst centuries c.e., rabbinic literature specically notes these two ocials in connection with this ceremony.86 Might such

84. L. Levine, Jerusalem, 21953. 85. M Yoma 7, 1; M Sotah 7, 8; see also M Tamid 5, 3. 86. T Megillah 3, 21 (p. 359); T Sukkah 4, 6 (p. 273). In the Temple setting, both the azzan knesset and rosh knesset were subordinate to the priestly hierarchy; in the newly developing synagogue, these roles were destined to become central. For literature in this regard, see above, note 63.


historical development of the synagogue

ocials have once held pivotal roles in city-gate aairs, including religious activities, and could these roles have then been transferred to the Temple and synagogue settings? Finally, our suggestion regarding the evolution of the synagogue has implications regarding its relationship with the Temple. As noted above, it is totally unwarranted to view the synagogue as a rival of the Temple.87 Throughout the Second Temple period, at least in Judaea, the synagogue was never endowed with any special sanctity or halakhic importance,88 and in several sources priests themselves were prominent in synagogue aairs as well.89 There is simply no evidence of any inherent conict between these two institutions, and priestly participation and leadership in synagogues simply reect the role they played as religious and communal leaders at the time. The supposition that the synagogue began emerging (socioreligiously, if not architecturally) in the course of the early Second Temple period, some time in the Persian or Hellenistic period, would also go a long way toward explaining the development of the Jewish proseuche in Egypt and some of its salient features, including the considerable inuence of Ptolemaic models. The Jews who settled in Egypt and elsewhere in the early Hellenistic period had no concrete, established framework for communal activities, which in Judaea had taken place at the city-gate or in a village public space. In adapting to a new environment and in seeking a setting for their communal functions, the Jews in Egypt and throughout the Diaspora looked to their immediate surroundings for suitable models.90 There is little to be gained in arguing whether the synagogue originated in Judaea or the Diaspora. In fact, the institution evolved in both places more or less simultaneously, under varying circumstances and for very dierent reasons. For the Jews of Judaea, there was a need to nd an alternative to the city-gate area, now fast disappearing, or to the village open-air setting, which was gradually being replaced by closed public buildings. For Jews in the Hellenistic and Roman Diaspora, use of the city-gate area in a pagandominated town or city was impossible. It was thus imperative to create a framework within which the Jewish community could preserve its communal identity and give expression to social and religious needs. Under these varying circumstances the synagoge and proseuche were born.
87. Contra Flesher, in Palestinian Synagogues, 2739. See my comments below, Chap. 3. 88. S. J. D. Cohen, Temple and the Synagogue, 15174. 89. See E. P. Sanders, Judaism, 17082; and below, Chaps. 35. 90. See below, Chap. 4.



o pre-70 source addresses the nature or functions of the Judaean synagogue systematically.1 In contrast to the Temple, the synagogue merited relatively little attention; we have few sources on how synagogues functioned, where they were located, or how they lookedaspects about which Josephus and the Mishnah supply a plethora of information with regard to the Temple.2

1. We have been using the name Judaea for the pre-70 period, as this was the ocial title of the Roman province at the time. Only in the wake of the Bar-Kokhba revolt in 135 c.e. did Hadrian rename it SyriaPalaestina, a change that will be reected in later chapters by use of the name Palestine. It should be noted that the name Judaea had a dual meaning in the pre-70 eraa limited reference to the southern part of the country (as against Samaria, Galilee, and Peraea) and a broader one relating to the entire province. See M. Stern, GLAJJ, I, 23334, 290; II, 1115, 16870, 21720; Feldman, Some Observations on the Name of Palestine, 614; and Schrer, History, I, 514. We shall distinguish between the two meanings by using the spelling Judea for the more limited geographical area, and Judaea for the province as a whole. 2. Josephus, War 5, 184237; idem, Antiquities 15, 380425; M Middot; M Sheqalim. Studies on the Jerusalem Temple based on these sources and archaeological evidence are abundant; see, for example, Hildesheimer, Die Beschreibung des herodianischen Tempels, 132; Hollis, Archaeology of Herods Temple; Simons, Jerusalem in the Old Testament; Vincent and Stve, Jrusalem de lAncien Testament; Busink, Der Tempel von Jerusalem; E. P. Sanders, Judaism, 47118; and L. Levine, Jerusalem, 21953 and bibliography cited therein. An interesting example of this imbalance is the Roman practice of oering gifts to the Jerusalem


historical development of the synagogue

As noted, the synagogue at this time had no halakhic or religious standing; it was a communal institution and, as such, merited no special status and consequently little attention.3 Nevertheless, the picture is not entirely negative. Almost a score of synagogues in rst-century c.e. Judaea are attested, especially in the literary sources; however, they are mentioned only en passant within the given agenda of each particular source.4 These include references in Josephus writings (Tiberias, Dor, and Caesarea), the New Testament (Nazareth, Capernaum, and Jerusalem), rabbinic literature (Jerusalem), and the Damascus Document (Qumran and presumably elsewhere as well). Six sites have yielded archaeological evidence for the rst century, the well-known remains from Gamla, Masada, Herodium, and Jerusalem, and two recently discovered buildings at Qiryat Sefer and Modiin in western Judaea (a possible additional site has been suggested at orvat Etri, south of Bet Shemesh). The designation of several other archaeological sites as synagogues is far less certain (Capernaum, Migdal, Chorazim, northern Jerusalem, and Jericho). Although this material is scattered throughout the entire province of Judaea, a concentration of nds exists in the Galilee (including the Golan), the Judaean Desert, and the Shephelah.5 Let us review this evidence region by region.

GALILEE Some Methodological Considerations

Because much of our information regarding the Galilee derives from the New Testament accounts of Jesus, determining the historical reliability of this material is of paramount importance. All the gospels refer to Jesus activity in Galilean synagogues. Mark and Luke speak of Jesus preaching and healing in synagogues throughout the region, Matthew of teaching, preaching, and healing there, and John of speaking openly in synagogues and in the Temple.6 The towns and villages that Jesus frequented (as has often

Temple; no comparable gesture is known regarding the pre-70 synagogue. See, for example, Philo, Embassy 157, 297, 319; Josephus, War 5, 562; idem, Antiquities 16, 14; 18, 122. 3. S. J. D. Cohen, Temple and the Synagogue, 15274. 4. It is dicult to view the term hieroi in War 4, 408, as referring to synagogues (per Binder, Into the Temple Courts, 12230). That the Sicarii would carry out a general destruction of synagogues would be strange indeed, as is the reference to cities attacked by the Zealots. Despite the diculty in assuming that there were such pagan temples in Judea, it is more likely that Josephus did, in fact, have such buildings in mind and that he was referring to greater Judaea, including pagan areas. Another possible reference to such structures may be found in War 1, 277, and its parallel in Antiquities 14, 374. 5. For most of the sites discussed below, see also Binder, Into the Temple Courts, 155204. 6. Evidence from each gospel is referred to respectively: Mark 1:2128, 39; 3:1; 6:2 (for other settings in Jesus ministry according to Mark, see Rhoads and Michie, Mark as Story, 6372); Matt. 4:23; 9:35; Luke 4:1544; 13:1021; John 6:3559; 18:20. See also Schwartzman, How Well Did the Synoptic Evangelists Know? 11532.

pre-70 judaea


been noted, he seems to have studiously avoided contact with cities)7 all appear to have had synagogues, whatever their size, location, or social-religious conguration. Jesus visits to these synagogues on the Sabbath were clearly timed so as to aord him maximum exposure among the local population, or at least the gospel narratives purport to convey this message. Unique among rst-century sources, the gospel narratives focus on the healing eected within the synagogue. Miracle-working activity was far from uncommon in antiquity generally and in Judaea specically.8 Exorcising demons, performing wonders, making miraculous signs, and healing the sick within the context of the synagogue would not have been considered unusual given the institutions centrality in Jewish communal life. What is highlighted in these narratives, however, is not only the nature and extent of this practice, but also the fact that it took place on the Sabbath. It is this timing that was controversial and thus became a bone of contention between Jesus and some of those present. In one case he is reportedly criticized by the archisynagogue, in another by the Pharisees.9 The gospel authors appear to have emphasized a common rural phenomenon, i.e., healing, in order to highlight Jesus charisma and popularity. The historical value of this New Testament literature regarding the rst-century Judaean scene has been called into question of late. It has been suggested that all the gospels, Matthew perhaps excepted, were written in the Diaspora and thus may reect a late rstcentury c.e. reality, i.e., the Diaspora synagogue familiar to the authors, who then projected this later setting onto the earlier rst-century Galilee.10 Moreover, the literary and theological agendas of each gospel have raised further doubts as to the materials historical credibility. However, in our case at least, such skepticism should be regarded with extreme caution. Although the gospels and Acts, those writings most relevant to the subject at hand, may have originated in the Diaspora, they are not so distant chronologically from Jesus setting; the dierence between the latter part of the rst century c.e. and ca. 30 c.e. is but a generation or two. Furthermore, it is not at all clear that a Diaspora setting had an impact on these accounts. All the gospels, no matter when or where they were written, report much the same information regarding the Galilean synagogues. How might we explain this coincidence? Are we to assume that all Diaspora synagogues, whether in Rome, Alexandria, Antioch, or elsewhere, shared identical features? This, as we will see below
7. Freyne, Galilee, Jesus and the Gospels, 13555; R. A. Horsley, Galilee, 158255; Edwards, SocioEconomic and Cultural Ethos, 5373. See also Freyne, Urban-Rural Relations, 7591. On urban-rural relations generally in the Empire, see MacMullen, Roman Social Relations, 2856. 8. Kee, Miracle in the Early Christian World, 14670; Freyne, Galilee, Jesus and the Gospels, 22739; and especially Aune, Magic in Early Christianity, 150757. See also E. P. Sanders, Jesus and Judaism, 157 73; M. Smith, Jesus the Magician, 6893; and Crossan, Historical Jesus, 30332. 9. Luke 13:14; Matt. 12:14, respectively. 10. See, for example, Kee, Transformation of the Synagogue, 18.


historical development of the synagogue

(Chap. 4), was almost certainly not the case; Diaspora synagogues varied considerably from region to region. Moreover, even were we to assume the existence of some sort of overriding unity among the far-ung rst-century Diaspora synagogues, might it not be warranted, then, to assume that Judaean synagogues shared in that commonality? Moreover, there is a signicant amount of corroboration between the New Testament evidence and other information relating to rst-century Judaea. For example, Acts refers to Diaspora Jewrys synagogues in Jerusalem, a presence reected in rabbinic literature and the Theodotos inscription as well (see below). The centrality of scriptural readings in the synagogue (viz., Luke 4) is echoed by the Theodotos inscription, Philo, Josephus, and several early rabbinic traditions (e.g., T Megillah 3 [p. 359]). Finally, we might wonder why New Testament writers would attribute a Diaspora synagogue setting to Jesus time if such an institution had no place in the Galilee. It appears rather unlikely that all the gospel writers would refer time and again to Jesus activity in an institution that never (or barely) existed in his time.11 The assumption, then, that there were no synagogue buildings in Galilean towns and villages in the rst century appears unwarranted.12 That such buildings existed throughout Judaea at the time is now well attested for Jerusalem and other parts of the province. Moreover, the recent archaeological discoveries at Qiryat Sefer and Modiin provide us with examples of village synagogues in western Judaea. Whether such evidence is to be found in every village is unknown, but this is probably not the case. Smaller settlements may have made do with a village square or someones home, but this is mere speculation.

Jesus appearance in his hometown synagogue is mentioned in each of the synoptic gospels.13 It was a memorable occasion, and it is precisely his appearance in his patris that determined the nature and course of this particular narrative. The accounts in Matthew (13:5358) and Mark (6:16)14 are similar: Jesus teaches at the local synagogue; those gathered are astonished by his words and deeds; he is immediately identied by those as-

11. That the gospel writers have been characterized of late less as historians and more as theologians and literary writers should not be given undue weight for our particular purposes; see Conzelmann, Theology of St. Luke, 3438; Talbert, Literary Patterns, contra H. Anderson, Broadening Horizons, 26174. It is rather safe to assume that even those writers using historical data in a supposedly biographical or historical narrative (e.g., Luke-Acts) would include as much reliable data as possible to make the account convincing. So, for example, Luke 1:14: It seems good to me . . . to write an orderly account. Regarding the inclusion of very early (and thus relatively authentic and historical) material in much of Lukes narrative, see Schurmann, Lukasevangelium, 223.; Ringe, Jesus, Liberation and the Biblical Jubilee, 4245, 1079; and Chilton, God in Strength, 12132. 12. See Groh, Stratigraphic Chronology, 5760; R. A. Horsley, Galilee, 22526. 13. Temple, Rejection at Nazareth, 22942. 14. See Perrot, Jsus Nazareth, 4049.

pre-70 judaea


sembled as the carpenters son, the son of Mary, brother of James, Joseph, Simon, Judas, and his sisters. We are told that those gathered were highly oended, although the reason for this is not made clear. Jesus reply that a prophet is not without honor except in his own country [i.e., hometown] and in his own house (Mark adds: among his own kin) may well be what triggered the above account. Both gospel pericopes add that owing to his townsmens disbelief, Jesus did not succeed in performing miracles there. Mark notes that Jesus nevertheless managed to heal some sick people. Lukes account of Jesus visit to the Nazareth synagogue is markedly dierent from the other gospel accounts, so much so that it has been suggested, in a harmonizing fashion, that Jesus made two dierent visits to Nazareth, one described in Mark and Matthew, the other in Luke.15 Whatever the case, the importance of Lukes pericope cannot be overestimated for our understanding of the rst-century Galilean synagogue.16
And he came to Nazareth, where he had been brought up: and, as was his custom, he went into the synagogue on the Sabbath day and stood up to read. And he was handed the book of the prophet Isaiah. And when he had opened the book, he found the place where it was written, The Spirit of the Lord is upon me, because He has anointed me to preach the gospel to the poor; He has sent me to proclaim release to the captives, and recover the sight to the blind, to set at liberty those that are oppressed, to proclaim the acceptable year of the Lord [Isa. 61:12].17 And he closed the book, and he gave it back to the attendant, and sat down. And the eyes of all those in the synagogue were xed on him. And he began to say unto them, Today this scripture is fullled in your ears. And all spoke well of him, and wondered at the gracious words which proceeded out of his mouth. And they said, Is this not Josephs son? And he said unto them, You will surely quote me the proverb, Physician, heal yourself : what we have heard you did in Capernaum, do also here in your own country. And he said, Truly, I say to you, no prophet is accepted in his own country. But I tell you in truth, many widows were in Israel
15. Luke 4:1630; Perrot, Jsus Nazareth, 47. See also Lagrange, Lvangile, 123, 201. The gospels themselves do not know of two visits. Alternatively, the account in Luke may be a rewriting or reworking of Mark 16 (see Leaney, Commentary, 5152; Bultmann, History of the Synoptic Tradition, 3132), honed to t Lukes theology (Conzelmann, Theology of St. Luke, 3438) and part of Lukes programmatic agenda (Brawley, Luke-Acts and the Jews, 1112; cf. J. T. Sanders, Jews in Luke-Acts, 16468). See also Kmmel, Introduction, 13051. 16. See, for example, Finkel, Jesus Sermon at Nazareth, 10615; idem, Jesus Preaching, 32541; H. Anderson, Broadening Horizons, 25975; Combrink, Structure and Signicance of Luke 4:16 30, 2747; Perrot, Luc 4:1630 et la lecture biblique, 32440 and bibliography in n. 11; J. T. Sanders, From Isaiah 61 to Luke 4, 75106; Chilton, Announcement in Nazara, 14772; C. A. Evans, Luke, 70 76; Monshouwer, Reading of the Prophet, 9099; Tyson, Images of Judaism, 5962. On the date and authorship of Luke, see Fitzmyer, Gospel according to Luke, 3562. 17. This passage is used also in Matt. 11:45 and Luke 7:22. The verses in our present context have been carefully edited by excluding some phrases, incorporating a clause from Isa. 58:6, and generally following the Septuagint version. See R. A. Horsley, Jesus and the Spiral of Violence, 25051.


historical development of the synagogue

in the days of Elijah, when the heaven was shut up three years and six months, when great famine was throughout the land; but to none of them was Elijah sent, except to Sarepta, in the territory of Sidon, to a woman who was a widow. And many lepers were in Israel in the time of Elisha the prophet; and none of them was cleansed, save Naaman the Syrian. And all those in the synagogue, upon hearing these things, were lled with anger, and rose up, and thrust him out of the city, and led him unto the edge of the hill on which their city was built, that they might cast him down headlong. But passing through the midst of them, he went away.18

This passage is far longer and more detailed than its parallels and seems to have been deliberately placed by Luke at the very beginning of Jesus career. Mark and Matthew, in contrast, place their versions of the Nazareth incident later in Jesus Galilean ministry.19 The positioning of this tradition is clearly of signicance for Lukes account; he obviously intends to use Jesus inaugural address in Nazareth to set forth the main themes of his gospel and its companion volume, Acts.20 As regards the synagogue itself in this account, a number of details are noteworthy. According to Luke, Jesus was accustomed to go to the synagogue on the Sabbath, either as an ordinary participant, a preacher, or a healer; other gospel traditions bear this out. Certain stages of the synagogue liturgy are noted: Jesus stood up to read from the Prophets, was handed the book of Isaiah, read several verses, returned the book to the synagogue ocial, sat down, and proceeded to address the congregation. Surprisingly, reading from the Torah is omitted. However, rather than conclude that this was an unusual practice, it is more reasonable to assume that Luke omitted the Torah reading and noted only that of the Prophets because this alone was congruent with his agenda (i.e., Jesus subsequent sermon).21 The actual selection of the prophetic reading described by Luke is enigmatic. Was it the prescribed reading for that particular Sabbath? Who made that decision? Lukes account seems to indicate that Jesus himself chose the passage. If so, was this Lukes invention, or was it indeed an accepted practice in the Galilee, and perhaps in most contemporary synagogues for that matter? Given the absence of parallels, however, no denitive answer is apparent. The sudden and dramatic change in the peoples attitude toward Jesus (vv. 22, 28) may
18. Luke 4:1630. 19. Finkel, Jesus Sermon at Nazareth, 115. 20. These themes can be summarized as follows: (1) Jesus message is rooted in Jewish tradition the synagogue setting, reading from Scriptures, and preaching; (2) his mission is of a decidedly social, humanitarian naturehelping the poor, releasing captives, curing the blind, and freeing the oppressed with a distinctly miracle-oriented component; (3) fulllment is to take place imminently, and Jesus himself is the messianic prophet; (4) having been rejected and persecuted by the Jews, Jesus mission will address the gentiles, as did those of Elijah and Elisha (see also Luke 17:724). See Combrink, Structure and Signicance of Luke 4:1630, 3942; Brawley, Luke-Acts and the Jews, 127; J. T. Sanders, From Isaiah 61 to Luke 4, 75106; Siker, First to the Gentiles, 7390. 21. In Acts 13:1516, the readings from both the Torah and the Prophets are noted.

pre-70 judaea


have resulted from the specic message delivered on this occasion, or it may reect the fellow-townsmens contrasting attitudes toward him, ranging from sympathetic acceptance to hostile rejection. Luke may well have chosen to condense these reactions into his account of Jesus programmatic sermon.22

Jesus activity in the Capernaum synagogue is mentioned in all the gospels.23 For Mark, followed by Matthew, this was, in essence, the beginning of his ministry. The synoptic gospels focus primarily on Jesus healing activity there while, according to John, Jesus delivered a long exposition regarding his divinity, the Eucharist, and Mystical Body clearly a speech setting forth the authors theological agenda. In contrast, nothing explicit is reported in the synoptic gospels about the content of his teaching in Capernaum, other than the fact that all who heard him were amazed. Jesus taught with authority and was, Mark adds, unlike the scribes. He reputedly exorcised demons and healed a withered hand there, and restored to full health a leper as well as a deaf and lame person.24 Whereas in Nazareth Jesus words caused an uproar, in Capernaum his deeds did so. A noteworthy aspect is the congregations reaction to Jesus act of healing on the Sabbath.25 The Pharisees (as well as the scribes, according to Luke, and the Herodians, according to Mark) found his behavior objectionable and sought ways to counter his inuence and activity.26 The historicity of any or all of these particular groups opposition is questionable but not impossible. Why such traditions developed specically with regard to Capernaum, rather than elsewhere, is unclear, unless we assume that this was Jesus base of operations throughout his Galilean ministry.27 Of further interest regarding Capernaum is the Lucan tradition that a Roman centurion stationed in the town built this synagogue.28 Although this claim is found only in Luke, Matthew, for his part, highlights the Roman ocers piety and faith; while Luke
22. Opinions vary regarding the reasons for this anger. See Brawley, Luke-Acts and the Jews, 1618 and bibliography cited there. See Bruce, Acts of the Apostles, 1518. 23. Mark 1:2129; Matt. 12:914; Luke 4:3138; John 6:3559. 24. Matt. 11:5; Luke 7:22; Mark 3:15. 25. Mark 3:16; Luke 6:611. See TDNT, VII, 2026. 26. On the identity of these groups and how they function in each gospel, see, inter alia, Schrer, History, II, 322.; Cook, Marks Treatment of Jewish Leaders; Saldarini, Pharisees, Scribes and Sadducees; Malbon, Jewish Leaders in the Gospel of Mark, 25981; D. R. Schwartz, Studies in the Jewish Background of Christianity, 89101. On the constantly recurring thesis that the Herodians were Essenes, see the judicious comments of W. Braun, Were the New Testament Herodians Essenes? 7588 and bibliography cited there. 27. See Iwe, Jesus in the Synagogue of Capernaum. 28. Luke 7:15. For possible archaeological remains of a rst-century synagogue building at Capernaum, see below.


historical development of the synagogue

emphasizes the mans worthiness as a gentile, Matthew focuses on his personal faith.29 Lukes reference to the Roman centurion ts neatly into his overriding interest in portraying the gentiles openness and receptivity to Jesus message, even from the very outset of his ministry. Thus, the centurions faith and humility, along with his generosity and support of the synagogue, were suitable traditions for Luke to include. Did Luke himself invent this account of the centurions building of the Capernaum synagogue? Was it a product of an earlier tradition that he inherited, or is it, in fact, a valid piece of historical evidence? 30Once again, the issue of historicity remains moot.

In anticipation of a Roman invasion following the outbreak of the revolt in Jerusalem, Josephus was sent to organize the Galilee in 6667 c.e. Tiberias gures prominently in Josephus writings owing either to the citys pivotal role in the region, the lengthy process by which the Tiberians decided whether or not to join the revolt, Josephus particular need in the 90s to refute personal attacks by Justus of Tiberias regarding his conduct of the war, or all of the above.31 Whatever the reasons, Josephus movements in and around the city, his meetings, speeches, and escapes, as well as the names of local leaders, leave us with a much more detailed picture of Tiberias than of any other Galilean city of the time.32 Josephus mentions a Tiberian proseuche on three occasions. The rst time it is described as a very large building ( ) that could accommodate a large crowd ( )33 and where deliberations were held on that Sabbath morning. Nothing is said about worship, which presumably took place beforehand. The meeting itself consisted of a series of speeches, and the participants disbanded only at midday, when, Josephus notes, the Sabbath meal was served.34 Another meeting was set for the next day, and very early that morning, at the rst hour,35 people again gathered in the proseuche to resume discussions. A third meeting was called for the following day, which was also proclaimed a day of public fast.36 The proceedings began with the usual ( ) service
29. Matt. 8:513. On this narrative in its Matthean and Lucan contexts, see R. P. Martin, Pericope of the Healing of the Centurions Servant/Son, 1422. 30. Luke may have exaggerated the size of the centurions gift, given the fact that it came from an army ocer; in fact, the Roman may have made only a modest contribution. For the acceptance of the historical validity of Lukes statement, see White, Building Gods House, 86. 31. S. J. D. Cohen, Josephus in Galilee and Rome, 11470. 32. Ibid., 21621; see also Schrer, History, II, 17882; Avi-Yonah, Founding of Tiberias, 16069; Hoehner, Herod Antipas, 91100; U. Rappaport, Tiberias and Her Role in the Great Revolt, 1223; Freyne, Galilee, Jesus and the Gospels, 12934; R. A. Horsley, Galilee, 7880, 16974, 27175. 33. Life 277. 34. Ibid., 279. 35. Ibid., 280. 36. Ibid., 290303.

pre-70 judaea


for a fast day, but soon thereafter a confrontation ensued, which quickly became heated and violent. Josephus references to the Tiberian proseuche and the events there are noteworthy on several counts. First is the term itself. This is the only instance in which a Judaean synagogue is referred to as a proseuche (although an equivalent Hebrew term does appear in the Damascus Documentsee below). In Josephus other accounts, as well as in the gospel traditions, the Theodotos inscription, and rabbinic literature, the term used for this Judaean communal institution is invariably synagoge. Why, then, was Tiberias dierent? If, indeed, proseuche was used primarily with regard to the Diaspora, then it is possible that there was some tie between this Tiberias building and the Diaspora. In the absence of rm evidence, we can only speculate. Perhaps it was owing to the fact that, in many ways, the city was modeled after a Hellenistic polis by its founder, Herod Antipas. It had been named for an emperor; its local government was structured as a polis, replete with archons, boule and demos; the city boasted a stadium and, later on, had been given an additional name, Claudiopolis, possibly following Claudius death in 54.37 The proseuche building may well have been constructed by Antipas himself or, alternatively, during the reign of Agrippa I. The latters contact with the Diaspora, and particularly with Alexandrian Jewry, is well known, especially with regard to the events surrounding the anti-Jewish outbreaks of 38 c.e. Finally, it may not be coincidental that a later midrash describes a third- or fourth-century Tiberian synagogue as a dyplastoon () the identical term used elsewhere by the Tosefta for the rst-century Alexandrian synagogue ( 83.)It is of interest that this term appears in rabbinic literature only in conjunction with these two synagogues. A second noteworthy feature of this Tiberian account is the purportedly large size of the proseuche. Josephus mentions this specically, and his statement is reinforced by the fact that when not meeting in the proseuche, Tiberians would gather in the local stadium, much as the demos of Ephesus and Antioch were wont to meet in their respective theaters.39 To date, monumental synagogue buildings are known only from the Diaspora: one (perhaps two) in Alexandria, described by Philo and the Tosefta, and another in Sardis, dating from the late third to seventh centuries c.e. and excavated in the 1960s.40
37. See, for example, Josephus, War 2, 641 ( of 600); idem, Life 69, 296 ( ); 134, 271, 278, 294; idem, War 2, 599 (); 2, 615 (); idem, Antiquities 18, 149 (). See also Avi-Yonah, Tiberias in the Roman Period, 15462; A. Kasher, Founding of Tiberias, 311; D. R. Schwartz, Agrippa I, 13740. 38. Tiberian synagogue: Midrash on Psalms 93 (p. 416). Alexandrian synagogue: T Sukkah 4, 6 (p. 273); and below, Chap. 4. 39. Tiberias: Life 91; 331. See also Hengel, Proseuche und Synagoge, 17778 (= Gutmann, Synagogue, 4748). Ephesus: Acts 19:29. Antioch: Josephus, War 7, 47. 40. See below, Chap. 4. On the Sardis synagogue, see Seager, Building History, 42535; Seager and Kraabel, Synagogue and the Jewish Community, 16890; Kraabel, Diaspora Synagogue, 48388; and below, Chap. 8.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

4. Model of the Gamla synagogue (Qatzrin Museum).

Finally, the Tiberias proseuche in this Josephan context served, inter alia, as a forum for discussing burning political issues of the day, thus indicating its pivotal communal role. In Josephus report, at least, its religious dimension was decidedly secondary, and on those occasions the Sabbath as well as fast-day rituals were apparently dwarfed by the pressures of the political agenda.

date.41 The building may have been built around the turn of the rst century c.e., although a mid-rst century b.c.e. foundation, some time between Alexander Jannaeus (10376 b.c.e.) and Herod (374 b.c.e.), has also been suggested. The Gamla building is architecturally impressive (g. 4). It is the only public building thus far excavated in that town and may well be the only one that ever existed there. Located adjacent to the eastern wall, the building runs on a northeastsouthwest axis and is 21.5 meters long and 17.5 meters wide. The hall itself measures 19.7 by 15.3 meters. In the northwestern corner of this hall is a niche that may have been used for storage. Two entrances are located to the southwest, one giving access to the northern aisle and a sec41. Gutman, Gamala, 46062; Maoz, Architecture of Gamla, 15254; and Z. Ilan, Ancient Synagogues, 7374; Syon and Yavor, GamlaOld and New, 233. The buildings dimensions dier somewhat in each of the above publications; I will follow those of Syon and Yavor.

Gamla is the earliest synagogue structure to have been discovered in Judaea to

pre-70 judaea


ond, larger one, leading directly into the main hall. Another entrance from the southeast opens onto the eastern aisle. Below these elevated aisles were four rows of benches running along all four sides of the main hall, and in front of each row of benches was a paved aisle with a row of columns. The columns surrounded an open space in the middle of the hall that was unpaved, with the exception of a row of stones running almost its entire width. A single bench ran along the eastern wall of the building, and similar ones may have run along the northern and southern walls as well. A small basin (for washing hands?) in the eastern aisle was fed by a channel that cut through the eastern wall. No inscriptions were discovered in the building, and the only depiction is a stylized palm tree carved in a stone block. A stepped cistern just west of the synagogues main entrance may have been used as a miqveh. East of the synagogues main hall are several rooms, one of which may have had some sort of opening into the main hall. This room also contains benches, which led the excavator to suggest that it may have served as a study hall. If our assumptions are correct, namely, that the synagogue at this time was rst and foremost a communal institution and that the structure at Gamla is the only public building in the town, then, indeed, it must have served as the local synagogue. The buildings internal plan is reminiscent of (although not identical to) the Hellenistic public hall, e.g., the bouleuterion or ecclesiasterion, and is similar in its overall plan to the synagogues found at other Judaean sites. To date, Gamla is the only building that can be identied with certainty as a synagogue from the pre-70 Galilee-Golan region.

The existence of a number of synagogues in Jerusalem is clearly attested in several New Testament passages. Paul, for instance, makes the following statement upon being apprehended by the Roman authorities: They did not nd me in the Temple disputing with anyone, or stirring up a crowd, neither in the synagogues nor in the city. 42 However, there is no more unequivocal testimony for the existence of synagogues in Jerusalem than Acts 6:9, which notes a series of such institutions, all associated with Diaspora Jewry.43 After describing a conict within the nascent Jerusalem church, between Greek-speaking (possibly also implying Diaspora-born) and Aramaic-speaking Jews (6:17),44 Acts goes on to note the opposition to Stephen (6:89):

42. Acts 24:12; see also ibid., 22:19 and 26:11. See Cadbury, Book of Acts, 8689. 43. See TDNT, VII, 83738. 44. Although the Greek speaks of Hebraisti, the term probably refers to a Semitic language, and in this case Aramaic is a far more likely candidate. The evidence for Aramaics widespread use in rstcentury Judaea is overwhelming; see Fitzmyer, Languages of Palestine, 50131; Gundry, Language Milieu, 4048; Rabin, Hebrew and Aramaic, 100739; Schrer, History, II, 2026; and L. Levine, Judaism and Hellenism, 8084.


historical development of the synagogue

Full of grace and power, Stephen worked great wonders and signs among the people. And some of those from the synagogue of the Freedmen [or Libertines] as it was called, and of the Cyreneans and of the Alexandrians, and of those from Cilicia and Asia came forward and disputed with Stephen.

Discussion of this passage has at times focused on the exact number of synagogues referred to, since the text itself is somewhat ambiguous. Opinions range from one or two to ve. However, given what we know of the extensive Diaspora presence in the city and given the signicant dierences between these various communities (see below, Chap. 4), the last option appears to be the most likely.45 The existence of Jews in Jerusalem from each of the above-mentioned communities nds conrmation elsewhere. That Jews of Cyrene frequented Jerusalem is attested in Mark and Luke as well as Acts.46 The presence of Alexandrian Jews in the city is documented in a variety of sources, including one rabbinic tradition which specically mentions an Alexandrian synagogue there.47 Paul himself hailed from Cilicia, and his nephew also may have lived in Jerusalem.48 Acts mentions Jews from Asia who visited the Jerusalem Temple. Funerary inscriptions take note of Jews from Cilicia and elsewhere in Asia Minor, as well as from Alexandria and Cyrene, who were buried in the city.49 The term Freedmen or Libertines in this passage has attracted much attention. Many have identied this synagogue with the Theodotos inscription (see below) on the basis of passages from Tacitus and Philo, which indicate that the Jews who had been brought to Rome in captivity were soon freed.50 Theodotos family apparently hailed
45. One synagogue: Jeremias, Jerusalem in the Time of Jesus, 6266; Bruce, Acts of the Apostles, 156; Conzelmann, Acts of the Apostles, 47; Riesner, Synagogues in Jerusalem, 2045. Two synagoguesthe Freedmen synagogue, composed of Cyrenean and Alexandrian Jews, and a synagogue of Jews from Asia Minor: Denton, Commentary on the Acts of the Apostles, I, 18788; Foakes-Jackson and Lake, Beginnings of Christianity, 1, 4, 66; TDNT, VII, 837 n. 252; Bruce, Commentary, 133 n. 24; G. A. Smith, Expositors Bible, ad. loc. Five synagogues: Schrer, History, II, 428 and n. 8; S. Safrai, Pilgrimage, 57. On the Diaspora presence in Jerusalem, see Jeremias, Jerusalem in the Time of Jesus, 5871; L. Levine, Jerusalem, s.v. Diaspora; and generally, S. Safrai, Relations, 184204. Individual Diaspora Jews were quite prominent in Temple aairs. A number of high priests families originally hailed from there (particularly Egyptsee M. Stern, Reign of Herod, 274), and a number of Diaspora Jews contributed signicant gifts to the Temple: Alexander, father of Tiberius (AlexandriaJosephus, War 5, 205); Nicanor (AlexandriaM Yoma 3, 10; T Kippurim 2, 4 [p. 231]; Horbury and Noy, Jewish Inscriptions, 24345); Helena and Monobaz (AdiabeneM Yoma 3, 10); see also T Kippurim 2, 5 [p. 231]. 46. Mark 15:21; Luke 23:26; Acts 2:10; 11:20. 47. T Megillah 2, 17 (p. 352) and parallels; see below. See, however, B Megillah 26a, which substitutes Tarsian for Alexandrian. 48. Acts 23:16. 49. Ibid., 21:27; Rahmani, Catalogue, 17. See also Kloner and Zissu, Necropolis of Jerusalem, 57. 50. Tacitus, Annals 2, 85; Philo, Embassy 155; Jeremias, Jerusalem in the Time of Jesus, 6566; Goodenough, Jewish Symbols, I, 179; TDNT, IV, 26566; Riesner, Synagogues in Jerusalem, 2046.

pre-70 judaea


from Rome, as the Roman name Vettenos seems to imply; in all probability they were descendants of Jews taken captive by Pompey in 63 b.c.e. 51 Alternatively, some scholars, following several ancient manuscripts and older commentaries of Acts, read Libyans instead of Libertines, a name that would t the North African setting of the following two geographical names on the list.52 Finally, in line with rabbinic evidence, Jeremias has suggested that the entire list in Acts reects one community and should be identied with the Synagogue of the Alexandrians (or Synagogue of the Tarsians according to certain manuscripts).53 The signicance of the Acts passage is twofold. The list of Diaspora Jewish communities in Jerusalem is impressive. In contrast to Acts 2:511, which attests to a large and multifarious gathering from every nation in the Diaspora on the festival of Shavuot,54 our passage speaks of the institutionalized presence of Diaspora Jews in the city. Who initiated the establishment of these synagogues? Were they the initiative of those who settled in Jerusalem, or were they sponsored (in whole or in part) by the various Diaspora communities both to serve their former members and to attend to the needs of those compatriots visiting the city on pilgrimage? 55 We simply have no way of knowing. Undoubtedly, the single most important piece of evidence relating to the pre-70 Judaean synagogues generally, and Jerusalem synagogues in particular, is the Theodotos inscription, found by Weill in 191314 during the City of David excavations (g. 5).56 Discovered in a cistern along with other building fragments, the stone slab bearing this inscription in all probability came from a nearby structure, traces of which were claimed
51. The Roman connection has been suggested by Clermont-Ganneau (Dcouverte, 19697) on the basis of: (1) gens Vettia or Vectia being associated with that city, and (2) a person by the name of Vettienus mentioned by Cicero. This suggestion has been accepted by some (e.g., Schwabe, Greek Inscriptions, 36364), though there have been reservations as well; see S. Safrai, Pilgrimage, 5657 and n. 147; RothGerson, Greek Inscriptions, 78. Cf. Riesner, Synagogues in Jerusalem, 198. 52. Bruce, Acts of the Apostles; idem, Commentary, 133 n. 24. 53. Jeremias, Jerusalem in the Time of Jesus, 6566. 54. On the possible origins of this list in either astrological or geographical circles, see Brinkman, Literary Background, 41827. 55. On Acts possible theological agenda in this account, see J. T. Sanders, Jews in Luke-Acts, 245. 56. FitzGerald, Theodotus Inscription, 17581; Deissmann, Light from the Ancient East, 10; as well as the thorough and comprehensive discussion by J. S. Kloppenborg Verbin, Dating Theodotos, 243 80. The inscription has usually been dated to the rst century c.e. (Schwabe, Greek Inscriptions, 362 65) and, of late, to the late rst century b.c.e. (Roth-Gerson, Greek Inscriptions, 7686). However, Kee (Transformation of the Synagogue, 124; idem, Early Christianity in the Galilee, 47; idem, Dening, 726), following Vincent (Chronique, 24777), has opted for a post-70 date, a position that has been severely criticized. See a thorough review of the evidenceand a refutation of Kees theoryin Oster, Jr., Supposed Anachronism, 178208; see also E. P. Sanders, Jewish Law, 34143; Riesner, Synagogues in Jerusalem, 179210; van der Horst, Was the Synagogue a Place of Sabbath Worship? 1843; Kloppenborg Verbin, Dating Theodotos, 24380; and L. Levine, First-Century Synagogue.


historical development of the synagogue

5. Theodotos inscription.

to have been found. The inscription, written in Greek and dating from the rst century c.e., is ten lines long and reads as follows:
Theodotos, the son of Vettenos, priest and archisynagogos, son of an archisynagogos, grandson of an archisynagogos, built the synagogue [ ] for reading the Law [i.e., the Torah] and teaching the commandments, and the guest chamber, the rooms, the water installations as an inn for those in need from foreign lands [i.e., the Diaspora], which [i.e., the synagogue] his fathers founded together with the elders and Simonides.57

As noted, the name Vettenos appears to place this inscription among Jews who came from Rome. Of singular importance in this inscription is the listing of three synagogue activities: reading the Torah, teaching the commandments, and providing rooms and water for itinerant pilgrims. Whether the hostel services were intended only for Jews from Rome in the context of a Landsmannschaft or whether they were available to others as well is unknown.58 The number of buildings referred to in this inscription is dicult to assess. Are we speaking of a single, all-inclusive structure, resembling those from later periods at Dura
57. See comments in Roth-Gerson, Greek Inscriptions, 7686; White, Social Origins, 29495. Hereditary positions such as those in the inscription are not uncommon in Jewish life in antiquity. The Hasmonean and Herodian dynasties, not to mention the high priestly oce (at least until the time of Herod), are cases in point, as is the family of Judah the Galilean, founder of the Fourth Philosophy (or Sicarii); see Urbach, Class Status and Leadership, 4345. From the rst century c.e. onward, the Hillelite (or perhaps Gamalielite) family of Patriarchs likewise constituted a hereditary oce, and this practice also seems to have been known among the sages in the talmudic era; see ibid., 6263; Alon, Jews, Judaism and the Classical World, 43657. This question has been addressed by Beer on several occasions; see his Sons of Moses, 14957; idem, Sons of Eli, 7993. An inscription from the Smyrna synagogue dating to the fourth century c.e. notes that one Irenopeus, a presbyter and pater of the community, was the son of Joub, likewise a presbyter in his time; see Goodenough, Jewish Symbols, II, 8081; Lifshitz, Donateurs et fondateurs, 2223. 58. See, for example, Schwabe, Greek Inscriptions, 363.

pre-70 judaea


and ammam Lif, for example? Or were there perhaps two separate buildings, one for worship and one for lodging? It is conceivable, though unlikely in light of the functions listed, that there were even more rooms involved (and perhaps an additional structure). The answer depends on how we understand the function of the water installations of this complex. Whom did they serve? If they were only meant to service the needs of the lodgers, then they may have been part of the hostel itself. If, perhaps, they were used for ritual and purication purposes of the general public, as those adjacent to the southern Temple entrance, then we may well be dealing with yet another structure. Also of interest in the above inscription is the nature of the institutions familial leadership, which is documented for over three generations. Those in charge were of priestly lineage, but it is very unlikely that this had anything to do with the synagogues proximity to the Temple. Rather, Vettenos was probably the head of a wealthy Roman Jewish family which had taken the initiative in building and maintaining this institution for several generations. The title of archisynagogue was clearly of importance in this particular institution and is known from other rst-century synagogues as well (see Chaps. 5 and 11). Finally, the concluding phrase of the Theodotos inscription has been taken to refer either to the building of this particular synagogue in Jerusalem several generations earlier by Theodotos ancestors, the elders, and Simonides, or to another, earlier, building of the original rst century b.c.e. congregation situated in either Jerusalem or Rome. Given the time frame involved, the Roman option is entirely conceivable. In that case, the Jerusalem synagogue referred to by the inscription would have been, in eect, the continuation of an earlier institution and was established for the purpose of serving visitors as well as the Roman Jewish community living in the city. Rabbinic sources have preserved a number of traditions of varying historical reliability with regard to the Second Temple synagogue, and specically those in Jerusalem. One of the earliest is a Toseftan tradition reported by R. Simeon b. Gamaliel of the second century concerning several disputes between the Houses of Shammai and Hillel (rst century c.e.) over synagogue practices permissible on the Sabbath: The House of Shammai says: Charity for the poor is not to be announced [or determined ]on the Sabbath in the synagogue, even [if it is a matter of collecting money in order] to arrange a marriage for orphans . . . and one does not pray for the sick on the Sabbath; and the House of Hillel permits [these things to take place]. 59 The House of Shammai seems to have made a clear and unequivocal distinction between what is proper and what is improper activity in the synagogue on the Sabbath; the Hillelites probably did not dier in principle but simply made allowance for extenuating circumstances, such as those noted.60 It seems safe to assume that the Houses of Shammai and Hillel were not conducting a
59. T Shabbat 16, 22 (p. 79); according to MS Erfurt, the tradent is R. Simeon b. Elazar, also of the second century. 60. Gilat, Development of the Shevut Prohibitions of Shabbat, 114.


historical development of the synagogue

theoretical discussion but rather were commenting on current practice in (some? all?) Judaean synagogues. Otherwise, it is dicult to understand why such an issue would arise.61 Healing the sick (and not merely reciting a prayer) on the Sabbath in some synagogues is clearly attested in New Testament pericopes (see above). Whether such a debate between these schools had any eect on the ordinary synagogue is unknown. Even within Pharisaic circles, the issue seems to have remained unresolved. Another Toseftan tradition reports that a rst-century synagogue of Alexandrian Jews located in Jerusalem was purchased by R. Elazar b. R. Zadoq for private purposes.62 This source appears in a relatively early rabbinic compilation dealing with laws and practices pertaining to the synagogue, many of which are clearly of older vintage. As already noted, the reality reected in this tradition, i.e., of Diaspora synagogues in rst-century Jerusalem, is corroborated in other sources as well. More problematic, however, are several rabbinic traditions that refer to a synagogue on the Temple Mount itself.63 The most explicit one in this regard is a reference in the Tosefta, which describes the Simat Bet Hashoeva ceremony on the holiday of Sukkot: R. Joshua b. anania said: We would not sleep all the days of Simat Bet Hashoeva. We would arise early for the morning Tamid sacrice and from there go to the synagogue and then to the additional sacrices. 64 In contrast to this tradition, which explicitly refers to a synagogue,65 the two parallel versions, in the Bavli (Sukkah 53a) and Yerushalmi (Sukkah 5, 2, 55b), do not. The former refers only to prayer, while the latter omits all reference to an interim stage between the morning and the additional sacrices. Although the Toseftan tradition is chronologically the earliest, and thus might have been assumed to preserve a more authentic account, these other conicting traditions tend to mitigateif not undermineits historical value.66
61. It is not clear whether the second issue discussed by the Houses in this source concerns a synagogue-related activity as well. 62. T Megillah 2, 17 (pp. 35253); Y Megillah 3, 1, 73d; B Megillah 26a. On the reading in the Babylonian Talmud, see above, note 47. 63. Hoenig, Supposititious Temple-Synagogue, 11531. See also Bacher, Synagogue, 620; G. F. Moore, Judaism, II, 12; E. Levy, Foundations of Prayer, 74. 64. T Sukkah 4, 5 (p. 273). On the Simat Bet Hashoeva festivities in the Second Temple, see Rubenstein, History of Sukkot, 13145. 65. Here, too, the manuscripts dier. MS London of the Tosefta reads academy (bet midrash) for synagogue, while the editio princeps and MS Erfurt add academy after synagogue; see Lieberman, TK, IV, 88889. In another Toseftan tradition (agigah 2, 9 [p. 383]) only the term academy appears in the Temple Mount ritual on Sabbaths and holidays. See parallels in T Sanhedrin 7, 1 [p. 425]; see also Y Sanhedrin 1, 4, 19c. 66. Chronological priority is, of course, no guarantee of historical accuracy. Early, even contemporary, sources can also be tendentious and incomplete. Nevertheless, being aware of what can happen to

pre-70 judaea


While the above tradition is inconclusive in and of itself regarding the presence of a synagogue on the Temple Mount, the fact remains that a synagogue is never mentioned in several important descriptions of the site. In Josephus two rather detailed descriptions of the Temple and Temple Mount,67 a synagogue is never noted. Nor does the Mishnah, with its very detailed listing of the dimensions of buildings, courtyards, and Templerelated appurtenances, ever make such a reference to a synagogue.68 Furthermore, the only time the Mishnah mentions a regular daily prayer service in the Temple, conducted by and for the administering priests, it never refers to a synagogue. Rather, these prayers appear to have been recited somewhere in the area of the Court of the Priests.69 Thus, taken together, the Mishnah and Josephus oer little basis for assuming the existence of a synagogue on the Temple Mount. Several rabbinic sources report on the number of synagogues in Jerusalem. One speaks of 480; another, probably a corruption of the rst, speaks of 460, and yet another notes 394.70 These numbersall appearing in later amoraic compilationsappear to be highly exaggerated, and in one case, clearly symbolic (viz., the number 480). Truth to tell, the unusual gure of 394 appearing in the Bavli is baing.71 What such traditions do evidence, however, is the assumption by later generations that late Second Temple Jerusalem abounded in such institutions. More than that, however, we cannot say.


Undoubtedly, the most famous synagogue remains from the Second Temple period are those from Masada (gs. 67). Although the original function of this building in Herods time is unclear, it was converted into a synagogue during the occupation of the

traditions in the process of being transmitted from generation to generation (and at times to entirely different cultural and social contexts), when judging a sources historicity, we ought to give some weight to its proximity to events or institutions. 67. War 5, 184237; see also Busink, Der Tempel von Jerusalem, 1062.; Hildesheimer, Die Beschreibung der herodianischen Tempels, 132. 68. M Middot 14. See Busink, Der Tempel von Jerusalem, 152974; Hollis, Archaeology of Herods Temple, 103231. 69. M Tamid 5, 1. 70. 480 synagogues: Y Megillah 3, 1, 73d; PRK 15, 7 (p. 257); Song of Songs Rabbah 5, 12; Lamentations Rabbah 2, 4 (p. 50b) and Proem 12 (p. 6a); Yalqut Shimoni, Isaiah, 390 (481 synagogues). 460 synagogues: Y Ketubot 13, 1, 35c. 394 synagogues: B Ketubot 105a. See Miller, On the Number of Synagogues, 5155. 71. On the tendency in rabbinic literature to exaggerate data from the Second Temple period, see T Pesaim 4, 15 (p. 166); and Lieberman, TK, IV, 568. See also B Pesaim 64b; Lamentations Rabbah 1, 49 (p. 23a); Josephus, War 6, 42327.


historical development of the synagogue

6. Masada synagogue, looking north. The room containing scroll fragments is to the left.

fortress by the Sicarii (an extreme revolutionary faction) between 66 and 74 c.e. 72 Surrounded on all four sides by benches arranged in four tiers (with a single bench adjacent to the northern room), the rectangular hall measured 15 by 12 meters and had a single door 1.35 meters wide in the middle of its southeastern wall. Unique to Masada is a small room (5.7 by 3.5 meters) protruding into the northern corner of the hall. It is in this room that fragments of scrolls from the books of Deuteronomy and Ezekiel were discovered; these contained passages undoubtedly used in synagogue liturgy at the time.73 It would appear that the main hall (compared by Yadin to an ecclesiasterion, by Avigad to a Hellenistic basilica, by Foerster to a pronaos of a temple at Dura Europos, and by Maoz to a hypothetical Alexandrian assembly house) served the revolutionaries of Masada for meeting purposes generally as well as for religious servicesi.e., it functioned as a rstcentury synagogue.74

72. Yadin, Masada, 18191; idem, Synagogue at Masada, 1923; and now Netzer, Masada, 40213. Yadin once opined that the hall also functioned as a synagogue in Herods time as well, but this suggestion has found few proponents. Foerster describes the room in this earlier stage as a vestibule and hall similar to those at Herodium, while Netzer identies it as a stable (see Yadin, Synagogue at Masada, 2021; Foerster, Synagogues at Masada and Herodium, 2429; Netzer, Masada, 41013). 73. The scrolls were found in pots, although it is not clear if they were put there for temporary storage or as a genizah; see Netzer, Masada, 410. 74. Yadin, Synagogue at Masada, 20 n. 1; Avigad, On the Form of Ancient Synagogues, 9598; Foerster, Synagogues at Masada and Herodium, 2629; Maoz, Judaean Synagogues as a Reection, 195. Thus, the positioning of the small room in the direction of Jerusalem, opposite the synagogues entrance, would have been merely coincidental, as Netzer also seems to indicate (Masada, 41012); nevertheless, this does not signify any intention of an orientation toward the city.

pre-70 judaea


7. Plan of the Masada synagogue.

The 15-by-10.5-meter hall at Herodium, which apparently functioned earlier as a triclinium in Herods fortress palace, was apparently used by the revolutionaries as a synagogue during the First Revolt against Rome.75 Four columns and several rows of benches were introduced into this rectangular hall. As at Masada, its plaster walls were simple and unornamented. Entrance was gained through three portals in the east, and just outside the hall was a stepped cistern that may have served as a miqveh, a juxtaposition similar to that at Gamla.

The situation at Qumran is indeed intriguing. The Qumran scrolls and various literary descriptions of the Essenes (and I am assuming that the Essenes and Qumran sect were closely associated, if not identical) indicate that the concept of worship was well developed by this group. On numerous occasions, community rules and practices relating to the form and content of the sects worship were prescribed, as this realm had a central role in the sects life. Proof of this can be found in the remains of over two hundred nonbiblical prayers, psalms, and hymns recovered from among the Qumran manuscripts.76
75. Foerster, Synagogues at Masada and Herodium, 24. 76. See, for example, 1QS 6:38; Josephus, War 2, 12829; Fraade, Interpretive Authority, 5658; Weinfeld, Prayer and Liturgical Practice, 16075; idem, Morning Prayers in Qumran, 48194; Chazon, Prayers from Qumran, 26584; Falk, Qumran and the Synagogue Liturgy, 40434.


historical development of the synagogue

Priests were pivotal in these worship settings, and collections of blessings and prayers seem to have been in circulation there.77 Various opinions have been put forth regarding the required number of times for daily prayer (ranging from two to six).78 Talmon, on the one hand, has argued that these liturgical modes were a continuation of biblical models rather than variations of forms that later found expression in rabbinic literature, a conclusion likewise reached by Nitzan in her study of Qumran prayer.79 Weinfeld, Schiman, and Chazon, on the other hand, have argued for a close tie between Qumran and later Jewish practices, particularly those recorded in rabbinic literature.80 Nevertheless, despite the centrality of liturgical patterns as reected in the scrolls, nothing whatsoever is said about the public reading of Scriptures. Could there have been a conscious aversion to imitating what was being done in contemporary synagogues yet another expression of the Qumran sects desire to maintain biblical precedents while rejecting models that had evolved within the Jewish community in the post-biblical era? Alternatively, did the ongoing practice of study within the Qumran community render such public readings superuous? Interestingly, it appears that another breakaway group of the Second Temple period, the Samaritans, did not include the reading of Scriptures as part of its communal ritual at this stage either. They, too, may have tried to remain within biblical parameters as much as possible or, alternatively, distance themselves from current Jewish practice.81 What was the setting for communal worship in Qumran? The Damascus Document mentions a =( house or place of prostration):
And no one entering a house of prostration shall come in a state of uncleanness requiring washing. And at the sounding of the trumpets for assembly, he shall have done it [i.e., the
77. On priests at Qumran, see, for example, 1QS 2:1920; 6:45, 8; 9:7; and comments in Cross, Ancient Library of Qumran, 128.; Gartner, Temple and the Community in Qumran, 415; Licht, Rule Scroll, 11015; Leaney, Rule of Qumran, 9195, 165, 184; Gaster, Dead Sea Scriptures, 33235; Gruenwald, From Apocalypticism, 12544; Weinfeld, Organizational Pattern, 19; Schiman, Eschatological Community, 68 71; Fraade, Interpretive Authority, 4669, esp. 5657; Newsom, He Has Established For Himself Priests, 10120. For a suggested additional dimension of Qumrans priestly orientation focusing on basic ideological issues, see D. R. Schwartz, Law and Truth, 22940. On the blessings and prayers found at Qumran, see Nitzan, Qumran Prayer and Religious Poetry, passim; Baillet, Qumran Grotte 4, 7386; Schiman, Dead Sea Scrolls and the Early History of Jewish Liturgy, 3542; Goshen-Gottstein, Psalms Scroll (11QPs 2), 2233. 78. Schiman, Dead Sea Scrolls and the Early History of Jewish Liturgy, 3940; Talmon, World of Qumran, 215. 4Q503 contains liturgical blessings for each day of the month, twice daily, in the evening and morning; see Chazon, Qedushah Liturgy, 1014. 79. Talmon, World of Qumran, 1152. 80. Weinfeld, Prayer and Liturgical Practice, 24158 and bibliography cited there; Schiman, Dead Sea Scrolls and the Early History of Jewish Liturgy, 3345; Chazon, Prayers from Qumran, 26584; eadem, Qedushah Liturgy, 717; eadem, When Did They Pray? 4251. 81. Weinfeld, Prayer and Liturgical Practice, 24142.

pre-70 judaea


washing] before or he shall do it later, but they (the impure) shall not interrupt the whole service; for it is a holy house [28.]

The term house of prostration, reminiscent of the Diaspora proseuche, may also hark back to biblical precedents, where such a term, together with the designation , refers to worship in general and possibly even to prayer.83 The nature and structure of this worship setting is never made explicit, but it undoubtedly consisted of prayers, blessings, and hymns described in other Qumran writings. It is such a setting that Josephus may have had in mind when he described the main Essene worship service (and communal meal) near midday.84 Where at Qumran would such a place have existed? The most reasonable guess is room 77 (according to de Vauxs numbering), the largest at the site (g. 8). This room may have functioned not only as a dining area but also as a place of worship. A platform at the western end of the room probably had some signicance.85 If indeed public worship at Qumran included hymns, prayers, and blessings, as well as a sacred communal meal, then paradoxically the synagogue at Qumranfor all its uniqueness in setting, function, and liturgyconforms in certain ways with the ordinary Judaean synagogue. The liturgical dimension was neither set apart nor assigned a special place; rather, it found expression in the main assembly hall of the building, as did other communal activities. Let us turn now to the literary evidence regarding the Essenes. Philo oers us an interesting description of their Sabbath worship in contrast to the description in the Qumran scrolls. Essene worship, according to Philo, consisted of the reading and expounding of
82. CD 11, 2112, 1; interpretation follows Steudel, Houses of Prostration, 4968, and especially 5052; but see also Vermes, Dead Sea Scrolls, 110, and Schiman, Reclaiming, 291, for somewhat dierent renditions of this passage. Talmon (World of Qumran, 24142) and Nitzan (Qumran Prayer and Religious Poetry, 6263) concur that the reference is to a place of worship in Qumran; Falk, however, demurs, claiming that the reference is to the Jerusalem Temple (Daily, Sabbath, and Festival Prayers, 24345). In general, he argues for close liturgical ties between Qumran and the Temple (ibid., 25355). 83. II Sam. 12:20; Jer. 26:2. 84. War 2, 12931. 85. Talmon, World of Qumran, 62. The fact that the pantry with a full array of dishes and bowls was located adjacent to room 77 supports this identication. The only other option for locating this gathering at Qumran in room 4, which was lined with benches, or room 30 next to it. However, room 4 is very small, thus precluding its use by the entire sect on a daily basis. Room 30 is immediately under the scriptorium and was probably used in connection with the preparation of scrolls. Alternatively, the author of the Damascus Document may not have had Qumran in mind, but rather the various sectarian communities scattered throughout the country that may have set aside a special expressly for liturgical purposes; see Schiman, Dead Sea Scrolls and the Early History of Jewish Liturgy, 35. For a recent attempt to view the Qumran texts as reecting the practices of a synagogue community closely related to Hellenistic associations, see Klinghardt, Manual of Discipline, 25167.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

8. Room 77 at Qumran, looking west, possibly used for meals and liturgical gatherings.

Scriptures in synagogues. Regarding their Sabbath worship, he notes: For that day has been set apart to be kept holy and on it they abstain from all other work and proceed to sacred spots which they call synagogues [ ]. There, arranged in rows according to their ages, the younger below the elder, they sit decorously as bets the occasion with attentive ears. Then one takes the books and reads aloud and another of especial prociency comes forward and expounds what is not understood. 86 Whether this description is reective of Essene settings generally, perhaps even in Qumran (although inexplicably ignored in the scrolls) or, alternatively, everywhere in Judaea except Qumran, is impossible to determine. It is quite possible that Essene groups throughout Judaea adopted certain practices regnant in the general community, and it is these that Philo highlighted.


The earliest mention of a Judaean synagogue in Josephus narrative is with reference to Dor, in the aftermath of Emperor Caligulas attempt to erect his statue in the Jerusalem Temple (4041 c.e.). Following this traumatic episode, and the subsequent ascent of Claudius as emperor, policy regarding the Jews reverted to its earlier, more tolerant, posture. Claudius promptly issued a series of decrees protecting Jewish rights and privileges, including their right to worship in accordance with their ancestral custom.87

86. Philo, Every Good Man Is Free 81. 87. Antiquities 19, 27991. On Claudius edict and other decrees, see Tcherikover, Hellenistic Civilization, 40915. See also D. R. Schwartz, Agrippa I, 3031. Imperial protection extended to holy objects as well. See Josephus, War 2, 231, and idem, Antiquities 20, 11517, where we are told that a Roman soldier was executed by the procurator Cumanus for tearing a Torah scroll.

pre-70 judaea


One very concrete expression of this Imperial support was awarding Agrippa I rule over Judaea and Samaria, in addition to the regions already under his control.88 It is in reaction to these developments that some young men in Dor desecrated the local synagogue by erecting a statue of the emperor there. Agrippa appealed to Petronius, governor of Syria, who quickly intervened by dispatching a harsh letter to the leaders of the city ( ) in which he demanded the apprehension of the perpetrators of this deed. In the course of this letter quoted by Josephus, Petronius makes the following points regarding the synagogue:89 ( ) the inviolability of the synagogue is considered a time-honored Jewish privilege, i.e., laws of the fathers ( . . . ); ( ) setting up a statue in a synagogue is prohibited (lit., is a sacrilege); ( ) the synagogue is the realm of the God of Israel; no other deity or statue can be put there without destroying the institution (each [god] must be lord over his own place, in accordance with Caesars decree).90 By responding quickly and rmly, Petronius hoped to squelch any overt hostilities between local Jewish youth and their pagan counterparts.91 We never learn what transpired thereafter, since Josephus narrative then moves on to other events. There can be little doubt, however, given the stability of Imperial rule and Agrippas rm control over his kingdom, that order was quickly restored and the statue removed.92 It is not dicult to imagine that a synagogue should be the target of attack by resentful pagan youth in the wake of Caligulas abortive attempt to desecrate the Jerusalem Temple, although it may be entirely fortuitous that such an incident took place in Dor and not elsewhere. In contrast to the mobs in Alexandria a few years earlier (38 c.e.), the pagan youth made no attempt to destroy the synagogue building at Dor, much less Jewish homes and shops.93 The Dor incident appears to have been the work of a small group of young men () who may have been resentful of Jewish particularism and its recognized status, frustrated by the failure of Caligulas plan, and perhaps apprehensive about the restoration of Jewish political sovereignty in the guise of Agrippa.94
88. Antiquities 19, 29296. See Schrer, History, I, 44254. 89. Antiquities 19, 300311. 90. Ibid., 305. 91. Ibid., 30911. On the history of such tensions between Jews and pagans during the Hellenistic and early Roman periods in Palestine, see J. H. Levy, Studies in Jewish Hellenism, 6078. 92. D. R. Schwartz, Agrippa I, 135. 93. Philo, Flaccus 4496; idem, Embassy 13237; Smallwood, Jews under Roman Rule, 23542. Likewise, extreme measures, such as the scourging, torturing, and hanging evidenced in Alexandria, were not invoked in Dor. 94. Josephus himself refers to them as young men of Dor, who set a higher value on audacity than


historical development of the synagogue

Some twenty-ve years later, ca. 6566, at the height of a conict over the status of the Jews in Caesarea, Josephus takes note of a synagogue in that city.95 It apparently adjoined land owned by a gentile; either the synagogue was located in a predominantly non-Jewish neighborhood, a non-Jew happened to own land in an area where many Jews lived, or there were no distinct neighborhood boundaries between Jews and non-Jews in rst-century Caesarea. Unfortunately, there is no way of determining which of the above options is correct, although the implications of each are intriguing. When the Jewish community attempted to buy the plot, the proprietor not only refused to sell but proceeded to build workshops on the site, leaving the Jews only a narrow, dicult, and perhaps noisy passageway to the building. Some Jewish youthreferred to by Josephus as hotheads ()decided to take matters into their own hands and tried to halt construction, but were prevented from doing so through the intervention of the procurator Florus.96 According to Josephus, when the Jews had congregated inside the synagogue on the following Sabbath, someone sacriced a bird on an overturned pot at the buildings entrance. The Jews were outraged; the mock sacrice was clearly to be perceived as a desecration of their building and may also have constituted an insidious reference to a wellknown pagan accusation that the Jews were expelled from Egypt because they were lepers.97 Although the Romans intervened and removed the oensive object, matters by then had gotten out of hand. Violence broke out and the Jews ed to nearby Narbatta.98

on holiness and were by nature recklessly bold (Antiquities 19, 300). By way of contrast, see the recently published honoric dedication from second-century Dor to a Roman governor, presumably of Syria, in Gera and Cotton, Dedication from Dor, 25866. 95. War 2, 26670, 28492; Antiquities 20, 17378, 18284. See also L. Levine, Jewish-Greek Conict, 38197. Opinions have varied regarding the nature of this conict and the Jews demands; see ibid.; A. Kasher, Isopoliteia Question in Caesarea, 1627; D. R. Schwartz, Felix and Isopoliteia, 26586. 96. War 2, 28588. To achieve the same end, Jewish leaders bribed the Roman ocial, whereupon Florus promptly ed the city, leaving the Jews to pursue their ends unobstructed. 97. Against Apion 1, 279 (in refutation of Manethos claim; see M. Stern, GLAJJ, I, 81). According to the Torah (Lev. 14:17), the purication rites for a leper included the sacrice of a bird on an earthen vessel. A similar calumny was also asserted by Pompeius Trogus and Lysimmachus; see M. Stern, GLAJJ, I, 33537, 38385, 533, and his comments on pp. 8485. On the accusations of leprosy made by Philo of Byblos, Ptolemy Chennus, and Helladius regarding Moses in particular, see ibid., I, 533; II, 14445, 149, 491. 98. According to the sixth-century historian Malalas, Vespasian destroyed a synagogue in Caesarea and built an odeum in its place (Chronicle 10, 261 [p. 138]), as he did in Antioch, replacing the synagogue there with a theater (see below, Chap. 4, note 232, as well as L. Levine, Roman Caesarea, 2526).

pre-70 judaea


9. Isometric reconstruction of the Qiryat Sefer synagogue. The entrance leads into a hall containing columns and an aisle. The structure is separated from the surrounding buildings.

Qiryat Sefer
In the course of building the modern town of Qiryat Sefer (near Modiin), an ancient village was discovered in the summer of 1995.99 Two clusters of buildings were found containing residential quarters, storerooms, and an olive press. Between these clusters and separated by an open space was a small square building measuring 9.6 meters on each side (g. 9). Unique to the site in terms of its size, location, plan, and the type of stone used, the building was oriented on a northwest-southeast axis, with its entrance facing northwest. The facade of the building was built of hewn stones with margins typical of the Herodian building style of the late Second Temple period. The oor of the hall was paved with large, well-tted agstones, and along three sides was a similarly paved elevation that presumably served as benches, with room for an aisle behind them, as at Gamla. Four columns with Doric capitals stood in the hall, and four pilasters against the northwestern and southeastern walls of the building continued the line of the columns. Potsherds and coins indicate that the site rst developed in the Hellenistic period, grew considerably in the rst century c.e., and continued into the second; it was abandoned some time during the rst half of the second century, quite likely as a result of the Bar-Kokhba revolt. Although a nal report has not yet been published, it seems quite certain that this building was a small synagogue that served the village and its immediate area. Given its central location in the settlement, its careful construction, and public character, as well as its overall similarity to the Gamla structure (albeit of considerably smaller proportions), this building seems to be a credible candidate for the rst village synagogue ever found in Judaea.

99. Magen et al., Kiryat Sefer, 2532.


historical development of the synagogue

10. Plan of Modiin synagogue.

Modiin (Khirbet Umm el-Umdan)

Discovered in excavations undertaken in 20002001 by A. Onn and others, this village existed from the Persian period onwards, with the main stratum dating from the Roman era.100 Residential homes, alleys, a miqveh, and remnants of a bath were found, in addition to a public building identied as a synagogue. The earlier phase of this building, a well-built rectangular structure (6.5 by 11 meters), with a agstone oor, benches, and fresco remains, dates from the Hasmonean era and may have served as a synagogue as well. The buildings second and main phase dates from the Herodian period and continued to the Bar-Kokhba revolt (132135 c.e.). The synagogue hall measured 10 by 12 meters (g. 10) and its main entrance was from the east; just outside the building was a courtyard surrounded by a stone bench. Another entrance was located on the north side of the building. A oorat rst of white chalk and later of thick gray plasterwas surrounded by stepped benches, and eight columns were arranged in two rows of four columns each. Fragments of painted plaster (red, white, and yellow) were found in the ll of the buildings next phase that lay on the stone pavement. A small square room (2.0 by 2.5 meters) was built just northeast of this hall with a connecting passageway. There is little doubt that this building served as a village synagogue and would be the earliest attested for the southern region of Judaea. It would be more or less contemporaneous with that of Gamla and the earlier structure noted in the Theodotos inscription, if indeed it was located in Jerusalem (see above).

100. Onn et al., Khirbet Umm el-Umdan, 64*68*, now superseded by Weksler-Bdolach et al., Identifying the Hasmonean Village of Modiin, 7276.

pre-70 judaea



Buildings at a number of sites have been identied as synagogues, but these identications are either tentative or have met with serious reservations. Capernaum. In the course of excavating the fourthfth-century c.e. synagogue, Corbo and Loreda noted black basalt walls under the buildings white limestone ones, although with a slight discrepancy in direction along the western side. In 1981, following the excavation of trenches to clarify the dating and function of these basalt walls and others under the later buildings stylobates, Corbo claimed to have discovered a rstcentury synagogue.101 Measurements of the earlier building approximate those of the later one, ca. 24.50 by 18.70 meters, and its walls are 1.201.30 meters thick and thus seem most likely to have been part of a public, rather than private, building. The dating of the earlier building is based on pottery found in and under a cobbled pavement of basalt, some 1.3 meters below the level of the later synagogue oor. It is thus possible that a rst-century building stood on this site. This assumes that the basalt walls were built not merely to support the later building, but belonged to an earlier structure later utilized to provide foundation courses for the limestone synagogue building. The excavators suggest that a layer of basalt stones constituted the oor of an earlier building, although no connection to the walls has been attested. No doors or benches have been identied. If, nevertheless, one assumes the existence of an earlier building, then its identication as a synagogue becomes compelling. A large building located in the center of town ts the bill for a rst-century synagogue, although it would have been dierent in its location and size from the structure at Gamla. According to the excavators, for example, this building would have been about 50% larger than the one at Gamla and more than twice the size of the Masada and Herodium buildings. However, given the meager remains, there is little to be learned about the overall plan of this building and its identication as a synagogue is at present rather tenuous. Migdal (Magdala). In the course of excavating Migdal or Magdala, north of Tiberias, in 197174, Corbo and Loreda discovered a building from the rst centuries c.e. which they identied as a synagogue in its rst stage.102 The building has an unusual shape; it is a small structure (8.2 by 7.2 meters, the inner hall measuring 6.6 by 5.5 meters) with ve rows of benches on the north and seven columns along the other three walls. Five columns were round and two were heart-shaped corner columns. The original oor was made of basalt slabs, but most of these were removed in the buildings second stage, when the oor was raised. Corbo identied this building as a rst-century synagogue that was converted into a water installation in the second century. Netzer, however, on the basis
101. Corbo, Resti dello sinagoga, 31357. For an English summary of these excavations, see Strange and Shanks, Synagogue Where Jesus Preached, 2531. Cf. also the reservations of Tsafrir, Synagogues of Capernaum and Meroth, 15557; Magness, Question of the Synagogue, 1920. 102. Corbo, La citt romana di Magdala, 36472; Groh, Stratigraphic Chronology, 5859.


historical development of the synagogue

of the published ndings, has concluded that the water channels on three sides of the hall were built at the same time as the earlier oor, i.e., that the original rst-century building was intended as a water facility that functioned as a nymphaeum in both stages. The oor of the second stage was raised to correct a problem of ooding.103 Chorazim. In 1926, Ory reported the discovery of a synagogue at Chorazim, some 200 meters west of the later building visible today at the site. Orys unpublished report states the following: A square colonnaded building of small dimensions, of a disposition similar to the interior arrangement of the synagogue, 7 columns, 3 on each side (the entrance was aorded through the east wall), were supporting the roof, and the whole space between the colonnade and walls on three sides was occupied with sitting benches in 5 courses. Unfortunately, these remains have never been veried. Visits to the site have yielded no clues as to the whereabouts of this structure. Orys report has been included on occasion in surveys of rst-century synagogues, but the use of such data seems unwarranted for the present.104 Northern Jerusalem. In 1991, while excavating an agricultural settlement in the northern part of the city, archaeologists found a large agricultural building complex. First reports spoke of a niche in the southern wall, benches along the walls, an adjacent courtyard also lined with benches, and nearby miqvaot. They suggested that this complex was, in fact, a synagogue or, at the very least, a prayer room. The complex was supposedly built in the rst century b.c.e. and abandoned following the earthquake of 31 b.c.e. 105 However, this identication has been greeted with general skepticism. The niche and benches were rather crudely made and did not seem to indicate any type of public building. Moreover, the claims made in a very brief report published several years later were severely reduced, and some details were enigmatically altered or eliminated.106 The case for a synagogue or prayer hall at this site appears to have evaporated. Jericho. In the winter of 199899, Netzer announced the discovery of a synagogue in Jericho dating from the rst half of the rst century b.c.e. (ca. 7570 b.c.e.).107 The build103. Netzer, Did the Water Installation in Magdala Serve as a Synagogue? 16572. Cf. also Maoz, Synagogue of Gamla and Typology of Second-Temple Synagogues, 39. 104. Foerster, Synagogues at Masada and Herodium, 26. 105. Josephus, Antiquities 15, 12122. 106. Onn and Rafyunu, JerusalemKhirbet a-Ras, 61: At the next stage [i.e., the rst century b.c.e.L.L.] the complex was renovated; along the southern wall of the northern tower, a new wing was built which spread over a large part of the courtyard. In the middle of this new wing, a miqve together with a courtyard in the shape of [the Hebrew letter] resh were discovered, and to the south a rectangular room (4m x 5m) divided by a low wall of hewn stones. In the eastern [sicL.L.] wall of this room was a square niche in front of which was a stone-slabbed oor. It is quite possible that at this stage, which ended with the destruction caused by the earthquake of 31 b.c.e., this complex served as an assembly place for worship purposes. See also the note on this excavation in Riesner, Synagogues in Jerusalem, 192. 107. Netzer, Synagogue from the Hasmonean Period, 20321; idem, Hasmonean and Herodian Palaces at Jericho, II, 15992.

pre-70 judaea


ing (28 by 20 meters in its nal stage) is located just east of the Hasmonean palace. The eastern part of the complex consisted of seven rooms and a small courtyard; to its west was a large hall (16.2 by 11.1 meters) lying on an eastwest axis and containing twelve pillars with benches between them, aisles along each side of the room between the pillars and walls, and a channel that conveyed water from an aqueduct in the north to a basin in the room, and then to a miqveh in the south. In the last stage of the building, a triclinium was added on the western side of the hall, and a niche in its northeastern corner was identied as a storage place for Torah scrolls and other books. The building, which has been compared with that of Gamla (thus establishing its identication as a synagogue), is said to have served the palace sta living in the buildings to the east. If authentic, this site would constitute the earliest synagogue ever found in Judaea. Unfortunately, however, this identication is not without its problems. When all is said and done, there is very little hard evidence on which to base such a conclusion. The addition of a triclinium in the last stage cannot be considered a decisive factor for such an identication, even though we know that some ancient synagogues featured triclinia. Roman villas often had such rooms, very similar, in fact, to the Jericho building, but there is no known synagogue parallel to the plan of the Jericho building. Neither is there any basis (or parallel) for the suggestion that the niche, tucked away in the northeastern corner of the hall, served as a storage area for Torah scrolls and perhaps other objects. The presence of a miqveh is interesting, but, again, of limited value as a basis for identication. As the nearby palace complexand the city of Jericho in generalserved as an important center for priests who were obliged to use such ritual baths regularly, the discovery of a miqveh is not uncommon and thus cannot be used as decisive evidence for the existence of a synagogue. The assertion that this building resembled the Gamla synagogue, a point that the excavator considers to be of prime importance, is likewise unconvincing; the few similarities (i.e., four rows of columns, benches, and a water channel) are far outweighed by the many signicant dierences, such as the overall plan of each, the number and nature of the adjacent rooms, the location and arrangement of the benches, and the absence of a triclinium in Gamla. Moreover, the location of this proposed synagogue is curious. Whom exactly did it serve? There was no community in the immediate area of the palace complex other than several possible villasnot yet fully excavatedlying further east, along the same ridge. The claim that such a synagogue might have served the palace sta is not particularly persuasive. We have noted that the plan of this building, including the triclinium, is similar to many Hellenistic-Roman villas;108 it is quite possible that this hall may have been part
108. See, for example, de Franciscis, Pompei, nos. 49, 73; Wallace-Hadrill, Houses and Society in Pompeii and Herculaneum, 4043, 81, 107; Zanker, Pompeii, 36 (House of the Faun), 170 (House of the Golden Cupids).


historical development of the synagogue

of such a villa, serving as a garden or atrium surrounded by a portico with a triclinium at one end.109 Thus, the claim that this rst-century b.c.e. Jericho building is the earliest synagogue ever found in Judaea should, for the moment at least, be held in abeyance. Perhaps further investigation and excavations will furnish more conclusive evidence. orvat Etri. In an excavation carried out by Zissu and Ganor in 19992000, a village dating from the Persian to Late Roman periods was discovered south of Bet Shemesh and northeast of Bet Guvrin.110 The village ourished in the rst centuries c.e., although it suered serious destruction in each of the revolts (6674 and 132135). The most prominent structure was a public building located on the northeastern outskirts of the village. Measuring 13 by 7 meters, the building was entered from one of its long walls (and thus it was termed a broadhouse-type building by its excavators); inside were remains of three columns across the width of the hall that were meant to support the roof. In front of the entrance was a courtyard with a stepped pool o to one side, identied as a miqveh. The building itself is dated to the period between the two revolts and is included here since it appears to be similar to the nearby Judaean village synagogues discussed above. At rst glance, it would seem that this public building is as good a candidate as any to be considered a rst-century synagogue. The building is clearly of a dierent order than those surrounding it, and the entranceway is a further indication of its public nature. Our main reservations involve the meager nds within the building itself. In contrast to the two other Judaean village synagogues from this time, the building has no benches or stone-paved oor, nor is the deployment of its columns (three in a single row in the middle of the hall) evidenced elsewhere, where two to four columns surround an open space in the middle. In sum, this building may well be a village synagogue but, admittedly, there remains an element of doubt.


Despite the scattered and fragmentary nature of the data, some interesting information may be gleaned regarding the rst-century Judaean synagogue. Certain features recur in many instances, others can be discerned in some, while some intriguing distinctions and dierences also appear, especially from region to region. The communal dimension is a prominent feature of the rst-century Judaean synagogue. Synagogue buildings excavated to date reect, each in its own way, an architectural

109. See Maoz, Synagogue that Never Existed, 12021; and the response of Netzer, Synagogue in Jericho, 6970. For other reservations as regards identifying this building as a synagogue, see Shanks, Is It or Isnt ItA Synagogue? 5355; Claussen, Versammlung, 18586; Schwarzer and Japp, Synagoge, Banketthaus oder Wohngebude? 27788. 110. Zissu and Ganor, orvat Etri, 1827. In support of this identication, see Maoz, Notes, 55.

pre-70 judaea


style betting a community-oriented framework. Gamla, Masada, Herodium, Qiryat Sefer, and Modiin each have a square or rectangular area with columns and benches, an arrangement facilitating communal participation for political, religious, or social purposes. The model chosen consciously or subconsciously for these settings approximated Hellenistic bouleuteria or ecclesiasteria, which likewise catered to an assembly of people empowered to make decisions. This communal dimension is likewise in evidence at Dor and Caesarea, where the synagogue became a target of anti-Jewish activity. As per one early rabbinic tradition, it was also a place where charity monies were collected. Furthermore, the events in Tiberias emphasize unequivocally the pivotal role that the local proseuche played as a setting for critical communal deliberations at the outbreak of the revolt. On this occasion, the synagogue lled several roles simultaneouslyas both a place of worship and a setting for political debates. The religious dimension was an important component on the synagogues agenda as well. However, it must be placed in perspective given the nature of our literary sources, which have a clear propensity to emphasize this aspect of the synagogues activities, often to the exclusion of others. The gospel accounts focus on Jesus preaching and teaching, and Lukes description of his Sabbath-morning appearance in a Nazareth synagogue is extremely valuable; the Sabbath assembly of Caesarean Jews in their synagogue provides the setting for a demonstrative anti-Jewish act; the Tiberian proseuche was the scene of Sabbath and fast-day worship within a period of three days; and even the Theodotos inscription mentions the religious-educational aspects of the synagogue before its socialcommunal ones. Given this proclivity on the part of the literary sources to highlight the religious aspects of the institution, archaeological remains from this period prove to be an important corrective. The buildings themselves are neutral communal structures with no notable religious componentsinscriptions, artistic representations, or a Torah shrine. Torah scrolls or the ark were introduced into the community hall only for the Torah-reading ceremony and were removed thereafter. The Tosefta, in one its many early sources, seems to reect this situation well.111 The rst-century synagogue did not have the decidedly religious prole that the institution was to acquire by Late Antiquity. However, one component of a religious nature, the presence of a nearby miqveh, seems to have been fairly common among Judaean synagogues. Such was the case at Gamla, Masada, Herodium, Qumran, Modiin, and Jerusalem (assuming that some of the water installations noted in the Theodotos inscription refer to this usage as well).112 While there
111. T Megillah 3, 21 (p. 360). 112. It should be pointed out that the closest miqveh to the Masada synagogue was some distance away, none has been found near the Qiryat Sefer building, and none is mentioned in conjunction with a rstcentury synagogue.


historical development of the synagogue

is nothing in Jewish law (then or now) that would require the use of a miqveh for synagogue worship (and, in fact, it becomes much less common in or near synagogues of Late Antiquity), its proximity does reect the communal, inclusive, nature of the building. Purity concerns were of enormous importance in Judaean society of the late Second Temple period, commencing in the mid-second century b.c.e. with the rise of the Hasmoneans. Early rabbinic legal traditions, the Qumran scrolls, and the discovery of some ve hundred miqvaot in and around Jerusalem and throughout much of Judaea all attest to this emphasis on purity, which was a prerequisite for entering the Temple precincts or selling agricultural products, wine, and oil to the Temple authorities.113 It is no coincidence, then, that the Mishnah mentions only two instances of the red heifer sacrice (for purifying those with corpse impurity) over a thousand-year period before the second century b.c.e., but seven instances (according to another tradition: ve) during the last 250 years of the Second Temple period (M Parah 3, 5). Despite the relative paucity of material, enough has survived to enable us to appreciate the diversity that characterized the pre-70 Judaean synagogue. Many synagogues in Jerusalem were linked to Jews from various Diaspora communities, which undoubtedly maintained diverse customs and practices. The proseuche of Tiberias, physically and functionally, stands in striking contrast to what we know about other Judaean synagogues. The Caesarea synagogue, located in the midst of other buildings (even those belonging to non-Jews), is clearly quite dierent from that of Gamla or Qiryat Sefer, not to speak of those at Masada and Herodium. And, of course, the Qumran worship setting was sui generis with regard to its name (house of prostration) and location, as indeed were most other aspects of this community. With regard to Jerusalem, given the presence of the Temple, there can be little doubt that what were considered usual synagogue activities elsewhere often found expression within the precincts of the Temple Mount. Moreover, synagogues in the city might have assumed additional responsibilities absent elsewhere, as, for example, in cultivating ties and serving the needs of pilgrims. This diversity holds true architecturally as well. Of the three best-known buildings from pre-70 Judaea commonly identied as synagogues (Gamla, Masada, and Herodium), the dierences are no less salient than the similarities. Nevertheless, several attempts have been made to delineate a typology for the Second Temple synagogue on the basis of these buildings: columns in the center, benches on all four sides with a focus in the center of the room, and the proximity of ritual baths.114 Such attempts at dening a typology, however, are unconvincing. Whatever elements

113. E. P. Sanders, Judaism, 21430; L. Levine, Jerusalem, s.v. ritual baths. 114. Foerster, Synagogues at Masada and Herodium, 2629; Maoz, Synagogue of Gamla and Typology of Second-Temple Synagogues, 3541; idem, Judaean Synagogues as a Reection, 1012; Netzer, Synagogues from the Second Temple Period, 27785; Strange, Archaeology and Ancient Synagogues, 3746.

pre-70 judaea


were held in common can be attributed to Greco-Roman architectural traditions and were adopted by synagogue builders because of their utility. Moreover, the dierences among these buildings are quite signicant, e.g., the location of entrances, internal plans, positioning and number of benches, shape of the hall, and overall setting vis--vis surrounding structures.115 Even the synagogues created by the revolutionaries at Masada and Herodium following the outbreak of hostilities in 66 dier from one another, although this is primarily due to the plans and functions of the earlier buildings that they replaced. Thus, it is unwarranted to speak of a single model of the Second Temple Judaean synagogue. Essentially, what we have is a variety of communal buildings that shared certain basic characteristics and served the myriad purposes of the rst-century institution. The diversity among synagogue buildings of Late Antiquity had a precedent in the rst century. A striking dierence between the status and role of Judaean synagogues may have been dueat least in partto regional diversity, and this is particularly evident when comparing the coastal area with the rest of the country. The synagogues at both Dor and Caesarea had extremely high proles: they were central institutions and were recognized as such by Jews and non-Jews alike. Such visibility had its disadvantages, for it made these buildings vulnerable to attack by hostile mobs. The events at Dor and Caesarea, as indeed at Alexandria at roughly the same time, point to a phenomenon that was to recur in times of political stress or in charged religious circumstances. Synagogues were desecrated and destroyed, and in Late Antiquity they could be outlawed and at times even converted into churches by local rebrands and agitated mobs.116 Thus, the coastal synagogues of Dor and Caesarea, located in pagan urban centers, had to deal with issues familiar to Diaspora Jewry but relatively unknown in areas where Jews formed the majority. In contradistinction to the Jews living in the interior of Judaea, those living in the coastal area as well as in the Diaspora ascribed to their synagogues a degree of sanctity, perhaps inuenced by the ubiquitous pagan temple. Petronius letter regarding the Dor synagogue certainly takes note of this characteristic, and apparently even those who perpetrated the desecration by placing the emperors statue there likewise assumed that there was some sort of sanctity inherent in this Jewish building. While the religious dimension of Judaean synagogues was not the decisive factor in dening the institution at the time, this may not have been as true of the hellenized, largely nonJewish coastal region. Living as they did among a dominantly pagan population, the Jews in these areas perhaps sought to enhance the status of their communal institution with a religious dimension, much as was being done by their Diaspora coreligionists (see below,
115. L. Levine, Second Temple Synagogue, 1013; Chiat, First-Century Synagogue Architecture, 4960; idem, Synagogue and Church Architecture, 4956. 116. Parkes, Conict of the Church and the Synagogue, s.v. synagogues; Simon, Verus Israel, 22433. See also below, Chap. 7.


historical development of the synagogue

Chap. 4). In the political realm, the Jews of Caesarea at least seem to have faced issues typical of a rst-century Diaspora setting.117 One rather well-documented activity of the synagogue was its setting for healing and miracles that purportedly occurred in rst-century Galilean synagogues. Other sources ignore this aspect of synagogue life, possibly because it was too common a phenomenon to require comment, too embarrassing, or simply one of the many items not addressed in these sources. Whatever the case, and despite its unusualness, there is no reason to question the reports validity. In fact, in diatribes delivered some three centuries later, the Antiochan John Chrysostom made the same points while attempting to dissuade Christians from attending synagogue services and following other Jewish practices (see below, Chap. 8). Nevertheless, it is dicult to assess the extent of such healings and miracle workings from the rst century. It seems doubtful that this activity was conned to the Galilee, to rural synagogues, or to the lower strata of society, but owing to the limited evidence at our disposal not much more can be said in this regard. A topic that has received a good deal of attention in recent publications is the relationship between the Jerusalem Temple and the rst-century synagogue. Some strikingly contrasting positions have been taken in this regard.118 On the one hand, a number of scholars have posited that the Temple inuenced the synagogue. Kasher has made this claim with respect to the Egyptian proseuche, and Strange regarding the Galilean synagogues.119 The most comprehensive presentation of this line of argument has been oered by Binder, who claims that the synagogue was, in fact, an extension of the Temple, asserting that the latters functions, ocials, liturgy, architecture, sanctity, and art shaped those of the former.120 Binder thus attempts to show that (1) synagogues of the rst century were considered sacred institutions; and (2) these synagogues were indeed patterned after the Temple. However, the evidence in this regard is, at best, partial. Synagogue sanctity seems to be indicated for only a small number of Diaspora sites and is never clearly attested for any Judaean ones, with the exceptions, perhaps, of Dor and the Tiberian proseuche. Moreover, Binders eorts to interpret a number of passages from Josephus that mention hiera (e.g., War 4, 4069) as referring to Judaean synagogues, and therefore as evidence for the existence of synagogue sanctity, are unconvincing. It is Binders second claim, namely, that synagogues everywhere were patterned after the Temple, that is far more revolutionary.121 Regrettably, he nds it dicult to substanti117. L. Levine, Jewish-Greek Conict, 38197. 118. For a fuller discussion of this issue and the various approaches, see my First-Century Synagogue. 119. A. Kasher, Synagogues as Houses of Prayer and Holy Places, 20520; Strange, Art and Archaeology, 7576; idem, Ancient Texts, Archaeology as Text, 2745. 120. Binder, Into the Temple Courts. 121. Ibid., 12326, 481.

pre-70 judaea


ate this claim, as solid corroborative evidence is almost nonexistent. When all is said and done, the Temple and the synagogue were very dierent in essence, function, and organization: One embodied the quintessence of holiness in Judaism, the other was called several centuries later only a diminished sanctuary (see above, Chap. 2); one focused on sacrice, the other on Torah reading and prayer; one demanded silence in the cultic ritual, the other public recitations; one required a priestly leadership, the other did not (in other words, a sacral hierarchical framework par excellence in contradistinction to a communal one). In light of the paucity of material, it is dicult to marshal enough evidence to substantiate the theory that synagogues were extensions of the Temple. Binders approach is countered by Flesher, who suggests that the synagogue and Temple were diametrically opposed religious institutions.122 Having accepted the theory of an Egyptian origin for the synagogue, and that the synagogue was later imported into Judaea, Flesher notes that synagogues in Judaea are attested only in literary and archaeological sources for the Galilee, but not for Jerusalem or Judaea. The reason for this, he claims, is that the synagogue was not able to strike roots in the latter because of the Temples overwhelming presence and prominence in this region, as well as the fact that the Judaisms of these religious institutions were strikingly dierent. Indeed, the only Jerusalem synagogues that we know of by name were those founded by, or catering to, Diaspora communities that brought these institutions with them from abroad.123 The phenomenon of only Diaspora-related synagogues being noted in our sources regarding Jerusalem is indeed worthy of consideration. Whether this should lead to Fleshers particular conclusion, or whether the paucity of sources should caution us against drawing far-reaching conclusions in this regard, involves a basic methodological issue over which opinions continue to be divided. The assumption that the Temple and synagogue represented two dierent Judaisms, per Flesher, is certainly questionable. The line between two Judaisms on the one hand, and two frameworks with dierent but complementary functions within one all-encompassing Judaism on the other, is a ne one that should be rigorously argued and not merely asserted. If the Judaism of the synagogue and Temple were indeed so dierent, it is surprising that no ancient source bothered to note this clash. Philo, Jesus, Paul, and Josephus do not seem to have been aware of two such Judaisms, and it would seem that no such dichotomy existed in antiquity. In summary, we nd in late Second Temple Judaean society two contrasting developments. On the one hand, the Temple was assuming an ever more central role in Jewish life, not only because of the growth of Jerusalem as an urban center and as a focus of signicant pilgrimage, but also because of the accruement of power by the priesthood and
122. Flesher, Palestinian Synagogues, 2829. 123. In this regard, Flesher is forced to dismiss the evidence of several New Testament references that speak of synagogues in Jerusalem generally; see, for example, Luke 4:44; Acts 22:24, 26.


historical development of the synagogue

the enhanced role of the Temple Mount (at least since Herods time) as the setting for a wide range of social, economic, religious, and political activities.124 On the other hand, the synagogue had been evolving as a distinct and dened institution, having fully assumed its role at the center of communal activity in Second Temple Judaea. Centralization in terms of the Temple was paralleled by a decentralization in the local synagogues. Prior to 70, the Temple was recognized as the central institution in Jewish life; nevertheless, the emerging synagogue had become the pivotal institution in local Jewish aairs. This parallel development in the rst century was indeed fortuitous. Though no one could have foreseen the outcome, the seeds of Jewish communal and religious continuity had already been sown well before the destruction of the Temple.
124. L. Levine, Jerusalem, 22653.



iaspora communities, particularly those of Alexandria and Egypt, have provided us with a signicant amount of material regarding the Hellenistic and early Roman synagogue, or proseuche. Epigraphical evidence hails from as early as the third century b.c.e., papyrological and archaeological data from the second century b.c.e., and literary sources from the rst century c.e. Together these sources aord an intriguing, if only partial, picture of this institution throughout the Hellenistic-Roman Diaspora. Regarding external appearance and internal organization, there were signicant dierences between the synagogues in Alexandria, Cyrene, Ostia, Delos, and Asia Minor. Even the various names by which communities referred to the synagogue may well reect dierent perceptions of the institution and its place in society. Nevertheless, the Diaspora synagogue fullled much the same function as a communal and religious center within each Jewish community, and Roman authorities clearly articulated the rights and privileges of this institution and the community in general in a number of contemporary decrees and edicts. By the rst century c.e., Jewish communities were to be found the length and breadth of the Roman Empire, with the possible exception of the northern and western provinces.1 Despite the well-known problems with regard to demographic estimates for an-

1. The widespread Jewish dispersion is attested, for example, by Diodorus Siculus (Bibliotheca Historica 40, 3, 8) for the Hellenistic period and, for the Roman era, by Strabo as quoted by Josephus in Antiqui-


historical development of the synagogue

tiquity, it appears quite certain that the Jewish population of the Diaspora, estimates of which ranged between two and ve million, outnumbered that of Judaea well before 70 c.e. 2 It is reasonable to assume that almost any Jewish community would have had its own place (topos per Josephus),3 i.e., a synagogue. Thus, the number of such institutions throughout the Empire undoubtedly reached into the many hundreds, if not thousands. However, the information available regarding the pre-70 Diaspora synagogue relates only to a very small percentage of these places and, what is more, varies greatly in what is presented, and how. Evidence for the geographical distribution of this institution is likewise imbalanced. Egyptian Jewry is relatively well documented; information about an important region such as Syria is practically nil; Asia Minor merits considerable attention in several sources, particularly Josephus and Acts, but only limited information is available regarding Greece, Italy, North Africa, and the Bosphorus region. Nevertheless, when taken together, what we have is far from negligible, and it is to an examination of this material that we now turn. Given the extensive geographical dispersion, the variety of sources, and the fact that the sources tend to focus on particular communities, each locale will be discussed individually.

EGYPT Epigraphical and Papyrological Evidence

Although no synagogue building has yet been discovered in Egypt, the epigraphical material that has been recovered, supplemented by a number of papyri,4 has contributed enormously to the study of this institution in the Ptolemaic-Roman era. Much of this material is considerably earlier than other Diaspora evidence.5 Altogether, the Egyptian synagogue is mentioned explicitly in fteen such sources and is implied in ve more.6
ties 14, 115; Philo, Moses 2, 232; idem, Flaccus 4546; idem, Embassy 214, 245, 28182. See also Acts 2:911; and Antiquities 4, 11516 (in Josephus recasting of Bilams prophecy). 2. A maximalist position is adopted by Baron, Social and Religious History of the Jews, I, 16771, 36972; a minimalist position is taken by McGing, Population and Proselytism, 88106. On the Diaspora generally, see M. Stern, Jewish Diaspora, 11783; Smallwood, Jews under Roman Rule, 35688; A. Kasher, Jewish Migration and Settlement, 6591. With regard to Diaspora synagogues, see Binder, Into the Temple Courts, 227341; Runesson, Origins, 40176; Gruen, Diaspora, 10523; Claussen, Versammlung, 83 112, 191208; Fitzpatrick-McKinley, Synagogue Communities, 5587. 3. Antiquities 14, 235, 260. 4. Tcherikover et al., CPJ, I, 8. 5. The evidence for a signicant Jewish presence in Egypt at the outset of the Hellenistic period is persuasive. What is less clear is whether this presence was forced, i.e., because of captivity, or the result of free choiceor both. There are conicting reports not only in the sources (Letter of Aristeas 1214 and Josephus, Antiquities 12, 1133; contra idem, Against Apion 1, 18689), but also among historians (Tcherikover, Hellenistic Civilization, 273; Modrzejewski, How to Be a Jew, 7576). See also L. Levine, Jerusalem, 4849. 6. The epigraphical material has been conveniently collected and extensively analyzed by Horbury

the pre-70 diaspora


Two inscriptions date from the reign of Ptolemy III Euergetes (246221 b.c.e.) and his wife Berenice, three from the reign of Ptolemy VIII (145116 b.c.e.) and his two wives named Cleopatra, three from the second or rst century b.c.e., and one from the rst century c.e. 7 Two other inscriptions are more dicult to date and stem from the late Hellenistic or early Roman eras.8 An inscription from Leontopolis may refer to a proseuche, and fragmentary remains of four inscriptions make mention of a temenos and probably also refer to a synagogue.9 Finally, four papyri dating from the late third century b.c.e. to the beginning of the second century c.e. note local synagogues in a variety of contexts.10 Altogether, this evidence sheds light on important aspects of the early Egyptian synagogue. The dedicatory inscription is the most common type, appearing (with minor dierences) some eight times throughout the Ptolemaic era. To cite two examples:
On behalf of king Ptolemy and queen Berenice his sister and wife and their children, the Jews [dedicated] the proseuche.11 On behalf of king Ptolemy and queen Cleopatra the sister and queen Cleopatra the wife, Benefactors, the Jews in Nitriai [dedicated] the proseuche and its appurtenances.12

Such inscriptions clearly reect the common Egyptian Jewish practice of dedicating synagogues to the ruling family. The geographical and chronological distribution of these inscriptions indicates that this practice was accepted by many segments of Egyptian Jewry. The implications of such a practice are fairly obvious: it expresses the loyalty and gratitude of the Jewish community toward the king and queen, as well as reects the Jews dependence upon them. The status of the Jews in Ptolemaic Egypt as part of the class of Hellenes (i.e., resident aliens and not native Egyptians) was due to their protection by and service to the king.13 In a strikingly similar fashion, Onias IV, who ed Judaea and sought asylum in Egypt, petitioned Ptolemy VI to build a temple to the God of Israel at Leontopolis in the likeness of that at Jerusalem and with the same dimensions on behalf of you and your wife and children. 14

and Noy, Jewish Inscriptions. Previously, most of these inscriptions had appeared in Freys CIJ and were later re-edited by Lewis in vol. III of CPJ. The last-mentioned work remains basic for papyrological material. 7. Horbury and Noy, Jewish Inscriptions, nos. 22, 117 (Ptolemy III); nos. 24, 25, 125 (Ptolemy VIII); nos. 13, 27, 28 (second to rst centuries b.c.e.); no. 126 (rst century c.e.). 8. Ibid., nos. 9, 20. 9. Proseuche: ibid., no. 105. Temenos: ibid., nos. 16, 17, 127, 129. 10. Tcherikover et al., CPJ, I, nos. 129, 134, 138; II, no. 432. 11. Horbury and Noy, Jewish Inscriptions, no. 22. 12. Ibid., no. 25. 13. See Bickerman, Jews in the Greek Age, 8385; Modrzejewski, Jews of Egypt, 7387. 14. Josephus, Antiquities 13, 67. On this episode, see Tcherikover et al., CPJ, I, 4446; Grabbe, Judaism, I, 26667; Modrzejewski, Jews of Egypt, 12133; Gruen, Origins, 4770.


historical development of the synagogue

Egyptian Jewry was dependent upon royal recognition and support for its communal institutions, its right to own and administer property and assets, as well as the legitimacy and authority of its communal activities and decisions. Such royal backing is reected in a number of inscriptions: the ruling couple is referred to as benefactors, they declare a synagogue inviolate (), and order an earlier dedicatory inscription to be restored.15 In terms of synagogue practice generally, the Egyptian Jewish custom of dedicating such a building to the ruler is most unusual, and only two other parallels are known: a dedicatory inscription from Qatzion in the Upper Galilee from 197 c.e. (although the identity of that building is far from clear) and a fragmentary inscription from late secondcentury c.e. Osijek, Hungary.16 Several instances from Italy approximate Egyptian practice in that a number of synagogues in Rome were named after prominent Romans, including Augustus.17 An Ostia inscription notes the well-being of Augustus, 18 and a late midrash speaks of a synagogue in Rome named after Severus.19 However, even these similarthough not identicalinstances are relatively few in number, and thus the concentration of dedicatory inscriptions in Egypt is indeed unique. This was undoubtedly due to the centralized control exercised by the Ptolemies; as a result, religious (and other) buildings often required the sanction and authorization of the ruler.20 Evidence of this pattern in pagan Egypt is not lacking, and the practice carried over to the Jewish community.21 It is important to note that despite the clear and unequivocal imitation of this Ptolemaic dedicatory norm, the Jews, nevertheless, adapted it so as not to compromise their
15. Horbury and Noy, Jewish Inscriptions, nos. 25 and 125, respectively. See also A. Kasher, Three Jewish Communities, 11516. 16. The inscription from Qatzion is quite explicit as to the dedicatees (Septimius Severus and his sons, Caracalla and Geta), the date (197 c.e.), and the donors (the Jews). However, it continues to be debated whether the building (as yet not fully excavated) in which it was found was a synagogue or some other building. Advocates of a synagogue identication are Schrer (History, III, 93), S. Klein (Galilee, 127), Avi-Yonah (In the Days of Rome and Byzantium, 49), and Roth-Gerson (Greek Inscriptions, 12529), while those who question this identication include Kohl and Watzinger (Antike Synagogen, 209) as well as Lifshitz, who excludes this inscription from his collection of Greek dedicatory inscriptions (Donateurs et fondateurs). Regarding the inscription from Hungary, see Scheiber, Jewish Inscriptions, 5355; and below, Chap. 8. 17. From the catacomb inscriptions, we learn of a synagogue of the Augustesians, Agrippesians, and perhaps also Volumnesians; see Leon, Jews of Ancient Rome, 14042, 15759; and below, note 134. 18. Dating from the rst or second century, the fragmentary inscription reads: pro salute aug[usti] (For the well-being of the emperor); see Noy, JIWE, I, no. 13, as well as Fortis, Jews and Synagogues, 118; White, Social Origins, 39294. 19. Genesis Rabbati 45, 8 (p. 209). 20. Fraser, Ptolemaic Alexandria, I, 190; A. Kasher, Jews in Hellenistic and Roman Egypt, 13438. 21. Pagans: Fraser, Ptolemaic Alexandria, I, 19091, 226., 28284. See also Tcherikover, Hellenistic Civilization, 349; Hengel, Die Synagogeninschrift von Stobi, 174 n. 97; idem, Proseuche und Synagoge, 159. Jews: Dion, Synagogues et temples, 4575.

the pre-70 diaspora


own religious sensibilities. In contrast to the pagans, for example, the avoidance of divine epithets (especially ) was an elegant way of not acknowledging royal divinity.22 Philo makes a point of noting Jewish sensitivities in this regard.23 According to Josephus, Onias IV was also careful in his expression of obeisance to the Egyptian king when negotiating the building of the Leontopolis temple, as were the Jewish and Samaritan protagonists whose case was brought before Ptolemy Philometor.24 The overwhelming majority of references in Egypt are to a proseuche, appearing ten times in the inscriptions and four times in the papyri.25 The word synagoge is used once, and a reference to ocers of this institution twice; the designation eucheion (a place of prayer) appears only on one occasion.26 The Jews of Ptolemaic Egypt also borrowed terminology associated with pagan contexts in other instances as well. The phrase used to describe the God of Israel ( theos hypsistos, the Most High God) is documented in pagan as well as Jewish contexts, as are various terms for synagogue ocers, such as the archisynagogue and nakoros.27 The religious dimension of these proseuchai is reected in the sanctity accorded to at least some of them. A number of inscriptions specically refer to the holy or great place;28 other sources associate the institution with the Most High God. 29 The sanctity of one proseuche was expressed as follows: On the orders of the queen and king, in place
22. Hengel, Proseuche und Synagoge, 16162; Fraser, Ptolemaic Alexandria, I, 283. 23. Philo, Embassy 13449. 24. Onias IV: Antiquities 13, 67. Jews and Samaritans: ibid., 13, 7476. 25. Horbury and Noy, Jewish Inscriptions, nos. 9, 13, 22, 24, 25, 27, 28, 117, 125, 126. For references in the papyri, see above, note 10. On the term proseuche with reference to the Jewish community, see Levinskaya, Books of Acts in Its Diaspora Setting, 20725. 26. Synagogue: Horbury and Noy, Jewish Inscriptions, no. 20. Papyrus no. 138 (Tcherikover et al., CPJ, I) seems to refer to a meeting of a Jewish (burial?) association in the proseuche. Synagogue ocials: Horbury and Noy, Jewish Inscriptions, nos. 18, 26. Eucheion: Tcherikover et al., CPJ, II, no. 432. 27. Theos hypsistos: C. Roberts et al., Gild of Zeus, 5572; Fraser, Ptolemaic Alexandria, I, 282; II, 440 nn. 76465; TDNT, VIII, 61419; Simon, Theos Hypsistos, 37285; Levinskaya, Book of Acts in Its Diaspora Setting, 51103; Mitchell, Cult of Theos Hypsistos, 81148, and esp. 11021. This phrase was already widely used in the Septuagint (e.g., Gen. 18:20; Ps. 7:8; 17:14), Egyptian Jewish inscriptions (Horbury and Noy, Jewish Inscriptions, 27475; and above), and contemporary Jewish Hellenistic literature (II Macc. 3, 31; III Macc. 7, 9). It appears also in the Delos synagogue (Frey, CIJ, I, nos. 72729; Schrer, History, III, 7071; and below), and the Bosphorus Kingdom (ibid., 72; Frey, CIJ, I, 690, 690a; Lifshitz, Prolegomenon, 67; Levinskaya, Book of Acts in Its Diaspora Setting, 22946; Goodenough, Bosphorus Inscriptions, 22145; and below). See also Kraabel, Hypsistos and the Synagogue at Sardis, 8193. Archisynagogos: Horbury and Noy, Jewish Inscriptions, no. 18, and comments on p. 29. See also Rajak and Noy, Archisynagogoi, 7593. Nakoros (attendant): Tcherikover et al., CPJ, I, no. 129; see also the material gathered in Dion, Synagogues et temples, 6573; G. H. R. Horsley, New Documents, IV, 4952. 28. Horbury and Noy, Jewish Inscriptions, nos. 16, 17, 127. Although the term proseuche does not appear in these fragmentary inscriptions, there can be little doubt that such a building was intended. 29. Ibid., nos. 19, 27, 105.


historical development of the synagogue

of the previous plaque about the dedication of the proseuche, let what is written below be written up: King Ptolemy Euergetes [proclaimed] the proseuche inviolate []. The queen and king gave the order. 30 The original inscription, usually dated to the latter part of the second century b.c.e., thus attests to the holy status enjoyed by an Egyptian proseuche in the Ptolemaic period.31 Such a status may well be paralleled in a papyrus from Alexandrou-Nesos in the Fayyum dated to 218 b.c.e., where it is stated that a Jew named Dorotheus was accused of stealing a cloak and took refuge in a proseuche (for purposes of asylum?). Only after the intervention of a third party did Dorotheus agree to leave the cloak with the nakoros of the synagogue until nal adjudication.32 Another indication of the synagogues sanctity, albeit indirect, is reected in the use of terms such as and for sacred precinct in connection with a proseuche.33 Furthermore, a second-century papyrus describes a plot of land attached to a proseuche in Arsinoe-Crocodilopolis as a sacred grove or garden ( ).34 The above clearly imply that in many places, at the very least, the synagogue was considered a sacred institution. Philo, too, alludes to the sacredness and inviolability of proseuchai on a number of occasions (see below).35 Epigraphical evidence makes it quite clear that the proseuche might include other buildings or structures in addition to the sacred precinct (i.e., land or courtyards) noted above. Several inscriptions mention , which seems to refer to ancillary buildings, annexes to the main building, or landholdings.36 Other structures may have in30. Ibid., no. 125; Tcherikover et al., CPJ, I, no. 125; Dion, Synagogues et temples, 5759; Modrzejewski, Jews of Egypt, 9798; Rigsby, Asylia, 57173. 31. Dion prefers to date this inscription to the days of Ptolemy VIII Euergetes (145116 b.c.e.) and notes the interesting but not particularly compelling parallel act of granting the right of asylum to the Jerusalem Temple by Demetrius in 152 (I Macc. 10:43). See also Horbury and Noy, Jewish Inscriptions, 214. 32. Tcherikover et al., CPJ, I, no. 129. See also A. Kasher, Synagogues as Houses of Prayer and Holy Places, 215. On the oce in general, see Llewelyn and Kearsley, New Documents, VI, 2036. 33. Horbury and Noy, Jewish Inscriptions, nos. 9, 129; Frey, CIJ, II, 1433; Lifshitz, Donateurs et fondateurs, no. 87; Dion, Synagogues et temples, 5960. 34. Tcherikover et al., CPJ, I, no. 134. See also A. Kasher, Jews in Hellenistic and Roman Egypt, 138 39. Cf., however, an alternative suggestion of Modrzejewski ( Jews of Egypt, 89), who views this term as a well-known technical term for one category of landed property. 35. Another indication of the holiness of a proseuche is reected in III Macc. 7:1920: And when they nished their voyage in peace with appropriate thanksgivings, there, too, in like manner they determined to celebrate these days also as festive for the duration of their community. They inscribed them as holy on a pillar and dedicated a house of prayer [ ] at the site of the banquet. That a proseuche was built as a memorial to the miraculous salvation of a community is noteworthy. Unfortunately, the historicity of much of this books narrative is questionable. See, for example, Nickelsburg, Stories, 8084; Modrzejewski, Jews of Egypt, 14153. 36. Horbury and Noy, Jewish Inscriptions, nos. 9, 25. On the various meanings of , see ibid., 14; A. Kasher, Three Jewish Communities, 121. On the appearance of this term in a fragmentary inscription from Cyrene, see Fraser, Inscriptions of Cyrene, 11516.

the pre-70 diaspora


cluded a gateway (), such as the one from second-century Xenephyris, which is noted as having been part of a proseuche, or an exedra, such as the one from second- or rst-century b.c.e. Athribis.37 One papyrus mentions what appears to be a rather high water bill owed by two local synagogues; we can only conjecture that this may have been due to the use of water for guests, communal needs (e.g., sacred meals), or ritual purposes. However, it is also conceivable that some water may have been used for domestic purposes by Jews whose homes were located near the proseuche.38 A number of other interesting details regarding Egyptian proseuchai emerge from these data. Dedicatory inscriptions are about evenly divided between the community as a whole and wealthy individuals. Proseuchai were built by the Jewish communities of ArsinoeCrocodilopolis, Schedia, Nitriai, Xenephyris, and Athribis; all of these originated in the Hellenistic period and were dedicated to the royal couple.39 Among the seven inscriptions mentioning individual donations, two speak of donating the entire building, the others of donating parts thereof: an exedra, a sundial, and a well.40 The remaining inscriptions are fragmentary and make no mention of the objects involved.41 Of these seven inscriptions, two were in honor of the royal couple.42 In only two cases is the donors name mentioned, while in two others the donors wife and children are also included.43 Although the dating of these dedicatory inscriptions is uncertain, they appear to range from the second century b.c.e. to the late Roman period (i.e., the secondthird centuries c.e.). The proseuches centrality to the Jewish community is reected not only in the number of dedicatory inscriptions, buildings, and property associated with it but also by the fact that it was the meeting-place for various Jewish associations (). So, for example, we read of one such group (a burial society?) meeting in a proseuche.44 Less clear
37. Xenephyris: Horbury and Noy, Jewish Inscriptions, no. 24. Athribis: ibid., no. 28. This was apparently an annex (partially open?) to the main hall or building, itself used for a variety of purposes. On the exedra, see ibid., 49; A. Kasher, Jews in Hellenistic and Roman Egypt, 117; and S. Krauss, Synagogale Altertmer, 34950. See also Goodenoughs suggestion (virtually ignored subsequently) to read cathedra instead of exedra, thus turning this into a reference to a bench or perhaps a Seat of Moses ( Jewish Symbols, II, 85). See Fraser, Ptolemaic Alexandria, II, 443 n. 773; Griths, Egypt and the Rise of the Synagogue, 910. 38. Tcherikover et al., CPJ, II, no. 432; Fuks, in ibid., 221; and A. Kasher, Jews in Hellenistic and Roman Egypt, 14044. 39. Horbury and Noy, Jewish Inscriptions, nos. 117, 22, 25, 27, and 24 (which mentions only the exedra). 40. Building: ibid., nos. 13, 126. Exedra: ibid., no. 28; for possible meanings of the term exedra, see Tcherikover et al., CPJ, III, 143, no. 1444; Horbury and Noy, Jewish Inscriptions, 4950; and, more generally, Modrzejewski, Jews of Egypt, 96. Sundial and well: Horbury and Noy, Jewish Inscriptions, no. 115. 41. Horbury and Noy, Jewish Inscriptions, nos. 16, 17, 20, 27, 129. 42. Ibid., nos. 13, 28. 43. Ibid., nos. 13, 20; and ibid., nos. 28, 126, respectively. See Noy, Jewish Place of Prayer, 11822. 44. Tcherikover et al., CPJ, I, no. 138. On the dekany as a burial association, see Noy, JIWE, II, no. 440. On the meaning of dekany in the Aphrodisias inscription, see Reynolds and Tannenbaum, Jews and God-Fearers, 2830; and below, Chap. 8.


historical development of the synagogue

is a reference to a meeting of a Sambathic association, perhaps in Naucratis.45 While the identication of this latter group (and if, indeed, it was even Jewish) and its precise venue remain unclear, from what we know of Ptolemaic Egypt generally, a templeor, in this case, the Jewish proseuchewould have been an obvious choice.46 Several papyri from Arsinoe contain some interesting details regarding local synagogues. One second-century c.e. document dealing with water distribution, referred to above, notes two institutions, one called a proseuche, the other an eucheion.47 The former is identied as having belonged to Theban Jews; the latter presumably belonged to the indigenous population. If this was the case, then we have here an interesting example of Theban Jews organizing their own house of worship that also served as their Landsmannschaft. A second papyrus, from the second century b.c.e., is a land survey noting that the synagogue was located on the outskirts of the town and bordered by private estates and a canalquite possibly indicating that a (the?) Jewish quarter of the town was there.48 One inscription refers to a gold crown, presumably a token of honor bestowed on someone.49 This well-known pagan practice was adopted by Diaspora Jews in both Cyrene and Asia Minor (see below), and apparently in Egypt as well. A papyrus notes a Jewish communal archive ( ) in Abusin el-Meleq that most likely was located in the local synagogue, and it is here that important documentscontracts, records of priestly lineage, wills, ocial statements, etc.were deposited.50 Local synagogues must also have had arrangements for the safekeeping of communal monies earmarked for local use or for the Jerusalem Temple.51 Finally, there is evidence that several Egyptian synagogues, one in Alexandria and the other in Naucratis, had statues. Statue bases were discovered in each, one with the explicit inscription to the synagogue ( ), the second mentioning a Sambathic association.52 Despite an attempt to explain away this phenomenon (e.g., both inscrip45. Horbury and Noy, Jewish Inscriptions, no. 26 and comments on pp. 4445. 46. C. Roberts et al., Gild of Zeus, 7287. Compare this to the meeting of a dekany in the Aphrodisias inscription; see below, Chap. 8; and White, Building Gods House, 88. 47. See above, note 38; and Tcherikover et al., CPJ, II, no. 432. 48. Tcherikover et al., CPJ, I, no. 134 and comments by Tcherikover on pp. 24748; A. Kasher, Jews in Hellenistic and Roman Egypt, 13839. 49. Horbury and Noy, Jewish Inscriptions, no. 129; see also Goodenough, Jewish Symbols, VII, 14871; and below. 50. Tcherikover et al., CPJ, II, no. 143; see also Frey, CIJ, II, no. 775. On the importance of documents proving priestly lineage that were stored in communal archives, see Josephus, Against Apion 1, 31 36; idem, Life 6. 51. See, for example, Philo, Embassy 15657, 216, 291, 31216; idem, Special Laws 1, 77. See also Josephus, Antiquities 14, 11213, 21416, 260, 261; 16, 16072. 52. Horbury and Noy, Jewish Inscriptions, nos. 20, 26.

the pre-70 diaspora


tions deal with Judaizers and not full-edged Jews: these were people who did not share the sensibilities of some Jews about images; this was a pagan institution), we may well have here evidence of communities whose Jewish practice condoned such images, not unlike those who built and attended the third-century Nehardea (Babylonia) synagogue that also had a statue.53

The writings of Philo are of inestimable importance as a source for Alexandrian Jewry generally and for the synagogue in particular. Living at the height of this Jewrys power and prosperity and through traumaticeven cataclysmicevents that shook the community to its foundations, Philo was far from being a dispassionate and objective bystander. This commitment, added to his natural penchant for conveying a denite religious and cultural message to readers and listeners, means that one must exercise caution in evaluating many of his claims.54 For example, Philo speaks of Jews in Rome conducting regular weekly meetings on sacred Sabbaths, when they are trained in their ancestral philosophy.55 He refers to proseuchai as schools () for the inculcation of virtue, emphasizing the instructional dimension of these synagogue gatherings, which were based on scriptural readings.56 These sessions were led by a priest or elder and may have lasted for a good part of the day;57 Philo himself mentions the late afternoon as a terminus ad quem.58 With regard to the Therapeutae, Philo describes the solemnity surrounding their Sabbath observance, which likewise featured an extensive discourse oered by the senior member of the group.59 Philo portrays the synagogues religious agenda as an intensive intellectual experience, and there may be a modicum of truth to his claim. Similar frameworks for serious philo53. B Rosh Hashanah 24b. See Rajak, Jews as Benefactors, 2728. 54. Opinions regarding the reliability of Philo as a historian and commentator on current events are seriously divided; see, for example, Smallwood, Philo and Josephus, 11429; and, for an opposite view, D. R. Schwartz, Josephus and Philo, 2645. See also idem, On Drama and Authenticity in Philo and Josephus, 11329. 55. Embassy 156. 56. Moses 2, 21516; Special Laws 2, 62, 63; Embassy 312. See also On Dreams 2, 127, and Leonhardt, Jewish Worship in Philo of Alexandria, 7495. 57. See Letter of Aristeas 310. 58. Hypothetica 7, 13: And indeed they do always assemble and sit together, most of them in silence except when it is the practice to add something to signify approval of what is read. But some priest who is present or one of the elders reads the holy laws to them and expounds them point by point till about the late afternoon, when they depart having gained both expert knowledge of the holy laws and considerable advance in piety. See also below, note 61. 59. Contemplative Life 31. On Therapeutae generally, see Schrer, History, II, 59197.


historical development of the synagogue

sophical discussions and study sessions were not an uncommon feature in the Roman world, and some Jewsparticularly in the various sectsmay well have created similar settings.60 The real question, however, is how widespread such a practice was. Did it engage only a Jewish intellectual elite in Alexandria, or was it typical of many Egyptian proseuchai, both in Alexandria and the chora? I am inclined to prefer the former alternative, as Philos emphasis is too unique and extreme: he alone calls the synagogue a didaskaleion. To assume that ordinary Jews would be interested in such intensive study sessions or would be willing to stay in the synagogue for much of the Sabbath day ies in the face of all we know of human nature and Jewish practice de facto.61 Philo also notes that the proseuche functioned as a repository, where funds for the Temple were collected and stored until their transfer to Jerusalem. Presumably, funds for local communal use were deposited there as well.62 In his dramatic account of the Alexandrian pogroms of 38 c.e. and their aftermath, Philo takes note of Alexandrian synagogues on a number of occasions.63 These buildings were located in every section of the city, and there was one particularly magnicent proseuche that he describes as the largest and most magnicent [ ] in the city. 64 The building was lavishly decorated with, inter alia, insignia, shields, golden crowns, stelae, and inscriptions honoring the emperor.65 By specically mentioning these accoutrements, Philo may have been indicating the loyalty of the Jewish community to Rome, thereby discounting one of the main charges brought by the Roman governor Flaccus and the Alexandrians against the Jews. Emphasizing the legal and recognized status of these buildings (Philo claims that only the Jews were so privileged by Augustus), he excoriates those perpetrating the violence and destruction as guilty of heretofore unheard of desecration and abominable acts.66 According to Philo, the desecration reached such proportions that, not only were the synagogues despoiled and in some cases
60. See Mason, Greco-Roman, Jewish, and Christian Philosophies, 1218. 61. In rst- and second-century Palestine, Jews abandoned liturgical study sessions or even intense political discussions for their Sabbath midday meal; see, for example, Josephus, Life 279; B Betzah 15b. Cf., however, A. Kasher (Synagogues as Houses of Prayer and Holy Places, 211), who suggests that the extended scriptural readings in Egyptian synagogues, as described by Philo, originated in the desire to imitate the original reading of the Septuagint as described in the Letter of Aristeas, implying that this was practiced widely. 62. Embassy 15657, 216, 31216. See also A. Kasher, Synagogues as Houses of Prayer and Holy Places, 217 n. 44. 63. On events in Alexandria, see Tcherikover et al., CPJ, I, 5574; Smallwood, Jews under Roman Rule, 22055; Modrzejewski, Jews of Egypt, 16183. 64. Embassy 134. 65. Embassy 133. 66. On the recognized status: ibid., 13839, 311. See also Josephus, Antiquities 14, 21316. On the violence: Philo, Flaccus 41.; idem, Embassy 132.

the pre-70 diaspora


destroyed, but statues of the emperor as well as other images (e.g., a bronze statue of a man riding a quadriga) were introduced in direct violation of the status quo enshrined by earlier Ptolemaic and Roman rulers.67 One nal comment on Philos terminology is in order. In line with Egyptian Jewish practice, as noted above, Philo almost always (nineteen times) uses the term proseuche or a derivative.68 Nevertheless, he does use the term synagoge on two occasions.69 Moreover, Philo, too, alludes to the institutions sanctity, invoking the terms and and .70

A Rabbinic Tradition
The number of sources in rabbinic literature relating to the pre-70 Roman Diaspora is almost negligible; even rarer are the references to the Diaspora synagogue. Nevertheless, we have one most unusual pericope, and if its historicity (or a signicant part thereof ) is upheld, it would constitute a source of major importance to the subject at hand. First appearing in the third-century Tosefta, this tradition is subsequently cited, with variations, in both the Yerushalmi and the Bavli.71 Owing to its importance, I quote the Toseftan version in full:
R. Judah [b. Ilai] said: Whoever has not seen the double stoa [i.e., colonnade] of Alexandria has never in his life seen the glory of Israel. It is a kind of large basilica, a stoa within a stoa, holding, at times, twice the number of those who left Egypt. And seventy-one cathedrae [i.e., honorary chairs or thrones] of gold were there for the seventy-one elders, each of them [worth] 25 talents [of gold], and a wooden platform [ ]was in the middle. And a azzan of the synagogue [lit., assembly] stood on it with kerchiefs in his hand. When one took hold [of the Torah scroll] to read, he would wave the kerchiefs and they [i.e., those congregated] would answer Amen for each benediction; and he would again [wave the kerchiefs] and they would [again] respond Amen. And they would not sit randomly, but goldsmiths would sit by themselves, silversmiths by themselves, weavers by themselves, Tarsian weavers by themselves, and blacksmiths by themselves. And why to such an extent [i.e., why the dierentiated
67. Flaccus 43; Embassy 13435, 138. It is not clear, however, whether the desecration caused by the introduction of statues meant that the proseuche per se was considered sacred by the Jews, as has sometimes been claimed. 68. Mayer, Index Philoneus, 247. See also Hengel, Proseuche und Synagoge, 169; A. Kasher, Synagogues as Houses of Prayer and Holy Places, 210. 69. Embassy 311; On Dreams 2, 127. With respect to the Essenes, see Philos Every Good Man Is Free 81; idem, Special Laws 3, 171. 70. Embassy 137; Flaccus 48. The term in Special Laws 3, 171, probably refers to a pagan temple, contra Binder, Into the Temple Courts, 12930. On the other hand, when Josephus refers to in second century b.c.e. Egypt (quoting a letter from Onias to Ptolemy), he seems to mean Jewish religious buildings (temples? shrines? synagogues?); Antiquities 13, 6566. 71. T Sukkah 4, 6 (p. 273); Y Sukkah 5, 1, 55ab; B Sukkah 51b.


historical development of the synagogue

11. Plan of the Sardis synagogue.

seating]? So that if a visitor comes he can [immediately] make contact with his trade, and thus he will be able to make a living [27 .]

According to the above tradition, this Alexandrian building was of colossal proportions. The statement that it could hold twice the number of those who left Egypt (i.e., 1.2 million people!) was, of course, never intended to be taken literally; it is a stock rabbinic phrase connoting a very large number of people.73 In this case, the reference is to an assembly hall of such monumental size that kerchiefs were required in order to signal the congregation when to respond. The description of the golden chairs, each worth twenty-ve talents, is probably exaggerated as well. Nevertheless, the above description is so detailed and unique that it perhaps ought not be rejected out of hand as totally fanciful, especially in light of the fact that archaeological excavations at Sardis have revealed a fourth-century and later synagogue building of monumental dimensionsits assembly hall and atrium measuring eighty meters in length (g. 11).74 This rabbinic tradition immediately calls to mind the large Alexandrian synagogue that Philo describes in his narrative of the events of 38 c.e., and it is quite plausible that both the rabbinic tradition and Philo refer to the very same building. The Toseftas description of the main hall as a kind of basilica, dyplastoon (a stoa within a stoa or a double stoa), is compatible with the architectural traditions of the period. The hall may have had rows of columns, perhaps two deep on two or four sides, thus forming a series of aisles, examples of which can be seen in the Basilica Aemilia and the Basilica Julia in Rome.75 Such
72. On this source, see the comments in Lieberman, TK, IV, 88992; S. Krauss, Synagogale Altertmer, 26163; Fraser, Ptolemaic Alexandria, I, 28485; A. Kasher, Jews in Hellenistic and Roman Egypt, 34955. See also Gordon, Basilica and the Stoa, 35962. 73. Lieberman, TK, IV, 890 n. 8. 74. Seager, Building History, 425; Seager and Kraabel, Synagogue and the Jewish Community, 169; While, Social Origins, 31024; and below, Chap. 8. 75. Boethius and Ward-Perkins, Etruscan and Roman Architecture, 19294. On monumental buildings in the East, particularly the Alexandrian kaisareion, see ibid., 45960.

the pre-70 diaspora


a multi-aisled synagogue was discovered in Byzantine Gaza (seventh century c.e.).76 As noted in Chap. 3, the rst-century proseuche in Tiberias must also have been of large proportions if meetings of the citys residents could be conducted there instead of the citys stadium.77 Another Tiberian synagogue of the third or fourth century (perhaps the same as the rst-century one) was described in rabbinic sources in a similar fashion as the one in Alexandria, namely, as a dyplastoon.78 Regarding the synagogue or community elders mentioned in the above Toseftan tradition, the number is far from unusual. While seventy clearly has a symbolic ring, the fact is that this number was adopted by many Jewish leadership bodies during this period.79 Rabbinic literature reports seventy (or seventy-one) members in the Jerusalem sanhedrin; Josephus appointed seventy leaders when organizing the Galilee in 6667 c.e.; there were seventy prominent people who represented the Jews residing in Batanaea, and the Zealots appointed seventy members to a high court in Jerusalem during the revolt.80 From an archaeological perspective, the only evidence that could possibly relate to this Alexandrian tradition of seventy-one elders comes, as mentioned, from fourth-century Sardis. At the western end of the hall were three semicircular benches, clearly intended for people of rank within the congregation. These Sardis elders, who sat on benches facing eastward, toward the center of the hall and the Torah shrine (or shrines), also may have numbered seventy, as the buildings excavators have estimated on the basis of the space available.81 Thus, the number of leaders in these two communities may have been identical, although the seating arrangement in Sardis was dierent from that indicated by the individual Alexandrian cathedrae described in the Tosefta. In the above-quoted Toseftan tradition, reference is made to a wooden platform (, bima) in the center of the hall that was used for the reading of Scriptures. Once again,
76. Ovadiah, Synagogue at Gaza, 195. 77. Josephus, Life 92, 27680, 331. 78. Midrash on Psalms 93, 8 (p. 208b). 79. Num. 11:16. On the council of elders () heading the Alexandrian community in the rst century, see Philo, Flaccus 74; Schrer, History, III, 9394; Smallwood, Jews under Roman Rule, 22733; M. Stern, Jewish Community and Institutions, 16869. For other representative bodies numbering seventy in the biblical period, see Judg. 9:2; II Kgs. 10:1. 80. M Sanhedrin 1, 6; War 2, 570; Life 14, 79; War 2, 482; Life 11, 56; and War 4, 336, respectively. Equally interesting is the fact that later on the Samaritans, too, had a governing council of seventy (lit., Family of Seventy), which was well established in Samaritan life at the time of Baba Rabbas reforms in the third century. See J. M. Cohen, Samaritan Chronicle, 70, 22829; and below, Chap. 6. On the number seventy in later rabbinic sources, see Ginzberg, Legends, VII, 429. 81. Seager and Kraabel, Synagogue and the Jewish Community, 169; Seager, Building History, 426. Interestingly, this seating arrangement is very dierent from that prescribed in T Megillah 3, 21 (p. 360), where the elders faced the congregation with their backs to the holy, i.e., the Torah shrine or the direction of Jerusalem.


historical development of the synagogue

the nds at Sardis prove enlightening, as a stone table was found in the middle of the hall, toward its western end, and what may have been traces of a platform or canopy were found toward the middle.82 This custom of having a table or bima in the center of the hall was thus not uncommon in synagogues of the Roman world. In synagogues of Second Temple Judaea, such as those at Gamla, Masada, Herodium, Qiryat Sefer, and Modiin, the reading of Scriptures would also have been carried out in the center of the hall since the benches and columns on all four sides left no room for a platform at one end.83 In third-century Dura Europos as well, benches on all four sides of the room would have required setting up a table or platform in the center of the hall. In fact, the nal excavation report notes a number of depressions found in the oor in the center of the room, perhaps made by the legs of a platform that once stood there.84 One element of the above tradition that has proven as intriguing as it has elusive is the concluding section dealing with the seating arrangements in the synagogue. Each professional group seems to have sat separately. Why should professional aliation have proven so critical in this synagogue? One can indeed point to the inscriptions in Rome, where at least one synagogue appears to have been organized around a professional group,85 although most Jews, both in Rome and elsewhere, seem to have based their aliation on other criteria. Several scholars have attempted to place this Alexandrian synagogue in a setting that would explain the economic element in its seating arrangements. Krauss, followed by Fraser, has suggested that the building itself was a basilica-marketplace, a merchants hall used mainly for economic purposes but also for worship and judicial proceedings.86 Alternatively, one might posit that this Alexandrian synagogue included members of all groups but that only the artisansfor whatever reasonsat in this fashion. Once again, the Sardis synagogue may oer an interesting parallel. Located on the main street of the city, it stood adjacent to a row of shops, many of which appear to have been owned by Jews.87 One side entrance of this synagogue even joined its atrium to these shops. Might there have been a similar situation in Alexandria, and might this in some way explain the unique seating arrangements specically aecting the artisan class there? Perhaps we ought to be looking elsewhereto the non-Jewish worldfor at least a
82. Seager and Kraabel, Synagogue and the Jewish Community, 16970. 83. See the articles of Yadin, Foerster, Gutman, and Maoz in L. Levine, Ancient Synagogues Revealed, 1941; and L. Levine, Second Temple Synagogue, 1019. 84. Kraeling, Excavations at Dura: Synagogue, 256. 85. Leon, Jews of Ancient Rome, 14244. 86. S. Krauss, Synagogale Altertmer, 26163; Fraser, Ptolemaic Alexandria, I, 285. Goodenough ( Jewish Symbols, II, 86) repeats this claim, although he is skeptical about the value of the entire tradition. 87. On representations of menorot on shops adjacent to the synagogue, see Seager and Kraabel, Synagogue and the Jewish Community, 17687; Hanfmann et al., Roman and Late Antique Period, 166.

the pre-70 diaspora


partial explanation of this phenomenon. There is a great deal of evidence, both literary and archaeological, attesting to highly structured and dierentiated seating arrangements in Roman public spaces, e.g., theaters and amphitheaters. Even during the late Republic, but especially after Augustus, there were set places for dierent Roman social and political groupings, not to speak of distinctions among foreigners, collegia, soldiers, circus factions, women, and others.88 It might well be, then, that as regards seating, the Alexandrian synagogue reected current Roman practice in large places of assembly and that rabbinic tradition noted this fact in its description. One nal issue to be addressed concerns the historicity of the above-quoted Toseftan tradition. It is quite obvious that many phrases in this source recall descriptions of the Jerusalem Temple found in Josephus as well as elsewhere in rabbinic literature. On the Temple Mount, we are told, there was a large area enclosed by a double stoa (a stoa within a stoa), at one end of which was a basilica (i.e., a royal stoa) of colossal proportions.89 The placing of the platform in the center of the Alexandrian synagogue and the custom of waving kerchiefs are likewise reminiscent of Temple practice. At the Haqhel celebration held in the Temple every seven years, a special wooden platform was constructed for the reading of the Torah;90 whenever the high priest would ociate in the daily ritual, a Temple functionary ( )would stand by the altar and signal by waving a kerchief.91 Even the opening phrase of the above-quoted tradition (Whoever has not seen the double stoa . . . of Alexandria has never in his life seen the glory of Israel) is remarkably similar to the hyperboles occasionally used in rabbinic literature when introducing Templerelated matters.92 Therefore, it is quite in place to ask whether these literary parallels do not, in eect, undermine the historical veracity of our source. Perhaps the transmitter of the Alexandrian synagogue description, R. Judah b. Ilai (or someone before him), had collected a series of phrases that originally related to the Temple and appended them to a description of the well-known Alexandrian synagogue, even though they had no basis in reality.93
88. See Suetonius, Augustus 44. For a discussion of the various social groups, see Rawson, Discrimina Ordinvm: The Lex Julia Theatralis, 83114; Small, Social Correlations to the Greek Cavea, 8593; Rouech, Aphrodisias in Late Antiquity, 21826; Edmondson, Dynamic Arena, 81111; Van Nijf, Civic World, 20940. My thanks to Zeev Weiss for bringing several of these references to my attention. See also MacMullen, Roman Social Relations, 7179. 89. Y Sukkah 5, 1, 55a; Y Taanit 3, 11, 66d; B Sukkah 45a. See also Josephus, War 5, 190; idem, Antiquities 15, 396, 41116. 90. M Sotah 7, 8. 91. M Tamid 7, 3. 92. For instance: Whoever has not seen the Temple standing has never seen a magnicent building B Sukkah 51b; Whoever has not seen Herods Temple has never seen a beautiful buildingB Bava Batra 4a. 93. The tendency to project descriptions of the Temple (not to speak of Temple practices) onto the synagogue is widespread in rabbinic literature. See, for example, R. Isaacs assertion that the phrase miq-


historical development of the synagogue

Perhaps, but I think not. In the rst place, why would someone (presumably in secondcentury Roman Palestine) want or need to invent such an exaggerated depiction? If it was, indeed, such a blatant fabrication, why would R. Judah or the editors of the Tosefta even bother to report it? Furthermore, the parallels noted above, from both Philo and Sardis, lend a measure of plausibility to the assumption that some such building might have existed in as powerful and wealthy a community as that of rst-century Alexandria. The fact that there are so many allusions to the Jerusalem Temple in the description of this synagogue may not necessarily be due to literary style but rather to a historical reality resulting from either a common architectural tradition that inuenced both Alexandrian and Jerusalem Jews or, what seems more likely, a conscious attempt by Alexandrian Jews to emulate the form and patterns of Herods Temple in their large synagogue. It should be remembered that this seems to be what Onias IV did when building his temple in Leontopolis in the second century b.c.e. 94 There may be other indications of the adoption of Temple-related practices by Egyptian synagogues,95 and if this was the case, then the Toseftan literary parallels point to a very signicant historical reality: the imitation of some Temple-related architectural forms in at least one important Diaspora synagogue (see above, Chap. 3).96

Three important communal inscriptions relating to the synagogue were found in this North African city, and together they contain not a few surprises.97 First and foremost is the very nature of these Greek inscriptions, which are decrees of the local Jewish polidash meat in Ezek. 11:16 refers to the synagogue (B Megillah 29a). The dierence, of course, is that in our case the reference relates to a particular place and time. 94. Josephus, Antiquities 12, 388; 13, 63, 67, 72; 20, 236; idem, War 1, 33. Cf., however, ibid., 7, 427. 95. See A. Kasher, Synagogues as Houses of Prayer and Holy Places, 20520; Binder, Into the Temple Courts, 23354; Runesson, Origins, 43659. Moreover, the claim that R. Judah was using a current Palestinian model anachronistically is dicult. It seems rather far-fetched, to say the least, that in the wake of the various unsuccessful rebellions and the resultant economic and social upheavals something as monumental as the building described here existed or was under construction in second-century Palestine. In all probability, Jews at that time would not have been able to aord such a structure, and the assumption that Galilean-type synagogues were being built around that time, in the mid second century, is no longer valid. Therefore, a claim for anachronism is hardly credible; see Krautheimer, Constantinian Basilica, 12324 and n. 22; Gutmann, Ancient Synagogues: Archaeological Fact, 22627; Fine, This Holy Place, 4345. 96. T Sukkah 4, 6. This source was not cited by Binder in his Into the Temple Courts, which is unfortunate. Since one of his aims was to show Temple inuence over synagogues everywhere, such a tradition might have been helpful. In any case, the thesis is very problematic in its own right. See above, Chap. 3. 97. On the history of the Jewish community of Cyrene generally, see Applebaum, Jews and Greeks, 130.; Hirschberg, History of the Jews in North Africa, I, 2186.

the pre-70 diaspora


teuma honoring various individuals who had beneted it in one way or another. These decrees not only refer to the same community but span a period of approximately sixty-ve years, thus oering a repeated glimpse into the workings and concerns of this synagogue. Moreover, the inscriptions refer to the synagogue as an institution (or, as we shall see, they also use an alternative term) and therefore furnish precious information in this regard.98 The earliest of these inscriptions, discovered several centuries ago, is the most poorly preserved of the three. It records a resolution of the Jewish community (referred to here as a politeuma) and its archons to honor one Decimus Valerius Dionysios in gratitude for his benefactions. The following is the text of the inscription:
In the year [?] 3, on the 5th of Phamenoth, in the archonship of Arimmas son of . . . , Dorion son of Ptolemaios, Zelaios son of Gnaius, Ariston son of Araxa . . . , Sarapion son of Andromachos, Nikias son of . . . , . . . son of Simon. Whereas Dec[i]mus Valerius Dionysios son of Gaius . . . remains a noble and good man in word and deed . . . , doing whatever good he can, both in a public capacity and as a private individual, to each one of the citizens, and in particular plastering the oor of the amphitheater and painting its walls, the archons and the politeuma of the Jews at Berenice resolved to register him in the . . . of the . . . and [resolved] that he be exempted from liturgies of every kind; and likewise [they resolved] to crown him with an olive wreath and a woolen llet, mentioning his name at each assembly and at the New Moon. After engraving this resolution on a stele of Parian marble the archons are to set it in the most visible place in the amphitheater. All [the stones cast were] white [i.e., the decision was unanimous]. Dec[i]mus Valerius Dionysios son of Gaius plastered the oor of the amphitheater and painted [its walls] at his own expense as a contribution to the politeuma.99

Dating to the end of the rst century b.c.e.,100 this inscription is remarkable on a number of counts. We learn that the Jewish community was led by archons and organized as a politeuma.101 At one stage it was assumed that this term reected the usual form of Jewish communal organization in the Greco-Roman Diaspora; however, this view has been called into question of late, particularly with regard to the Alexandrian and the larger Egyptian community.102 Nevertheless, the evidence is clear-cut, at least from Berenice;
98. Published originally by Roux and Roux (Dcret de politeuma des juifs, 28196), these inscriptions have been analyzed by Reynolds (Inscriptions, 24247) and Lderitz (Corpus jdischer Zeugnisse aus der Cyrenaika, 14758). See also G. H. R. Horsley, New Documents, IV, 2029; White, Social Origins, 296300; Binder, Into the Temple Courts, 25763. 99. Translation based on G. H. R. Horsley, New Documents, IV, 203, with some changes. 100. Roux and Roux (Dcret de politeuma des juifs, 28889) date this inscription to 86 b.c.e., assuming the missing letter is iota (= 10) or, more likely, kappa (= 20) and that the era is that of Actium (31 b.c.e.). 101. Cf., however, the problematic suggestion by Lderitz (What Is the Politeuma? 183225), who denes politeuma here as the Jewish oligarchy or ruling body, and not the community as a whole. 102. See, for example, Tcherikover et al., CPJ, I, 6, 9, 32, 61; Smallwood, Jews under Roman Rule, 225


historical development of the synagogue

such a form of communal organization did, in fact, exist there. It is well attested throughout antiquity, in all parts of the Diaspora, that archons often stood at the head of a Jewish community.103 Four of the six preserved personal names of these archons are Greek; of the patronyms, four are in Greek, two are entirely missing, and one (Simon) is of Jewish derivation. The name of the honoree, Decimus Valerius Dionysios, son of Gaius, bears a Roman stamp, and he appears to have been a Roman citizen. If this is the case, then a number of other issues connected with his name are unclear: Why is no tribal status noted? Was his father also a Roman citizen? Was he a freedman? What was his connection (ocial or otherwise) to the Jewish community? 104 In fact, it appears certain that Decimus was a member of the politeuma, since the decree notes that he was to be exempt from communal liturgies. The honor accorded him consisted of an olive crown and a llet, and the mention of his name at each assembly and on the New Moon. This assembly may well refer to the Sabbath gathering, well known in rst-century sources. However, taking special note of a monthly meeting is unusual, as New Moon celebrations are unknown elsewhere in the Diaspora.105 The reference to Decimus signicant impact on many people (the citizens of Berenice generally? the Jewish community only?) may have been due to his position as a public ocial. Specically, he is recognized as having contributed to the Jewish community by plastering the oor of the amphitheater and painting its walls. One of Decimus benefactions, the painting of walls, is most intriguing. The Greek word conveys two possible meanings: to paint generally or to paint gures (human or animal). If the former was intended, then the paintings may have been similar to more or less contemporary ones at Pompeii, Jerusalem, or Masada.106 If the latter was intended, then the amphitheater would have boasted more striking decorations (gural images?); were it a synagogue, as will be argued below, then the decorations might have been similar to what has been found in other Diaspora synagogues, such as the third-century walls at Dura Europos or the sixth-century mosaic oor at ammam Lif in North Africa.107 At this juncture, however, certitude is elusive.
33, 35960; A. Kasher, Jews in Hellenistic and Roman Egypt, 29. See the recent discussion and bibliography on this issue in Zuckerman, Hellenistic Politeumata and the Jews, 17185; Hnigman, Birth of a Diaspora, 9398; Lderitz, What Is the Politeuma? 183225. 103. See below, Chap. 11. 104. See Reynolds, Inscriptions, 24647; Lderitz, Corpus jdischer Zeugnisse aus der Cyrenaika, 151. 105. The juxtaposition of regular (i.e., Sabbath) meetings and those of the New Moon in this and the following inscription (see below) is intriguing, but elusive; see Judith 8:6; McKay, Sabbath and Synagogue, 1142. 106. Pompeii: Brion, Pompeii and Herculaneum, 20130. Jerusalem: Avigad, Discovering Jerusalem, 81. Masada: NEAEHL, III, 973., and generally Hachlili, Ancient Jewish Art and ArchaeologyIsrael, 6583. 107. See below, Chap. 8. On the murals in the rst-century synagogue at Acmonia (Asia Minor), see

the pre-70 diaspora


Perhaps the most engaging detail in this inscription is the thrice-mentioned term amphitheater: To what does it refer? Was it a civic building that served all citizens as a place of sports, entertainment, or assembly? Or was this the synagogue of the Jewish politeuma? Those who have addressed the issue have formulated three distinct approaches. Schrer, Goodenough, Caputo, Gabba, Horsley, and Zuckerman assume that this was a regular Roman amphitheater;108 Robert, Applebaum, Cohen, M. Stern, Rajak, and Barclay assume it was a Jewish public building;109 Hirschberg, Roux, Reynolds, Lderitz, and White are less committal.110 The major arguments in favor of regarding the amphitheater as a civic building are as follows: the term itself suggests that interpretation; the decorations involved are attested in public entertainment buildings elsewhere in North Africa; Jews are known to have frequented theaters and amphitheaters and thus may well have contributed funds to them; the inscription also notes Decimus Valerius Dionysios benevolence to all citizens of the city; the Jews of Cyrene were quite hellenized, and therefore erecting a stele in a civic building might constitute but one more example of their acculturation; the amphitheater was probably a well-enough-established institution in the late rst century b.c.e. as to preclude applying the name to a dierent type of building. On the other hand, the following considerations argue for identifying the amphitheater as a specically Jewish building: Why should a prominent Jew have been responsible for or have seen t to repair and plaster the oor of a city amphitheater? Moreover, the usual Roman amphitheater is not known to have had plastered oors. Would the Jews have been able to place their own stele in a civic institution, and why would they have wanted to place their communal inscriptions in a public arena? And if Dionysios benecence was not related to them, why would the Jewish community have honored him for it? The amphitheater was still not a functionally well-dened institution at this time, and consequently the name could well have been used for other institutions, in this case a synagogue.111 Moreover, the last lines of the inscription indicate that Decimus Valerius
below. The same root, , appears on a third- or fourth-century sarcophagus from Rome to designate the profession of Eudoxius (Noy, JIWE, II, no. 277). 108. Schrer, History, III, 104; Goodenough, Jewish Symbols, II, 14344; XII, 52 n. 11; Caputo, Nota sugli edici teatrali della Cirenaica, 28385; Gabba, Iscrizioni Greche e Latine, 63.; G. H. R. Horsley, New Documents, IV, 2089; Zuckerman, Hellenistic Politeumata and the Jews, 179. 109. Robert, Gladiateurs, 34 n. 1; Applebaum, Jews and Greeks, 16467; idem, Organization of Jewish Communities, 48688; S. J. D. Cohen, From the Maccabees to the Mishnah, 10910; M. Stern, Jewish Diaspora, 135; Rajak, Jews as Benefactors, 2830; Barclay, Jews in the Mediterranean Diaspora, 237; Baldwin Bowsky, M. Tittius, 507; Binder, Into the Temple Courts, 14045. 110. Hirschberg, History of the Jews in North Africa, I, 26; Roux and Roux, Dcret de politeuma des juifs, 29092; Reynolds, Inscriptions, 247; Lderitz, Corpus jdischer Zeugnisse aus der Cyrenaika, 155; White, Social Origins, 297 n. 36. Cf., however, Lderitzs comments in What Is the Politeuma? 213, where he seems much more comfortable with a Jewish identication. 111. On the architectural uidity of the Roman amphitheater prior to the Flavian era and the building


historical development of the synagogue

Dionysios gift was given to the politeuma, thus conrming that this was most probably a Jewish building. A second inscription from Berenice (see below) likewise associates the Jewish community with an amphitheater, further reinforcing this connection. If, on the other hand, the amphitheater was a civic institution, then why were the Jews using it regularly for their communal purposes? 112 Certitude in this matter is impossible. We read of no remotely similar occurrence elsewhere in antiquity. As we are dealing here with obviously Jewish communal matters, the use of the term amphitheater for a Jewish building is most unusual, to say the least. However, to assume that this inscription refers to honors bestowed by the Jews on one of their own in return for benefactions to the citys amphitheater, and then to call this a contribution () to the politeumata, would require an enormous stretch of the imagination. It seems, therefore, that we are indeed dealing here with a Jewish institution.113 We cannot be sure why exactly it was called an amphitheater. The most likely explanation is that the name was related to the shape of the building. The word seems to indicate a circular or elliptical structure where the audience sits in the round, or, per Dionysios of Halicarnassus, it could refer to a U-shaped building with seating on three sides.114 Serving as a meeting place for the community, this amphitheater qua synagogue had to have adequate seating arrangements. We possess no information regarding its location in the city. Applebaum once suggested a site outside the city wall to the south, but later retracted this in light of subsequent excavations.115 The second Berenice inscription, from 2425 c.e.,116 contains a further resolution of

of the colosseum, see Golvin, Lamphithtre romain, I, 26872. Humphrey (Amphitheatrical HippoStadia, 122) writes: In the Late Republic and early Empire, the terminology for Roman entertainment buildings, and especially for the building that we would later know as the amphitheater, was still in ux. Welch, in Roman Amphitheatres Revived, 273, notes: The Colosseum canonized the Roman amphitheatre as an architectural form. Amphitheatres securely dated after it (e.g., Capua) self-consciously refer to it in the same way that circuses throughout the empire looked back to the Circus Maximus. On the early evolution of the amphitheater, however, Welch adopts a dierent approach than that of Golvin; see idem, Arena in Late-Republican Italy, 69. 112. On the other hand, theaters, amphitheaters, and stadiums were often used for meetings of the citizenry as a whole, as in AlexandriaJosephus, War 2, 49091; Antiochibid., 7, 47; EphesusActs 19:29; TiberiasJosephus, Life 331. 113. See M. Stern, Jewish Diaspora, 135. 114. See, for example, his description of the old Circus Maximus as a U-shaped building in Roman Antiquities 3, 68, 3; and comments by Cary, LCL, II, 24243 n. 2. See also Humphrey, Amphitheatrical Hippo-Stadia, 12223. It is interesting to note that in the fourth century c.e. Epiphanius speaks of Samaritans imitating the Jews by building a synagogue which is shaped like a theater and thus is open to the sky (Panarion 80, 1, 6). 115. Applebaum, Jewish Community, 15962; idem, Jews and Greeks, 194. 116. For an earlier dating, see Baldwin Bowsky, M. Tittius, 5046.

the pre-70 diaspora


the communitythis time in the name of nine archons and the politeuma at largetaken on the festival of Sukkot. The inscription notes the honors bestowed upon a Roman ocial, Marcus Tittius, for his support of the Jewish community, as well as for his kindness to the Greek citizens of the city.
Year 55. Phaoph 25. At the gathering of the Festival of Tabernacles during the terms of oce of the archons: Cleandros son of Stratonicos Euphranor son of Ariston Sosigenes son of Sosippos Andromachos son of Andromachos Marcus Laelius Onasion son of Apollonios Philonides son of Hagemon Autochles son of Zenon Sonicos son of Theodotos Josephos son of Straton Since Marcus Tittius 117 son of Sextus [from the tribe of ] Aemilia, a goodly and worthy man, who has assumed the responsibility of government in public aairs and who has exercised management of these matters benevolently and rightly, and has always displayed in his conduct a gentle character on all occasions; and not only does he give of himself unstintingly in these matters to those citizens [of the city generally] who entreat him in private, but also to the Jews of our politeuma, publicly and privately, he has been supportive in his governance and has not ceased, in his own noble goodness, behaving in a worthy manner. Now, therefore, the archons and politeuma of the Jews in Berenice have decided to praise him by name and dedicate to him at each assembly and new moon [celebration] a wreath of olive branches and a woolen llet; and [it has been decided] that the archons are to record this resolution on a stele of Parian marble and set it up in the most prominent place in the amphitheater. All [the stones cast] were white.118

The dating of this document seems certain: the year 55 apparently relates to the Actium era, i.e., 2425 c.e. 119 The document is clearly a Jewish one. It begins with a list of archons and a decree taken on the Sukkot holiday.120 The benefactions of Marcus Tittius to the Jews may have been appreciably more signicant than those he bestowed on the Greeks; at least the Jews seem to have thought so. That such a declaration was made by the Jew117. On the spelling of this name, see Reynolds, Inscriptions, 245. On the honoree, see Baldwin Bowsky, M. Tittius, 495. 118. For the Greek text, see the references in note 98. 119. However, Lderitz (What Is the Politeuma? 212) has suggested an earlier date43 b.c.e., as has Baldwin Bowsky13 b.c.e. (M. Tittius, 5046). 120. Might this Sukkot meeting time be related to a reference in a later Christian source, stating that Jewish archons were elected annually in September (Tishri)? See Schrer, History, III, 99; and below, Chap. 8. See also the reference to Sukkot by Plutarch (Quaestiones Convivales, IV, 6, 2 (GLAJJ, I, 557); as well as Burtchaell, From Synagogue to Church, 235 n. 61.


historical development of the synagogue

ish community at its regular weekly (?) and monthly meetings in the amphitheater once again suggests that this was a Jewish building. The inscription provides clear evidence of the salutary rapport enjoyed by the Jewish community with this Roman ocial. Moreover, the communitys willingness to publicly proclaim these excellent relations, orally and in writing, is striking evidence of its self-condence and rapport vis--vis the authorities. The fact that the favors bestowed by Marcus Tittius do not appear to have been at the expense of, or opposed by, the local citizenry but were part of his overall policy reinforces the impression thatat this juncture, at leastthe Jews were on good terms with their Greek neighbors and Romes representatives. This had not always been the case, as a letter from Marcus Agrippa to the polis of Cyrene some four decades earlier attests.121 The use of wreaths and llets indicates the imitation of the Greek custom in bestowing honors, as does dedicating a stele in a public building. There is no more striking evidence of this acculturation than the names of the nine archons and their fathers. Of the twenty names, nineteen are Greek or Roman and only one (Josephos) is Jewish. It is noteworthy that these Greek names (as in the other two inscriptions) were popular in pagan Cyrene.122 Together with the Greek language, the other elements in this inscriptionthe year and month, the voting process (casting white stones), its formulary announcement, the designation politeuma, and the ways in which honor was bestowedall point to a community comfortably ensconced in its larger Greco-Roman milieu.123 Once again, we are informed of the communitys appointed times for honoring Tittius. It would seem that the regular assembly was on the Sabbath, and additional meetings were held at the beginning of each month. At least part of these assemblies undoubtedly included a liturgical element; to what extent, if at all, these gatherings dealt with secular communal matters is unknown. If they did, then such a combination would be reminiscent of Josephus account of events in Tiberias at the outbreak of the revolt in 66 c.e. 124 It is of interest that major gatherings of the community took place on the Sukkot festival, when important decisions such as the one recorded in the above inscription were made. A third inscription, from 55 c.e., commemorates a series of donations made by at least eighteen individuals (part of the slab is broken and part is missing) for restoring their synagogue (here referred to as a ):

121. Josephus, Antiquities 16, 16061, 16970; Pucci Ben Zeev, Jewish Rights in the Roman World, 27380. Hirschbergs suggestion, that Tittius was honored because he implemented the Imperial edicts recorded by Josephus, is problematic owing to the time gap between the decree and the writing of the inscription; see his History of the Jews in North Africa, I, 2526. 122. Reynolds, Inscriptions, 24445; Applebaum, Jewish Community, 163. 123. See Fraser, Ptolemaic Alexandria, II, 484 and n. 781. 124. Life 276303; and above, Chap. 3.

the pre-70 diaspora


The year 2 of Nero Claudius Caesar Drusius Germanicus Imperator on 6 Choiak: It has been decided by the community [] of Jews in Berenice that those donating towards the restoration of the synagogue building [] be inscribed on a stele of Parian marble: Zenion son of Zoilos archon 10 dr[achmae] Isidoros son of Dositheos archon 10 dr. Dositheos son of Ammonios archon 10 dr. Pratis son of Jonathan archon 10 dr. Karnedas son of Cornelius archon 10 dr. Heracleides son of Heracleides archon 10 dr. Thaliarchos son of Dositheos archon 10 dr. Sosibios son of Jason archon 10 dr. Pratomedes son of Socrates archon 10 dr. Antigonos son of Straton archon 10 dr. Kartisthenes son of Archias priest 10 dr. Lysanias son of Lysanias 25 dr. Zenodoros son of Theuphilos 28 dr. Mar . . . [son of ] 25 dr. Alexander son of Euphranor 5 dr. Isidora daughter of Serapion 5 dr. Zosima daughter of Terpolios 5 dr. Polon son of Dositheos 5 dr.125

Compared with the two previous inscriptions, this one is unique in a number of ways. The list of donors reveals a wealth of names unmatched in the other Cyrenian inscriptions. Once again, Greek names predominate, with many characteristic Greek Cyrenian (Karnedas, Kartisthenes, Pratis, Pratomedes), Egyptian (Ammonios, Serapion), Roman (Cornelius), and Hebrew ( Jonathan) names. Of the eighteen donors, the rst ten are archonsas compared to seven and nine in the two earlier inscriptionsa factor that may indicate growth in the local Jewish community or perhaps only an administrative reorganization that had taken place in the three decades between the second and third inscriptions. Either of these alternatives is possible, although the latter option is supported by the fact that the word synagoge appears twice here, once denoting the community (instead of the previously used term, politeuma) and once with reference to the building (instead of amphitheater). The above terms ( politeuma-synagogue; amphitheater-synagogue) may have been synonymous,126 and thus no great signicance is to be attributed to the change in nomenclature.
125. For the Greek text, see above, note 98; and Lifshitz, Donateurs et fondateurs, no. 100. See Reynolds, Inscriptions, 244; Applebaum, Jewish Community, 163 and esp. n. 152. 126. See A. Kasher, Jews in Hellenistic and Roman Egypt, 181.


historical development of the synagogue

It is less likely that these terms refer to two distinct Jewish frameworks existing side by side in the city.127 We can only speculate as to the reasons for the communitys decision to change its nomenclature. Was it assuming a more Jewish posture; i.e., were these new terms now considered more Jewishly identiable and signicant? This line of reasoning, however, is not borne out by the names of the community leaders or the language of the inscription, where the Greek component remains strong. Perhaps the Jews right to be called a politeuma was revoked? Was the term amphitheater for the building now associated more with a sports arena and thus became inappropriate for a Jewish public edice? It may also be the case that the Jewish community underwent some farreaching changesinternally and externallyin the rst century c.e.; however, any such changes remain unknown.128 In any case, we can assume rather condently that in all three inscriptions the main body of Berenice Jewry, and not some marginal group, is the subject at hand; each appears to speak on behalf of the citys entire Jewish community. The third inscription also raises several points of interest with regard to the people and sums of money mentioned. First among the non-archons on the list is a priest, perhaps owing to his lineage and prominent status in the community; this is reminiscent of Philos reference to a priest presiding over Sabbath instruction as well as of Jerusalems Theodotos inscription.129 Moreover, several Jewish women are mentioned as donors, a phenomenon attested for rst-century Asia Minor and one that recurs with greater frequency in later antiquity, especially in the Diaspora.130 Finally, the sums of money recorded are generally quite modest, although it is impossible to know whether this is signicant regarding the economic and political status of the Jewish community in the mid-fties.

Although the Ostia synagogue as it stands today dates from the fourth century c.e., there are several earlier stages in the buildings history.131 Most of the extant building existed for at least several centuries: the main hall, adjacent areas to the east, the triclinium, and parts of other walls. All underwent extensive renovations over time, including those sections that were added to the original structure. Nevertheless, there is gen127. Perhaps, as in second-century b.c.e. Memphis, where the terms politeuma and synagoge appear simultaneously; see Rappaport, Idumens en Egypte, 7382. See also Applebaum, Jews and Greeks, 162 n. 149. 128. See Applebaum, New Jewish Inscription, 172. 129. See above; and Chap. 3. 130. See below, Chap. 14. 131. On the earlier stages of this building, see Squarciapino, Synagogue of Ostia, 25; Fortis, Jews and Synagogues, 12425; Kraabel, Diaspora Synagogue, 49899; White, Building Gods House, 69; idem, Social Origins, 37982; idem, Synagogue and Society, 2738.

the pre-70 diaspora


eral agreement that the earliest traces of the building hail back to the rst century c.e.; whether or not it was a synagogue at this stage is a hotly contested point of dispute among scholars.132 The synagogue itself was located on the outskirts of the city, near the city wall and close to the sea, a pattern adopted by other Diaspora Jewish communities as well (see below, Chap. 9).

Our knowledge of synagogues in Rome derives primarily from the rich epigraphical evidence found in the local Jewish catacombs. Originally attributed to the rst centuries c.e., these nds have been reevaluated over the past decade, resulting in a general consensus that dates them from the third to fth centuries c.e. 133 Nevertheless, some of the synagogues referred to probably existed already in the rst century c.e. There are four likely candidates, three of which were apparently named after prominent rst-century Romans: Augustus, Agrippa, and Volumnius. The fourth is called the synagogue of the Hebrews, perhaps so-named because it was the rst Jewish congregation in the city.134 In addition, one literary source clearly relates to the rst-century synagogues of Rome Philos description of the Roman community in his Embassy. In noting the benecence of Augustus to the Jews generally, and to those in the capital city in particular, he writes:
How then did he [Augustus] show his approval? He was aware that the great section of Rome on the other side of the Tiber is occupied and inhabited by Jews, most of whom were emancipated Roman citizens. For having been brought as captives to Italy they were liberated by their owners and were not forced to violate any of their native institutions. He knew therefore that they have houses of prayer [ proseuchai] and meet together in them, particularly on the sacred sabbaths when they receive as a body a training in their ancestral philosophy. He knew, too, that they collect money for sacred purposes from their rst-fruits, and send them to Jerusalem by persons who would oer the sacrices. Yet, nevertheless, he neither ejected them from Rome nor deprived them of their Roman citizenship because they were careful
132. On the buildings history and the alternative reconstructions, see below, Chap. 8. 133. Rutgers, berlegungen zu den jdischen Katakomben, 14057. 134. La Piana, Foreign Groups, 35456; Leon, Jews of Ancient Rome, 14042, 14749, 15759; M. Stern, Jewish Diaspora, 16667; van der Horst, Ancient Jewish Epitaphs, 8689. Synagogues of the Hebrews were also found in Philadelphia in Lydia (see Frey, CIJ, II, no. 754; Trebilco, Jewish Communities, 162) and Corinth (Frey, CIJ, I, no. 718), and a synagogue of the Jews is attested in Reggio di Calabria (ibid., no. 635b; Noy, JIWE, I, no. 139). For a review of the interpretations of the term Hebrews, see Harvey, Synagogues of the Hebrews, 13247. It is indeed unusual that Jews would name their synagogues after prominent Romans; even in Ptolemaic Egypt, proseuchai were dedicated to rulers but not named after them. The high regard in which the Jews held these rulers (and perhaps were dependent upon them) is certainly reected in this practice, just as it is in Philos encomium to Augustus cited above. See now Richardson (Augustan-Era Synagogues in Rome, 1729), who suggests that there were perhaps ve early synagogues, including that of the Herodians.


historical development of the synagogue

to preserve their Jewish citizenship also, nor took any violent measures against the houses of prayer, nor prevented them from meeting to receive instructions in the laws, nor opposed their oerings of the rst-fruits. Indeed so religiously did he respect our interests that supported by wellnigh his whole household he adorned our temple through the costliness of his dedications, and ordered that for all time continuous sacrices of whole burnt oerings should be carried out every day at his own expense as a tribute to the most high God.135

Philos description of the Roman proseuchai is fascinating. The problem, of course, is determining how much here is Philos projection of Alexandrian practices and terminology onto Romes Jewry and how much actually accords with the Jewish scene in the Imperial capital during the rst century c.e. If later catacomb evidence can provide a clue, it would seem that Philos use of the term proseuche is out of place here. Epigraphical evidence from the catacombs indicates that the Jews of Rome used the term synagoge with but rare exception.136 But since these inscriptions date by and large from the third to fth centuries, it is also possible that the nomenclature had changed by then. In this vein, we should also note that Juvenal, in the early second century, likewise used the term proseuche in discussing the local Jewish community.137 Nevertheless, Philos description of these Roman synagogues and their practices seems to correspond with what we know about other Jewish communal frameworks at the time: the Jews possessed a communal building, met regularly, especially on the Sabbaths, designated study as the focal liturgical activity, and sent funds (i.e., the rstfruits) to Jerusalem.138 In short, although we cannot be entirely certain of Philos accuracy, this information should not be rejected out of hand. In fact, Philos description of the status of Romes Jewish community, in its general outline, is corroborated by Josephus. In a decree directed to the citizens of Delos, Josephus records the following: For example, Gaius Caesar, our consular praetor, by edict forbade religious societies to assemble in the city [i.e., Rome itself ], but these people alone he did not forbid to do so or to collect contributions of money or to hold common meals. 139 The rights thus enjoyed by Romes Jewish communityto assemble, collect monies, and hold communal mealswere rmly xed and were the basis of Jewish communal life. Moreover, the granting of such privileges was an exception to general policy. According to this document at least, only the Jews were accorded such recognition, at rst

135. Embassy 15557. 136. Frey, CIJ, I, no. 531; Leon, Jews of Ancient Rome, 139. 137. Satires III, 296 (M. Stern, GLAJJ, II, 99). 138. Philo may have accorded his last point special prominence given the overall agenda of Embassy, i.e., to demonstrate the sanctity and centrality of the Jerusalem Temple for Jews everywhere. Still, it is doubtful whether Philo would have invented such practices rather than merely highlight certain existing customs for his purposes. See also Embassy 31216. 139. Antiquities 14, 215; and comments in Smallwood, Jews under Roman Rule, 135 n. 52; Rajak, Jewish Rights, 2324.

the pre-70 diaspora


by Julius Caesar and later by Julius Gaius.140 In reality, however, many groups in Rome functioned in a similar fashion in the rst century c.e., whether ocially or unocially. Their rights, however, were not always guaranteed and their fortunes might uctuate over time; some political and religious collegia were forbidden, others were declared licita, and still others existed surreptitiously.141 Of interest regarding Romes synagogues is another statement made by Philo, that the Jews tended to live together in the Transtiberine (Trastevere) region, much as Alexandrian Jewry concentrated in two of that citys ve districts.142 It appears that of the eleven or more synagogues mentioned in the catacomb inscriptions, at least seven were located in the Transtiberine area.143 But just as some Alexandrian Jews were to be found in other quarters of the city as well,144 such was the case in Rome, too. In fact, the discovery of Jewish catacombs in dierent directions outside the city walls might argue for such a spread, at least at a later period.145

Discovered in the early part of the twentieth century, the building at Delos, an Aegean island lying to the south and east of the Greek mainland, has been a subject of debate for decades. Only since the 1970s has a consensus emerged that the building was a synagogue, the earliest known to date and the only building complex securely identied as such from the pre-70 Diaspora.146 It is unclear precisely when the local Jewish community built or acquired this building. It is generally agreed that the site functioned as a synagogue at least from the rst century b.c.e. until the second century c.e., and that the structure did not undergo any far-reaching alterations during this period. There is little agreement, however, as to when it rst became a synagogue. The controversy involves the degree to which the structure
140. See below, note 167. On Caesars attitude toward collegia generally, see Suetonius, Iulius 42, 3: He dissolved all guilds, except those of ancient foundation. 141. See La Piana, Foreign Groups, 183. On Jewish rights, especially in comparison with those of other peoples, see Pucci Ben Zeev, Caesar and Jewish Law, 2837. 142. Philo, Flaccus 55. Cf., however, War 2, 495, where Josephus claims that the Jews were particularly concentrated in the Delta quarter, a claim indirectly borne out in the papyri; see Tcherikover et al., CPJ, II, nos. 194, 200, 202, 209, 212, 213, 221. 143. M. Stern, Jewish Diaspora, 167; and below, Chap. 8. 144. Philo, Embassy 132. 145. M. Stern, Jewish Diaspora, 16568; and below, Chap. 8. 146. Overviews of the history of this debate have been oered on a number of occasions; see, for example, Bruneau, Recherches sur les cultes de Dlos, 486.; idem, Les Isralites de Dlos, 48995; Kraabel, Diaspora Synagogue, 491; White, Delos Synagogue Revisited, 13740; idem, Social Origins, 332 42. See also Goodenough, Jewish Symbols, II, 7175.


historical development of the synagogue

12. Plan of the Delos synagogue. Room A functioned as the main assembly hall. Note the benches along two walls and the chair in the middle of the western wall.

suered damages, if at all, following Mithridates conquest of the city in 88 b.c.e. On the one hand, some believe that the rebuilding was substantive, and that only afterwards was a synagogue established there. Beforehand, it is opined that the building was used for other purposes, either as a private home (White) or a pagan cultic hall (Runesson).147 On the other hand, some believe that the building functioned as a synagogue even earlier, and thus dates from the second or perhaps even the late third, century b.c.e. (Trmper).148 Still others have left the issue open (Bruneau, Binder).149 The synagogue was located on the eastern shore of the island, far from the main harbor and city center.150 The complex, measuring 28.30 by 30.70 meters (g. 12), comprised three parts: (1) a tristoa (or atrium) facing eastward with remains of a marble stylobate and marble benches (C); (2) a large room subsequently divided into two (A and B). Room A had a triportal entrance, marble benches, as well as a carved marble chair and niche on its western wall; and (3) a series of small rooms in the southern part of the building (D) under which lay a large water reservoir.151
147. Private quarters: Kraabel, Diaspora Synagogue, 493; White, Social Origins, 33637; Hachlili, Ancient Jewish Art and ArchaeologyDiaspora, 3539, and others. Pagan cultic hall: Runesson, Origins, 186 87. 148. Trmper, Oldest Original Synagogue, 51398. 149. Bruneau, Les Isralites de Dlos, 49599; Binder, Into the Temple Courts, 31415. 150. Bruneau, Recherches sur les cultes de Dlos, pl. A; idem, Les Isralites de Dlos, 466; White, Delos Synagogue Revisited, 156. 151. For a description of the remains, see Plassart, Synagogue juive de Dlos, 2015; Bruneau, Re-

the pre-70 diaspora


The identication of the Delos building as a synagogue is based on a number of factors, some of secondaryif not negligibleimportance, others more consequential. The former category includes the following considerations: Jews were already living in Delos in the later Hellenistic period; the building was located close to the sea, as was customary in many communities; there was a large hall with benches; the marble chair was possibly a Seat of Moses; the building plan has some characteristics reminiscent of Galilean-type synagogues (e.g., a triportal entrance); the building faced east, toward Jerusalem; fragments of water basins were found; rosette decorations were used; and the cistern found there might have functioned as a miqveh. These above considerations, however, are not of equal weight; some are insignicant, others speculative or simply wrong. The most telling evidence with respect to the buildings identication as a synagogue comes from the inscriptions found in or near the building. Four were inscribed on column bases of votaries found in Rooms A and B,152 each mentioning Theos Hypsistos:
[ ] Zosas of Paras to Theos Hypsistos [gave this in fulllment of ] a vow. [ ] Laodice to Theos Hypsistos, who saved him from his inrmities, [gave this in fulllment of ] a vow. [ ] Lysimachus, on his own behalf, [made] a thanks-oering to Theos Hypsistos. [ ] To Hypsistos, a vow, Marcia.153

A fth inscription, found in a house nearby, mentions a proseuche: Agathocles and Lysimachus [have made a contribution] to the proseuche. 154 As noted above, both termsproseuche and theos hypsistoscould have been used in a pagan context, but this was not usually, or likely to be, the case.155 Thus, when combined with some of the ancillary considerations mentioned above, the identication of the building as a Jewish proseuche becomes compelling.156 The absence of Jewish symbols and a Torah shrine has rightfully been dismissed as irrelevant. There is no reason to assume that such an early synagogue building would have had them; places like Gamla, Masada, and Herodium did not.157
cherches sur les cultes de Dlos, 48184; White, Delos Synagogue Revisited, 14752; Binder, Into the Temple Courts, 297317. With regard to architectural matters relating to the synagogue, see the comprehensive and thorough treatment of Trmper, Oldest Original Synagogue, 51398. 152. There is speculation as to what these bases once carried: gurines? incense altars? 153. These inscriptions appear in Frey, CIJ, I, nos. 72730; Lifshitz, Donateurs et fondateurs, nos. 47; Bruneau, Recherches sur les cultes de Dlos, 484; White, Delos Synagogue Revisited, 139 n. 25; idem, Social Origins, 33840; Binder, Into the Temple Courts, 303. 154. Frey, CIJ, I, no. 726; Lifshitz, Donateurs et fondateurs, no. 3; White, Delos Synagogue Revisited, 140 n. 28. A sixth, very fragmentary inscription (Frey, CIJ, I, no. 731) contributes nothing in this regard. 155. See, for example, M. Stern, GLAJJ, II, 29495, 569. See also C. Roberts et al., Gild of Zeus, 5572. 156. See Hengel, Proseuche und Synagoge, 175; Robert, Inscriptions grecques de Sid, 44. 157. See above, Chap. 3; and the summaries of Bruneau, Goodenough, and Kraabel cited above, note 146.


historical development of the synagogue

The likelihood that this building was, in fact, a proseuche is signicantly enhanced by the discovery and publication in the early 1980s of two inscriptions found about one hundred meters to the north. Inscribed on marble stelae, they reveal the existence of a Samaritan community in the third and second centuries b.c.e. Calling themselves the Israelites on Delos, who make oerings to the sacred Mount Gerizim (lit., Argarizein), these Samaritans honored several benefactors of their community:
[ ] The Israelites on Delos, who make oerings on hallowed Argarizein, crown with a gold crown Sarapion, son of Jason, of Knossos, for his benefactions toward them. [ ] The Israelites [on Delos], who make oerings to hallowed, consecrated Argarizein, honor Menippos son of Artemidoros, of Herakleion, both himself and his descendants, for having constructed and dedicated at their expense the proseuche of God . . . and they crowned him with a gold crown and [. . .].158

It is not clear whether these benefactors were fellow-Samaritans honored by their coreligionists on Delos or prominent pagans who had contributed to the local Samaritan community.159 In the rst inscription, the donation of Sarapion of Knossos is unspecied; in the latter, Menippos of Herakleion built the proseuche and perhaps other parts of the complex as well, though this part of the inscription is mutilated. If, indeed, these men were pagans, then the reason for their benefaction is unclear: Were they maintaining political or economic ties with the Samaritans? Or perhaps they had other kinds of bonds (familial?) with members of this community? The fact that pagans would contribute to synagogues is not, in and of itself, unusual. We have already come across one such example (e.g., Capernaum according to Luke 7:5), and others will follow below. Interestingly, the Samaritans, like numerous Diaspora Jewish communities, chose to honor those benefactors with crowns or wreaths as well as public inscriptions. No less important for our purposes is the mention of a Samaritan proseuche. This would
158. For interpretations of the proseuche of God, see Bruneau (for a vow Les Isralites de Dlos, 488) and White (to the proseucheDelos Synagogue Revisited, 14244). I have adopted Whites interpretation. These inscriptions were rst published in Bruneau, Les Isralites de Dlos, 46775. See also Kraabel, New Evidence, 33132; idem, Synagoga Caeca, 22024; White, Delos Synagogue Revisited, 141 44; idem, Social Origins, 34042; Llewelyn, New Documents, VIII, 14851. Use of the name Israelites by the Delian Samaritans is undoubtedly part of the same phenomenon as the appearance of the name Jeroboam on coins from Samaria at about this time (Meshorer, Ancient Jewish Coinage, I, 3132, 160). In both instances, the implicit claim is that the Samaritans are the successors of the true Israel, i.e., the northern kingdom. For a Jewish polemic against the Samaritans contrasting the virtuous woman of Judah with the compromised daughters of Israel, see C. Moore, Daniel, Esther and Jeremiah: Additions, 114. 159. Bruneau prefers to consider them Samaritans; Kraabel leaves the question open, though he leans toward a Samaritan identication; White concludes that they were, in fact, pagans (Bruneau, Les Isralites de Dlos, 481; Kraabel, New Evidence, 334; and esp. idem, Synagoga Caeca, 22223; White, Delos Synagogue Revisited, 142 n. 36, 144; and, for a more decisive view, idem, Building Gods House, 6667).

the pre-70 diaspora


seem to clinch the fact that the other Samaritan inscription refers to such a building as well. The fact that Samaritans lived in the vicinity may strengthen the likelihood that this residential area served Jews as well. In second-century Ptolemaic Egypt, both of these communities lived side by side and even shared the same charity funds.160 Whether these were two separate proseuchai (one Jewish and one Samaritan), one proseuche serving both communities at one and the same time, or a proseuche that served both communities in successive eras are all intriguingalbeit elusivepossibilities.161 The main reservation in identifying the Delos building as a synagoguebesides the absence of explicit evidenceis based on the lamps found there, some of which are decorated with pagan motifs. Of late, this evidence has been either ignored or dismissed. One might, indeed, argue that they were not found in situ and therefore may well have been deposited there at a later stage. Secondly, it was once assumed that Jews throughout antiquity were united in their opposition to gural representation. Over the past generation, however, this view has been dramatically modied by both the endless stream of archaeological nds and a more nuanced understanding of the literary sources. Jewish communities of Late Antiquity were far from monolithic in this regard, and there is no reason to assume that all Diaspora Jews in the late Hellenisticearly Roman periods were of one mind. Delos may very well reect a dierent cultural and artistic norm from that of Judaea or elsewhere in the Diaspora. Taken in conjunction with the statue bases from Egypt and the possible evidence from the synagogues of Berenice and Acmonia (see below), which included wall decorations, perhaps even depictions of animals and humans (not to mention the third-century c.e. Dura Europos synagogue), gural representations may have been quite acceptable, at least in some parts of the Diaspora.162 A signicant degree of hellenization within the local Jewish community is also attested by the language and prosopography evidenced in the above inscriptions. Like the Jews of Egypt and Cyrene, those of Delos used Greek exclusively (at least as far as our limited evidence indicates), and their Greek names likewise reect a hellenized cultural milieu. From Josephus we learn that many Delian Jews were also Roman citizens (see below).163 The building used by the Jews for communal purposes followed the pattern established by other Delian associations, whose premises often included a large porticoed courtyard and a lavish marble chair.164 The chair, similar to those reserved for priests in pagan tem-

160. Antiquities 13, 7479. 161. See Kant, Jewish Inscriptions, 7078 n. 9. 162. Lamps with pagan ornamentation were found in the Jewish catacombs of Rome; see Rutgers, Jews in Late Ancient Rome, 8588. 163. Antiquities 14, 23132. This phenomenon was so widespread that their exemption from military service became an issue for other Delians. 164. Mazur, Studies on Jewry, 1819. It should be noted that such models diered considerably in elaborateness, depending on the material circumstances of each group; see Bruneau, Recherches sur les cultes de Dlos; White, Building Gods House, 26.


historical development of the synagogue

ples, may be our earliest evidence of the Seat [cathedra] of Moses, a special chair or bench found in a number of synagogues of Late Antiquity and mentioned in both the New Testament and rabbinic literature.165 Finally, some information regarding the Delos community generally, and its synagogue indirectly, has been preserved among the Roman decrees relating to a number of Diaspora communities and included by Josephus in his Antiquities. Pertinent to our present discussion is the following:
Julius Gaius, Praetor, Consul of the Romans, to the magistrates, council and people of Parium,166 greeting. The Jews in Delos and some of the neighboring Jews, some of your envoys also being present, have appealed to me and declared that you are preventing them by statute from observing their national customs and sacred rites. Now it displeases me that such statutes should be made against our friends and allies and that they should be forbidden to live in accordance with their customs and to contribute money to common meals and sacred rites, for this they are not forbidden to do even in Rome. For example, Gaius [Julius] Caesar, our consular praetor, by edict forbade religious societies to assemble in the city, but these people alone he did not forbid to do so or to collect contributions of money or to hold common meals. Similarly do I forbid other religious societies but permit these people alone to assemble and feast in accordance with their native customs and ordinances. And if you have made any statutes against our friends and allies, you will do well to revoke them because of their worthy deeds on our behalf and their goodwill toward us.167

The above document reveals a number of important matters. As noted, the Jews of Rome seem to have enjoyed extensive communal privileges; and Roman ocials expected this practice to be followed elsewhere as well.168 Secondly, the Delos Jewish community had apparently encountered considerable diculties with its neighbors. The Jews were forced to appeal to Rome because Delian authorities were perceived to be undermining Jewish communal life. Presumably, laws had been enacted to prevent the Jews from observing their traditions and collecting monies to nance activities such as common meals and sacred rites ( ). The severity of the matter is underscored by the simultaneous appearance in Rome of a counterdelegation that presented arguments against the Jews. In a second decree from about the same time, Delian ocials acknowledged (though not very happily, it seems) that Jews were to be exempt from military service by orders of a Roman ocial.169 Privileges of this sort could not but have had a souring eect on the
165. See below, Chap. 9. 166. On suggestions for the identication of this site, see comments by Marcus in Josephus, Antiquities, LCL, VII, 561 n. f. 167. Antiquities 14, 21316. Pucci Ben Zeev (Who Wrote a Letter? 23743; eadem, Jewish Rights, 10718, and especially 115) dates this decree to 43 (possibly 42) b.c.e., and claims that it was authored by Octavian. On the many issues involved with this document, see Gruen, Diaspora, 9293 and n. 52. 168. See above, notes 139 and 140. 169. Josephus, Antiquities 14, 23132: Decree of the Delians. In the archonship of Boeotus, on the

the pre-70 diaspora


relations between the Jews and their neighbors, with the latter, in turn, attempting to reduce (if not eliminate) the Jews preferred status. In light of these documents, the location of the Delian Jewish community in a relatively isolated part of the island takes on additional signicance. Not only could it have served Jewish interests to be somewhat isolated, but it may already reect (or have subsequently contributed to) a degree of social alienation, and perhaps even hostility, between the Jewish and pagan residents on the island.


We are probably as well informed about the Jewish communities and their synagogues in Asia Minor and Greece as in any other part of the Diaspora.170 Not only does the pre-70 material begin to rival that of Egypt, but archaeological data (buildings and inscriptions) from this area supply extensive information about these communities in Late Antiquity as well. These two regions will be treated together since the sources of information are the same for both. Josephus and Acts supply the bulk of data for the pre-70 era, while practically all of the archaeological remains come from Late Antiquity alone. The material below is organized by source and not by individual locale.

Josephus has preserved a series of Imperial edicts relating to the rst century b.c.e., each rearming the rights and privileges of various Jewish communities in Asia Minor in the face of local opposition and hostility.171 Each edict presumes that the Jews were
twentieth day of the month of Thargelion, response of the magistrates. The legate Marcus Piso, when resident in our city, having been placed in charge of the recruiting of soldiers, summoned us and a considerable number of citizens and ordered that if there were any Jews who were Roman citizens, no one should bother them about military service, inasmuch as the consul Lucius Cornelius Lentulus had exempted the Jews from military service in consideration of their religious scruples. You must therefore obey the magistrate. Similar to this was the decree that the people of Sardis passed. See Pucci Ben Zeev, Jewish Rights, 16872. 170. On the numerous Jewish communities in this area, see, inter alia, Philo, Embassy, 245, 281; M. Stern, Jewish Diaspora, 14355; Trebilco, Jewish Communities, 536; Levinskaya, Book of Acts in Its Diaspora Setting, 13766; Lichtenberger, Organisationsformen, 2327. 171. Josephus, Antiquities 14, 21364; 16, 16073. On these edicts, see Schrer, History, III, 114.; Juster, Les Juifs, I, 391.; II, 127; La Piana, Foreign Groups, 34851; Smallwood, Jews under Roman Rule, 12043; Applebaum, Legal Status, 42063; Moehring, Acta pro Judaeis, 12458; Rajak, Was There a Roman Charter? 10723; eadem, Jewish Rights, 1935; Pucci Ben Zeev, Caesar and Jewish Law, 28 37; eadem, Jewish Rights, 3953; eadem, Jewish Rights, passim. These decrees stem from the basic recognition extended by Julius Caesar to the Jews in the Roman Empire, in no small part in gratitude for the support he received from Antipater, Hyrcanus II, and the Jews of Judaea in his struggle against Pompey ( Josephus, Antiquities 14, 21112). At the same time, these


historical development of the synagogue

well organized, having formed their own communal framework, and possessed their own social and religious mores (see below). Undoubtedly, this also meant the presence of a place of worship, i.e., a proseuche or synagogue, although, admittedly, in only a few of these edicts is such a place specically mentioned. It is to this evidence that we now turn. Josephus records a decree regarding the Jews of Halicarnassus,172 in which it is carefully noted that the community was assured the inviolability of their holy days and gatherings: We have also decreed that those men and women who so wish may observe their Sabbaths and perform their sacred rights in accordance with the Jewish laws, and many built places of prayer [ proseuchai ] 173 near the sea, in accordance with their native custom. Of particular interest here is the explicit statement that many Jews built synagogues near bodies of water, a phenomenon we have already encountered in Egypt, Delos, and Ostia.174 A similar reference appears in Acts.175 The reason for this practice is not entirely clear, although one obvious possibility is the need to be close to water for purication purposes, a practice already attested in the Letter of Aristeas.176 There may also have been other reasons for this preference, e.g., the Jews desire to distance themselves from the pagan city generally in order to avoid, or at least reduce, tensions with their neighbors stemming from their dierent practices and behavior, or to allow for a less polluted worship environment, far from pagan places of idolatry.177 Josephus has preserved several decrees regarding Sardis that are likewise pertinent to our discussion. One document explicitly notes that the Jews had a place () where they decided their aairs, being organized, as they were, in an association () governed by ancestral laws.178 In another decree, the Jews were granted both the right to orgaprivileges also continued the earlier Hellenistic tradition of religious and ethnic tolerance with regard to the Jews. Josephus makes this claim in an exaggerated fashion on several occasions; see his Antiquities 12, 11920; 16, 16061; and below. 172. Antiquities 14, 258. 173. E. P. Sanders, Jewish Law, 259, 341 n. 26; Pucci Ben Zeev, Jewish Rights, 20616. 174. See above; see also Lauterbach, Tashlik, 207.; Goldin, Studies in Midrash, 346. 175. Acts 16:13. See also Philo, Flaccus 12223. 176. Letter of Aristeas 3046. The idea that the sea was valid for purication purposes is reected in the following: All seas are valid as a miqveh . . . so [says] R. Meir. R. Judah says: The Great Sea is valid as a miqveh. . . . R. Yosi says: All seas render clean by virtue of being owing waters (M Miqvaot 5, 4); and And it happened that Rabban Gamaliel and Onqelos the Proselyte would come to Ashkelon, and Rabban Gamaliel immersed himself in a bath and Onqelos the Proselyte in the sea. R. Joshua b. Qabusai said: I was with them and Rabban Gamaliel immersed himself in the sea (T Miqvaot 6, 3 [p. 658]). Regarding the waters of a river, see T Makhshirin 2, 12 (p. 674); T Miqvaot 4, 5 (p. 656). See also E. P. Sanders, Judaism, 22324; Pucci Ben Zeev, Jewish Rights, 21516. 177. See Mekhilta of R. Ishmael, Bo, 1 (p. 2); and Sukenik, Ancient Synagogues, 49., where being near water is associated with the rabbinic concept of impurity of gentile lands. See also Alon, Jews, Judaism in the Classical World, 14689. 178. Antiquities 14, 235.

the pre-70 diaspora


nize their communal life, including self-adjudication, and a place () for the practice of their ancestral customs, which included prayer and sacrice.179 A general edict of Augustus to the Jews of Asia Minor from 12 b.c.e. is of especial importance: And if anyone is caught stealing sacred books or sacred monies from a Sabbath-house [] or a banquet hall [], he shall be regarded as sacrilegious, and his property shall be conscated to the public treasury of the Romans. 180 Many, if not most, of the synagogues had a banquet hall in addition to a place for Sabbath assembly. Sacred communal meals are documented in a number of rst-century Jewish contexts (Pharisaic avurot, Essenes [Qumran], Therapeutae, and the early church), mostly in connection with religious associations. However, it would appear that such communal meals were not foreign to the wider Jewish community, as reected in the decree regarding Delian Jews.181 Such banquet halls could have served other functions as well, and smaller facilities may have consisted of a single room utilized for several purposes. Finally, the reference to a synagogue as a repository for sacred books and sacred monies is invaluable. These monies were probably donated for local use as well as for the Jerusalem Temple, a practice that, as we have seen, was widespread among rst-century Diaspora communities.182

New Testament
As was the case with the Second Temple Judaean synagogue, the New Testament has preserved invaluable material relating to the Diaspora synagogue in Asia Minor and Greece. The account in Acts of Pauls journeys attests to the density of Jewish settlement and the development of Christianity throughout the eastern Mediterranean in general and in Asia Minor and Greece in particular. In all, Acts mentions the synagogue nineteen times, almost always referring to the institution as synagogue, with but one exception, Philippi, where the term proseuche is used. Acts informs us of synagogues in Damascus
179. Ibid., 260: they may, in accordance with their accepted custom, come together and have a communal life and adjudicate suits among themselves, and that a place be given them in which they may gather together with their wives and children and oer their ancestral prayers and sacrices to God. The reference to Jewish sacrices is enigmatic; see below, Chap. 5, note 33. 180. Josephus, Antiquities 16, 164. See M. Stern, Jewish Diaspora, 146. On a similar term in Syriac, , see Payne Smith, Thesaurus Syriacus, I, col. 497. There is also a reference to a sabbateion in an early second-century inscription from Thyatira in Asia Minor, where a synagogue is probably intended (Frey, CIJ, II, no. 752; van der Horst, Ancient Jewish Epitaphs, 15051). For a reference to a house of the Sabbath in Rabbat Moab that was destroyed by Barsauma between 419 and 422 c.e., see Nau, Deux pisodes, 188. On the andron generally, see C. Roberts et al., Gild of Zeus, 4748. 181. See above, note 167. 182. Antiquities 14, 21316. Regarding the sending of money to Jerusalem by the Jews of Ephesus, see Philo, Embassy 315. On the Jerusalem church and contributions from various diaspora churches, see, inter alia, Rom. 15:2526; I Cor. 16:14; Meeks, First Urban Christians, 110.


historical development of the synagogue

(9:2, 20), Salamis (13:5), Antioch of Pisidia (13:14), Iconium (14:1), Thessalonica (17:1), Berea (17:10), Athens (17:17), Corinth (18:7, 8), Ephesus (18:19.), and Philippi (16:13). The frequent reference to synagogues in Acts is not fortuitous. According to Luke (author of Acts), this institution was a critical factor in the spread of Christianity in its early stages. Almost every reference to a synagogue is related to Pauls missionary activity;183 at rst he addresses the Jews and only later the gentiles. The pattern appearing in Acts is almost inexorable: visit to a synagogue, eective preaching, Jewish hostility, and expulsion.184 This recurrent phenomenon goes to the heart of Acts theological and political message. Paul is rebued time and again by the Jews, and only then devotes himself fully and unequivocally to the gentile mission. The theological basis of this schema is clearly spelled out in Acts 13:46: And Paul and Barnabas spoke out boldly, saying, It was necessary that the word of God should be spoken rst to you. Since you thrust it from yourselves, thereby judging yourselves unworthy of eternal life, behold, we turn to the gentiles. 185 Much has been written about the historical reliability of Actsfrom the more skeptical to the largely accepting. Theological agendas aside, one may assume that the specic events reported, especially those relating to the synagogue, are largely credible. The author was certainly familiar with the Jewish Diaspora and wrote for Christian Diaspora communities. It is hard to imagine that he would invent accounts for a population that knew a great deal about the synagogue, its workings, and Pauls activities. At the very least, even were one to doubt the specic details included in Acts, one would have to admit that such events could well have taken place, even if not precisely in the manner recorded. Many interesting and important details regarding the synagogue emerge from the accounts in Acts. For instance, the antiquity of the custom of reading Scriptures in the synagogue on the Sabbath is considered here, as in other sources, to derive from Moses himself (Acts 15:21). In fact, it is Sabbath worship that regularly provides the setting for Pauls encounters with the local community (e.g., Acts 13:42; 16:13; 17:2; 18:4). Clearly, the Sabbath was the primary occasion for the community to congregate, particularly in a worship context.186 As noted, this phenomenon may be alluded to in two Berenice inscrip183. The one exception is the reference to an Alexandrian Jew named Apollos, who spoke in the synagogue of Ephesus (Acts 18:26). 184. So, for example, Acts 9:2022; 13:4448; 14:16; 14:19; 17:19, 16.; 18:1821. 185. See also Kee, Jews in Acts, 18395. Despite a rather skeptical approach to the historical validity of many New Testament traditions concerning the synagogue, McKay (Sabbath and Synagogue, 16571) seems to accept most of Acts accounts as reliably reecting the synagogues that Luke knew. See also A. F. Segal, Paul, 26773. 186. Regarding the Christians of Berea, Acts notes that some Jews would gather daily to study Scriptures, particularly in response to Pauls message. Clearly, the intent here is to describe an emerging Christian community whose fervor and commitment led to daily study, reminiscent of the practice among

the pre-70 diaspora


tions, where the regular (as opposed to monthly) gathering seems to refer to the Sabbath, as well as in Josephus reference to the of Asia Minor Jewry.187 While most cities appear to have had one synagogue, the plural synagogues is used on several occasions, probably reecting a large local Jewish population that required more than one building.188 The only synagogue ocial specically named in these accounts is the archisynagogos. In one instance, this ocial invited Paul to speak to the congregation in Antioch of Pisidia following the reading of the Torah and Prophets; clearly the position entailed a degree of responsibility and authority.189 In a second instance, Acts notes that Crispus, the archisynagogue of the Corinthian synagogue, became a believer (18:8), and it is not surprising that his whole household followed suit. However, the fact that immediately afterwards many other Corinthians, having heard Paul, began to believe in Jesus and were then baptized may attest to the impact of Crispus conversion on his fellow-citizens and thus to his prominence in the community generally. Two synagogue scenes described in Acts are especially noteworthy. One (16:1213) has to do with Pauls rst encounter with the Jews of Philippi in Macedonia. He came to the city in midweek and waited several days for the Sabbath and his rst encounter with the local Jewish community. He then went to the riverside, where we supposed there was a place of prayer [ proseuche], 190 and there met a group of women. The presence of women, some of high standing, in Pauls audience at Berea is also noted.191 Acts singles out gentiles, both men and women, as also having frequented Diaspora synagogues.192 The attraction of many pagans to Judaism in antiquity is well documented.193 Although the nature of this missionary phenomenon has been vigorously debated of late, the issues in dispute concern its extent in the rst century and the degree of active missionizing on the part of the Jews.194 What is to be noted in the present convarious sects such as the Essenes, Therapeutae, and, undoubtedly, the Pharisees. The author of Acts describes these Jews as more noble than those of nearby Thessalonica (Acts 17:11). 187. Antiquities 16, 164. 188. Acts 9:2, 20; 13:5. 189. Ibid., 13:15. See Chap. 5. 190. Alternatively: where prayer was carried out. 191. Acts 17:12. The special attraction of women to Judaism is attested elsewhere as well (e.g., War 2, 560); see below, Chap. 14. 192. So, for example, Acts 13:43; 14:12; 17:4, 12; 18:4; as well as Josephus, War 7, 45. See also ibid., 2, 463; idem, Against Apion 2, 123; and Trebilco, Jewish Communities, 14566, esp. 16466. 193. So, for example, Josephus, Antiquities 14, 110; see Schrer, History, III, 15076; Juster, Les Juifs, I, 253337; M. Stern, Sympathy for the Jews, 15567 (= Studies, 50517); Feldman, Proselytes and Sympathizers, 265305; idem, Jew and Gentile, 177445; McKnight, Light among the Gentiles; Figueras, Epigraphic Evidence, 194206. 194. See, for example, S. J. D. Cohen, Was Judaism in Antiquity a Missionary Religion? 1423 (cf. Feldmans very dierent emphasis in the same volume, Was Judaism a Missionary Religion in Ancient


historical development of the synagogue

text is the central role played by the synagogue in this regard. It was not only the place where Judaism was most visible, thus drawing the sympathetic and curious, but, once they were won over, it became a focus of identication and aliation. These gentiles, of many stripes and referred to by dierent names, may have played a prominent role in some Diaspora synagogues (see below, Bosphorus), and this appears to have been the case in Late Antiquity as well.195 In fact, the status enjoyed by many Diaspora communities may have been due, at least in part, to the presence and support of a large number of pagan sympathizers. The second account of importance described in Acts (13:15) oers us a eeting glimpse at the Sabbath-morning liturgy in the Antioch of Pisidia synagogue. Four elements are featured in this schema: a selection from the Torah is recited; then a selection from the Prophets is read; the archisynagogue invites Paul to speak; and Paul addresses the congregation. This order of events generally parallels Lukes earlier description of the synagogue service at Nazareth.196 Perhaps the remarkable fact in this account is the receptivity of the local community to the participation of outsiders. Pauls appearance in Antioch was unannounced; he was, for all intents and purposes, a stranger. Nevertheless, he was asked to address the congregation. How widespread this custom was is impossible to assess, although we may note that something similar happened at Ephesus. According to Acts 18:2426, an Alexandrian Jew named Apollos came to the local synagogue there and spoke eectively and fervently about Jesus from a distinctively non-Pauline perspective. Once again, the synagogue served as an open forum for Jews of dierent backgrounds and persuasions.

The Acmonia Inscription

The region of Asia Minor and Greece has left us with an unusually rich trove of epigraphical evidence. While the total number of Jewish inscriptions stands at about one
Times? 2437); Goodman, Jewish Proselytizing, 5378; idem, Mission and Conversion, 6090. For a more strident denial of the sebomenoi phenomenon, now seriously undermined by the publication of the Aphrodisias inscription, see MacLennan and Kraabel, God-Fearers, 4653; Kraabel, Disappearance of the God-Fearers, 11326. See also Gager, Jews, Gentiles, and Synagogues, 9199; Feldman, Jew and Gentile, 34282. 195. On the gentiles and Diaspora synagogues, see S. J. D. Cohen, Respect for Judaism, 40930; idem, Crossing the Boundary, 1333; see also Trebilco, Jewish Communities, 14566, esp. 16466; Georgi, Opponents of Paul, 83117. Regarding Late Antiquity, see Wilken, John Chrysostom and the Jews, 6694; Meeks and Wilken, Jews and Christians, 83127; Smelik, John Chrysostoms Homilies. On the Aphrodisias inscription in this regard, see Reynolds and Tannenbaum, Jews and God-Fearers, 4392. See below, Chap. 8. 196. Luke 4:1621. It has been claimed that this similarity indicates Lukes projection of the synagogue liturgy he knew from the late rst-century Diaspora onto Jesus Galilee; however, such skepticism seems unwarranted. See my comments in this regard, above, Chap. 3.

the pre-70 diaspora


hundred, the inscriptions relating to the synagogue or its ocials comprise almost half this number. Most are dedicatory inscriptions from synagogue buildings; a few are epitaphs that mention a synagogue aliation. Although practically all these inscriptions date from Late Antiquity, one of the most important among them comes from rst-century c.e. Acmonia. Located inland in Phrygia, this city has an importance in large measure due to its strategic position on the Persian Royal Road. The inscription reads as follows:
The edice was constructed by Julia Severa. Publius Tyrronios Clados, archisynagogos for life, Lucius son of Lucius, archisynagogos, and Popilios Zoticos, archon, have renovated [the building] from their own funds and from the community treasury. They decorated the walls and the ceiling, and they made the windows secure and [made] all the rest of the decoration. The synagogue honors these individuals with a gold shield on account of their excellent leadership and their kindly feelings toward and zeal for the congregation.197

The items of interest here are manifold. Most striking, of course, is the fact that the synagogue building itself was built by one Julia Severa a number of years prior to the date of this inscription, which itself deals with the restoration of the structure. Even more unusual than the nature of this womans benefaction is the fact that she was a well-known pagan who came from a nexus of leading families. The local coinage celebrates Julia Severa as politically active in the mid-rst century, holding the positions of agonothete and (high priestess) of the local Imperial cult.198 Pagan donations to synagogues are known elsewhere in Asia Minor as well, but donating an entire building was indeed
197. The translation of this inscription has been adapted from White, Social Origins, 30810, and Trebilco, Jewish Communities, 5859. See also Frey, CIJ, II, no. 766 (= MAMA, VI, no. 264); Lifshitz, Donateurs et fondateurs, no. 33. A major issue here is whether Julia Severa donated a building to the Jews for use as a synagogue or simply built an edice that later was transferred to, or bought by, the Jewish community. The term (lit., house or building) has been used in both ways (see the Phocaea and Stobi inscriptions in White, Social Origins, nos. 68, 73). From the inscription here, it seems most likely that a synagogue building was intended. Why else should the fact that Julia erected a building be mentioned at all in this context? If one assumes that the building was originally earmarked for some other purpose, and only later was given or sold to the Jews, why is this fact not noted at all? It seems very plausible that the edice had been intended from the outset as a gift of a synagogue to the Jewish community and only later was renovated by the three named leaders. See the comments of Rajak, Synagogue within the Greco-Roman City, 16173, as well as Matthews, Ladies Aid, 199218; Walton, Oriental Senators in the Service of Rome, 4445. 198. Levick, Roman Colonies, 107. Julias husband, Lucius Servenius Capito, was a decurion in Acmonia in the time of Nero, and her son, L. Servenius Cornutus, held many oceshe was, for example, a senator under Nero and a legatus to the proconsul of Asia. For the numismatic evidence, see Ramsay, Cities and Bishoprics, 63839; see also Schrer, History, III, 31. On the positions Julia held in the Imperial cult, see Ramsay, Cities and Bishoprics, 639; Trebilco, Jewish Communities, 59.


historical development of the synagogue

rare.199 It was once suggested that Julia was, in fact, a Jewess; however, this claim has been controverted by her now well-documented pagan aliations.200 Some timeperhaps several decades after the initial contribution by Juliarepairs of the synagogue were undertaken by three leading ocials, two archisynagogoi and one archon.201 How the archon Popilios Zoticos was related to the synagogue is unknown. The two archisynagogoi, bearing Greek and Latin names, may not have held identical positions. One was head for life ( ); whether this was a purely honorary title following years of service or, indeed, a reection of a continuous term of oce is dicult to determine. The second one, Lucius, is simply noted as an archisynagogos. Whether his father, of the same name, also held this position is unknown; it will be remembered that retaining such an oce for generations within a single family is attested in the Theodotos inscription from rst-century Jerusalem.202 The funds used for restoration of the building appear to have been matching grants (whatever the relative percentages) from these three leaders and the community at large. The wall and ceiling paintings are noteworthy, though the nature of these paintings geometrical, oral, or gural motifsis unknown. Depending on the lavishness of the ornamentation, this synagogue may have been similar to those in contemporary Berenice, or Sardis and Dura Europos later on. The three major donors were honored in ways typical among Greek and Roman donors generally. Not only did they merit the above inscription, but they were awarded a gold shield.203

199. See below, Chap. 8. A parallel phenomenon from the fourth century may be the building of a proseuche by the Imperial governor in Panticapaeum (Levinskaya, Book of Acts in Its Diaspora Setting, 229 31). It is interesting to note that relations between the Jews and their neighbors in this region appear to have been particularly close; see Sheppard, Jews, Christians and Heretics, 16980; Crawford, Multiculturalism at Sardis, 3847. On this phenomenon generally, see M. Stern, Sympathy for the Jews, 15567 (= Studies, 50517). 200. See, for example, White, Social Origins, 3089; Ramsay, Cities and Bishoprics, 64851, 67375; and the somewhat ambiguous note in Juster, Juifs, I, 43031. On the opposite phenomenon, namely, Jews contributing to pagan shrines, see the examples cited in Barclay, Jews in the Mediterranean Diaspora, 32122. 201. A similar synagogue inscription, of undetermined date, was found in Olbia, along the shores of the Black Sea. It, too, speaks of synagogue ocials restoring the building: The society under the presidency of [? . . .] Pourthaios Achilleus son of Demetrius, Dionysodoros son of Eros [?], Zobeis son of Zobeis, the archons, have restored by their own care the synagogue from the foundations till . . . and have roofed it. See the translation and comments of Lifshitz, Prolegomenon, 64; idem, Donateurs et fondateurs, no. 11; Frey, CIJ, I, no. 682. 202. See above, Chap. 3. White (Social Origins, 30910 n. 48) has speculated that the synagogue ocials named Tyrronios and Lucius were Jewish freedmen (or their descendants) in the service of Julia Severa, and this would therefore explain her involvement with the Jewish community. 203. See Robert, Inscriptions grecques de Sid, 41 n. 1. On the phenomenon of such dedications in Asia Minor and some of their social and political implications, see Rogers, Gift and Society, 18899.

the pre-70 diaspora


13. The Bosphorus Kingdom.


Jews probably reached the Kingdom of Bosphorus on the northern shore of the Black Sea via Asia Minor owing to the political and economic ties between these two regions (g. 13).204 By the rst century c.e., with Bosphorus serving as a vassal kingdom of Rome, the presence of Jews as well as the inuence of Judaism are well attested. This religious inuence continued to grow until the fourth century at least, when the cult of Theos Hypsistos became one of the most popular in the region.205 Excavations now under way in Chersonesus in southwestern Crimea have uncovered remains of what may have been a synagogue from the rst century c.e. 206 Throughout the region, nine inscriptions dating from the rst and early second centuries c.e. and linked to the Jewish community have been found,207 each referring to the manumission of slaves. The proseuche is explic204. Minns, Scythians and Greeks, 62022. 205. Schrer, History, III, 3738; Goodenough, Bosphorus Inscriptions, 22145; Lifshitz, Jewish History in the Bosphorean Kingdom, 13033; Levinskaya, Book of Acts in Its Diaspora Setting, 10516; Levinskaya and Tokhtasyev, Jews and Jewish Names, 5573; Gibson, Jewish Manumission Inscriptions, 1529; as well as Finley, Black Sea and Danubian Regions, 5159. See also the comments in Nadel, Bosporan Jewish Manumissions Reconsidered, 45163. 206. MacLennan, In Search of the Jewish Diaspora, 4951; Edwards, Jews and Christians, 165. 207. According to Nadel (Slavery, 197233), ten out of twelve manumission inscriptions from the region relate to a Jewish setting.


historical development of the synagogue

itly mentioned on eight occasions, and it would seem that this procedure was regularly carried out there;208 in ve instances, the synagogue (here the term refers to the congregation) was appointed guardian.209 These inscriptions are unique not only because the ceremony took place within the proseuche, but also because the congregation per se was involved in the legal proceedings and the freed slave was committed to serving it in some capacity (see below). To date, three inscriptions, from 41, 59, and 67 c.e., have been discovered in Gorgippia (modern Anape). The rst of these translates as follows:
To the Most High God, Almighty, blessed, in the reign of the king Mithridates, the friend of [?] and the friend of the fatherland, in the year 338 [= 41 c.e.], in the month Deios, Pothos, the son of Strabo, dedicated to the proseuche, in accordance with the vow, his female slave [],210 whose name is Chrusa, on condition that she should be unharmed and unmolested by any of his heirs under Zeus, Ge, Helios.211

Noteworthy, in the rst place, is the threefold invocation of God in typical Jewish form ( ).212 The Greek names of the Jews, both father and son, should not surprise us, for elsewhere in the Diaspora, as we have seen, Jews, as a matter of course, adopted the nomenclature regnant in their surroundings.213 The fact that this and similar manumission ceremonies were performed in the synagogue is indeed unusual. Clearly, this was an act with religious as well as social implications, as the manumission formula itself attests. At rst there was some skepticism about the Jewishness of this text owing to the pagan formula summoning Jupiter, the earth, and the sun to witness the transaction.214 Such
208. Gibson, Jewish Manumission Inscriptions, 12452. Several other inscriptions do not specically note a proseuche (or Jews), but they may well refer to a Jewish context (ibid., 1972). 209. These inscriptions are to be found in one or more of the following collections: Frey, CIJ, I, nos. 683, 684, 690; Lifshitz, Prolegomenon, 6569; Levinskaya, Book of Acts in Its Diaspora Setting, 23142; and especially Gibson, Jewish Manumission Inscriptions, 12452, 15967. Following Gibsons presentation (ibid., 11213), we should probably also include two other inscriptions (CIRB 1125, 1126); see ibid., 124. For a manumission document from Roman Egypt in 291 c.e., see Tcherikover et al., CPJ, III, no. 473. In this case, Jewish slaves were freed and the Jewish community (the synagogue of the Jews) paid their ransom. See also Harrill, Manumission of Slaves, 17278. On the manumission of Jewish slaves at Delphi, see Frey, CIJ, I, nos. 70911; on the manumission of slaves in Roman Italy, see Dyson, Community and Society, 199203. With regard to Delphi, see Hopkins, Conquerors and Slaves, 13371. 210. See A. Cameron, and Related Terms, 2753. 211. CIRB 1123 in Gibson, Jewish Manumission Inscriptions, 99100. See also Frey, CIJ, I, no. 690, with some minor variations; Levinskaya, Book of Acts in Its Diaspora Setting, 23940; Overman, Jews, Slaves, and the Synagogue, 14951; and comments there of Edwards, Jews and Christians, 16366. On the term Highest God, see Gibson, Jewish Manumission Inscriptions, 10923. 212. See also Horbury and Noy, Jewish Inscriptions, no. 116, and comments, p. 200. 213. Lifshitz, Prolegomenon, 68. 214. Westermann, Slave Systems, 12426; Gibson, Jewish Manumission Inscriptions, 1057, 11921.

the pre-70 diaspora


usage should not be overly surprising, as similar formulas appear frequently in Bosphoran documents (and in other Jewish ones as wellsee below) and undoubtedly had become so common that they had lost all overtly pagan connotations.215 Moreover, in many other instances, from fth-century b.c.e. Elephantine through Josephus writings and down to third-century c.e. Bet Shearim and Hungary, we nd Jews utilizing formulas with distinctly pagan elements.216 Three other inscriptions from the rst and early second centuries come from the area surrounding the ancient city of Panticapaeum (near modern-day Kerch).217 While similar in many ways to the inscriptions from Gorgippia (i.e., with a manumission ceremony in the synagogue and a promise that the slave will not be reclaimed), they nevertheless display several unique features. To cite one example: I release in the proseuche, Elpias the son [?] of my slave, bred in my house; he shall remain undisturbed and unassailed by any of my heirs, except for [his duty] to serve the proseuche regularly; the community [synagoge] of the Jews and the God-fearers [?] will be guardian [of the enfranchised]. 218 We note here use of both proseuche and synagoge; the rst clearly refers to the building, as in other inscriptions from the region, the latter to the community. Unique to these Panticapaeum inscriptions are two stipulations: that the emancipated slave is to frequent the synagogue in the future and that the synagogue (i.e., the community) is to assume responsibility as a guardian () for the act of manumission. All of these elements appear in each of the Panticapaeum inscriptions, as well as in a more fragmentary one from Phanagoria, dating, perhaps, to 16 c.e. 219
215. Nock, Conversion, 63; Lieberman, Hellenism in Jewish Palestine, 214; Minns, Scythians and Greeks, 516, 616; Lifshitz, Jewish History in the Bosphorean Kingdom, 130; MacLennan, In Search of the Jewish Diaspora, 48; M. Stern, Jewish Diaspora, 156. 216. Schrer, History, III, 37; Lieberman, Hellenism in Jewish Palestine, 214; Schwabe and Lifshitz, Beth Shearim, II, no. 127; Levinskaya, Book of Acts in Its Diaspora Setting, 222; Scheiber, Jewish Inscriptions, 37. 217. Frey, CIJ, I, nos. 683, 684; Lifshitz, Prolegomenon, 6566; MacLennan, In Search of the Jewish Diaspora, 4447. 218. CIRB 71; Lifshitz, Prolegomenon, 66; Trebilco, Jewish Communities, 15556; Levinskaya, Book of Acts in Its Diaspora Setting, 7481; and especially Gibson, Jewish Manumission Inscriptions, 14044. Scholars have long debated the precise nature of the obligation of the freed slave to the community: Was it a religious duty (to attend services regularly or to convert), an economic obligation (to perform services for the institution), or a commitment of service (to be involved or maintain some sort of identication with it)? See Gibson, Jewish Manumission Inscriptions, 13750; Overman, Jews, Slaves, and the Synagogue, 15051; and Gruen, Diaspora, 3012 n. 33. On the status of the community in these legal proceedings, see Overman, Jews, Slaves, and the Synagogue, 15157. 219. Another Panticapaeum inscription, dating to 80 c.e., reads as follows: During the reign of King Tiberius Julius Rhescuporius, pious friend of Caesar and friend of the Romans, in the 377th year the 12th of Peretou, I Chreste, formerly wife of Drusus, release in the proseuche my slave Heraclas, a free person once and for all according to my vow; he shall remain untouched and undisturbed by any heir, as I have vowed. He may go where he wants without restraint except that he must adhere devoutly [Gibson, 160:


historical development of the synagogue

Of interest here as well is the possible reference to a group of God-fearers ( ) mentioned together with the Jews. If this reading is correct, it would indicate that God-fearers held a legally recognized position in the synagogue alongside the regular Jewish community, a presence even more institutionalized then than later on, in Late Antique Aphrodisias.220 Such a situation has far-reaching implications regarding these God-fearers numbers as well as their social and political standing.221

Despite its large size and relatively long history, the Jewish community of Syria generally, and of Antioch in particular, is only very partially known.222 Ironically, of all the Diaspora communities, this one oers us some important literary data from Late Antiquity (Chrysostom, Malalas, and several rabbinic traditionssee below, Chap. 8); for the pre-70 period, however, we must rely almost exclusively on Josephus and several traditions preserved by the sixth-century chronicler Malalas.223 Antioch and its kings played a central role in the Jewish aairs of Judaea during its Seleucid period, particularly in the
except for submissiveness and service] to the proseuche [see previous footnoteL.L.]. This agreement is made by my heirs Heraclides and Heliconias, and under the joint guardianship of the community of the Jews. See Frey, CIJ, I, no. 683; Gibson, Jewish Manumission Inscriptions, 160, as well as Levinskaya, Book of Acts in Its Diaspora Setting, 23132. On the fragmentary inscription from Phanagoria, see Frey, CIJ, I, no. 691; Lifshitz, Prolegomenon, 69; Levinskaya, Book of Acts in Its Diaspora Setting, 23637, where there is a slightly dierent rendition. Lifshitz, following Nadel, has suggested that these emancipated slaves may have been obligated to work (in the elds?) for the synagogues (Jewish History in the Bosphorean Kingdom, 128). 220. See Bellen, , 17176, as well as Lifshitz, Prolegomenon, 66. See also Figueras, Epigraphic Evidence, 2023. Note the Miletus inscription, which arguably may be read: A place for Jews and God-fearers (Frey, CIJ, II, no. 748); see below, Chap. 8. 221. Levinskaya (Book of Acts in Its Diaspora Setting, 7476) rejects this reading and suggests the following toward the end of the inscription: that he (the freed slave) works for the prayer-house under the guardianship of the Jewish community, and honours God. She thus assumes that there is no reference here to a class of God-fearers, although she herself accepts the fact that they existed in large numbers in the Bosphorus Kingdom at the time (ibid., 11316). Most scholars, however, have accepted the suggested reading. See, for example, Schrer, History, III, 16668; Reynolds and Tannenbaum, Jews and God-Fearers, 5456; Trebilco, Jewish Communities, 15556; and especially Overman, Jew, Slaves, and the Synagogue, 15556; Gibson, Jewish Manumission Inscriptions, 14044. 222. Perhaps the size and importance of this community inuenced Herods generous contribution to the city; see Josephus, War 1, 328, 425. 223. On Jews of Antioch in this period, see S. Krauss, Antioche, 2749; Kraeling, Jewish Community at Antioch, 13060; Meeks and Wilken, Jews and Christians, 213; M. Stern, Jewish Diaspora, 137 42; Levinskaya, Book of Acts in Its Diaspora Setting, 12735. On Jewish-Christian relations in rst-century Antioch, see Hahn, Judaism and Jewish Christianity, 34166. See also Downey, History of Antioch.

the pre-70 diaspora


century following Jasons Hellenistic reforms in Jerusalem (ca. 17575 b.c.e.),224 and it is in this context that some information is forthcoming. Josephus writes the following: For, although Antiochus surnamed Epiphanes sacked Jerusalem and plundered the Temple, his successors to the throne restored to the Jews of Antioch all such votive oerings as were made of brass, to be laid up in their synagogue, and, moreover, granted them citizen rights on an equal footing with the Greeks. Continuing to receive similar treatment from later monarchs, the Jewish colony grew in numbers, and their richly designed and costly oerings formed a splendid ornament to the holy place [ ]. Moreover, they were constantly attracting to their religious ceremonies multitudes of Greeks, and these they had in some measure incorporated with themselves. 225 Given Josephus claim that Jews lived in Antioch from its very foundation under Seleucus I or, at the latest, from the reign of Antiochus III,226 the existence of a synagogue in this community at an early date is more than likely. In the above source, Josephus claims that one of Antiochus IVs successors gave the citys synagogue all the brass ornaments and gifts that had been plundered from the Jerusalem Temple. If this is true, then this must have occurred between 163 and 65 b.c.e., i.e., after the death of Antiochus IV and prior to the Roman conquest of Syria by Pompey.227 If Josephus is correct that later Seleucid kings showered further benefactions on the Jews and their synagogue, then the original gifts must have been granted early on in this period, probably in the mid-second century, under Alexander Balas or Demetrius, when relations with the Jews of Judaea had improved considerably. The above source is also noteworthy in what it tells us about the centrality and prominence of this synagogue; for whatever reason, when a Seleucid king wished to restore the Temple treasures to the Jewish people, he turned to the Antioch synagogue, and not to Jerusalem itself. Thus, the term undoubtedly refers to the local synagogue and not, as is sometimes assumed, to the Jerusalem Temple. As a result of these gifts, and perhaps also the local communitys contributions, the Antioch synagogue became quite ornate and lavish, perhaps rivaling that of Alexandria as described by Philo and the Tosefta (see above). As in Asia Minor and Greece, Greeks in Syria were drawn to Judaism in signicant numbers. After describing the grandeur of the local synagogue, Josephus notes the at224. See, for example, II Macc. 4, 4.; I Macc. 11, 4244; Josephus, Antiquities 13, 377., 387. 225. War 7, 4445. On the tension between the Syrian pagan and Jewish populations throughout this period, see Roth-Gerson, Anti-Semitism in Syria, 30121. 226. Antiquities 12, 119; Against Apion 2, 39 (Seleucus I). Cf., however, Josephus statement in War 7, 4445 (Antiochus IV). See also Kraeling, Jewish Community at Antioch, 13839. 227. See, however, comments by Downey (History of Antioch, 109), who has doubts regarding the reliability of this report.


historical development of the synagogue

traction of Jewish religious ceremonies for Antiochan gentiles. He seems to allude to the fact that many of these Greeks identied with Judaism, but either they had not fully committed themselves to the religion or the Jews had not fully accepted them: the Jews had in some measure incorporated [these Greeks] with themselves. 228 Malalas reports the existence of a tomb of the Maccabean martyrs in Antioch;229 a medieval source (Nissim ibn Shahin of Qairuon) notes the existence of a ashmunit synagogue in the city that was later converted into a church.230 The synagogue was allegedly named after the mother of the Maccabean martyrs.231 Finally, Malalas reports that, during the rst Jewish revolt, Vespasian had destroyed a synagogue at Daphne and replaced it with a theater: He built the theater of Daphne, inscribing on it, ex praeda Iudaea (from the spoils of Judaea). The site of the theater had formerly been a Jewish synagogue. 232 The report, at rst blush, appears strange indeed. Interestingly, Malalas reports a similar instance in Caesarea: Vespasian also built in Caesarea in Palestine out of the spoils from Judaea a very large odeum, the size of a large theater; its site, too, had formerly been that of a Jewish synagogue. 233 Might we have in these two reports evidence for Roman practice in several places following their successful resolution of the Jewish revolt of 66? We have additional information regarding Syrian synagogues from Damascus. Acts (9:12) speaks of the existence of a budding Christian community within the citys synagogues. Paul was incensed by what he considered a perversion and gained permission to bring these Christians to Jerusalem. Following his conversion, he preached in these very synagogues (9:2023). Moreover, the synagogues of Damascus, as in Antioch and elsewhere, were known for attracting pagan sympathizers and converts. Of especial noteworthiness in Josephus account is the fact that women were singled out as being particularly prominent among these Damascan pagans.234
228. War 7, 45. For a strikingly similar phenomenon in Antioch several centuries later, in the time of John Chrysostom, see below, Chap. 8. 229. Malalas, Chronicle 8, 2067 (pp. 1089). According to Malalas, Antiochus IV brought the Maccabean remains to Antioch, and a few years later Judah Maccabee (sic) requested and received the bones from Demetrius; he then proceeded to bury them near a synagogue in the Kerateion section of Antioch. See Obermann, Sepulchre of Maccabean Martyrs, 25065; Bickerman, Les Maccabes de Malalas, 6383; Simon, Recherches, 14753; Downey, History of Antioch, 11011. The source for the tradition of the Maccabean martyrs is II Macc. 6:187:42. See also Schatkin, Maccabean Martyrs, 97113. 230. Kraeling, Jewish Community at Antioch, 140; Downey, History of Antioch, 544; Roth-Gerson, Jews of Syria, 24750. For a Samaritan tradition describing a synagogue built over the tomb of Baba Rabba in Constantinople, see Chronicle II, 25, 16 (p. 109). 231. See Nau, Un Martyrologe, 19, 52, 106, 123, 131 (PO 10.1). 232. Malalas, Chronicle 10, 261 (p. 138); Downey, History of Antioch, 2067. See also Malalas report (Chronicle 10) that Titus had erected a gate of cherubs in Antioch from the spoils of the Jerusalem Temple. 233. Malalas, Chronicle 10, 261 (p. 138). See also L. Levine, Roman Caesarea, 2526. 234. War 2, 560. As often happened, widespread attraction was matched by extensive hostility. This,

the pre-70 diaspora



As we have seen, the range of sources relating to pre-70 Diaspora synagogues is varied. Three major categories are representedliterary, archaeological, and epigraphicaland within each there are substantial dierences in the nature of the evidence and its historical value. The literary material, for example, ranges from references in Philonic religious monographs and historical accounts, to edicts cited by Josephus, and, nally, to Acts accounts of Pauls synagogue visitations. Similarly with regard to the epigraphical material: some inscriptions are major communal documents (Berenice) or shorter contracts of manumission (Bosphorus), others are brief statements of individual (Delos, Acmonia, Egypt) or communal (Egypt) contributions, while still others are epitaphs noting synagogue aliation (Rome). Of the two building remains, one dates to the rst centuries b.c.e. and c.e. (Delos), while the other is a fourth-century c.e. structure that originated in the rst or second century c.e. (Ostia). Each Diaspora community discussed above has left us a dierent variety of sources. Epigraphical evidence is our sole source for Cyrene and the Bosphorus kingdom; literary and epigraphical testimony is available for Egypt, Asia Minor, Rome, and Greece; literary evidence for Syria; and archaeological, epigraphical, and limited literary material for Delos and Italy. Owing to this broad range of primary material at our disposal, a number of features common to the pre-70 Diaspora synagogue or proseuche become quite evident. All sources are in agreement regarding the centrality of this institution among Jewish communities throughout the Roman world. As in Judaea, this point is driven home by the fact that no other Jewish communal institution or building is ever noted in these sources. And while a number of these communities boasted larger communal frameworks, such as a politeuma or gerousia (Berenice, Alexandria), no specic place is ever mentioned as housing these bodies; it therefore seems safe to assume that they met in the synagogue. Each locale may express the centrality of the synagogue in a dierent way, given the particular medium, the subject at hand, and the local context, but the implication is always the same. The synagogue was the Jewish communal institution par excellence everywhere.235 As the focus of the local community, the synagogue was geared to fulll its needs and serve as a setting for all facets of communal life. For those seeking to preserve their Jewish identityand most Jews in antiquity wished to do so, however they might dene that task such an institution was a sine qua non. Moreover, it served to distinguish them from the surrounding society. Familial and ethnic ties were deepened by common historical roots and memories, enhanced by the special ties to Jerusalem and its Temple, strengthened by
too, found expression in Damascus at the outbreak of the revolt in 66 c.e. See ibid., 2, 25961; 7, 368; Josephus, Life 27. 235. See, however, the reservations of Rajak (Synagogue and Community, 2238), who prefers to emphasize structural variety in this regard. I do not nd her arguments or reservations persuasive.


historical development of the synagogue

a network of customs and ceremonies, and bolstered by a set of beliefs that at times contrasted sharply with those of their neighbors.236 The Diaspora synagogue paralleled the contemporary Judaean one in its myriad functions while, at the same time, sharing many characteristics of non-Jewish institutions as well. It served the multiple purposes that a religious and ethnic minority such as the Jews would have neededreligious, educational, social, political, and economic. Nevertheless, although linked by a distinct (though not always easily dened) religious and ethnic heritage, these Diaspora communities reected a striking degree of diversity. The synagogues various names may indeed reect diverse perceptions of just what this institution was and how it was to function within the community. The most widespread terms, proseuche and synagoge, have been noted and may well point to varying emphases in each, at least in their formative stages.237 However, as we have already had occasion to note, other terms were being used in the rst century: (holy place), (place of prayer), (Sabbath meeting-place), and (place of instruction).238 More unusual terms, such as amphitheater and templum, were also invoked, as was the word .239 The synagogues of Romesome early, others lateare of a unique order, having been named after famous people, professions, or places of origin.240 A striking example of the unity and diversity among these synagogues may be found in even the small amount of archaeological evidence hailing from this period, i.e., the buildings of Delos and Ostia. On the one hand, both structures were close to the sea, far from the city center; both were oriented (in some fashion) toward Jerusalem and exhibited certain similarities to other structures in the vicinity. On the other hand, each had it own unique stamp, both vis--vis one another and with regard to their surroundings. The Ostia synagogue plan, for example, is far dierent from that of Delos. Moreover, the types of inscriptions and artistic representations found at Delos are not found at Ostia. Delian inscriptions feature the term theos hypsistos and have blatantly pagan symbols on the lamps, whereas an Ostia inscription notes the welfare of the emperor, and the artistic

236. Gafni, Punishment, Blessing or Mission, 22950. 237. L. Levine, Second Temple Synagogue, 1314. The situation may be somewhat analogous to the contemporary American Jewish scene, where a synagogue might be called a temple, synagogue, shul, or community center. While to most Jews there may be little or no dierence between these various names, they can reect very dierent ideological and functional notions about the institution in the eyes of the founders and those it served in subsequent stages. 238. : Josephus, War 7, 4445; III Macc. 2:28. : Tcherikover et al., CPJ, II, 223. : Josephus, Antiquities 16, 164; Tcherikover et al., CPJ, III, 46. See also S. Krauss, Synagogale Altertmer, 2627; and above, note 180. : Philo, Special Laws 2, 62. 239. Amphitheater: see above, notes 10815. Templum: Tacitus, Hist. 5, 5, 4; see M. Stern, GLAJJ, II, 43. : see above, note 225, as well as later examples; cf. Lifshitz, Donateurs et fondateurs, nos. 13, 21, 22, 61. 240. Leon, Jews of Ancient Rome, 13566.

the pre-70 diaspora


expression there is almost nonexistent. Finally, of course, the names used in connection with these two buildingsproseuche and synagogueare quite dierent.241 One of the major reasons for this pronounced diversity stems from the fact that the Jews who established these early Diaspora communities had no set models of what a community center facility should look like. Furthermore, in the Diaspora, powerful forces impacted upon each community, resulting in numerous instances of adopting patterns of the wider culture. Years ago, Kraabel called attention to this phenomenon, and with the passage of time and new discoveries and studies, this perception has only been strengthened.242 The names used by members of the community often imitate those generally popular on the local scene. Julia Severas contribution of a synagogue building seems to reect a patronage unique to this part of Asia Minor; the organization and functioning of the Jewish politeuma in Berenice may well have derived, in part at least, from Cyrenian models;243 the type of building used by Delian Jews bore similarities with other buildings on that island; and the manumission decrees from Bosphorus, with their formulary components, are well known in that particular region. Studies focusing on the Egyptian synagogue further conrm this perception, highlighting the many links between the Jewish proseuche on the one hand and the surrounding Greco-Egyptian culture on the other. Such parallels include dedications on behalf of the ruling family, the proseuches status as a place of asylum, the names and functions of synagogue ocials, and various architectural components. Signicant local inuence points to another characteristic of these Diaspora synagoguesnamely, a high degree of hellenization. Diaspora synagogues employed Greek terms for their institutions and ocers, and they often referred to the God of Israel as the Greeks did to Zeus (i.e., theos hypsistos). They almost always wrote in Greek, bore Greek and, at times, Latin names, honored fellow-Jews and benevolent pagans with crowns, shields, woolen llets, and inscriptions, and built and decorated their buildings in ways customary in Hellenistic-Roman society. This assimilation of outside patterns came quite naturally, nurtured as it was within a heterogeneous society.244 Throughout the course of Jewish history, the more diverse and pluralistic a society, the greater its acceptance of the Jewish community within it. The Roman Empire provided such a multicultural setting,
241. That dierent Diaspora communities may have had dierent calendars, see S. Stern, Jewish Calendar Reckoning, 10716. 242. Kraabel, Diaspora Jews and Judaism, 25767 (= Social Systems, 7991). See also Price, Jewish Diaspora, 17677. More recently, White has carried this argument even further, suggesting that almost every Diaspora synagogue was a private home converted into a communal institution; see his Building Gods House, 6061, 64, 78; and esp. his Delos Synagogue Revisited, 13536. 243. See Lderitz, What Is the Politeuma? 21922. 244. On the multicultural, social, and religious heterogeneity of Ptolemaic Egypt, see, for example, Fraser, Ptolemaic Alexandria; Walbank, Hellenistic World, 99102; Samuel, From Athens to Alexandria, 10517.


historical development of the synagogue

as did Ptolemaic Egypt throughout much of the Hellenistic period. In the latter, as we have seen, the Jews dedicated building after building in honor of the reigning king and queen, as did their neighbors. Besides borrowing specic practices and architectural elements, some scholars have suggested far more profound connections between the synagogue and the Greco-Roman world. According to them, the Jews patterned the synagogue and its activities on outside models, the one most frequently mentioned in this regard being the Hellenistic-Roman private association. These associations ranged from the more ocially recognized politeuma and collegium to less-dened groupings (synodos, koinon, thiasos, communitas) that might have been based on common geographical origins, commercial interests, religious aliation, mutual aid, or dining and burial needs.245 The subject of Greco-Roman models inuencing the synagogue has been addressed of late in a collection of studies, Voluntary Associations in the Graeco-Roman World, and particularly in an article by Richardson.246 His claim, that the synagogue was dened as a collegium, is based on two considerations: (1) the use of the term in Roman documents referring to the Jewish community; (2) the fact that the synagogue functioned as a social and religious association, as did the collegium. Regarding his rst point, we might ask whether the use of this terminology by the Romans with reference to the synagogue was merely a question of convenience, whereby the Roman authorities merely used a term familiar to them without attempting to be exact and precise.247 In other words, did the term collegium actually reect the legal status of the synagogue? And, if so, did it apply to all synagogues throughout the Mediterranean or did the use of several dierent terms, as noted above, reect alternative models? Nevertheless, it must be remembered that the synagogue diered from the above Greco-Roman frameworks in many and signicant ways.248 The Jewish community operated with a far greater range of activities and rights than the ordinary collegium, and the Romans were far more tolerant of the Jewish communal framework than of others; to wit, collegia were often banned by the authorities while the Jewish community remained unaected.249 The Jews had the right to maintain their own courts, attend to their own food requirements, avoid worshipping the civic deities or appearing in court on Sabbaths
245. See Schrer, History, III, 87137; Waltzing, tude historique, passim; Juster, Les Juifs, I, 41324; La Piana, Foreign Groups, 34851; Smallwood, Jews under Roman Rule, 13338; Kraabel, Diaspora Jews and Judaism, 2326 (= Unity and Diversity, 5154); Meeks, First Urban Christians, 3436; Rabello, Legal Condition, 71920; Rajak, Was There a Roman Charter? 10723. 246. Richardson, Early Synagogues as Collegia, 90109; idem, Architectural Case, 90117. See also Guterman, Religious Toleration and Persecution, 13056. 247. As, for example, in noting the right of the Jews to oer sacrices in an edict to the Jewish community of SardisJosephus, Antiquities 14, 261. 248. See Gruen, Diaspora, 12123. 249. Smallwood, Jews under Roman Rule, 22430.

the pre-70 diaspora


and festivals; they were exempt from serving in the army, sent monies to Jerusalem, and conducted a wide range of communal aairs. In certain places, the Jewish community was recognized as a politeuma, a civic framework attested in Alexandria and Berenice. Most of the above rights and privileges were not applicable to contemporary collegia or thiasoi. Thus, the term collegia with regard to the synagogue seems to have been one of Roman convenience and not in any way reective of a specic legal framework. Certainly, the Jews never viewed their synagogue or community in this light, and this term never appears in any document or inscription ascribed to Jews.250 Flesher takes a very dierent approach in his search for Greco-Roman models of the synagogue, classifying the institution as a type of Greco-Roman temple:
The research discussed in this paper reveals that the Graeco-Romans saw the synagogue as belonging to the genus of temple, even though it was not a perfect t. This should not be surprising since scholars have long noted similarities between synagogues and GraecoRoman temples in terms of architecture, artwork, and activities practiced in the buildings. These similarities appear in matters we think of as specically Graeco-Roman as well as in matters we usually associate with Judaism but which were also practiced in Graeco-Roman temples. When analyzed in taxonomic terms, it becomes clear that the similarities between synagogues and Graeco-Roman temples are not random and ad hoc, but indicate that the synagogue belonged to the genus of Graeco-Roman temple. 251

According to Flesher, this is the way non-Jews and, consequently, Jews viewed the synagogue. Both institutions had many activities in common, including those that the synagogue clearly borrowed from the pagan temple. However, despite its apparent attractiveness, this theory presents certain problems. First and foremost is the fact that the pivotal activity in each institution was strikingly dierent, while each relegated to a secondary role (if at all) what was central to the other. For the pagan temple, it was the sacricial act under priestly auspices, for synagogues it was the reading of Scriptures that involved congregational participation. Sacrice was unknown in a synagogue setting (the problematic Sardis evidence aside), as was the public reading of a sacred text in a Greco-Roman temple. In any case, most of the activities that the synagogue had in common with the temple may have evolved independently; there is little reason to assume that the former was specically inuenced by the latter. Did the Diaspora synagogue need the pagan temple to learn about prayer ritual or about its use as a meeting place for communal councils and courts, the collection of donations, etc.? The few persuasive instances of inuence (e.g., manumissions and asylum) are so minimally attested for synagogues as to make any generalization in this direction risky.252 The interest of pagans in the synagogue is indicative of the institutions accessibility
250. See also my more detailed comments in First-Century c.e. Synagogue. 251. Flesher, Prolegomenon, 12153 (cited from pp. 12324). 252. See also my more detailed comments in First-Century c.e. Synagogue.


historical development of the synagogue

as well as importance and centrality in the Jewish community. Evidence for pagan sympathizers and converts has been noted throughout the Diaspora, and in many instances these people chose to be actively supportive of the local Jewish community. The Godfearers of Bosphorus are an interesting example of this, as are Julia Severas involvement and benefaction. Especially noteworthy in this regard is the presence of women among those attracted to Judaism, a phenomenon attested to throughout the Roman world and in a variety of sources.253 Acceptance by many elements in the wider society notwithstanding, the Jews of the Diaspora were continually seeking to gain and maintain their rights, which included conrmation of the status of their central institution. We have also noted numerous occasions on which the Jews were forced to seek ocial Roman validation of these rights in face of attacks and hostility from their neighbors. In these cases, their minority status and distinctive customs proved as irritating and intolerable to some as they were attractive to others. Such a reality may well lie behind much of Philos apologetics, as well as Josephus decision to include numerous edicts issued by the Romans on behalf of Diaspora communities. Such tensions may have played a role in fostering a sense of Jewish marginality, which found expression in the not-uncommon location of the synagogue (and presumably also the community itself ) on the outskirts of a city. Such was the case over a wide geographical area, as we have noted with regard to Delos, Ostia, Macedonia, and Egypt.254 This would seem to have been at least part of the reason why the Acmonian community gave prominence to Julia Severas gift yearsif not decadesafter the original donation, and why the Jews of Berenice honored Marcus Tittius. Let us pursue this last example a bit further. The synagogue inscriptions from Cyrene indeed reect the communitys integration of synchronic and diachronic dimensions. The award ceremony for a Roman ocial was conducted on the Sukkot holiday and included praise of the honoree and a gift of a wreath to be awarded at each meeting (Sabbath?) and New Moon. This award, memorialized on a stele erected in the synagogue, was the unanimous decision of the entire congregation. Jewish communal and religious dimensions come into play in these inscriptions. The ceremony was initiated by the congregation, which decided whom to honor and how. All the events mentioned were conducted on dates from the Jewish calendar: Sukkot, the New Moon, and perhaps the Sabbath. Nevertheless, together with these Jewish components, there were also very denite and discernible Greco-Roman inuences. The leadership bore ocial Greek titles, and the
253. See, for example, Josephus, War 2, 56061 (Damascus); Acts 13:50 (Antioch, Pisidia); ibid., 16:13 (Philippi, Macedonia); Martial, Epigrammata, IV, 4 (M. Stern, GLAJJ, I, 524). See also van der Horst, Ancient Jewish Epitaphs, 10913; and below, Chap. 14. 254. Even in Alexandria, the major Jewish residential area, Delta, was on the coast, in the northeastern section of the city.

the pre-70 diaspora


personal names listed were almost all Greek as well. The same holds true of the communal institutions noted (i.e., politeuma, amphitheater). Moreover, the forms of bestowing honor (a wreath and woolen llet, a public inscription) and the voting procedure (casting white or black stones) are likewise well-attested Greek practices.255 When all is said and done, the Diaspora synagogue was indeed a creative synthesis of Jewish tradition, the requirements of each community, and the inuence of the surrounding culture. Far from constituting an isolated and insulated minority, or the opposite on the threshold of full assimilationthe Jews succeeded in creating an institution that expressed and reected their needs both as individuals and as a community, and did so within the connes of the cultural and social contexts in which they found themselves. They borrowed, yet within limits; Ptolemaic proseuchai were not dedicated to Ptolemy but on behalf of the king. They honored the ruler as was customary in other dedicatory inscriptions to Greek and Egyptian deities at the time.256 The Jewish place of worship did not resemble the pagan sanctuary or any other place of sacrice; rather, it was a proseuche or synagoge, a place for Torah reading, prayer, and communal activity. Thus, for all its borrowing and diversity, the Jewish communal institution remained quintessentially Jewish. It served the Jewish community and housed its rites and observances, which were inuenced rst and foremostthough far from exclusivelyby a common Jewish past and present. The Jews had brought their own patria to the Diaspora, a cultural and religious tradition that pagans could either respect, resent, or ignore but of which the Jews themselves were proud. They were committed to honoring and perpetuating this heritage, and, for the most part, the surrounding world was supportive. To safeguard and transmit ones traditional customs was an undisputed value in Roman society,257 and on the communal level the synagogue was the main vehicle for achieving this goal. Many Jewish communities even regarded their synagogues as holy places. The very term proseuche may be indicative of this fact, but even more so are specic references to the synagogue as a place of asylum, a sacred precinct (Egypt), or a holy place (Antioch; Philos reference to Essenes).258 The manumission of slaves in the synagogues of the Bosphorus Kingdom may also indicate the degree of sanctity associated with these buildings. The intention seems to have been that this ceremony be carried out not only in the

255. The awarding of a wreath and llet bears further comment. This type of honor is well known in Greco-Roman society and is documented already in classical Greece. Cyrenian Jewry did not hesitate to appropriate this practice, but did so in its own way. Whereas the general practice was to perpetuate or acknowledge such awards on an annual basis, the Jews, for whatever reason, decided to recall this ceremony every month, and perhaps even once a week. 256. Fraser, Ptolemaic Alexandria, 22627, 282. See also Nock, Conversion, 6162. 257. See, for example, Antiquities 16, 44; 19, 290; Against Apion 2, 23235, and comments in A. Kasher, Josephus, Against Apion, II, 51921; and, generally, MacMullen, Paganism, 24. 258. See Goodman, Sacred Space, 46.


historical development of the synagogue

presence of the community, but also in the presence of the Jewish God, as was the case with regard to Apollo in Delphi.259 The reasons for this attribution of sanctity to some Diaspora synagogues are worthy of comment. It seems most likely that such status was an attempt on the part of Jewish communities to accord their synagogues and proseuchai the prestige enjoyed by temples throughout the Hellenistic and Roman worlds. In fact, it was precisely at this time that granting temples the title of sacred and inviolable ( ) increased dramatically and was viewed as a mark of high honor.260 Consequently, some Diaspora communities also adopted this status (whether formally or not), thus enhancing, in their own eyes as well as in the eyes of others, the prestige of their main communal institution. The tenacity with which the Jews defended the integrity of this institution, its functions, and their rights generally is a reection of these commitments and loyalties. From within and without, attempts to undermine what they perceived as their fundamental interests and rights were met head-on. So, for example, when Paul was looked upon as threatening the status quo, he encountered erce resistance in Diaspora synagogues,261 and Jews frequently appealed to Rome in order to counter attempts by municipal authorities to undermine their status. The fact that Diaspora Jewry continued to thrive for centuries in many of these cities and regions attests to the overall success of its eorts.262
259. Frey, CIJ, I, nos. 70911; Gibson, Jewish Manumission Inscriptions, 3649, 13740, although, truth to tell, some pagan deities, but not the God of Israel, are mentioned in these documents. 260. Rigsby, Asylia, 129. 261. Acts 18:6 and esp. 19:9. 262. See, however, the reservations of Bohak (Ethnic Continuity, 18591), at least as regards the Egyptian chora.



y the rst century c.e., the synagogue was playing a pivotal institutional role within the Jewish communities of Judaea and the Diaspora.1 This centrality is reected in the wide range of activities that took place there. Though many rstcentury sources focus on particular events relating to a specic synagogue, severale.g., the Theodotos inscription, the New Testament, and a number of documents cited by Josephuslist a number of functions that transpired there. Rabbinic traditions speak of the synagogue as the venue for various educational and other activities, and while this material primarily reects the circumstances of Late Antiquity, some information is probably relevant for the rst century as well.2 The synagogue was also the logical setting in which a Jewish community would honor one of its members or a prominent non-Jew, often via a dedicatory inscription.3 A parallel of sorts exists between the range of functions within the synagogue and those that found expression in some contemporary pagan temples. Frequently surrounded by courtyards and ancillary rooms, these buildings or complexes might at times function

1. For recent comprehensive treatments of the rst-century synagogue in Judaea and the Diaspora, See Binder, Into the Temple Courts, passim; Claussen, Versammlung, passim; Runesson Origins, 169235. 2. See below, Chap. 10. On recent developments in the study of the social setting of early Christianity, see Barton, Communal Dimension, 399427. 3. As in the case of Decimus Valerius Dionysios at Berenice; see above, Chap. 4.


historical development of the synagogue

as libraries, markets, banks, and even as venues for study and learning. As a meeting place for collegia or sodalitates, the temple served some of the religious, social, political, and economic needs of members of these associations.4 The synagogues centrality was also recognized by the outside world. Hostility of pagans toward Jews was often vented through violent attacks on local synagogues.5 In contrast, when a ruler wished to honor or express support for a Jewish community, this, too, was often done within the context of the local synagogue. Thus, Seleucid rulers are reported as having donated spoils from the Jerusalem Temple to a synagogue in Antioch,6 and the many privileges enjoyed by Jewish communities throughout the Diaspora were invariably linked, directly or indirectly, to this institution. Jewish communities, for their part, often chose to honor kings and emperors in the synagogue, as, for example, the Egyptian proseuchai, which were dedicated to Ptolemaic rulers, and the Alexandrian synagogue, which displayed tributes to the emperor.7 Romes Jews named synagogues after Augustus and his viceroy, Agrippa; the Jews of Berenice honored a Roman ocial in their amphitheater at regular meetings; and the Jewish community of Acmonia recalled the benefactions of the pagan noblewoman Julia Severa.8 Moreover, the fact that these various activities are documented for synagogues and proseuchai throughout the Empire argues for a basic similarity between the roles of both institutions at this time. If, as we have argued, these institutions served, rst and foremost, the needs of a community, then it is most likely that such needsbe they of an economic, social, political, or religious naturedid not dier all that signicantly from Judaea to the Diaspora, nor among the various communities in Egypt, Asia Minor, or Rome. Some degree of commonality among rst-century synagogues is evident, for example, in the leadership positions within this institution. While information is severely limited owing to the scarcity of epigraphical evidence, two kinds of leaders appear to have been especially prominent in the rst century: the priest and the archisynagogos.9 The Theo4. Stambaugh, Functions of Roman Temples, 58091. See also C. Roberts et al., Gild of Zeus, 76 79, 83; Meeks, First Urban Christians, 3132. According to Flesher, the synagogue was, in fact, patterned after the Greco-Roman temple; see above, Chap. 4. 5. Josephus, Antiquities 19, 299305; idem, War 2, 28592; Philo, Embassy, 13234. On the nature and extent of anti-Jewish sentiments in antiquity, see Daniel, Antisemitism, 4565; Gager, Origins of AntiSemitism, 35112. 6. Josephus, War 7, 44. In the early third century c.e., an emperor referred to in rabbinic literature as Antoninus (in all probability, Caracalla, 211217 c.e.) allegedly donated a menorah to a synagogue; see Y Megillah 3, 2, 74a. 7. Horbury and Noy, Jewish Inscriptions, nos. 22, 24, 25, 27, 28, 117; Philo, Embassy 133. 8. Leon, Jews of Ancient Rome, 14042. The Roman ocial was Marcus Tittius, son of Sextus Aemilia; see above, Chap. 4; Lifshitz, Donateurs et fondateurs, no. 33. 9. However, we have also had occasion to note other ocial titles associated with the synagogue: (= ), ,and (Egypt); (Berenice, Antioch); and (Gali-

the second temple synagogue


dotos inscription presents the succession of these two leadership roles over a period of three generations, and a similar instance of a priest who also functioned as a community leader reappears at third-century Dura Europos, where the priest Samuel bar Yedayah served as archon and elder.10 Other rst-century evidence focuses on one role or the other. Philo speaks of priests in Egyptian synagogues who instructed the congregation during Sabbath-day meetings, and an earlier instance of priestly leadership in Egypt may be reected in III Macc. 7, 13. One of the Berenice inscriptions mentions a priest, Cartisthenes son of Archias, at the head of a list of donors who were not community ocers.11 We have noted the centrality of priests at Qumran generally, especially with regard to the ritual of prayer and study.12 Thus, these examples drawn from Judaea and the Diaspora may indeed be reective of the role that priests played in other rst-century synagogues as well. The title archisynagogos is well documented in the literary and epigraphical material, particularly with regard to the post-70 era.13 Also used by pagans,14 this title is attested in a variety of rst-century sources and locales. Besides appearing in the Theodotos inscription, the term is used by Luke with regard to the Galilee (13:14) and Antioch in Pisidia (Acts 13:15), in the Julia Severa inscription from Acmonia, and in an inscription from Egypt of less than certain rst-century dating.15 Here, too, we are dealing with a series of examples that together embrace urban and rural communities in both Judaea and the Diaspora. The prominence of this oce is fairly well attested in early rabbinic tradition. A number of sources refer to the Hebrew equivalent of archisynagogos, the rosh knesset, with respect to the Torah-reading ceremonies in the Temple as well as with regard to the secondand third-century synagogue (and possibly the rst-century synagogue, too).16 On the basis of the evidence at hand, we may conclude that the archisynagogos was not only a com-

lee)all of which have parallels in contemporary pagan contexts; see Oster, Jr., Supposed Anachronism, 2024. 10. Frey, CIJ, II, nos. 82829. 11. Philo, Hypothetica 7, 13. Berenice inscription: above, Chap. 4; and Applebaum, Jews and Greeks, 163 64. In the post-70 era as well, priests appear to have played a prominent role in many synagogues over a wide geographical areaSusiya, Eshtemoa, Naaran, and Naveh in Palestine; Dura Europos (as noted) and Sardis in the Diasporaas well as in the towns and villages mentioned in the list of priestly courses. See Roth-Gerson, Greek Inscriptions, 7879 n. 11; Seager and Kraabel, Synagogue and the Jewish Community, 189; and below, Chap. 15. 12. See above, Chap. 3, note 77; and, more generally, Burtchaell, From Synagogue to Church, 25356. 13. See Binder, Into the Temple Courts, 34852; Claussen, Versammlung, 25664; and below, Chap. 11. 14. TDNT, VII, 84445. See also G. H. R. Horsley, New Documents, IV, 21416. 15. Acmonia: Lifshitz, Donateurs et fondateurs, no. 33. Egypt: Horbury and Noy, Jewish Inscriptions, no. 26. An archisynagogos appears in a later Alexandrian inscription from the third century c.e.; see ibid., no. 18. 16. M Yoma 7, 1; M Sotah 7, 78; T Megillah 3 (pp. 35364).


historical development of the synagogue

munal leader (Acmonia, Galilee), but also might have been a wealthy member of the community (Acmonia) who participated in the ritual, administrative, and nancial aspects of the institution.17 The apparent commonality shared by these far-ung Jewish communities did not preclude the fact that a great deal of diversity also prevailed. As we have had occasion to note, local inuences often shaped the synagogue in its physical appearance, leadership positions, legal standing, and specic customs and practices. Were we better informed, we would undoubtedly detect many subtleand not so subtlecultural and religious dierences among the various communities. Yet, notwithstanding this diversity, there clearly was a broad-based common agenda and range of activities for most, if not all, of these rst-century institutions. Diversity among Diaspora and Judaean synagogues has usually been claimed on the basis of the dierent nomenclature used for the synagogue building. In the former, the term proseuche was prominent (though not exclusively); in the latter, synagoge was almost totally universal. Indeed, all but two references to the Judaean institution use the term synagoge.18 Of the fty-nine references to the Diaspora institution, thirty-one, i.e., some 53%, refer to a proseuche.19 It has been claimed that not only was the geographical factor of importance, but the chronological one as welli.e., the term proseuche predominated until the rst century c.e. and was replaced thereafter by the term synagoge.20 This last claim, although having some basis in fact, rests on somewhat shaky ground owing to the paucity of evidence. True enough, the term proseuche is used almost exclusively in Hellenistic Egypt, the Bosphorus, and Delos, and this accounts for a large percentage of our epigraphic evidence for the rst-century c.e. Diaspora and earlier. Nevertheless, one should be cautious about overgeneralizing on the basis of such admittedly limited data. These three regions may have preferred such nomenclature; elsewhere (e.g., Rome,
17. Others contributed to the synagogue as well, although material regarding individual gifts is limited. Aside from the Theodotos inscription from Jerusalem, two inscriptions from rst-century Egypt refer explicitly to Jews (Alypus and Papous) who built proseuchai (Horbury and Noy, Jewish Inscriptions, nos. 13, 126); several others, only partially preserved, seem to refer to some sort of gift (ibid., nos. 16, 20, 127, 128, 129). Of interest in this regard is a second-century b.c.e. inscription that speaks of a military commander named Eleazar who contributed a sundial, probably to a synagogue (ibid., no. 115 and commentary; see also no. 129). On the rst-century Acmonia inscription referring to Julia Severas gift of a building to the local Jewish community and its restoration by synagogue ocials, see above, Chap. 4. 18. These two exceptions are the Tiberian proseuche and the Qumran . 19. Oster, Jr., Supposed Anachronism, 186. On the suggestion that the proseuche in these and other instances refers to a Jewish institution, see Lifshitz, Prolegomenon, 6469; M. Stern, Jewish Diaspora, 15557; Levinskaya, Jewish or Gentile Prayer House? 15459; idem, Book of Acts in Its Diaspora Setting, 20725. Some, however, have expressed reservations; see, for example, G. H. R. Horsley, New Documents, I, 2528; Kraemer, Jewish and Christian Fish, 14447. 20. Hengel, Proseuche und Synagoge, 16978; Horbury and Noy, Jewish Inscriptions, 33; G. H. R. Horsley, New Documents, IV, 21920.

the second temple synagogue


Greece, Asia Minor, and Cyrene), synagoge and other terms appear to have been more in vogue.21 It may well be that the terms proseuche and synagoge reect a dierent emphasis in each institution. The former seems to indicate a desire by some Diaspora synagogues to highlight their religious dimension, and some of these institutions even acquired a measure of sanctity unknown in most contemporary Judaean synagogues.22 Such a development may have been due to the Diaspora synagogues unique context, being distant from the Jerusalem Temple and surrounded by pagan religious models. Moreover, the greater need of Diaspora Jews to dene themselves in religious-communal terms vis--vis their pagan surroundings may also have had a bearing on the nature of and emphasis within their synagogues.23 The Judaean synagogue, on the other hand, was unique not only because it was referred to almost exclusively by one term, synagogue, but also because, at this stage, the term seems to have been bereft of any religious connotation. The synagogue in Judaea was thus designated by a term denoting a place of gathering; it was primarily a communal institution whose religious prole was less pronounced than that of its Diaspora counterpart.


The role of the synagogue as the focal communal framework is reected in its venue for public gatherings. Tiberias in 6667 c.e. provides a striking example of this function. In debating whether to join the rebellion then in progress, the populace convened in either the citys stadium or the local proseuche.24 In addition, several decrees quoted by Josephus regarding Jewish rights and privileges in Asia Minor make the association of a synagogue as a community center quite clear:

21. On other names for the synagogue, see above, Chap. 4, notes 23839. 22. As we noted in Chap. 3 with respect to the Dor incident, it is quite likely that coastal synagogues may have been inuenced by Diaspora models, as, for example, in the sanctity ascribed to it by both Jew and non-Jew. 23. L. Levine, Second Temple Synagogue, 2122. What is the signicance of the term proseuche? The word itself appears some 114 times in the Septuagint as a translation of the Hebrew word for prayer, tellah ( .)However, it is doubtful, as we shall see below, whether the prayer component was the dominant feature of Egyptian Jewrys liturgy. As in Judaea, the Torah-reading ceremony appears to have been central. Thus, it is unlikely that the term proseuche would have been chosen owing solely to this factor. Another consideration that may have come into play here in opting for a term such as proseuche instead of synagoge is the fact that the latter had been (or was being) appropriated in the Septuagint to designate the community, a translation of the Hebrew terms qahal and edah. Therefore, not only was the term synagogue more neutral or secular (to invoke modern parlance), but it was also used for other biblical concepts in the Greek-speaking, Septuagint-reading Diaspora. 24. Josephus, Life 27198, 331; and above, Chap. 3.


historical development of the synagogue

Jewish citizens of ours have come to me and pointed out that from the earliest times they have had an association of their own in accordance with their native laws and a place of their own, in which they decide their aairs and controversies with one another.25 It has therefore been decreed by the council and people that permission shall be given them to come together on stated days to do those things which are in accordance with their laws, and also that a place shall be set apart by the magistrates for them to build and inhabit, such as they may consider suitable for this purpose.26

There is also some indirect evidence that Egyptian synagogues served a wider communal purpose. As will be recalled, several inscriptions mention when referring to the proseuche.27 What precisely is intended remains unclear, but it may well be a reference to multipurpose ancillary rooms such as those in contemporary pagan temples.28 Moreover, the building(s) mentioned in the Theodotos inscription, together with the Gamla, Delos, and perhaps Ostia buildings, as well as the inscriptions from Cyrene seem to indicate that these structures functioned in a similar capacity. In addition to serving as a meeting place for the community as a whole, the synagogue also hosted various subgroups within the community. Several papyri from Hellenistic Egypt indicate that a burial society once met in a local proseuche; another papyrus indicates that some sort of association in Apollonopolis Magna organized a series of banquets, probably in the local synagogue.29 This may also have been the case with regard to the Sambathic association in Naucratis, although the precise nature of this group (Jews? sebomenoi? ) remains unclear.30 The various professional guilds mentioned in the Tosefta 31 in connection with the Alexandrian synagogue may have used the premises for professional gatherings as well as for worship purposes. One of the most intriguing functions of the ancient synagogue that is mentioned on a number of occasions is its use as a place for communal meals. Once again, Josephus provides us with the clearest statements of this activity. In an edict issued by Julius Caesar (cited above), the following rights are conrmed: Now it displeases me that such statutes
25. Antiquities 14, 235; see also the decree of the people of Halicarnassus, where the religious component of Jewish rights is emphasized (discussed above, Chap. 4). 26. Antiquities 14, 25961. Communal use of the synagogue is likewise attested in a later period; see B Shabbat 150a; B Ketubot 5a; and Chap. 10 below. 27. Horbury and Noy, Jewish Inscriptions, nos. 9, 25. 28. Tcherikover et al., CPJ, III, no. 1433; Lifshitz, Donateurs et fondateurs, no. 87. See also Hengel, Die Synagogeninschrift von Stobi, 165 n. 68; Dion, Synagogues et temples, 60. 29. Tcherikover et al., CPJ, I, nos. 13839. Similar societies or individuals are mentioned in rabbinic literature with regard to Second Temple Jerusalem; see T Megillah 3, 15 (p. 357); Tractate Semaot 12, 5 (ed. Higger, 19596); and Ginzberg, Commentary, II, 5556. It is not clear, however, whether they convened in a synagogue setting. 30. Horbury and Noy, Jewish Inscriptions, no. 26. 31. T Sukkah 4, 5 (p. 273).

the second temple synagogue


should be made against our friends and allies, and that they should be forbidden to live in accordance with their customs and to contribute money to common meals and sacred rites. . . . I forbid other religious societies but permit these people alone to assemble and feast in accordance with their native customs and ordinances. 32 The edict is a general recognition of two basic rights that the Jews enjoyed: to assemble according to their ancestral tradition and to collect monies for communal meals. Similarly, in writing to the Jews of Asia Minor, Augustus makes reference to their sacred books and sacred monies that are stored in the synagogue (here referred to as a sabbateion) and banquet hall ().33 It is dicult to determine the nature of these meals. Were they holiday feasts, meals for transients and visitors, or events sponsored by local Jewish associations or individuals? Alternatively, they might have been regular communal gatherings on the Sabbath and perhaps New Moon. Or perhaps they were all of the above in various permutations over time and place. Whatever the case, one fact remains eminently clear from the documents quoted by Josephus: These meals were recognized by Romans and Jews alike as important communal activities that played an integral part in the corporate life of the Jews.34 The centrality of the meal in ancient Judaism generally is well attested. Many Jewish sects in the Second Temple period, such as the Pharisaic avurah, the Essene or Qumran sectarians, and the Therapeutae in Egypt, featured communal meals;35 one mishnaic
32. Antiquities 14, 21416. See Pucci Ben Zeev, Jewish Rights, 10718. 33. Antiquities 16, 164. See comments in Marcus, Josephus, Antiquities, LCL, VIII, 273 n. c, and especially those of Pucci Ben Zeev, Jewish Rights, 23556. It is possible that an edict issued to the Jews of Sardis, assuring them of their right to oer prayers and sacrices to God (Antiquities 14, 26061), may likewise point to a communal banquet. As noted above (Chap. 4), the reference to sacrices is baing. Does it mean oerings in general, and does the word sacrices reect a pagan misunderstanding of Jewish ritual; does it perhaps refer to actual sacrices made by gentile worshippers to the God of Israel; or does it point to what we are discussing, i.e., communal meals or banquets of the Jewish community or parts thereof? On the above, see Marcus, Josephus, Antiquities, LCL, VII, 589 n. d; Bickerman, Altars of Gentiles, 151; S. J. D. Cohen, Pagan and Christian Evidence, 166; Leonhardt, (A 14:260), 189203. The word sacrices may refer simply to the monies sent by Sardis Jews to Jerusalem, as per Josephus, Antiquities 16, 171. Might this interpretation nd substantiation in ibid., 12, 10? More recently, Runesson has interpreted this reference to sacrice as reecting what he considers to have been normative throughout the Diaspora, namely, that synagogues there were initially Jewish temples replete with sacrices and other cultic functions and appurtenances (Origins, 40176; and with respect to Sardis in particular, 46466). 34. An interesting twist to the association between synagogue and sacred meal, if indeed historical, would be the account in III Macc. 7, 1920, which tells of the proseuche that was allegedly built on the site of the festive banquet commemorating the salvation of the Alexandrian community. 35. Pharisaic avurah: Lieberman, Discipline, 199200 (= Texts and Studies, 200207); Neusner, Fellowship, 12542; Oppenheimer, Am Ha-aretz, 11869. Essenes or Qumran sectarians: War 2, 12833; 1QS 6:45; 1QSa 2:1721, CD 13, 23. Regarding the Essenes, Philo writes: They live together, formed into clubs, bands of comradeship, with common meals, and never cease to conduct all their aairs to serve the general weal (Hypothetica 11, 5). See Delcor, Repas cultuels, 40125. See also van der Ploeg, Meals


historical development of the synagogue

passage may also indicate that meals were eaten regularly in synagogues.36 At times, these repasts were intimately connected to cultic acts, as was the Passover sacrice in the Jerusalem Temple. Such meals, of course, became a central feature in the nascent Christian church as well.37 Although Jewish literary sources tend to ignore this aspect of synagogue life, comparative pagan material is not lacking. Communal meals within temple precincts are well known in the Roman world.38 Moreover, temple settings were well suited to provide space for either large family aairs (space not ordinarily available in private homes) or various kinds of fraternities.39 As in the Ostia synagogue, temples, too, were at times appointed with kitchens for such purposes. Archaeological remains of banquet areas within temple precincts have been found throughout Greece, Asia Minor, and Syria, in Nabatean temples, and in the mithraeum discovered in Caesarea Maritima.40 In addition to the pagan temples, Hellenistic and Roman religious and other associations likewise sponsored such meals. Aelius Aristides describes a gathering of Serapis worshippers at which the icon of the god was brought into the banquet hall and set on a chair in order to participate in the festivities.41 Devotees of Isis, Asclepius, Jupiter, Heracles, and others are noted as having incorporated communal meals into their religious practice.42 Thus, it would appear that Jewish communities throughout the Empire were adopting (and undoubtedly adapting) widespread practices of the Hellenistic and Semitic worlds, incorporating them in one way or another into their communal life. So central
of the Essenes, 16375; Klinghardt, Manual of Discipline, 26162. On the general identication of the Essenes with the Qumran sect, despite recent reservations by some, see Vermes and Goodman, Essenes According to the Classical Sources, 1214. Therapeutae: Philo, Contemplative Life 6490. 36. M Zavim 3, 2. See comments of Rabin, Qumran Studies, 34. The mishnah of the Bavli and of Lowes MS reads instead of . See also Alon, Studies, I, 28691. Inscriptions from later synagogue buildings in Caesarea and Stobi specically mention a dining hall (triclinium); see Roth-Gerson, Greek Inscriptions, 11517; Lifshitz, Donateurs et fondateurs, no. 10; and Hengel, Die Synagogeninschrift von Stobi, 16768. It is interesting to note that the building at Ostia (from its fourth-century stage, at the very least) had kitchen facilities; see White, Building Gods House, 69 n. 24; and Hengel, op. cit., 16772. On Netzers claim to have discovered a triclinium in a Jericho synagogue, see above, Chap. 3. 37. See, for example, I Cor. 11:1734, as well as Meeks, First Urban Christians, 15762; Kuhn, Lords Supper and Communal Meal at Qumran, 6593; M. Black, Scrolls and Christian Origins, 10218. 38. Nock, Early Gentile Christianity, 7276; C. Roberts et al., Gild of Zeus, 7779; Kane, Mithraic Cultic Meal, 31351; Meslin, Convivialit ou communion sacramentelle? 295306; Lipiski, Le repas sacr, 130*34*; Burkert, Oriental Symposia, 724. 39. MacMullen, Paganism, 3639, esp. n. 24. 40. See above, notes 35 and 36; Negev, Nabataean Necropolis, 111, 12729; Bull, Mithraeum of Caesarea, 79. 41. Oratio 8, 54, 1. 42. C. Roberts et al., Gild of Zeus, 4748; MacMullen, Paganism, 3839. At Petra, meals were held in rock-hewn chambers, and a participant was referred to as hbr; see Cantineau, Nabaten, II, 63.

the second temple synagogue


were these meals in Jewish life that they, along with several other activities, became subjects of conict and tension between the Jews and their neighbors (see above, Chaps. 3 and 4). Among its communal functions, the synagogue also served as a place for administering justice. Attested already in the Septuagint version of Susannah (see Chap. 2), adjudication was one of the basic privileges granted to Diaspora communities,43 and in Judaea this right was exercised for the most part on a local level. New Testament evidence is helpful in this regard. One tradition, appearing in each of the synoptic gospels, makes this point rather clearly: Beware of men; for they will deliver you up to councils and og you in their synagogues; and you will be dragged before governors and kings for my sake, to bear testimony before them and the Gentiles. 44 Two stages of adjudication are indicated; sentences were meted out in the synagogue, and, the trial, in all probability, was conducted there as well. At times, such procedures may have constituted only a rst stage. More serious cases might then have been appealed to higher courts, while others might have been heard from the outset by these higher authorities, whether Jewish or Roman. Flogging carried out regularly in the synagogue is further indicated by another passage in Matthew, following Jesus famous Woe speech: Therefore I send you prophets and wise men and scribes, some of whom you will kill and crucify, and some you will scourge in your synagogues and persecute from town to town. 45 Pauls speech on the steps of the barracks in Jerusalem (Acts 22:19) once again recalls this phenomenon. His words are especially poignant, for he is presumably speaking from personal experience: I said: Lord, here of all places they know that I imprisoned those who believed in You, or beat them in the synagogues. 46 The synagogue as a court and place of punishment in later centuries is explicitly indicated in rabbinic sources as well as by Epiphanius.47 Undoubtedly, all sorts of judicial matters were addressed within the synagogue framework, one of whichthe manumission of slavesis well attested in the epigraphical evidence, albeit primarily in one particular region. As noted above (Chap. 4), a number of rst-century inscriptions from the Bosphorus relate to this public procedure in a proseuche setting, some stipulating that the freed slave must continue to visit the synagogue and that the community would protect him or her from re-enslavement.48
43. On Susannah, see Chap. 2, notes 7576. See also C. A. Moore, Daniel, Esther and Jeremiah: Additions, 104: So they went to the synagogue of the city where they were living, and all the sons of Israel who were there sat in judgment. On these rights, see Antiquities 14, 235, 25961. 44. Matt. 10:1718; Mark 13:9; Luke 21:12. See also Luke 12:11. 45. Matt. 23:34. 46. Acts 22:19. Elsewhere (II Cor. 11:24), Paul speaks of being ogged by Jews thirty-nine times on ve dierent occasions; see Trebilco, Jewish Communities, 2021. 47. M Makkot 3, 12; Y Bikkurim 1, 3, 64a; Epiphanius, Panarion 30, 11; and below, Chap. 10. 48. See above, Chap. 4, notes 21819.

144 h i s t o r i c a l d e v e l o p m e n t o f t h e s y n a g o g u e

The rst-century synagogue also provided for visitors of all stripes, perhaps indigents as well. While most evidence in this regard comes from the post-70 period,49 the one clear and explicit pre-70 source is the Theodotos inscription, which notes that this synagogue also functioned as a hostel ( ).50 How unique this function was to Jerusalemthe focus of Temple pilgrimageis unclear. The fact that synagogues elsewhere also served such a purpose later on may point to other needs in the pre-70 era as well. It appears that a wide range of charitable activities were housed in the synagogue. As noted in Chap. 3, the two Pharisaic schools, Bet Hillel and Bet Shammai, disagreed over the propriety of making decisions in the synagogue on the Sabbath regarding charity to the poor, and Matthew may refer to the same phenomenon when he has Jesus caution his followers not to trumpet their gifts as do the hypocrites in the synagogues and streets that they may be praised by men. 51 In all likelihood, these funds were kept somewhere on the synagogue premises, and the actual distribution of monies likewise took place there again, as was the case later on.52 We have seen that in the Diaspora, the right to collect funds for a variety of purposes was specically guaranteed to the Jews, as reected in the general privilege issued by Julius Caesar.53 Although evidence for schools in synagogue buildings is negligible for the pre-70 period, there is reason to believe that synagogue premises served in such a capacity in many, if not most, places in this era. One late rabbinic tradition speaks of 480 synagogues in pre-70 Jerusalem, each of which had a primary school ( ) and an advanced school ( 45.) Despite this exaggerated and schematic number, the assumption that many children learned in some sort of formal setting at this time should not be construed as far-fetched. In the Greco-Roman world, educational frameworks for children were well known.55 Moreover, rabbinic material attributes the introduction of some sort of public schooling to several personalities in the Second Temple period, Simeon b. Shata (rst century b.c.e.) and the high priest Joshua b. Gamla (rst century c.e.).56 Even though such attributions are highly questionable (although, truth to tell, it is hard to fathom why Joshua b. Gamla, a relatively unknown gure, would have been singled out for no reason whatsoever), some sort of educational apparatus may already have been functioning (at
49. For example, B Pesaim 100b101a; Y Megillah 3, 74a; the multiroom synagogue complexes as at Dura, Ostia, En Gedi, and ammat Tiberias may indicate a provision for lodging accommodations. 50. Frey, CIJ, II, no. 1404; Lifshitz, Donateurs et fondateurs, no. 79; Roth-Gerson, Greek Inscriptions, 7686; and above, Chap. 3. 51. T Shabbat 16, 22 (p. 79); Matt. 6:2. 52. See, for example, T Terumot 1, 10 (p. 109); T Bava Batra 8, 14 (p. 158). 53. Antiquities 14, 215. 54. Y Megillah 3, 1, 73d; Y Ketubot 13, 35c; B Ketubot 105a. 55. Marrou, History of Education, 199209, 33536. 56. Y Ketubot 8, 32c; B Bava Batra 21a. See also Schrer, History, II, 41721; Ebner, Elementary Education, 3850, 1057; S. Safrai, Education, 946., and esp. Goodblatt, Sources, 83103.

the second temple synagogue


least in certain places) in this pre-70 period. The well-established and widespread local educational system already reected in rabbinic discussions from the second century onward (see below, Chap. 10) was probably not created overnight. Such frameworks were not necessarily under rabbinic auspices; these schools seem to have been a communal responsibility and thus might well have existed much before the time the editors of rabbinic literature began including such material in their compilations. Josephus, for his part, emphasizes the instruction received by Jewish children, presumably reecting, at the very least, the situation in the rst century.57 Even discounting his proclivity to exaggerate in matters of Jewish piety, there may well be some truth in his statements. Although Josephus does not specify a public setting, one seems to be implied by his far-reaching claims for Jewish literacy; if a public institution was involved, and it was not only a private matter as in the Greco-Roman world, it in all probability was the local synagogue. The Theodotos inscription as well may allude to such a framework of instruction when it refers to study of the commandments, but we cannot be sure of its precise implication. Other forms of instruction within the context of the synagogue took place on the Sabbath and included the community as a whole (see below). Until now, we have focused on the communal non-worship dimension of the early synagogue. Although information in this area is limited, there can be little doubt that this aspect of the synagogue constituted the very heart of the institution at this time. Serving as a place of gathering on a regular (i.e., daily) basis, the synagogue functioned rst and foremost as a community center with a broad and varied agenda.58


In the rst century, worship in the synagogue seems to have taken place only on Sabbaths and holidays. Pre-70 sources speak almost exclusively about the Sabbath, and in several instancesboth in Bereniceto New Moon and Sukkot gatherings as well. Other holiday gatherings, although unattested, may be assumed, although we have no evidence regarding the nature of their observance.59 Most extant literary sourcesgiven their par57. Against Apion 2, 204: Again the Law . . . orders that they [i.e., the children] shall be taught to read and shall learn both the laws and the deeds of their forefathers, in order that they may imitate the latter, and, being grounded in the former, may neither transgress nor have any excuse for being ignorant of them, and comments of A. Kasher, Flavius Josephus, Against Apion, II, 49496. See Josephus further comments in this regard in his Against Apion 1, 4243; 2, 151.; as well as A. Baumgarten, Torah as a Public Document, 1724. 58. On the range of the synagogues communal functions, see Binder, Into the Temple Courts, 41550; Claussen, Versammlung, 20923; Httenmeister, Die Synagoge, 35769. 59. What follows is an implicit rejection of the main thesis of McKays Sabbath and Synagogue, wherein she argues that Sabbath worship made its appearance in a synagogue setting only from the third century c.e. onward. Her argument relies on a series of methodological and conceptual errors: making an arbi-


historical development of the synagogue

ticular agendasfocus on the religious dimension of the synagogue, and it is to this aspect that we now turn.60 Information regarding the religious agenda of the synagogue is uneven. Some activities are well documented, and there is a virtual consensus as to their centrality; others remain more elusive with respect to both their form and content, while still others are the subject of scholarly debate as to whether they, in fact, existed at all in this period and, if so, in what form. The rst category for which documentation is comparatively rich includes the reading and study of Scriptures (the Torah and Prophets) on Sabbaths and holidays. The second includes related Torah-reading activities such as the sermon and targum. The last category refers to communal prayer, the existence of which at this timeespecially in Judaeais far from clear. We will examine each separately.

Torah Reading
By the rst century c.e., the Torah had become the holiest object in Judaism outside the Temple itself and its appurtenances. As early as the second century b.c.e., the Letter of Aristeas, composed in Egypt, had spelled out the sanctity of the Torah, in this case in its Greek translation. When the seventy-two emissaries from Judaea had completed their translations and presented Ptolemy Philadelphus with a copy of the Torah, the king reputedly stood for a moment in silence and then bowed seven times, saying: I thank you, good sirs, and him [i.e., the Jerusalem high priest] that sent you even more, but most of all I thank God, whose holy words these are. 61 Whether Ptolemy actually said this or not, the author succeeded in conveying, at the very least, the awe and sanctity of this text. At about the same time, I Maccabees notes that on several occasions prior to battle, the Maccabees had the Torah read aloud in order to invoke divine aid and gain scriptural
trary and articial distinction between the terms proseuche and synagoge and thus not relating to the plain meaning of the former (i.e., a place of prayer); explaining away contradictory evidence as not reecting contemporary Jewish norms (e.g., Qumran); positing a dierence between weekday and Sabbath rituals (thus dismissing the Tiberias evidence); ignoring the sacred status attributed to a number of pre-70 buildings (which would obviously be connected with their religious dimension), and more. Most crucial of all is her narrow denition of what exactly constituted worship in the ancient synagogue. I would claim evidence for prayer asidethat the reading of sacred texts and the accompanying instruction were recognized and valid forms of worship in the Jewish community. To dene worship simply as the recitation of prayers and psalms is quite arbitrary. For more extensive critiques of her thesis, see Reif, Sabbath and Synagogue (review), 61012; and esp. van der Horst, Was the Synagogue a Place of Sabbath Worship? 60. On the diering foci between the archaeological data and literary sources regarding the nature of the rst-century synagogue, see my First-Century Synagogue. 61. Letter of Aristeas 177. The letter goes on to describe the elaborate reception accorded the completed Greek translation of the Torah by the Alexandrian Jewish community and Ptolemy himself (ibid., 30821). It has been suggested that the Torah enjoyed such a status already in the biblical period, according to Deuteronomy; see van der Toorn, Iconic Book, 22948.

the second temple synagogue


support; a letter from Jonathan to Sparta makes reference to the Holy Books in our possession. 62 Josephus relates two incidents in the mid-rst century that highlight this very special status. Around the year 50, while searching a Judaean village, a Roman soldier found a Torah scroll that, according to War, he tore and threw into a re. Enraged by this blasphemy, the Jews demanded (and received) retribution.63 The second instance is a fascinating aside that Josephus relates while reporting the events in Caesarea ca. 66 c.e. He notes that when eeing the city, the Jews made sure that they took their Torah scroll with them. However, this angered the governor, Florus, and the Jews were arrested and put into chains because they had dared to remove this object from the city.64 This account clearly reects the sanctity in which the Torah scroll was held by both Jews and non-Jews. But why were the Romans outraged by the removal of a Torah scroll (or scrolls) from Caesarea? What dierence would it make to them? Their reaction can be understood only by assuming that the Torah was regarded as the holiest object that the local Jewish community possessed, and for the Romans it was the Jewish equivalent of a statue of a pagan deity. Its presence served as added insurance for the protection of the city, and its absence increased its vulnerability. Associating the sanctity of the Torah scroll with that of a statue of a pagan deity is further indicated in twoalbeit laterDiaspora sources, one literary and one architectural. The rst is a papyrus from second-century c.e. Egypt describing an event from the time of Trajan; we learn of two disputing delegations, one representing the Jewish community, the other the pagan population, which were dispatched from Alexandria to Rome, each to present its case before the emperor. The pagans, we are told, brought with them a statue of their god, the Jews some other object. Here, unfortunately, the papyrus is damaged and the object that the Jewish delegation brought is unknown. Tcherikover has posited, with a good deal of plausibility, that the Jews brought a Torah scroll as their source of divine protection and support, functioning in much the same way as the statue that accompanied the pagan delegation.65 The second example comes from third-century Dura Europos, whose synagogue, well known for its stunning wall paintings, was heavily inuenced in its architecture and art by
62. I Macc. 3, 48; II Macc. 8, 23; I Macc. 12, 9. 63. War 2, 22831: At that the Jews were roused as though it were their whole country which had been consumed in the ames; and, their religion acting like some instrument to draw them together, all on the rst announcement of the news hurried in a body to Cumanus at Caesarea, and implored him not to leave unpunished the author of such an outrage on God and on the law. See also the parallel account in Antiquities 20, 11317. On this incident and its larger context during Cumanus governorship, see Schrer, History, I, 45860; Smallwood, Jews under Roman Rule, 26369; Aberbach, Conicting Accounts, 1.; M. Stern, GLAJJ, II, 7880. 64. War 2, 292. See also L. Levine, Roman Caesarea, 2233. 65. Tcherikover et al., CPJ, II, 8287 and also 8586 and n. 17.


historical development of the synagogue

other Duran and eastern religious buildings.66 One clear link is the centrality and prominence of the shrine (i.e., aedicula) in the naos of the temples and in the synagogue. The aediculae of the temples each contained a statue of a deity; the aedicula of the synagogue, a Torah scroll.67 Finally, as in Caesarea, the Torahs sanctity was clearly acknowledged by Rome itself, and in a very public and ocial manner. Josephus notes a number of spoils of war (many appearing on the Arch of Titus as well) in the triumphal procession in Rome following the victory over the Jews and Jerusalem in 70 c.e. Preceding images of victory and the Roman conquerors Vespasian, Titus, and Domitian were a series of Temple objects (the golden showbread table and menorah) and, lastly, a Torah scroll.68 Thus, the centrality of the Torah-reading ceremony in Jewish worship of the pre-70 synagogue should come as no surprise.69 In fact, almost every source from the Second Temple period indicates this importance explicitly. Josephus:
He [Moses] appointed the Law to be the most excellent and necessary form of instruction, ordaining, not that it should be heard once for all or twice or on several occasions, but that every week men should desert their other occupations and assemble to listen to the Law and to obtain a thorough and accurate knowledge of it, a practice which all other legislators seem to have neglected.70

He [Augustus] knew therefore that they have houses of prayer [ proseuche] and meet together in them, particularly on the sacred Sabbaths when they receive as a body training in their ancestral philosophy.71
66. Rostovze, Dura and Its Art, 6899; Kraeling, Excavations at Dura: Synagogue, 34849; Goodenough, Jewish Symbols, IX, 2937; Perkins, Art of Dura, 10., 3369, 11426. 67. Rostovtze, Dura and Its Art, s.v., aedicula; Kraeling, Excavations at Dura: Synagogue, 16; Goodenough, Jewish Symbols, IX, 6567. 68. See War 7, 14852, and depictions of the Arch of Titus. 69. On the public dimension of the Torahthat it belonged to the Jewish people as a wholesee A. Baumgarten, Torah as a Public Document, 1724. An interesting account of the utilization of a Torah scroll that would lend dramatic eect to a speech is oered by Josephus when describing one of his archrivals who harangued the Tiberian populace: The principal instigator of the mob was Jesus, son of Sapphias, at that time chief magistrate of Tiberias, a knave with an instinct for introducing disorder into grave matters, and unrivalled in fomenting sedition and revolution. With a copy of the laws of Moses in his hands, he now stepped forward and said: If you cannot, for your own sakes, citizens, detest Josephus, x your eyes on your countrys laws, which your commander-in-chief intended to betray, and for their sakes hate the crime and punish the audacious criminal (Life 13435). On Torah scrolls in the rst centuries c.e., see Haran, Torah Scrolls, 93106; S. Schwartz, Imperialism, 4966. 70. Against Apion 2, 175. See also Antiquities 16, 43, a passage purportedly taken from a speech delivered by Nicolaus of Damascus before Agrippa on behalf of the Jews. See also H. Weiss, Sabbath in the Writings of Josephus, 36390, and bibliography listed therein. 71. Embassy 156. See also Philos comment in his On Dreams 2, 127.

the second temple synagogue


He [Moses] required them to assemble in the same place on these seventh days and, sitting together in a respectful and orderly manner, hear the laws read so that none should be ignorant of them.72

Philo also describes Essene practice on the Sabbath:

They use these laws [of the Torah] to learn from at all times, but especially each seventh day, since the seventh day is regarded as sacred. On that day they abstain from other work and betake themselves to the sacred places which are called synagogues. They are seated according to age in xed places, the young below the old, holding themselves ready to listen with the proper good manners. Then one of them takes the books and reads. Another, from among those with most experience, comes forward and explains anything that is not easy to understand.73

New Testament:
And he came to Nazareth, where he had been brought up; and he went to the synagogue, as his custom was, on the Sabbath day. And he stood up to read; and there was given to him the book of the prophet Isaiah. He opened the book and found the place where it was written . . .74 But when they departed from Perga, they came to Antioch in Pisidia, and went into the synagogue on the Sabbath day, and sat down. And after the reading of the law and the prophets the rulers of the synagogue sent unto them.75 For Moses of old time has in every city those that preach him, being read in the synagogues every Sabbath day.76

Rabbinic literature:
And a wooden bima was to be found in the center [of the hall, referring to an Alexandrian synagogue], and the azzan of the synagogue would stand in the corner [of the bima] with kerchiefs in his hand. When one came and took hold of the scroll to read [a section from the Torah], he [the azzan] would wave the kerchiefs and all the people would answer Amen for each blessing. He would [again] wave the kerchiefs and all the people would respond Amen. 77

Theodotos, son of Vettenos, priest and archisynagogos, son of an archisynagogos and grandson of an archisynagogos, built the synagogue for reading the Law and studying the commandments.78
72. Hypothetica 7, 12. 73. Every Good Man Is Free 8182. Suetonius (Tiberius 32:2) notes one Diogenes who lectured every Sabbath in Rhodes. 74. Luke 4:1622. 75. Acts 13:1415. 76. Ibid., 15:21. 77. T Sukkah 4, 6 (p. 273); and Lieberman, TK, IV, 89192. 78. Frey, CIJ, II, no. 1404; and above, Chap. 3.


historical development of the synagogue

Thus, there can be little question that scriptural readings constituted the core of contemporary Jewish worship in the synagogue. A remarkablethough very laterabbinic tradition articulates the highest possible religious value of such activity. On the verse and Moses gathered all the congregation of Israel (Exod. 35:1), the midrash continues: The Holy one, Blessed be He, said to Israel: If you congregate each Sabbath in your synagogues and read Torah, I will accredit you as if you bore witness on me that I am your King. 79 As already noted, we have no rm evidence for the initial stages of this custom.80 Whether this was a strictly internal development or one stimulated from without, or both, must remain moot for the present. The form and frequency of the Torah-reading ceremony are likewise uncertain. Did Torah reading, once introduced, constitute a regular Sabbath practice, as rabbinic literature would have it and as a number of modern scholars hold? 81 Or did the custom begin as Sabbath readings in preparation for the festivals, especially Passover, a theory that seems to be indicated in a number of rabbinic sources and that has found modern adherents as well? 82 Or perhaps, at rst, sections of the Torah deemed especially meaningful were chosen, irrespective of the calendar or a specic order? 83 Whatever these early stages of Torah reading were, the question arises as to when this ceremony became institutionalized as the central component of the synagogue. As discussed above (Chap. 2), the chronological parameters are probably to be xed between the fth and third centuries b.c.e. On the one hand, the terminus post quem is undoubtedly the period of Ezra and Nehemiah, when the rst public Torah-reading ceremony (albeit a
79. Midrash Hagadol, Exodus 35, 1 (p. 722). 80. See above, Chap. 2. 81. Rabbinic literature ascribes the weekly Sabbath readings to Moses and the weekday readings (Mondays and Thursdays) to Ezra or the prophets and elders (Y Megillah 4, 1, 75a; B Bava Qama 82a). See also S. Safrai, Synagogue, 91213. 82. M Megillah 3, 6; Sifra, Emor, 17 (ed. Weiss, p. 103a); SifreDeuteronomy 127 (p. 185). See also B Megillah 32a, and Elbogen, Jewish Liturgy, 13031. According to this theory, the earliest readings were taken from Pentateuchal sections relating to a specic Sabbath or festival and deal with the laws and rituals associated with each. This seems to be reected in the earliest stage of festival readings, as preserved in the Mishnah, where, inter alia, the four special Sabbaths before Passover are noted (M Megillah 3, 45). Similarly, on the Day of Atonement, the high priest would not only perform the prescribed ritual, but also read the scriptural section detailing these ceremonies (M Yoma 1, 3). Thus, Thackeray has suggested (Septuagint and Jewish Worship, 43.) that holiday readings per Leviticus 23, as well as those relating to the four special Sabbaths, eventually led to regular weekly readings and, ultimately, the triennial cycle. Assuming a gradual introduction of these readings, one might guardedly suggest that, of the special prePassover Sabbath readings, Shabbat Sheqalim (dealing with the half-sheqel tax) and Shabbat Parah (dealing with purity matters) were introduced in the Hasmonean period, when priestly and Temple concerns came to the fore in Jewish society. 83. For example, Deuteronomy 32, as suggested in McNamara, New Testament and the Palestinian Targum, 112. See also Elbogen, Jewish Liturgy, 132. Gilat (Studies in the Development of the Halakha, 35657) opts for the last two possibilities.

the second temple synagogue


one-time occurrence) was held. On the other hand, by the third century the existence of a regular communal Torah-reading framework was probably a prime factor, though not necessarily the only one, in the creation of the Septuagint. Such a translation would have served, inter alia, the liturgical needs of the Egyptian Jewish community.84 This liturgical practice of reading Scriptures was probably brought to Egypt from Judaea rather than created by the edgling Diaspora community; there is no evidence of such a custom in the earlier Elephantine community, which ourished in Upper Egypt in the fth century b.c.e. The brief account of the high priest Ezekias reading to his friends from the Torah (lit., scroll) upon his arrival in Egypt may indeed point to such a practice.85 By the rst century, a weekly ceremony featuring the communal reading and study of holy texts had become a universal Jewish practice. It was a unique liturgical feature in the ancient world; no such form of worship was known in paganism, yet certain mystery cults in the Hellenistic-Roman world produced sacred texts that were read on occasion.86 However, it was indeed sui generis for an entire Jewish community to devote regular meetings to such an activity. This, it appears, is the context in which to understand the above-quoted sources (especially Josephus and Philo). While their tone is manifestly selflaudatory, there was cause. These authors were indeed trumpeting a form of worship that set the Jewish community apart from the surrounding cultures.87 It has been generally acknowledged that the prevalent custom, in Judaea at least, was to complete the Torah-reading cycle within a three- to three-and-a-half-year period.88 This is explicitly documented only for later Roman and Byzantine Palestine but is presumed to have been operative in the pre-70 period as well.89 These later sources regularly contrast
84. Schrer, History, III, 47476; Kahle, Cairo Geniza, 13233. Cf., however, Bickerman ( Jews in the Greek Age, 1025), who views the Ptolemaic courts initiative as the main impetus for the translation. 85. Against Apion 1, 18389, and GLAJJ, I, 42. 86. See, for example, Nock, Conversion, 2632; Momigliano, On Pagans, Jews and Christians, 8991 n. 10; Patte, Early Jewish Hermeneutic, 36; Thackeray, Septuagint and Jewish Worship, 43. On the Torah reading generally, see Elbogen, Jewish Liturgy, 12942. 87. See Momigliano, On Pagans, Jews and Christians, 8991. This unique practice was already adumbrated in Deut. 31:913; see Tigay, Deuteronomy, 499502. 88. For a discussion of this issue and the various approaches (old and new), see below, Chap. 16. 89. On the triennial Torah-reading cycle in pre-70 Palestine, see, for example, the classic studies of Bchler, Reading of the Law and Prophets, 42068; Heinemann, Triennial Lectionary Cycle, 4148; Perrot, La lecture de la Bible, 147.; idem, Reading of the Bible, 13759. According to Guilding (Fourth Gospel and Jewish Worship, 2457, 22933), not only is the triennial cycle an early phenomenon, but the literary form of the gospel of John, and even the Pentateuch itself, was based on such a cycle. For a critique of this thesis, especially as it relates to the Pentateuch, see Porter, Pentateuch and the Triennial Lectionary Cycle, 16374. For a later dating of the triennial cycle, see Wacholder, Prolegomenon, xviixliii. According to Modrzejewski ( Jews of Egypt, 9596), the rst-century Cairo scroll of Deuteronomy (Papyrus Fouad 266) may indicate a triennial reading cycle. On the triennial Torah-reading cycle in Roman-Byzantine Palestine, see B Megillah 29b; Dierences in Customs, no. 48 (ed. Margalioth, pp. 172 73; ed. Lewin, pp. 9899). Indirect rabbinic evidence for this practice may be found in M Megillah 3, 4;


historical development of the synagogue

the Palestinian practice with that of Babylonia, where the Torah reading was completed in one year and was marked by the holiday of Simat Torah. Fleischer, however, has suggested that the annual cycle originated in Palestine as well and that it actually predated the triennial practice that, according to him, developed only in the tannaitic perod.90 Even though his suggested sequence rests on a series of arguable assumptions, an original Palestinian provenance for the annual cycle is certainly conceivable.91 Information available regarding Torah reading in specic settings is spotty. Presumably, the Torah was read on at least two occasions in the towns and villages whose local priestly courses were ociating in the Jerusalem Temple. To the best of our knowledge, regular readings did not take place in the Temple, even at the end of the Second Temple period, or at Qumran as part of the communal liturgy.92 Philo describes Essene Sabbath gatherings in synagogues replete with Torah reading and instruction 93 and is apparently referring to Essene communities throughout Judaea and not specically to the Qumran sectarians;94 as was the case in other local synagogues, Sabbath worship included scriptural readings. Presumably, these village- or town-dwelling Essenes did not have the same time available on a daily basis for continuous study as did their peers at Qumran, and thus the Sabbath oered an opportunity for more intensive study. If this assumption is correct, then we have evidence here for signicant liturgical variation, even within the sect, that was linked to geographical and social considerations.
Leviticus Rabbah 3, 6 (p. 69); Tractate Soferim 16, 8 (pp. 29192); see also Y Shabbat 16, 5, 15c; Esther Rabbah, Proem 3; comments by A. Epstein, Meqadmoniot Hayehudim, 54.; and below, Chap. 16. Traces of the triennial cycle have been detected in the presumed divisions of various midrashic works and targumim; see, for example, Albeck, Midrash Wayyiqra Rabba, 2543; Theodor and Albeck, Genesis Rabbah, 97. (introduction); J. Mann, Bible as Read, 11 (introduction); Elbogen, Jewish Liturgy, 132; Fried, List of the Sedarim, 10313; Shinan, Numerical Proem, 89. Ancient piyyutim likewise reect a triennial division; see Heinemann, Triennial Lectionary Cycle, 4148; Zulay, Studies on Yannai, 213. Later reports of this practice, from twelfth-century Egypt, are to be found in the Itinerary of Benjamin of Tudela (pp. 6263), Genizah fragments (Fleischer, EretzIsrael Prayer and Prayer Rituals, 220), and Maimonides, Laws of Prayer 13, 1. 90. Fleischer, Annual and Triennial Torah Reading, 2543. 91. Admittedly, assuming the existence of two parallel and contemporaneous systems for Torah reading (whether pre- or post-70) would raise a number of intriguing questions: Why two customs? Did they always exist simultaneously? If not, which came rst, and why did the other develop? Fleischer, for example, has suggested that the dierence between the two practices is that the system of shorter readings (i.e., the triennial cycle) was adopted in places that had a more expansive Torah-related liturgy, i.e., sermons and targumim. The longer, annual, reading held sway where there were fewer concomitant activities. This is an interesting but highly speculative theory even for the later period (Fleischer, Annual and Triennial Torah Reading, 27). 92. On the recitation and study of Torah at Qumran, see 1Q 6, 68; Schiman, Early History of Public Reading of the Torah, 4546. 93. Every Good Man Is Free 8183. 94. Ibid., 7576; Hypothetica 11, 1; Josephus, War 2, 124.

the second temple synagogue


Similar to the Essene practice throughout Judaea is that of the Therapeutae in Egypt. As Jews likewise devoted to a life of piety, purity, and learning, they spent their Sabbaths studying scriptural readings.95 The same also may have been true of the Samaritans, but reliable contemporary sources are unavailable in this regard. Not until the third and fourth centuries c.e., with the communal and religious revival under Baba Rabba, do we have substantive material on which to base historical assessments. At that time, some Samaritans spent the entire day in the synagogue, where the reading and study of Scriptures had become a xed routine. If we assume, as appears likely, that Baba Rabba basically revived older customs rather than instituting totally new ones, then we may conjecture that such scriptural readings existed beforehand, perhaps as early as the rst centuries c.e. 96 This, however, is speculative. Regarding Samaritan practices, anything earlier than the third or fourth centuries c.e. becomes a matter of guesswork.97

Reading from the Prophets (Haftarah)

By the rst century c.e., at the very latest, readings from the Prophets also became normative. The New Testament evidence here is crucial; Jesus was asked to read from the book of Isaiah when attending a Nazareth synagogue, and Paul delivered a sermon in the synagogue in Antioch in Pisidia following readings from the Torah and the Prophets.98 The third-century Tosefta has preserved a list of haftarah readings for (and prior to) holidays, a list of passages not to be read in public at all, and a series of norms governing such readings.99 How much of this Toseftan material is relevant to the pre-70 era is dicult to assess. Probably some of it is; much of the material in this particular tractate appears to reect early practice, i.e., the rst centuries c.e. Regarding the actual procedure for reading from the Prophets, it appears that only Luke might shed some light on the custom: And there was given to him the book of the prophet Isaiah. And when he had opened the book, he found the place where it was written. . . . And he closed the book, and he gave it again to the attendant, and sat down. And the eyes of all those in the synagogue were fastened on him. And he began to say unto them, This day is this Scripture fullled in your ears. 100 According to Luke, following
95. Philo, Contemplative Life 3031; idem, Every Good Man Is Free 8184. See also Vermes, Essenes and Therapeutai, 495504; Vermes and Goodman, Essenes According to the Classical Sources, 1517. 96. Samaritan Chronicle 16, 12 (p. 85); Crown, Byzantine and Moslem Period, 5759; Boid, Use, Authority and Exegesis, 6045; Pummer, Samaritan Rituals and Customs, 67778. 97. The problem of determining rst-century Christian liturgy is likewise formidable; see Meeks, First Urban Christians, 14450. 98. Luke 4:1719 and Acts 13:1415, respectively. On the regularity of the prophetic readings in the context of the rst-century Sabbath liturgy, see also Acts 13:27: For those who live in Jerusalem and their rulers, because they did not recognize him nor understand the utterances of the prophets which are read every Sabbath, fullled these [words] by condemning him. 99. T Megillah 3, 19 (pp. 35355). 100. Luke 4:1721.


historical development of the synagogue

the Torah reading Jesus was handed a scroll containing the book of Isaiah. Later rabbinic literature assumes that the Prophetic readings were xed and usually related in some fashion to the Torah reading.101 Was this true even at this early stage? The impression from the above passage is that Jesus chose at least the particular passage to be read, and perhaps the book as well.102 Furthermore, the Prophetic reading seems to have consisted of only a few verses, and those quoted in Luke might indeed be indicative of the usual portion that was read.103 Nevertheless, it is clear from Luke that, at least in this case, it was the Prophetic readingand not that from the Torahthat determined the nature of the sermon subsequently delivered.104 It is impossible to say when these readings from the Prophets were introduced into synagogue worship. Since they followed and presumably related to the Torah portion,105 they would then have postdated the introduction of the Torah-reading liturgy that, as we have seen, was introduced in the third century b.c.e. at the latest. Abudraham (fourteenth century c.e.) dates the institution of the Prophetic reading to the time of Antiochus IVs persecutions.106 While this medieval source has little historical value in and of itself, the period singled out may, in fact, not be far o the mark. Both Ben Sira and II Maccabees already speak of books of the Prophets as sacred literature alongside the Torah.107 The Hasmonean erawith its many upheavals and dramatic political, military, social, and
101. B Megillah 29b: one that resembles it [i.e., the Torah reading]. For a suggestion regarding the relevance of Philonic writings for determining the haftarah cycle, see N. Cohen, Earliest Evidence, 22549. 102. Scholarly opinion is divided over the readers degree of freedom in this regard. Some opine that the selections were predetermined, others that the reader had autonomy in this matter, and still others that the book read from was chosen earlier but the choice of the exact passages was left to the reader. For a summary of such opinions, see Crockett, Luke IV: 1630 and the Jewish Lectionary Cycle, 2627. 103. Luke 4:1819. The Tosefta (Megillah 3, 18 [p. 358]) speaks of three, four, or ve verses for the haftarah. Amoraic sources expand this (in theory at least) to twenty-one verses; see Y Megillah 4, 2, 75a; B Megillah 23a. See also Tractate Soferim 13, 15 (pp. 25051); Bchler, Reading of the Law and Prophets, 7, 13. According to M Megillah 4, 24, the reading from the Prophets was to consist of at least three verses, the Torah reading itself of at least ten (on weekdays) or twenty-one (on Sabbaths). 104. It was this very idea that formed the basis of J. Manns monumental Bible as Read. Despite much criticism, this approach has been rened and nuanced in Bregman, Triennial Haftarot, 7484. Elbogen ( Jewish Liturgy, 143) suggests that the term haftarah refers to the end of the scriptural-reading segment of the liturgy, parallel to the term =( completion). It has also been suggested that the term means dismissed; this prophetic reading, along with its related homily, may have concluded the communal worship service. On the haftarah generally, see Perrot, La Lecture de la Bible, 17593; J. Mann, Bible as Read, I, 55560; S. Rappaport, Erech Millin, I, 32850; Schrer, History, II, 452, following Bacher, Exegetische Terminologie, II, 14. 105. B Megillah 29b: [one] that resembles it. 106. Siddur Abudraham (p. 172). On persecution as a factor in medieval (and modern) explanations of liturgical changes, see J. Mann, Changes in the Divine Service, 241302. 107. Ben Sira, Prologue; II Macc. 2, 13; 15, 9.

the second temple synagogue


religious developmentsgave rise to messianic expectations in certain circles as well as hopes of renewed grandeur; apocalyptic speculation ourished, and eschatological groups such as the Dead Sea sect combed the Prophets for contemporary allusions.108 The use of the prophetic corpusor variations of it, as the apocalyptic mode appears to beseems to have been widespread at the time, and it may well have been this climate that gave rise to such institutionalized recitations, even among non-apocalyptic groups.109

Study and Instruction

It is eminently clear from all sources that Torah reading in the early synagogue was more than the recitation of a holy text. No matter what the provenanceJudaea or the Diasporathe Torah reading (and the reading from the Prophets) served as a springboard for further instruction and edication. Philoas might be expected, given his philosophical bentaggrandizes these Sabbath sessions, claiming that a great deal of time was spent in expounding the portion read, so much so that the synagogue was aptly described as a school () for the learning of virtues:
For it was customary on every day when opportunity oered, and pre-eminently on the seventh day, as I have explained above, to pursue the study of wisdom with the ruler expounding and instructing the people what they should say and do, while they received edication and betterment in moral principles and conduct. Even now this practice is retained, and the Jews every seventh day occupy themselves with the philosophy of their fathers, dedicating that time to the acquiring of knowledge and the study of the truths of nature. For what are our places of prayer throughout the cities but schools of prudence and courage and temperance and justice, and also of piety, holiness, and every virtue by which duties to God and men are discerned and rightfully performed.110

Or, as he states elsewhere:

So each seventh day there stand wide open in every city thousands of schools of good sense, temperance, courage, justice and the other virtues in which the scholars sit in order quietly with ears alert and with full attention, so much do they thirst for the draught which the teachers words supply, while one of special experience rises and sets forth what is the best and sure to be protable and will make the whole of life grow to something better. But among the vast number of particular truths and principles there studied, there stand out practically high above the others two main heads: one of duty to God as shown by piety and holiness, one of duty to men as shown by humanity and justice, each of them splitting up into multi-

108. Schrer, History, II, 488.; M. Stone, Apocalyptic Literature, 383441. See also J. J. Collins, Apocalyptic Literature, 34570; idem, Jewish Apocalypses, 2159. On the Dead Sea sect, see Dimant, Qumran Sectarian Literature, 51422. 109. See P. R. Davies, Scribes and Schools, 12225. 110. Moses 2, 21516.


historical development of the synagogue

form branches, all highly laudable. These things show clearly that Moses does not allow any of those who use his sacred instruction to remain inactive at any season.111

Philo also addresses the procedures and educational-religious goals of these gatherings, each of which might last for some time:
And indeed they do always assemble and sit together, most of them in silence except when it is the practice to add something to signify approval of what is read. But some priest who is present or one of the elders reads the holy laws to them and expounds them point by point till about the late afternoon, when they depart, having gained both expert knowledge of the holy laws and considerable advance in piety.112

Clearly, the instruction accompanying the reading of Scriptures was the main focus, lasting, according to Philo, almost the entire day. In contrast to the episodes recorded in the New Testament, where the preacher is described as having spoken spontaneously, here Philo talks about a priest (), elder (), or leader () who fullled this function. The goals of such expositionsthe acquisition of moral principles, a knowledge of the laws, and pietyare singled out.113 Josephus also attests to a similar devotion to study when quoting a speech of Nicolaus of Damascus delivered to Marcus Agrippa on behalf of Ionian Jewry. In an attempt to explain certain Jewish customs viewed unfavorably by their neighbors, Nicolaus proclaims: There is nothing hostile to mankind in our customs, but they are all pious and consecrated with saving righteousness. Nor do we make a secret of the precepts that we use as guides in religion and in human relations; we give every seventh day over to the study of our customs and law, for we think it necessary to occupy ourselves, as with any other study, so with these through which we can avoid committing sins. 114 The statements of Philo and Josephus indeed should not be taken as evidence of what happened in all synagogues in their day, neither in Alexandria nor elsewhere. The nature and extent of biblical exposition on any given Sabbath undoubtedly varied from place to place. The educational bent of Galilean villagers or of Jews from the Egyptian chora was undoubtedly much less than that of the Essenes, Therapeutae, or Alexandrian intellectuals, whose entire Sabbath might have been spent in either self-study or communal study.115
111. Special Laws 2, 6264. 112. Hypothetica 7, 13. 113. For comparative material from the Greco-Roman world, see Malherbe, Moral Exhortation. Note should be taken of A. Kashers suggestion (Synagogues as Houses of Prayer and Holy Places, 211) that much of Philos description and a good part of Egyptian synagogue practice derived from the memories and traditions surrounding the ceremony that concluded the translation of the Torah into Greek centuries earlier, an event of major historical and religious importance to that Jewish community. 114. Antiquities 16, 4243. 115. On the philosophizing aspect of ancient Judaism, see Mason, Greco-Roman, Jewish, and Christian Philosophies, 1218; on the study dimension as crucial to the Qumran sects self-identity, see

the second temple synagogue


Nevertheless, what is clear is that in some places, and probably not only in certain Alexandrian circles, Sabbath morning worship developed into a serious learning session. For the ordinary Jew who attended Sabbath worship, however, this kind of rigor seems most improbable. The element of study in the typical synagogue was undoubtedly much more curtailed in scope.

The New Testament evidence makes it crystal clear that the sermon (i.e., the exposition of an idea that appears in the scriptural reading) was a recognized component of the Sabbath service. Jesus preached in Nazareth, as did Paul in Antioch in Pisidia.116 As noted above, Philo has left us several descriptions of communal Sabbath observances among at least some Alexandrian Jews, and on each occasion he focuses on the sermon or exposition of the Torah. The focus of Sabbath worship for the Therapeutae of Egypt was a sermon delivered by their leader and teacher. Philo oers a brief description of the setting,117 although he avoids touching upon the contents of these homilies. Nevertheless, given the fact that members of this community spent the entire week studying Scriptures individually, there can be little doubt that the Sabbath sermon dealt with matters pertaining to their studies. Philo has the following to say: Then the senior among them, who also has the fullest knowledge of the doctrines which they profess, comes forward and with visage and voice both quiet and composed gives a well-reasoned and wise discourse. He does not make an exhibition of clever rhetoric like the orators or sophists of today but follows careful examination by careful expression of the exact meaning of the thoughts, and this does not lodge just outside the ears of the audience but passes through the hearing into the soul and there stays securely. All the others sit still and listen, showing their approval merely by their looks or nods. 118 The sermon or exposition of scriptural readings became central to Jewish worship in Egypt and elsewherein the rst century and appears to be reected in an interesting remark made by Josephus when recounting the story of the translation of the Torah into Greek. The Letter of Aristeas (3058) reports that its focus was a festive reading of Scriptures; Josephus remarks that the text was also explained by the elders (as in a sermon?) for the benet of those assembled.119
Fraade, Interpretive Authority, 5169; on the intellectual climate of Alexandria at this time, see H. A. Wolfson, Philo, I, 5586; Mendelson, Secular Education in Philo, 8184. 116. Luke 4:2021 and Acts 13:15, respectively. 117. Contemplative Life 28. 118. Ibid., 31. Interestingly, this scenario took place in the sects sanctuary (), in which there was a strict separation of men and women (ibid., 3233). The term likewise referred to individual cells used by members of the sect during the week (ibid., 25). See Daniel-Nataf, PhiloWritings, 188 n. 25. 119. Letter of Aristeas 3058; Antiquities 12, 1078.


historical development of the synagogue

We have no way of knowing for certain the forms and styles of rhetoric used by preachers in the rst century. On the basis of later analogies, and assuming that some speeches in rst-century sources may derive fromor were at least inuenced bya synagogue setting, various suggestions have been proposed. Some have assumed that IV Maccabees is, in fact, a series of synagogue sermons, that Philos questions and answers originated in a synagogue setting, or that the speeches in Acts are testimony to Hellenistic Jewish sermonic patterns and presage later homiletical ones.120 All such hypotheses, however intriguing, remain inconclusive. In conclusion, let us comment on several other aspects of the sermon. From the accounts concerning Jesus and Paul, it is clear that the sermon probably followed the Prophetic reading and related to it. This would seem to have been the norm with oppositional, messianically oriented groups, such as early Christians and members of the Qumran sect. The books of the Prophets lent themselves to revolutionary messages, be they of a political, social, or religious nature. More mainstream synagogues undoubtedly focused (though by no means exclusively) on the Torah reading itself, as per Philo and Josephus, but we have no information in this regard.121 Later synagogue sermons often preceded the scriptural readings, but this type of sermon remains unattested for the rst century. However, it is safe to assume that whenever a sermon preceded the readings, it invariably related to, and prepared the way for, the Torah reading itself, as was the case in Late Antiquity. Even within the limited material at our disposal, we nd contrasting practices with regard to the location of the preacher in the synagogue when delivering his sermon. Luke describes Jesus as sitting when he preached; Paul stood.122 Might this reect a change in the position of the preacher between the early and mid rst century? Or perhaps this marked a dierence in practices of delivery between Judaea and the Diaspora? While certitude in this matter is elusive, neither of these options is compelling. Indeed, we may simply have two alternate practices with no geographical or chronological implications. This last possibility nds some conrmation in the fact that, subsequently, we nd both practices in use within synagogue settings.123

120. IV Maccabees: Freudenthal, Die Flavius Josephus beigelegte Schrift; see also Siegert, Drei hellenistischjdische Predigten, II. Philo: Runia, Structure of Philos Allegorical Treatises, 230; idem, Further Observations, 107, 112; and H. A. Wolfson, Philo, I, 9596. Acts: Wills, Form of the Sermon, 27799; C. C. Black II, Rhetorical Form of the Sermon, 118; Bowker, Speeches in Acts, 96111. 121. Hengel (Scriptures and Their Interpretation, 159) has estimated that 96% of Philos quotations come from the Torah, an emphasis characteristic of other Jewish-Hellenistic writings as well. If there is a correlation between Philos writings and his synagogue sermons, then the centrality of the Torah for sermonic material is clearly established, in his case at least. 122. Luke 4:20 and Acts 14:1416, respectively. 123. See below, Chap. 16.

the second temple synagogue


Another activity that seems to have accompanied the Torah reading was the recitation of a targum. It is generally assumed that the custom of translating Scriptures into the vernacular at the synagogue service already existed in the Second Temple period.124 Given the widespread usage of Aramaic throughout the Near East from the Persian period onward, and the concomitant fact that this language was used in books, ocial documents, personal names, and the speech of Jews in the late Second Temple period, it is not at all surprising that there was a need to make the reading and study of Scriptures accessible to the people at large.125 This phenomenon was preceded by similar, though far more accelerated, developments within the Jewish communities of Greek-speaking countries during the Hellenistic and early Roman periods. It was in Egypt, in the wake of the largescale settlement of Jews there, that the Torah was soon translated into Greek, and then closely followed by the translation of other biblical books.126 It is possible, however, that targum was unnecessary in many, if not most, places in the Roman Diaspora, as the readings themselves may have been in the vernacular. This is far from certain, and the Septuagint translation (or variations thereof ) may have been used after the Hebrew reading, parallel to the Aramaic targum practice in Judaea.127 It is impossible to assess when Aramaic targumim rst made their appearance, although both G. F. Moore and R. Bloch have suggested dates in the early Second Temple period.128
124. See the introductory remarks on the subject in McNamara, Palestinian Judaism and the New Testament, 1789, 171210; le Daut, Introduction la littrature targumique; Alexander, Jewish Aramaic Translations, 24250; Grossfeld, Targum Onqelos, 24146; S. A. Kaufman, Dating the Language of the Palestinian Targums, 12930; Wilcox, Aramaic Background, 36278. See, however, Z. Safrai (Origins of Reading the Aramaic Targum, 18793), who, following rabbinic evidence closely, dates this practice to the second century c.e. 125. Rabin, Hebrew and Aramaic, 100739 and the extensive bibliography there; Fitzmyer, Languages of Palestine, 50131; Schrer, History, II, 20 n. 68 and bibliography cited there; McNamara, Palestinian Judaism and the New Testament, 5468. 126. The Jews of Alexandria, at least, viewed their translation as divinely inspired, celebrating the anniversary of its completion with an imposing public ceremony (Letter of Aristeas 310; Philo, Moses 2, 4142; Antiquities 12, 1078). 127. See generally on this issue Perrot, La lecture de la Bible dans la diaspora hellnistique, 118 21. On Greek in Diaspora liturgy, see Fraser, Ptolemaic Alexandria, I, 284; Schrer, History, III, 14042; Tcherikover, Prolegomenon, 3032; van der Horst, Neglected Greek Evidence, 27796. On the suggestion that the various Greek versions of the Septuagint are, in fact, Greek targumim, see Kahle, Cairo Geniza, 21314; Modrzejewski, Jews of Egypt, 100101. Cf., however, the objections raised in Wevers, Barthlemy and Proto-Septuagint Studies, 5877. This may have been the case in certain locales in Judaea as well, even in the pre-70 era. 128. See G. F. Moore ( Judaism, I, 3023), who suggests that such translations may have been coterminus with the scriptural reading itself. R. Bloch (Methodological Note, 6061) opines that the targum genre is much closer to the Midrash. . . . It is even probable that it originally was a homiletic midrash, or simply a series of homilies on Scripture, read in the synagogue after the public reading of the Torah.


historical development of the synagogue

The diculty is that all extant manuscripts of the targumim (Onqelos, Targum Jonathan of the Prophets, Targum Pseudo-Jonathan, Neoti, Genizah fragments, etc.) derive from anytime between the seventh and the sixteenth centuries c.e. 129 Moreover, there is evidence of late interpolation, such as the oft-quoted reference to Muhammads wife and daughter (Targum Pseudo-Jonathan of Gen. 21:21), the mention of Constantinople (ibid., of Num. 24:24), and reference to the six orders of the Mishnah (ibid., of Exod. 26:9). Given the presence of clearly late material, why should one assume the antiquity of the targum genre at all? The response to this is fourfold. (1) We have evidence of written targumim for the pre-70 period. Fragments of an Aramaic targum of Leviticus (4Q tgLev) and Job (11Q tgJob; 4Q tgJob) have been found at Qumran; and rabbinic literature as well makes reference to a targum of Job found in the time of Rabban Gamaliel II, at the turn of the second century.130 This last source also cites an earlier tradition that a fragment of the targum of Job was found in the days of Rabban Gamaliel the Elder (ourished ca. 2545 c.e.) during repair (or reconstruction) work on the Temple Mount.131 (2) Over the last century, many studies have compared targumic traditions with those found in Josephus, the writings of the Qumran sect, the Septuagint, apocryphal literature, the New Testament, and rabbinic literature.132 The similarities are often striking and have led some scholars to the conclusion that certain targum traditions date to the same (i.e., the late Second Temple) period as well. (3) Various studiesfocusing especially on the targumim, on the one hand, and on Qumran Aramaic, on the otherhave demonstrated linguistic anities between these
129. Kahle, Masoreten des Westens, II, 3. For an overall introduction to targumic studies, see McNamara, New Testament and Palestinian Targum, 537. See also M. Black, Aramaic Studies and the Language of Jesus, 1728. 130. T Shabbat 13, 2 (p. 57) and parallels in Y Shabbat 16, 1, 15c; B Shabbat 115a; and Tractate Soferim 5, 17 (p. 161). See also comments by Lieberman, TK, III, 2034; and Sokolo, Targum to Job. 131. Sokolo, Targum to Job. Rabbinic claims to the antiquity of targum in the days of Ezra (e.g., B Megillah 3b; B Nedarim 37b; B Sanhedrin 21b) are not a serious factor in this debate. 132. Josephus: see references in McNamara, New Testament and the Palestinian Targum, 269. In the classic work of S. Rappaport (Agada und Exegese, xxxxii), the author discusses the many parallels between rabbinic tradition and Josephus, suggesting that an early written Aramaic targum stood behind both. Qumran sect: A number of scholars have pointed out the anities between the targum genre and other Qumran writings, i.e., the pesher commentariesespecially Habbakukand the Genesis Apocryphon. See, for example, Vermes, A propos des commentaires bibliques, 95103; Brownlee, Habakkuk Midrash and Targum of Jonathan, 16986; Wieder, Habakkuk Scroll and the Targum, 1418; Lehmann, 1 Q Genesis Apocryphon, 25152; Fitzmyer, Genesis Apocryphon, 2634. Septuagint: already suggested in Z. Frankel, Vorstudien zur der Septuaginta, 18591; idem, ber den Einuss der Palstinischen Exegese, 81, followed by Churgin, Targum and the Septuagint, 4165. Apocrypha: Marmorstein, Studien zum Pseudo-Jonathan Targum. New Testament: McNamara, New Testament and the Palestinian Targum, 70.; idem, Palestinian Judaism and the New Testament, 91169; Wilcox, Aramaic Background, 377. Rabbinic literature: McNamara, Some Early Rabbinic Citations, 115.

the second temple synagogue


dierent types of literature.133 To explain such ties, a common historical setting is both logical and justiable. (4) The targum as an integral part of the synagogue liturgy was well known in Late Antiquity. Detailed laws relating to this practice, as well as accounts of people who functioned in the capacity of delivering the targum, along with the translations and interpretations on such occasions, abound in the rabbinic sources.134 Even second-century authorities were already well aware of this practice, and many regulations associated with it are discussed in this contemporary literature.135 It seems that tannaitic sages were not initiating a new practice in this regard, but rather commenting upon, critiquing, and dening (for themselves? for others?) the existing institution. Thus, it is quite likely that the use of targum goes back at least to the late Second Temple period. Moreover, the reference in Targum Jonathan to the Hasmonean ruler John Hyrcanus is often cited as proof of the existence of early material.136 On the basis of the above considerations, it is clear that the targumic compilations are multilayered with material from dierent periods; the editing process continued over centuries throughout Late Antiquity and beyond, into the early Middle Ages. Much material was added and adapted over time, although a few traditions may possibly be early, deriving from the rst century c.e., if not beforehand.137 While dierences among scholars still exist regarding how much early material is embedded in the targumim, there is a consensus that the targumim already existed in the rst century c.e. in both written and oral form,138 and that an Aramaic targum of Job found on the Temple Mount would indicate that other such works of the Torah and the Prophets were undoubtedly also in circu133. See Diez-Macho, Recently Discovered Palestinian Targum, 60. For a suggestion that a priestly legacy informed later Palestinian targumim, see Flesher, Literary Legacy, 467508. 134. See below, Chaps. 13 and 16. 135. For example, T Megillah 3, 3141 (pp. 36264). 136. The historical reference to the Hasmonean John Hyrcanus is from Targum Jonathan of Deut. 33:11. 137. See Schrer, History, I, 99114; Kahle, Cairo Geniza, 191208; McNamara, New Testament and the Palestinian Targum; idem, Palestinian Judaism and the New Testament; Diez-Macho, Recently Discovered Palestinian Targum, 60. See also Alexander, Targumim and Early Exegesis, 6071; Bowker, Targums and Rabbinic Literature; Bengtsson, Passover, 2131; and above, note 124. 138. See, for example, in this regard the statement by Goshen-Gottstein, Aspects of Targum Studies, 36: I am afraid that in spite of the Qumran material of targumic character and more sophisticated approaches in Aramaic dialectology, the gap between dicta on targumic origins and extant Targumim remains unbridged. It is one thing to talk about Targum as an institution, possibly hailing back to the early times of the Second Templei.e., its prehistory being shrouded in the general traditions about the Men of the Great Assembly. It is another to analyse actual Targumim. More than ever we are aware that the institution in the abstract and exegetical traditions must not be mixed up with an actual protoformulation in literary standard Aramaic, say rst-century b.c.e., even less so with the nal xation of the text on Babylonian soil, four centuries later. If we have learned anything in the past quarter of a century it is this: we may at best connect isolated exegetical traditions; we can never overcome a gap of eight centuries.


historical development of the synagogue

lation, and not only at Qumran. Moreover, much of the material in the extant targumim originated in the synagogue setting.139 While some traditions may have originated in a later, literary (bet midrash), context, others are quite dierent in both form and content 140 and may have come from a more popular synagogal Sitz im Leben. Thus, we are left with both certitude and perplexity. Targumim were in use in the rst century,141 and most probably (though by no means exclusively) in a synagogue setting. However, the content of these targumim, how they were delivered, how literal or exible their renditions were, and, more importantly, what ideological viewpoints they promoted are all issues that, for the present, must remain unanswered.

Communal Prayer
The most problematic component of synagogue worship in the Second Temple period is that of communal prayer. On the one hand, private prayer was a well-known phenomenon in biblical and Second Temple times;142 on the other, it is universally acknowledged that steps were taken within rabbinic circles at Yavneh soon after the destruction of the Temple to institutionalize communal prayer.143 The question today, however, is whether communal prayer as a regular and obligatory worship framework already existed in the pre-70 era and, if so, to what degree. For the most part, discussion in this regard has focused on the Shemoneh Esreh or Amidah, the central prayer in Jewish liturgy. Over the last century and more, there has been a general consensus that the activity of Rabban Gamaliel and his colleagues in Yavneh was one of editing and organizing an already extant public prayer. Scholars dier on the degree of editing involved, from a minimalist position of touching up or lightly editing an already existing version to a position advocating a serious reworking of earlier materials by the Yavnean sages.144
139. See R. Kasher, Aramaic Targumim and Their Sitz im Leben, 7585; York, Targum in the Synagogue and the School, 7486. 140. York, Targum in the Synagogue and the School. See Shinan, Aggadah of the Palestinian Targums, 20317. 141. For a more cautious approach regarding the early dating of targumic traditions, see York, Dating of Targumic Literature, 4962; Grabbe, Jannes/Jambres Tradition, 393401; as well as the more general critiques in S. A. Kaufman, On Methodology in the Study of the Targums, 11724; idem, Dating the Language of the Palestinian Targums, 11841. 142. M. Greenberg, On the Renement of the Conception of Prayer, 5792; idem, Biblical Prose Prayer; idem, Tella, cols. 896922; Kaufmann, Religion of Israel, 16061, 36667; Weinfeld, Deuteronomy, 3245. See also Johnson, Prayer in the Apocrypha and Pseudepigrapha; Flusser, Psalms, Hymns and Prayers, 55177; Charlesworth, Prolegomenon, 26585; idem, Jewish Hymns, Odes and Prayers, 411 36. 143. See B Berakhot 27b28b; B Megillah 17b18a; Heinemann, Prayer, 1336; Elbogen, Jewish Liturgy, 195203; G. F. Moore, Judaism, I, 292. 144. See the range of positions, from, for example, the minimalist position of N. Cohen (Nature of

the second temple synagogue


Many have attempted to determine the earliest versions of the Amidah, usually by focusing on sources that purportedly preserve embryonic forms of this prayer. A unilinear view of the development of Jewish prayer was once axiomatic among scholars, and attempts have frequently been made to trace the evolution of this prayers urtext. Its layers have been attributed to various historical contexts, ranging from the Persian period to the post-70, Yavnean, era.145 Earlier traces of the Amidah were supposedly detected in a wide range of sources: Nehemiah, Psalms, Ben Sira, the Psalms of Solomon, and the Dead Sea Scrolls.146 Bickerman went a step further, suggesting that many of the Amidahs formulations, rst appearing in Ben Sira, were subsequently incorporated into a civic prayer of Jerusalem and were nally adopted and adapted by the Yavnean sages following the destruction.147 Common to all the above is the assumption that the Amidah crystallized as a communal prayer at some point in the late Second Temple period. Even the pioneering study of Heinemann, which posited a multiplicity of orally transmitted forms at this stage, assumed that the basic outlines of the Amidah prayer (the number of blessings, their content, and order) had taken shape before 70.148 The work at Yavneh was thus one of editing and perhaps reformulation. Until recently, the lone dissenting voice to this consensus was that of Zeitlin. Decades ago, he argued that, in fact, no public prayer was known in Judaea in the pre-70 period and that the institution of communal prayer in the synagogue was a post-70 development.149 Of late, this line of argument has been adopted by me and, subsequently, by Fleischer and Reif.150 The case against the existence of institutionalized communal prayer in the Second Temple synagogue rests squarely on the evidence at hand (or lack thereof ) for communal Jewish prayer-worship in the pre-70 period. With all their diversity, extant sources are unanimous in this respect; as we have seen above, Philo, Josephus, the New Testament, the Theodotos inscription, and what appear to be early rabbinic traditions speak only of
Shimon Hapekulis Act, 54755) through the intermediary ones of Elbogen ( Jewish Liturgy, 2012) and Heinemann (Prayer, 13.) to the maximalist position of Zahavy (Studies in Jewish Prayer, 95101). 145. So, for example, Kohler, Origins of the Synagogue and the Church, 18., 206.; idem, Origin and Composition, 41025; Finkelstein, Development of the Amidah, 143, 12770. 146. Mirsky, Piyyut, 1829; Liebreich, Impact of Nehemiah 9:537, 22737; Liber, Structure and History, 35357; Marmorstein, Oldest Form of the Eighteen Benedictions, 13759; Lvi, Les dix-huit bndictions, 16178; and Weinfeld, Prayers for Knowledge, Repentance and Forgiveness, 186200; idem, Traces of Kedushat Yozer, 1526; idem, Morning Prayers in Qumran, 48194; Talmon, World of Qumran, 20043; idem, Manual of Benedictions, 475500; Flusser, Second Benediction, 33134. 147. Bickerman, Civic Prayer, 16385; idem, Jews in the Greek Age, 280. See also Baer (Israel among the Nations, 3236), who claimed that Greek prayers provided the model for the Amidah. 148. Heinemann, Prayer, 1336. 149. Zeitlin, Tellah, 20849 (= Studies, I, 92133). 150. L. Levine, Second Temple Synagogue, 1920; Fleischer, On the Beginnings of Obligatory Jewish Prayer, 397425; Reif, Judaism and Hebrew Prayer, 4452, 8287.


historical development of the synagogue

scriptural readings and sermons.151 None mentions public communal prayer. Moreover, the few extant buildings usually identied as synagogues would seem to indicate not only a range of plans and styles, but the lack of any discernible or distinctive orientation toward Jerusalem.152 The Judaean examples all focused on the center of the hall and had benches along the four walls. This is signicant, as it was the physical orientation of the devotee, i.e., facing Jerusalem, that was associated with prayer early on and with all prayer halls in Late Antiquity.153 Recently, Fleischer has pursued this line of reasoning to an extreme, arguing against the existence of any formal public prayer either in Second Temple Judaea or in the Diaspora.154 On the face of it, the basis for his argument would appear self-evident; the abovenoted written sources invariably speak of a Torah-reading ceremony and never of prayer. According to Fleischer, the evidence indicates that the Diaspora situation was much the same as that in Judaea, namely, that prayer was not yet a recognized communal form of worship. However, Fleischers claim is too radical and monolithic, leaving little room for the necessary diversity and nuances that characterized all aspects of rst-century synagogues. Regarding the Diaspora, the diculty with Fleischers claim rests primarily in the very name used for many Diaspora synagogues: proseuche, literally, house of prayer. On the basis of the name alone, it is rightfully assumed that prayer was a signicant element of Jewish worship in such places. Nevertheless, on the basis of the repeated references to Torah reading in Diaspora sources, Fleischer dismisses the prayer factor, claiming that the name proseuche was invoked by Diaspora Jews to ascribe a measure of sanctity to their institution, thereby asserting its inviolability in the face of pagan attack.155 However, to dismiss a name used for hundreds of years in a wide variety of geographical locales as being a kind of ruse to demonstrate self-condence and mislead the gentiles is most problematic. Non-Jews visited Diaspora synagogues in large numbers throughout the rst century c.e. and undoubtedly were familiar with what went on inside. To assume that the institution was called by a name that had nothing to do with what actually transpired therein is stretching credulity to the limit. Indeed, if Jews desired to create an aura of sanctity for their place of worship, many other terms were available (and some were,
151. It may be of signicance that Luke, who incorporates prayer pericopes more than any other gospel writer, speaks of prayer in the Temple and in private homesbut never in the synagogue. See Falk, Jewish Prayer Literature, 26976. 152. L. Levine, Second Temple Synagogue, 1019. 153. See Dan. 6:11; and below, Chap. 9. 154. Fleischer, On the Beginnings of Obligatory Jewish Prayer, 402. and esp. 42425. Note, furthermore, Fleischers rather rigid denition of communal prayer (ibid., 401, 414, 426). 155. Ibid., 409. The argument, that the term proseuche was invoked to oer an extra measure of protection to the synagogue, seems to be undermined by the fact that it was used as early as the third and second centuries b.c.e., when Jewish life ourished in Hellenistic Egypt.

the second temple synagogue


in fact, used) that would have conveyed this message no less poignantly.156 Moreover, if Fleischer is correct, why is it that only some Diaspora communities invoked the term proseuche and not others? The Jews of Berenice apparently had no qualms about using the name amphitheater and later synagogue, nor did the Jews of Rome and Asia Minor about regularly using the term synagoge. Nevertheless, Fleischer has rightly pointed to a sharp discrepancy between the name proseuche, on the one hand, and the activities recorded in Diaspora sources as having transpired therein, on the other. The former emphasizes prayer, the latterreading the Torah. Perhaps an answer is to be sought not in denying the prayer dimension, but rather in seeking to explain why Diaspora writers singled out the Torah-reading ceremony specically as so central an activity. Several factors seem to have been involved. Firstly, the Torah service may indeed have been the primary focus of Jewish worship in the Diaspora, as it was in Judaea. Secondly, this part of the service was undoubtedly the most dramatic and participatory component of Jewish worship and thus the one most likely to be described. Thirdly, not only was the Torah reading important in its own right, but it also served as a focus around which most of the other liturgical elements revolved, i.e., the targum, sermon, and haftarah. Finally, and as noted above, the Torah-reading ceremony and its related components reected what was most unique and distinctive in the synagogue worship context, especially when compared to other religious institutions in the Greco-Roman world. In its religious dimension, the synagogue was fundamentally a place of study and instruction; since such an emphasis was sui generis in antiquity, it was only natural that ancient writers would highlight this aspect. Thus, precisely because of its centrality, ancient authors made use of the Torah-reading ceremony, each for his own purpose: the author of Acts, to set the stage for Pauls preaching to Jews and non-Jews; Philo, to expound on the didactic, philosophical, and moral lessons to which the Jews were exposed on the Sabbath; and rabbinic tradition, to describe a hall so huge and magnicent that the only way the congregation could respond was by the azzan waving kerchiefs. In sum, it would appear unwarranted to deny the existence of prayer as an integral part of Diaspora worship, although, admittedly, it was not the dominant element. The name proseuche, associated with many Diaspora institutions, is simply too telling to be summarily dismissed.157 Nevertheless, we have no way of determining the nature, composi156. For example, the synagogue could have been referred to as (holy place), as seems to have been the case on occasion; see Josephus, War 7, 45 (Antioch); III Macc. 2, 28 (Egypt). See also Josephus, Against Apion 1, 209 on Agatharchides reference to in Jerusalem; however, the reference here is undoubtedly to the Jerusalem Temple itself (despite the plural). 157. Moreover, it may be of more than passing interest that in the only Palestinian synagogue that Josephus refers to as a proseuche, namely, the one in Tiberias, the recitation of prayers is specically mentioned. However, in this case Josephus was speaking not of an ordinary Sabbath but of a fast day, which may explain the predominance of prayers there (Life 29095).


historical development of the synagogue

tion, and extent of communal Jewish prayer in the Diaspora during the Second Temple period. Some have tried to identify specic compositions in extant Jewish and Christian literature that may have stemmed from a pre- or post-70 Diaspora proseuche setting.158 Such suggestions, however, remain tentative, at best.159 We have already oered several reasons as to why communal prayer played a more prominent role in Diaspora synagogues than in those of Judaea. The use of hymns and prayers was central to pagan religious frameworks, much more so than in the Jerusalem Temple, where silence accompanied the actual sacricial rite and only the levites provided a hymnal component.160 Thus, exposure to pagan forms may well have stimulated local Diaspora communities to imitate these practices in some way. Moreover, the distance from the Jerusalem Temple and its manifold ceremonies may have played a role as well. Diaspora communities perhaps felt a need to compensate for this remoteness by embellishing their own liturgy.161 In fact, the absence of (or in this case, the distance from) the Temple may have provided a powerful incentive to develop a prayer format, as happened at Qumran and in the post-70 era (see below). If, then, Torah reading was indeed the dominant activity, even in the proseuche setting, why did so many of these communities choose the term proseuche to designate their communal building? As noted some time ago 162and Fleischer has reemphasized this point as wellit is reasonable to assume that the name proseuche bestowed an aura of holiness and sanctity on these Diaspora institutions. Given their minority status and the distance from the Jerusalem Temple, Diaspora Jews might well have desired to enhance their local institutions with this additional religious dimensionin name as well as in fact. On the Judaean scene, the issue of communal prayer fares dierently and requires a
158. Letter of Aristeas 305. See A. Kasher (Synagogues as Houses of Prayer and Holy Places, 211), who claims that the story provided an important paradigm for imitation. See also Charlesworth, Old Testament Pseudepigrapha, II, 67197; idem, Jewish Hymns, Odes and Prayers, 41136. See also Flusser, Psalms, Hymns and Prayers; Fiensy, Prayers Alleged to Be Jewish; idem, Hellenistic Synagogal Prayers, 1727. 159. The one possible example of a prayer from a proseuche setting may be preserved in the famous Nash papyrus from second-century b.c.e. Egypt. The prayers listed there (the Ten Commandments and the Shema) possibly reect Hebrew prayers current in at least some Egyptian proseuchai, but there is no way to prove this. See below, Chap. 16, note 87, as well as Albright, Biblical Fragment, 1539; Fraser, Ptolemaic Alexandria, II, 443 n. 777; Bickerman, Jews in the Greek Age, 86. Cf. Lacheman, Matter of Method, 1539. See also Liber Antiquitatum Biblicarum 11, 8: Take care to sanctify the Sabbath day . . . to praise the Lord in the assembly of the elders and to glorify the Mighty One in the council of the older men. 160. Pagan ritual: MacMullen, Paganism, 16, 44. Temple ritual: for example, Letter of Aristeas 92, 95. For diering interpretations of the reason for the silence in Jerusalems two Temples, see Kaufmann, Religion of Israel, 3014; Knohl, Sanctuary of Silence, 14852; idem, Between Voice and Silence, 1730. 161. Hengel, Proseuche und Synagoge, 3335. 162. L. Levine, Second Temple Synagogue, 22.

the second temple synagogue


nuanced approach. While there is no evidence for communal prayers in the typical Judaean synagogue,163 prayers settings did, in fact, exist in certain cases. It is clear from literary sources and from the Qumran scrolls that the Essenes and related groups conducted regular communal prayer sessions. Moreover, there is some indication that certain prayers found in the scrolls may have originated before Qumran was established, among groups with which the sect had anities.164 The Essenes throughout Judaea, according to Philo, would gather in their own synagogues for regular worship, and at Qumran a is mentioned as the setting for worship services.165 It is understandable that a group that consciously distanced itself from Jewish society generally, and from the Temple specically, might nd it necessary to develop an alternate religious mode (i.e., prayer) as its primary form of worship. The ecacy and validity of the sacricial cult in Jerusalem being denied by this sect, substitute worship elements were introduced.166 In addition, priests ociating in the Temple held prayer services every morning; many elements of this liturgy found their way into later normative Jewish prayer. The Mishnah records the following: The appointed [ocial] said to them: Recite one blessing. And they blessed and read the Ten Commandments, the Shema [Deut. 6:49], If, then, you obey [ibid., 11:1321], And He spoke [Num. 15:3741]. They blessed the people with three blessings: True and certain, Temple sacrice [Avodah], and the priestly blessing. On the Sabbath, they would add a blessing for the departing priestly course. 167 We also nd hints of group prayer among the rst-century Pharisees. Bet Hillel and Bet Shammai are said to have disputed the precise number of blessings to be recited on holidays and Rosh Hashanah that fell on the Sabbath. In the context of that discussion, the Tosefta refers to a specic incident that took place in the presence of the elders of Bet Shammai and oni Haqatan.168 Acts speaks of ninth-hour prayers in the Temple; other occasionspublic and private,
163. Ibid., 1920. 164. See Chazon, Prayers from Qumran, 27173; idem, On the Special Character of Sabbath Prayer, 121; Knohl, Between Voice and Silence, 30. 165. Judaea: Philo, Contemplative Life 8183. Qumran: CD 11, 21; Talmon, World of Qumran, 24142; Steudel, Houses of Prostration, 4968; Chazon, On the Special Character of Sabbath Prayer, 121; and above, Chap. 3. 166. According to Talmon (World of Qumran, 2026, 237.), Qumran was inuenced by biblical verses stressing the importance of prayer, while its communal prayer stemmed from the commune ideology and corporate personality of the sect. 167. M Tamid 5, 1. See comments of Zeitlin, Morning Benediction and the Readings in the Temple, 33036; Hammer, What Did They Bless? 30524; Kimelman, Shema and Its Rhetoric, 13543; and below, Chap. 16. In the case of this Temple prayer as well, Fleischer attempts to neutralize the evidence of communal prayer by assuming that this tradition is anachronistic (On the Beginnings of Obligatory Jewish Prayer, 41415, 41924). For a recent suggestion that the recitation of the Shema was rst introduced in the Hasmonean era, see A. Baumgarten, Invented Traditions, 2029. 168. T Rosh Hashanah 2, 17 (pp. 32021); and below, Chap. 16.


historical development of the synagogue

daily, and holidaysare mentioned in a variety of sources.169 The evidence in Matthew for prayer in the synagogue and on street corners is equivocal: And when you pray, you must not be like the hypocrites; for they love to stand and pray in the synagogues and on the streetcorners, that they may be seen. 170 It would appear that Matthew was not being critical of communal prayer in synagogues, but of the ostentatious display of piety by individuals; he was advocating devotions in private prayer, and not demonstrative prayer in public, which he associates with hypocrites. Nevertheless, it is quite possible that some blessings were recited in synagogues in Judaea as well as in the Diaspora. Although only nding expression in the Mishnah and Tosefta, blessings before and after the Torah reading may have already been a customary practice in the rst century.171 In addition to the Torah blessings, there is some evidence that the priestly benediction was recited in pre-70 synagogues. The Mishnah records the following: In the province it was pronounced as three blessings, but in the Temple as a single blessing; in the Temple they [i.e., the priests] pronounced the Name as it was written, but in the province by substituted terms; in the province the priests raised their hands only as high as their shoulders, but in the Temple above their heads, except for the High Priest who raised his hands only as high as the frontlet. 172 According to this tradition, the priestly blessing was recited not only in the Temple itself,173 but also outside Jerusalem. The term used in this last regard is somewhat vague; in juxtaposition with the Temple, this tradition speaks of outlying areas ( )or, in a parallel version, borders ( 471.)This source, however, poses a number of questions. Can we regard it as historically reliable, and, if so, to which communal framework does it refer? Assuming a modicum of credibility for this tradition does not appear entirely unwarranted, and it would seem that the synagogue is the venue being discussed. What is not at all clear, however, is whether the two practices noted as having taken place in the Temple and synagogue were contemporaneous. Were there similar rituals being conducted simultaneously in the Temple and synagogue, or does this tradition in reality compare pre-70 Temple practice with that of the post-70 synagogue? The answer is not clear. Only by assuming the former alternative can we posit the recitation of priestly blessings in the Second Temple synagogue. However, since this mishnaic pericope may well derive from the Ushan era (ca. 140180 c.e.), some three generations after the Temples destruc169. Acts 3:1; Falk, Jewish Prayer Literature, 28598. See the interesting though problematic tradition in T Sukkah 4, 5 (p. 273); and comments in Lieberman, TK, IV, 88889; and above, Chap. 3. 170. Matt. 6:5; McKay, Sabbath and Synagogue, 172. 171. M Megillah 4, 2; T Sukkah 4, 6 (p. 273). 172. M Sotah 7, 6. See also M Tamid 7, 2. 173. M Tamid 5, 1. 174. Sifre Zuta 6, 27 (p. 250). On as outlying areas, see B Berakhot 12a.

the second temple synagogue


tion, it is very possible that the comparison being made is between pre-70 Temple practice and that of contemporary synagogues, with which these early sages were familiar.175 Thus, we may conclude that the place of prayer in the Second Temple synagogue in Judaea varied considerably. Prayer appears to have played little or no role in the typical Judaean synagogue.176 Whether some places distant from Jerusalem (e.g., in the Galilee) may have introduced this component in some limited fashion is a moot issue. There was a short, xed morning prayer service for ociating priests in the Temple but, interestingly enough, this framework was separate from their sacricial routine.177 The various sects seem to have developed communal prayer patterns; evidence regarding Qumran, the Essenes, and the Therapeutae is clear-cut thanks to the Dead Sea scrolls, Josephus, and Philo. Moreover, there is some indication that the Pharisees knew of a xed prayer mode (possibly communal), at least on holidays. Outside Judaea, however, the prayer component seems to have been considerably more developed, although, unfortunately, we are in no position to determine its nature, form, or content.


On the basis of our discussions here and in Chapters 3 and 4, a number of interesting aspects of the rst-century synagogue have emerged. In the rst place, the institution was essentially a communal one. This is patently evident from the archaeological remains; each of the six buildings that can be dated to the rst century c.e. with certainty (Delos, Gamla, Masada, Herodium, Qiryat Sefer, and Modiin) served as communal frameworks. These were neutral buildings, bearing no indication whatsoeverarchitectural, artistic, or epigraphicalthat they were intended to serve a religious purpose. The buildings featured benches around the main hall and were similar in plan to assembly halls of the Hellenistic-Roman world. Finally, Josephus has preserved edicts that record the range of communal functions that took place in Diaspora synagogues and proseuchai. The epigraphic material as well emphasizes the institutions communal status. The proseuchai of the Bosphorus Kingdom played a central role in Jewish life (and perhaps for non-Jews as well) not only as a place where manumission ceremonies were held, but also because the Jewish community served as a guardian to the agreement and the synagogue itself was the place where the freed slave had to oer services in the future. The Cyrene inscriptions similarly bear witness to the centrality of the local synagogue (referred to as an amphitheater in several instances). It was here that the community honored Jews and non-Jews alike and perpetuated these honors via inscriptions and repeated public mention
175. On the use of the term for post-70 practice, see T Kippurim 1, 9 (pp. 22324); B Yoma 19b. 176. See Talmon, World of Qumran, 26768. 177. M Tamid 4, 35, 1.


historical development of the synagogue

of them. This is most reminiscent of Philos claim that tributes of shields, gilded crowns, slabs, and inscriptions to the emperor were found in Alexandrian proseuchai.178 Notwithstanding, the religious component indeed played a part in synagogal activity. There is no question that it was an important element, although the degree of its priority may have varied from place to place. The religious aspect was but one of many functions that transpired in this institution. Thus, the synagogue was a communal institution par excellence, and indeed the only one of its kind attested in either Judaea or the Diaspora throughout antiquity. The proclivity among many scholars today to describe the synagogue in religious terms derives not only from our own associations with contemporary synagogues as religious institutions, but also from the fact that ancient literary sources tend to refer to the synagogue in religious terms. Despite this emphasis, while it was the communal feature that characterized the synagogue in the rst century c.e., its religious component gradually gained in importance in Late Antiquity until it eventually dominated the prole of this institution. In Judaea, this process commenced after the destruction of the Temple. The situation in the Diaspora was somewhat dierent, as the synagogues religious and sacred character was discernible even before 70 c.e. For a few of these institutions, a term indicating holiness is used, and some are said to have contained holy objects or to have held sacred activities (e.g., meals, books, monies for the Temple and community, etc.). True enough, Philo also attributes sanctity to Essene synagogues of Judaea, but if the accuracy of this characterization (as against its being the projection of a Diaspora writer) is admitted, it could still be claimed that, in this respect as in many others, the Essenes did not reect the more general Judaean practice. In the course of our discussions of the Second Temple synagogue, we have noted that in a number of areas a phenomenon appeared in the Diaspora before it did in Judaea. There are three salient examples of this chronological priority. The rst is the very existence of the institution; as noted, the earliest evidence for a synagogue appears in Hellenistic Egypt, several centuries before a comparable evidence surfaced in Judaea. A second example has to do with the appearance of regular public prayer, again attested in Ptolemaic Egypt earlier than in Judaea. Thirdly, the evidence that exists regarding synagogue sanctity derives almost exclusively from the Diaspora. Since all of the above examples are documented for Egypt, a logical conclusion might be to assume that all were initially a product of that Diaspora community.179 Let us examine this claim of Diaspora precedence, item by item. The evidence for the existence of the synagogue clearly speaks for itself. The proseuche rst appeared in Egypt, and thus many scholars have followed this evidence and concluded that it is, in fact, a
178. Embassy 133. 179. Maoz raises an interesting though highly speculative suggestion that the Judaean synagogue was inuenced by contemporary Alexandrian architecture (Synagogue in the Second Temple Period, 512).

the second temple synagogue


Diaspora creation.180 However, there are very basic methodological issues at stake here that require consideration. Are these data indicative of the institutions very beginning or are they merely fortuitous in their preservation? Moreover, given the fact that the evidence relating to synagogues before the rst century c.e. is so sparse, how much weight should be accorded to the rst appearance of this phenomenon? Or, to phrase this issue dierently, does the absence of data necessarily mean that something did not exist? Caution is especially required since we are dealing with material from Egypt, where much more material has been preserved than elsewhere because of the countrys unusually dry climate. Finally, when the synagogue does begin to appear in other locales in the rstcentury b.c.e. (e.g., Delos, Cyrene, and Asia Minor), it is depicted as a well-established and universal institution that had already acquired a status and centrality in the eyes of both the Roman authorities and Jewish communities throughout the Empire. Clearly, its history has considerable roots, whether we think in terms of decades or, more likely, in terms of centuries. If the above-noted lines of reasoning are valid and the evidence that Egypt was necessarily the cradle of the synagogue for the Diaspora may not be foolproof, could this be argued for Judaea as well, i.e., that the synagogue evolved there independent of the Diaspora? The situation is somewhat dierent with regard to communal prayer. The widespread Diaspora use of the term proseuche (as against the Judaean synagoge) clearly indicates that there was a substantive dierence in liturgy between the two geographical areas. While the Temple stood, Judaean communities seem to have limited their liturgical expression to the cluster of activities that accompanied the Torah-reading ceremony. Outside of Judaeafar from the dominance (psychological as well as physical and political) of the Temple, and perhaps stimulated (or challenged) by their surroundingsmany Jewish communities developed prayer and even emphasized its importance by naming their communal institution accordingly. Whatever factors lie behind this development, it is important to bear in mind that not all Diaspora communities necessarily moved in this direction. Many did not use the term proseuche but preferred synagoge or some other term; thus, there is no compelling reason to assume that, in light of the many dierences between these institutions, prayer necessarily played a central role everywhere. It should be noted that clear and explicit statements regarding sanctity are rare. Those who tend to generalize regarding the extent of this phenomenon rely, in many cases, on shaky evidence that requires substantiation. Only with regard to Antioch (where the Jeru180. Hengels pioneering study was the rst serious attempt to establish this claim (Proseuche und Synagoge, 15784); later on, this line of reasoning was developed by Dion, who demonstrated how indebted the Egyptian proseuche was to Egyptian temple models (Synagogues et temples, 4575). An even more revolutionary thesis was put forth by Griths (Egypt and the Rise of the Synagogue, 115), who sought the Egyptian roots of the proseuche in the Per Ankh, which was associated with local temples. For a critique of this theory, see my First-Century Synagogue.


historical development of the synagogue

salem Temple vessels were deposited, having been appropriated by Antiochus IV during the period of his persecutions) is there some basis for assuming that the word hieron (holy place) reects a historical reality. Several other terms (e.g., a sacred precinct or grove) that may imply a buildings sanctity refer to a number of Egyptian Jewish locales, and III Maccabees associates a proseuche with the holy commemoration of the local communitys salvation.181 Admittedly, use of the term proseuche may indicate a degree of holiness, and the references in Roman edicts to sacred meals, sacred books, and sacred funds are all presumably to be associated with the synagogue. It is not clear, however, how much the presence of these objects or activities in the building, or even the name proseuche, may actually reect, or contribute, to the sanctity of the place. This having been said, it is nevertheless clear that the element of sanctity indeed played a role in some Diaspora synagogues, whereas this was rarely, if ever, the case in contemporary Judaea. Here, too, as in the case of prayer, it is undoubtedly the social and religious contexts of Diaspora communities that contributed greatly to this development. In summary, the claim that the Diaspora pioneered in various matters relating to the synagogue is complex. There is no one simple answer, armative or negative, and each case must be judged on its own merits. The claim that the very existence of synagogues was based on an Egyptian model is most problematic, resting, as it does, on slim evidence. As for communal prayer and sanctity, the evidence is quite conclusive; the Diaspora, or at least parts of it, did take the lead. Whether or not this inuenced Judaea is another issue; in these two realms, the development there was much later and under very dierent historical circumstances (see below). In previous chapters, we noted the extensive diversity among rst-century synagogues in Judaea and the Diaspora that stemmed from diering regional contexts and local inuences, as evidenced in both the physical remains of synagogues (Delos, Gamla, and Masada) and its various names (proseuche, synagoge, didaskaleion, amphitheater, etc.). Recognition of the diversity in Second Temple synagogues is certainly a valid and, indeed, crucial component for understanding their nature at this time. However, such a claim should be regarded as necessarybut not sucientfor a full appreciation of the institution. The present chapter has highlighted the commonality within this diversity. The rst-century synagogue, for all its variety, exhibited many shared features, both communal and religious; these common denominators, despite all the nuances from one community to the next, were far from inconsequential. With the notable exception of the pre-70 Jerusalem Temple, the synagogue encapsulated Jewish communal life within its wallsthe political, liturgical, social, educational, judicial, and spiritual. It is this inclusiveness that made the rst-century synagogue a pivotal institution in Jewish life that played a major role in enabling communities throughout the world to express their Jewishness, preserve their Jewish identity and communal
181. III Macc. 7, 1920.

the second temple synagogue


cohesion, and eventually negotiate the trauma and challenges created by the Temples destruction in 70 c.e. For all the engaging comparisons that have been made between the synagogue and comparable Greco-Roman associations (thiasos, koinon, collegia, etc.),182 many of them cogent to some degree, no analogy can do justice to the unique role of this institution. Given the Jews special needs, together with the willingness of Roman society to tolerate and often support such dierences, the synagogue assumed a key role within Jewish society.
182. Surveyed by Poland in his Geschichte des griechischen Vereinswesen.



he impact of Jerusalems destruction in 70 c.e. on the Jews of Roman Palestine, and the concomitant elimination of the leadership and institutions associated with the city, were traumatic indeed.1 Suddenly, the major national and religious focus of Jewish lifethe Jerusalem Templehad been eliminated, along with the rituals and ceremonies that had constituted the warp and woof of divine worship in Israel. True enough, there was already a historical precedent for coping with such a loss. The destruction of Jerusalem and the First Temple in 586 b.c.e. did not spell the demise of Judaism or the disappearance of the Jewish people; life went on, adaptations were made, and the Temple was eventually restored.2 Moreover, Diaspora communities long before 70 c.e. had come to terms with their geographical distance from

1. On the response to 70 c.e., see L. Levine, Judaism from the Destruction of Jerusalem, 12548, 33840, and literature cited therein; Baron, Social and Religious History of the Jews, II, 11028; S. J. D. Cohen, Signicance of Yavneh, 4551; idem, From the Maccabees to the Mishnah, 21431; Goldenberg, Broken Axis, 86982; Bokser, Wall Separating God and Israel, 34974; idem, Rabbinic Responses to Catastrophe, 3761; Kirschner, Apocalyptic and Rabbinic Responses, 2746; Avery-Peck, Judaism without the Temple, 40931. 2. Whether and to what degree the destruction of the First Temple was remembered and commemorated in the Second Temple period are intriguing questions with little scholarly consensus. Discussion often centers around the nature and extent of observance of the Ninth of Av and other related fasts, a practice apparently indicated by M Rosh Hashanah 1, 3; T Taanit 3, 6 (p. 338); Y Betzah 2, 2, 61b (= B Taanit 13a). See also J. Rosenthal, Four Commemorative Fast Days, 44659.

late roman pa le st i ne


the Temple, which, for all practical purposes, no longer impacted on their daily lives,3 and the same may have been true, to a much lesser extent, for Galilean Jews. Some sects of the Second Temple period had created alternative forms of worship to supplement and, at times, even replace Temple ritual. Nevertheless, the tragedy of 70, following an era of dramatic Jewish demographic, religious, and social growth, undoubtedly caused serious reverberations throughout the Jewish world. A new reality had now emerged. Generally speaking, one can assume that those living closest to Jerusalem and the Temple were most aected by the destruction, the Jews of the region of Judaea proper more so than those of the Galilee, and the latter more so than those living in the Diaspora. In most cases, geographical propinquity probably correlated with a sense of attachment, dependence, and loss. No other Jewish institution was more aected by the events of 70 than the synagogue. In a religious vein, the synagogue had come to play an important, though limited, role on the local level; now, in the post-70 era, it would begin to acquire an increased centrality in Jewish religious life. Here, too, however, geographical distinctions are in order. The religious component of the Second Temple synagogue diered from place to place, and it appears that there may have been a correlation between distance from the Temple and degree of liturgical development in the synagogue. Diaspora synagogues seem to have had a more elaborate ritual than their Palestinian counterparts, and there may have been some dierences in worship between the Galilean and Judaean synagogues as well. Synagogues located in Jerusalem itself probably had a more limited religious agenda than elsewhere, given the presence of the Temple. Nevertheless, there is little question that Palestinian synagogues as a whole, whatever the dierences between them, were deeply aected by the destruction and that the institutions subsequent religious development was in no small measure a response to this catastrophe. Although the year 70 thus provides us with a clear watershed for the synagogue, the subsequent periodization of the post-70 era is less easily dened. The fourth century has been chosen as a convenient dividing line for a number of reasons, both external and internal; the former relates to more general historical circumstances, the latter to the primary source material at our disposalboth written and archaeological. The fourth century marked the ocial demise of the pagan Roman political order and its replacement by Christianity, beginning with Constantine and culminating under Theodosius I. The passing of the relatively benign pagan rule, with its generally tolerant attitude toward the Jews and Jewish tradition, and the concomitant rise of its Christian successor under the inuence, in varying degrees, of an often hostile clergy, is an obvious benchmark that could not but aect Jewish life.4 There is no precise date for this change of attitude and policy. Formally, we can point
3. Goodman, Diaspora Reactions to the Destruction of the Temple, 2738. 4. See discussion on Christian hostility below, Chap. 7. For an example of pagan tolerance, see Minucius Felix, Octavius 6, 13. On the issue of periodization in the post-70 era, see my article, Between Rome and Byzantium, 748.


historical development of the synagogue

to 324 c.e. and Constantines ocial recognition of Christianity. However, this event did not immediately inaugurate a dramatic shift in the attitude toward the Jews and Judaism. The fourth century was a period of transition and uidity, when the fate of the Jews and Jewish institutions was far from being xed. On occasion Jews were adversely aected but often found support, and in some respects even gained in stature, under the Imperial authorities.5 The status of the Patriarchate at this time is a case in point.6 Thus, the full import of the Christianization of the Empire on the Jews became clear only at the end of a process that began in the early fourth century but required a number of generations to complete (on the part of the church and Imperial authorities) and be assimilated (on the part of both Christians and Jews). Between such dates as 324 (Constantines recognition of Christianity), the 330s (the reshaping of Jerusalem as a Christian city and the beginning of extensive church building throughout the country), 363 (the abandonment of Julians plans to rebuild the Temple), and 380 (Christianitys becoming the ocial religion), most Jews, it is safe to assume, became fully aware of the momentous changes that were taking place and possibly even cognizant of some of their long-range implications. The second major change of the fourth century was the close of the talmudic era in Palestine. The last sages mentioned in the Yerushalmi and midrashim lived in the third quarter of the fourth century and the nal editing of the Talmud took place soon thereafter. Thus, some time in the second half of the fourth century, all recording of specic attributions and names in the Palestinian rabbinic corpus ceased.7 With but rare exception, then, every attributed saying or story regarding Palestinian sages in rabbinic literature predates the mid fourth century, with the overwhelming bulk of the material associated with sages from the second and third centuries. The same holds true for aggadic midrashim; although these were not compiled until the following centuries, the ascribed material, i.e., names of people, places, and events, refers to the fourth century and earlier. Thus, specic references in the most important literary sources at our disposal regarding the ancient synagoguei.e., rabbinic literatureterminate some time in the latter half of that century. As a result, historiographically as well, the fourth century appears to have been a watershed in the Jewish narrative. This phenomenon is emphasized further by the massive appearance of archaeological evidence in the form of Palestinian synagogues, beginning in the fourth century. Based on the data at hand, it seems that post-70 synagogues were rst constructed in the late third and fourth centuries. Excavations carried out over the past generation indicate that the rst stages of a synagogue edice can be dated no earlier than the mid third century,
5. Goodman, State and Society, 11618; Stemberger, Jews and Christians, 2247, 23097; Linder, Jews in Roman Imperial Legislation, 6778; S. J. D. Cohen, Pagan and Christian Evidence, 17075; Millar, Jews of the Graeco-Roman Diaspora, 97112. 6. L. Levine, Status of the Patriarch, 2832; Stemberger, Jews and Christians, 23068. 7. Ginzberg, On Jewish Law and Lore, 2627; Sussmann, Once Again Yerushalmi Neziqin, 1013 and esp. 13233. See also Strack and Stemberger, Introduction to Talmud and Midrash, 18889.

late roman pa le st i ne


while most fall in the fourth century and later.8 Such is the case, for example, in the Golan at Qatzrin; in the Galilee at orvat Ammudim, Gush alav, Khirbet Shema, Meiron, Nevoraya, ammat Tiberias, ammat Gader, Maoz ayyim, Reov, Bet Shean (north), Chorazim, Baram, and possibly Merot; in the Shephelah at orvat Rimmon; and in Judaea at En Gedi, Eshtemoa, orvat Anim, Maon, and Susiya.9 The fth century, with but few exceptions (e.g., Capernaum and Sepphoris), is less well attested with regard to building activity. Few sites oer a rm dating for this period, partly because of the uncertainty of the numismatic and ceramic evidence. A renewed surge in synagogue building is in evidence for the sixth century.10 Of no less import is the fact that the rich artistic and epigraphic material from the Palestinian synagogues dates almost entirely from the fourth century onward. In short, the information to be garnered from the above elds of researcharchitecture, art, and epigraphyis far more detailed and comprehensive in the Byzantine era (fourth to seventh centuries) than for the later Roman period (i.e., the second to early fourth centuries), thus warranting a separate treatment. Our focus in this and the next chapter is the synagogue of Late Antique Palestine. The extensive literary evidence for this institution dwarfs Diaspora-related material for Late Antiquity; the hundreds of literary references to the synagogue as an institution, together with the many hundreds, if not thousands, of sources in Palestinian rabbinic legal and homiletic works referring to its liturgical functions, far outweigh comparable references to the Diaspora synagogue, Babylonia included. The archaeological material from Roman-Byzantine Palestine is likewise predominant, though to a somewhat lesser degree. To date, remains of well over 100 buildings throughout Roman Palestine have been identied, with some estimates going as high as 180 (g. 14).11 With regard to the epigraphical evidence, Palestinian synagogue remains have yielded some 130 inscriptions in Aramaic
8. The recently published synagogue from orvat Sumaqa on the Carmel has been dated by Dar to the later second or early third century, although the evidence for this dating is thin, per the excavators own admission; see Dar, Sumaqa, 73; Dar and Mintzer, Synagogue of Horvat Sumaqa, 162. In the same vein, Maoz has revived the Kohl-Watzinger position, dating all Galilean synagogues to the era of R. Judah I in the early third century (When Were the Galilean Synagogues First Constructed? 416 26), yet oers no new solid evidence for his view. 9. Groh, Stratigraphic Chronology of the Galilean Synagogue, 6069; Foerster, Dating Synagogues, 8794. See also Bloedhorn (Capitals of the Synagogue of Capernaum, 4954), who suggests a late third-century date for the early stage of the Capernaum synagogue, as does Fischer (Das korinthische Kapitell ). On the issues surrounding the dating of Galilean synagogues and several recent suggestions for an early third-century dating of Galilean synagogues (Dar, Maoz), see below and Chap. 9. 10. On the above sites, as well as others referred to below, seeunless otherwise indicatedthe relevant entries in E. Stern, NEAEHL, to be supplemented now by Amit, Synagogues, passim. On renewed synagogue construction in the sixth century, see Foerster, Basilical Plan as a Chronological Criterion, 17379; idem, Dating Synagogues, 92; and Yeivin, Synagogue at Korazim, 53*54*. 11. For the higher estimate, see Z. Ilan, Ancient Synagogues.

14. Map of excavated synagogues in Roman-Byzantine Palestine.

late roman pa le st i ne


and Hebrew and about 70 in Greek, as compared to 13 buildings and over 200 inscriptions from the Diaspora.12 To the Diaspora gure we should add well over 100 funerary inscriptions, deriving mostly from the catacombs of Rome, which refer to deceased synagogue ocials.


In addressing the history of the synagogue in late Roman Palestine, we are confronted by a number of baing issues. The literary and archaeological sources at our disposal are problematic in their own rightbeing both plentiful and scanty, in evidence for one period and absent for another, at times corroborating each other and on occasion at striking variance. Let us consider these sources one by one.

Rabbinic MaterialAbundance and Scarcity

For all the richness of rabbinic material regarding the synagogue, it is devoid of any overall picture as an institution. Instead, these sources focus mainly on the liturgical components of synagogue (and private) worship, i.e., key prayers and the Torah reading, as well as the use of the building as a place of study. Information relating to other aspects of the institutioncommunal functions, synagogue ocials, benefactors, interior and exterior design, furnishings, relatively sparse and usually noted only en passant. Given the institutions centrality in the post-70 era, one might expect it to have received great attention in rabbinic writings. The Temple, for example, merited a mishnaic tractate on its physical dimensions alone, in addition to other tractates or parts thereof that describe ceremonies and rituals conducted therein.13 Moreover, the Sanhedrin, another pre-70 institution, was accorded an entire tractate although it was not a Pharisaic institution.14 Even a facility such as the miqveh, which played a signicant role in Temple
12. Regarding epigraphical remains from Palestine, see Naveh, On Stone and Mosaic. Roth-Gerson lists thirty inscriptions, but does not include in her calculations sixteen inscriptions from the Beth Shearim synagogue (Greek Inscriptions, 13444; Schwabe and Lifshitz, Beth Shearim, II, 9094) or three Greek inscriptions from ammat Gader (Roth-Gerson, op. cit., 13133). She also refers to the eight dedicatory inscriptions found at the northern end of the ammat Tiberias synagogue as if they were one. Finally, nine Greek dedicatory inscriptions have been discovered in the Sepphoris synagogue (Di Segni, Greek Inscriptions, 20916) and several more, as yet unpublished, in Tiberias (forthcoming). 13. Aside from Middot describing the Temple and Temple Mount, the rest of the order of Qodashim (another ten tractates) is devoted to Temple-related matters. In addition, many tractates in the orders of Moed (e.g., Yoma, Sukkah, Rosh Hashanah, Sheqalim, Pesaim), and Zeraim (Bikkurim) focus in varying degrees on Temple ritual. 14. On the diculties in dening the pre-70 Sanhedrin, its activities, composition, and authority, see Mantel, Studies, 54101; Efron, Studies, 287338; Bickerman, On the Sanhedrin, 35659; L. Levine, Judaism and Hellenism, 8790; idem, Rabbinic Class, 7683; and Goodblatt, Monarchic Principle, 10330.


historical development of the synagogue

and sectarian ritual in the pre-70 era, merited tractates in the Mishnah and Tosefta, although not in either talmud.15 Thus, for all its centrality in Jewish life following the destruction of the Temple, the synagogue per se merits minimal attention in the Mishnah and in tannaitic midrashim, though somewhat more in the Tosefta, the two talmuds, and later midrashim. Prayer (Tractate Berakhot) and, to a lesser extent, the Torah and haftarah readings (Tractate Megillah) are discussed more fully than other worship activities, but even here the former often does not necessarily refer to a synagogue setting. In Mishnah Berakhot, for example, the synagogue is never mentioned in relation to the two main prayers (the Shema and Amidah) that are discussed in the tractates rst ve chapters. The prayer settings noted include riding on an animal, traveling on a ship, and working in a tree or on a wall.16 The prescriptions in rabbinic literature regarding prayers may equally have applied to settings such as an academy, a private home, an open square, or a eld. Indeed, obligatory prayer at this time may have been, to a large degree, personal. While the apparent lack of attention given to the synagogue in rabbinic literature is rather surprising, upon further consideration the phenomenon should not be regarded as all that strange. Bearing in mind that rabbinic literature was never intended to be a balanced documentation of Jewish life in antiquity, we may regain some perspective not only on how to evaluate what has been preserved, but alsoand no less importanthow to judge what, in fact, has been omitted. Rabbinic literature is, rst and foremost, an account of laws, homilies, and stories that the rabbis deemed important to discuss and transmit and that later generations of sages saw t to preserve. From this mass of material, one can cull a great deal of information about contemporary Jewish society and even about certain aspects of the Roman world generally, though this clearly was not the intention of the editors of these compilations. If information extraneous to the rabbinic agenda does, in fact, appear, it is usually because it was in some way linked to a halakhic or homiletic context, or a rabbinically focused anecdote. What has been omitted in this literature regarding Jewish and even rabbinic society far outweighs what has been preserveda truism, indeed, but one that must always be borne in mind. Moreover, even what has been preserved should be treated with a measure of caution. This material has been ltered through many hands before reaching its nal form in a particular compilation, and it often reects the worldview and value system of those who transmitted and edited the material.17
15. See Reich, Synagogue and the Miqweh, 28997. 16. These examples raise an important question regarding the degree to which mishnaic rulings represent a systematic treatment of a given subject and thus deal with fundamental issues or the degree to which they deal with exceptions to the rule and problematic situations. It would appear that both dimensions are in evidence with regard to prayer. The presentation of the principal prayers in the rst part of Berakhot is general and universally prescriptive, although some pericopes may indeed deal with very specic and problematic issues whose more general applicability can only be guessed. 17. For various perspectives on the historical value of rabbinic literature and how this corpus should

late roman pa le st i ne


It is clear, therefore, that the rabbis were not interested in reporting on the synagogue per se, nor, for that matter, on any other communal framework. Regarding pre-70 institutions, except for the Temple, what little has been preserved are merely random traditions, some more idealized and tendentious than others, that the sages considered to be of signicance. Perhaps some of this material was intended to serve as a blueprint for the future. However, even with respect to their own day, i.e., the second to fourth centuries c.e., rabbinic sources regarding non-rabbinic frameworks are of limited value. We know very little about Roman and Jewish municipal institutions,18 and nothing systematic has been recorded regarding schools and charitable frameworks of the Jewish community two institutions certainly of interest to the sages. What does exist is at best anecdotal and fragmentary.19 Even as regards the rabbinic academy, an institution of prime importance and concern to the sages, our information happens to be even more limited than that relating to the synagogue.20 We know next to nothing as to where rabbinic academies were located, how they looked inside and out, how they functioned, who was in charge, how learning frameworks were organized, who the students were, and what role, if any, the institution played in the larger community. Nevertheless, for all the limitations in the nature and variety of sources on the synagogue, there is a real distinction to be made between second-century tannaitic material on the one hand and that of the third- and fourth-century amoraic sources on the other.21 The former corpus is generally much more restricted in scope, and its agenda is xed by either the halakhic emphasis and parameters set by R. Judah I in his Mishnah, on which the Tosefta expands greatly, or the biblical text that the tannaitic midrashim address.22 Amoraic material, which includes the talmuds as well as aggadic midrashim, is far richer in evidence regarding the synagogue.23 This increase in material may not be coincidental. For one, it reects a very profound transformation that was taking place within rabbinic circles in the third and fourth centuries. More rabbis were now involved in issues and institutions relating to the Jewish
be used, see, inter alia, Herr, Conception of History, 13242; Neusner, Formative Judaism, 99144; Green, Context and Meaning, 97111; Goodblatt, Towards a Rehabilitation of Talmudic History, 31 44; Goldenberg, History and Ideology, 15971; L. Levine, Rabbinic Class, 1622; Gafni, Concepts of Periodization and Causality, 2138; and Shapira, Deposition of Rabban Gamaliel, 538. 18. Juster, Les Juifs, I, 43856; II, 243.; Baron, Jewish Community, I, 13340. 19. Baron, Jewish Community, I, 12426, 13031; S. Safrai, Education, 94570; Bergman, Charity in Israel, 1338. 20. Baron, Jewish Community, I, 15055; Urman, House of Assembly and the House of Study, 238 55. See also idem, On the Question of the Location of the Academy, 16372. 21. On this distinction and its ramications, see Strack and Stemberger, Introduction to Talmud and Midrash; EJ, II, 86575; XV, 798803; Halivni, Midrash, Mishnah and Gemara, 3875; Ab. Goldberg, Mishna, 21151; idem, Palestinian Talmud, 30319. 22. B. Cohen, Mishnah and Tosefta, 3758; EJ, XII, 93110; Ab. Goldberg, Tosefta, 283301. 23. Albeck, Introduction to the Mishnah, 79143.


historical development of the synagogue

community generally and, as a result, became more responsive to its needs.24 This attitudinal and behavioral change on the part of the sages, responding here to a changing socioeconomic reality, is reected in a statement made by the third-century R. Isaac with regard to the teaching required of sages in his generation: At rst, when money was available [lit., a perutah was to be found], one would desire to study Mishnah and Talmud [i.e., halakhic material]; now that money is not available, and, what is more, we suer from the [gentile] kingdoms, one desires to hear Bible or aggadic teachings. 25 In this light, then, it is not surprising that we nd much more material relating to various facets of the synagogue as an institution included in later sources. However, there may have been another factor at play as well, and that is the increased rabbinic interest and involvement in the Late Antique Palestinian synagogue as it gradually acquired a more salient religious prole. Thus, the relative absence of second-century synagogue material in rabbinic literature may not reect the institutions minor importance at that time, but rather the relative indierence of contemporary rabbinic attitudes toward the synagogue and the rabbis circumscribed involvement in it; these factors determined the limited selection of synagogue-related material by tannaitic editors in the early third century.

Post-70 Synagogues: The Archaeological Evidence?

The archaeological issues concerning synagogues in this period are as fascinating as they are perplexing. To date, there is meager evidence, at best, of synagogue buildings for almost two centuries following the destruction of the Temple. The assumption made by earlier archaeologists, that many of the monumental Galilean synagogue buildings are
24. I have written elsewhere about the greater involvement of the rabbis in the Jewish community in the third and fourth centuries. The rabbis of Yavneh inherited a basically sectarian tradition, which was gradually transformed only in subsequent generations. Throughout much of the second century, the rabbis involvement in the larger society, their acceptance of other forms of behavior, and their involvement with other institutions were generally quite limited. A watershed was reached in the lifetime of R. Judah I, whose inuence on Jewish society generally and the rabbis in particular was profound. The rabbis became more and more involved with their surroundings, their academies became xed institutions in various locales and were not dependent on any one particular sage, and more rabbis began undertaking communal tasks and relating to the public at large. Certain classic sectarian interestsstrict observance of tithing and sabbatical year laws, hostility toward other Jews with dierent types or degrees of observance, and disdain for the less educatedwere modied by the new reality. Within this context, therefore, it is not at all unexpected that some rabbis, at least, would begin addressing communal issues more earnestly and that the editors of rabbinic compositions would subsequently choose to include some of this material. The increased data available from the third and fourth centuries should be understood against this backdrop. See L. Levine, Rabbinic Class, 2342; idem, Sages and the Synagogue; S. J. D. Cohen, Place of the Rabbi, 15773; idem, Rabbi in Second-Century Jewish Society, 92290; Goodman, State and Society, 93118; and below, Chap. 13. 25. PRK 12, 3 (p. 205).

late roman pa le st i ne


to be dated to the period of R. Judah the Prince and the Severan dynasty, has yet to be stratigraphically substantiated.26 With but one or two possible exceptions, none of the scores of post-70 buildings excavated to date stems from this period,27 and this applies to archaeological data from all parts of Roman Palestine: the Galilee and Golan, the Bet Shean region, the Jordan Valley, southern Judaea, and the coastal plain. No less striking than the absence of archaeological evidence following the destruction in 70 is the burst of synagogue building in the Galilee and elsewhere from the mid third century onward.28 As noted, excavations over the past generation have indicated that numerous synagogues were erected at this time. What is particularly unusual about this phenomenon is that it seems to have occurred when least expected. Given the heretofore generally accepted picture of the third century as a time of economic, social, and political instability, one would not have anticipated so intense a period of synagogue building.29 In fact, MacMullen has claimed that it is precisely at this time that pagan culture was on the wane, owing in part to the inability or unwillingness of wealthy benefactors to subsidize the pomp, performances, sacrices, and public feasting that had been the norm for
26. This date was already posited in the monumental study by Kohl and Watzinger (Antike Synagogen, 20418) and became axiomatic among archaeologists through the writings of Sukenik (Ancient Synagogues, 68) and Avi-Yonah (Synagogue Architecture, 6773). See also Avigad and Foerster (in L. Levine, Ancient Synagogues Revealed, 4243 and 4748, respectively), Httenmeister and Reeg (Antiken Synagogen, I, viiiix), and Tsafrir (On the Source of Architectural Design, 7079), who tend to prefer a thirdcentury dating. As noted above, Maoz (When Were the Galilean Synagogues First Constructed? 416 26) has attempted to revive the Kohl and Watzinger thesis, but cites no new hard evidence. 27. The earliest stratum at Nevoraya (north of Safed) is dated by E. M. Meyers (Second Preliminary Report on Excavations at en-Nabratein, 40; E. M. Meyers et al., Nabratein, 107778; Torah Shrine, 31718) to the mid second century. However, these remains are fragmentary and the published report is preliminary; the evidence has yet to be presented fully and systematically. Thus, there is some hesitation to view this phase as denitive evidence for a second-century Galilean synagogue. Another possible candidate for a second-century synagogue is a public building that was erected by the Jews of Qatzion, in the eastern Upper Galilee, in honor of Septimius Severus and his sons in 197 c.e. The nature of this building, however, remains a mystery. The preserved dedicatory inscription in stone may have once been axed to a synagogue, but it also could have belonged to some other public building; see Roth-Gerson, Greek Inscriptions, 12529. For a second- to third-century dating regarding the orvat Sumaqa synagogue, see Dar and Mintzer, Synagogue of orvat Sumaqa, 162; Dar, Sumaqa, 73. The synagogues of Qiryat Sefer and Modiin that date from the rst and early second centuries should be regarded as continuations of pre-70 Judaean structures; see above, Chap. 3. 28. So, for example, at Gush alav, Nevoraya, Khirbet Shema, Meiron, orvat Ammudim, and En Gedi. A similar third century and later dating occurs at this time with respect to the Jewish catacombs in Rome. Those discovered date from the late second through fth centuries. Where, then, are the remains of Jews from the rst century b.c.e. or the rst and second centuries c.e.? And why were the existent catacombs of Late Antiquity in use precisely at this time? No convincing answers to these questions have been forthcoming to date. See Rutgers, Jews in Late Ancient Rome, 9699, 26768. 29. A classic description of this era, assuming it to be one of instability and anarchy, can be found in Avi-Yonah, Jews of Palestine, 89114.


historical development of the synagogue

centuries. Such contributions were essential for the maintenance of pagan temple ritual. Third-century instability claimed a heavy toll, and across the Empire, whether in Italy, Asia Minor, or Syria, this century has invariably been characterized as one of decline and decay.30 Let us address each of the above two issues separately. The fact that practically no synagogue remains have been found for some two centuries after 70 is indeed surprising. It has been suggested that the absence of synagogues may have been caused by their destruction in the wake of the various revolts throughout Palestine between 66 and 135 c.e., especially that of Bar-Kokhba (132135 c.e.). We have already taken note of Malalas report of Vespasians (perhaps Titus) conversion of the synagogues of Caesarea and Daphne into an odeum and theater respectively, and to this we might also add the destruction of a major Alexandrian synagogue as a result of the 115117 Diaspora revolt.31 Rabbinic literature knows of one such case in Roman Palestine. The following is recorded in the Bavli:
R. Ami and R. Asi came to visit him [R. Yoanan, and] said: Was there not an incident in the Tiberian synagogue [regarding] a lock with a fastening device [ ]at its tip that was a subject of dispute between R. Elazar and R. Yosi [second century] until they tore a Torah scroll in their anger? Do you actually imagine [that they, in fact, tore a scroll]? Rather, say: A Torah scroll was torn in their anger. And R. Yosi b. Qisma was there [and] said: I would be surprised [i.e., I predict] if this synagogue [where such an incident occurred] did not become [a place of ] idolatry. And so it happened. 32

Despite the chronological issues connected with this source, as well as its citation only in the Bavli, there may be embedded here some memory (following R. Yosi b. Qismas statement) of a Tiberian synagogue that was converted into a pagan temple some time in the early second century. Might this possibly be a reference to the building of a Hadrianeum in Tiberias, as reported by Epiphanius? 33 If so, then this source would be an indication of the destruction of a second-century synagogue or its conversion into a pagan temple. The issue must remain moot for the present. Regarding the post-135 Hadrianic persecutions, there is no way of verifying whether these decrees included the obliteration of existing synagogues; no substantiation of such a claim exists in either rabbinic or archaeological material, and no other source mentions

30. MacMullen, Paganism, 12730; see also, for example, Meiggs, Roman Ostia, 8385 (deterioration and decline); Magie, Roman Rule in Asia Minor, 688723 (decay and chaos); Downey, History of Antioch, 25253 (strife, uncertainty, a disorderly and dicult period). On the social-religious crisis in Egyptian villages that gave rise to the monastic movement, see Brown, Making of Late Antiquity, 81. 31. On Malalas, see above, Chap. 4, notes 229, 232, and 233. On the destruction of Alexandrian synagogue(s) as a result of the revolt in 115117 c.e., see Y Sukkah 5, 1, 55b; Tcherikover et al., CPJ, I, 93. 32. B Yevamot 96b, and variations in Y Sheqalim 2, 7, 47a, as well as the comments in Eliav, Sites, Institutions and Daily Life, 6871. On the word ,see Jastrow, Dictionary, loc. cit. 33. Epiphanius, Panarion, 30, 12.

late roman pa le st i ne


or even alludes to the closing or dismantling of synagogues;34 furthermore, no archaeological excavation has uncovered a destruction layer clearly related to the second century. Even if we were to ignore this negative evidence, there is little to substantiate the claim that the Bar-Kokhba revolt or its aftermath aected the Galilee in any measurable way.35 Thus, the Hadrianic persecutionswhile perhaps explaining the lack of evidence in several localesclearly do not account for the absence of remains throughout the country. A second theory attributes the scarcity of archaeological remains for this period to the fact that the synagogue, like the contemporary church, was not an identiable public building, but rather a private home, a domus ecclesiae,36 and hence the diculty in identifying such remains as synagogues. However, to compare the synagogue with the domus ecclesiae is to assume that the congregation invariably met in an ordinary home. While this may have been the case at times in the Diaspora, it would contradict everything we know about Palestinian synagogues. On the one hand, all rst-century synagogues were clearly communal buildings, constructed (or restored) as separate edices with columns as well as benches on three or four sides, as found at Gamla, Masada, Herodium, Qiryat Sefer, and Modiin. Literary sources likewise conrm the public character of synagogue buildings. At Caesarea, Dor, Tiberias, and Jerusalem, the synagogues noted appear to have been separate, identiable buildings. On the other hand, synagogues from the late third century onward were also public buildings with a presence and prominence analogous, in large measure, to those of the rst century. Rabbinic sources as well as archaeological nds conrm that the synagogue was a public edice (see below). Moreover, there is no reason to assume that synagogues went underground from the late rst to third centuries and were ensconced in private dwellings, as the church was forced to do owing to its status as a religio illicita. A hiatus of approximately two hundred years in the building of public Jewish communal institutions appears most unlikely. Indeed, as noted, rabbinic literature indicates just the opposite. The evidence points to the continued centrality of the synagogue as a multifaceted communal institution.
34. On the Hadrianic persecutions following the Bar-Kokhba revolt, see Lieberman, Persecution of the Jews, 21345; Herr, From the Destruction of the Temple, 36567; Mor, Bar-Kochba Revolt, 23840. The comments of R. Judah b. Ilai (mid second century c.e.) regarding what is to be done with a synagogue that has been destroyed may be relevant here (M Megillah 3, 3); however, this comment stands alone and indeed may be only theoretical. 35. Goodman, State and Society, 13738; Mor, Bar-Kochba Revolt, 10321; Oppenheimer, Galilee, 3744; and the discussion of this issue in Oppenheimer et al., Jewish Community in the Galilee, 5183. 36. Tsafrir, On the Source of Architectural Design, 7980. The term domus ecclesiae in this context is somewhat problematic. It may be understood as a religious community building, much as was found at Dura or Capernaum, or it may refer to a house church that did not entail interior alterations for use by the community and was thus architecturally indistinguishable from other private buildings. The Dura church is thus not a domus ecclesiae, but rather a private dwelling that was converted into an identiable church building; see White, Social Origins, 25. On the conversion of private homes into synagogues in many Diaspora communities, see idem, Building Gods House, 6277.


historical development of the synagogue

A third approach, advocated by Meyers and Strange, suggests that the absence of synagogues was due to the fact that the inux of Judaeans to the Galilee took place only in the late rst and second centuries. Since these people acquired the means to construct large buildings over the course of the next century, it was only then that the earliest Galilean synagogues were erected.37 This explanation assumes that synagogue building was a distinctly Judaean phenomenon and that the Galileans themselves were not interested, willing, or capable of undertaking such projects. Such an assumption is wholly gratuitous, as we have seen above (Chap. 3), for the Galilee boasted synagogues as early as the rst century c.e. A fourth explanation, however, may be the most plausible. At site after site, in cities or villages, public or private buildings, the extensive rebuilding activity from Late Antiquity or, for that matter, from any period thereafter has almost entirely obliterated earlier remains. For example, our knowledge of pre-Severan Sebaste, Bet Shean, or Gerasa, and to a lesser extent pre-Byzantine Caesarea, is limited and fragmentary.38 As regards Second Temple Jerusalem, the extensive and monumental building program carried out by Herod and his successors during the last one hundred years of this era almost entirely obliterated earlier remains. Given the large-scale construction of synagogues and other structures in Late Antiquity, a similar situation may have held true for earlier synagogue structures as well. Only the synagogue remains at Nevoraya and perhaps Capernaum (at least as interpreted by the Franciscan excavators) may possibly be construed as examples of later structures replacing earlier ones lying beneath them. In other words, the absence of second- and third-century remains may have been the result of their elimination by later construction.39 Nevertheless, despite the lacunae in our sources, both literary and archaeological, the synagogue undoubtedly continued to function as a central communal institution in Jewish communities everywhere after 70 and continued to do so throughout the ensuing centuries. Moreover, the centrality and prominence of the synagogue in the rst century are matched by a similar status in the third and fourth centuries (albeit with an enhanced religious component); there was hardly time for dramatic changes in such a relatively short interim. The few tannaitic sources we have are too explicit in this regard for us to assume otherwise. Take, for example, the revealing exchange between two rabbis in Lydda in the rst half of the third century c.e.:

37. E. M. Meyers and Strange, Archaeology, the Rabbis, and Early Christianity, 141. 38. Interestingly, the same holds true for Diaspora synagogues. Those of Late Antiquity cannot be traced before the second century or third centuries c.e. Regarding rst-century Judaea, see E. P. Sanders, Historical Figure, 100, and above, Chap. 3. 39. On a possible reference to a second-century synagogue in En Gedi, see Lewis, Documents from the Bar Kokhba Period, 8485. My thanks to Prof. H. Eshel for calling my attention to this document. On the use of spolia from earlier buildings in constructing later ones, see below, Chap. 9.

late roman pa le st i ne


And R. ama bar anina and R. Hoshaya were walking among the synagogues of Lydda. R. ama bar anina said to R. Hoshaya: How much money have my ancestors [lit., my fathers] invested here [in these buildings]? The other responded: How many souls have your ancestors lost here [lit., have they sunk here]? There are no people to study Torah! 40

Here, then, we have two contrasting rabbinical reactions to the impressive synagogue buildings in early third-century Lydda: one takes pride in the material remains; the other is sharply critical of this use of communal funds. Of interest to us is the fact that synagogue structures were standing at this time, and it is clear from the exchange that we are dealing with imposing edices. Moreover, this report is far from sui generis; other sources report many synagogue buildings in third-century Palestinee.g., the reputed eighteen synagogues in Sepphoris and its environs at the time of Rabbi Judah Is funeral (ca. 225), or the thirteen structures in Tiberias (see below). Broader contextual circumstances would also argue for a similar conclusion, i.e., an increased role, prestige, and sanctity for the synagogue following the Temples disappearance. Thus, any attempt to base far-reaching conclusions regarding the nature and importance of the synagogue in the post-70 period based on the presence or absence of archaeological material is fraught with danger. Only an assessment that balances the archaeological and literary remains within the historical context of the late Roman era can hope to achieve some degree of credibility.

Synagogue Building in the Mid Third Century

Just as the absence of archaeological material in the post-70 era raises perplexing questions, so does the unexpected reappearance of evidence in the third century.41 Prior to the wave of excavations over the past few decades, the mid to late third century would not have seemed a likely time for such a spurt in synagogue building. Nevertheless, the archaeological evidence exists and continues to grow. In addition, there are several rabbinic traditions from precisely this time that attest to contemporary synagogue building:
The people of Bet Shean asked R. Ami [late third century]: Is it permissible to take stones from one synagogue in order to build another? He said to them: It is forbidden. R. elbo said: R. Ami only forbade this because of the sadness [that would be caused]. R. Gorion said: The people of Migdal [north of Tiberias] asked R. Simeon b. Laqish [mid third century; died ca. 280]: Is it permissible to take stones [of a synagogue] in one city and build in another? He said to them: It is forbidden. R. Ami had taught: Even from

40. Y Sheqalim 5, 6, 49b; Y Peah 8, 9, 21b. On the town of Lydda in this period, see J. J. Schwartz, Jewish Settlement in Judaea, 6980; Oppenheimer, Jewish Lydda, 11536; and Rosenfeld, Lod and Its Sages, 71. 41. So, for example, ammat Tiberias (IIb); Khirbet Shema, Nevoraya, Meiron, Gush alav, Bet Shearim, orvat Rimmon, orvat Ammudim, Maoz ayyim, ammat Gader, and En Gedi. Several of the above were excavated decades ago, rendering their dating insecure.


historical development of the synagogue

the eastern to the western [side of the city] it is forbidden because of the destruction [that would be evident] at that site [where a synagogue had once stood]. 42

These traditions are indeed unique. Only for this periodand only in these pericopes does Palestinian rabbinic literature exhibit a concern about rebuilding or relocating synagogue buildings.43 For our purposes, these sources from the mid to late third century are relevant in that they attest to actual cases of Jewish communities interested in building new synagogues with the remains of old ones. Is it merely coincidental that these particular sources were preserved, and were there, in fact, other instances in Late Antiquity that simply do not appear in rabbinic literature? Or, alternatively, are these sources reective of a unique situation in third-century Palestine, in which older synagogues were being abandoned and new ones required elsewhere? If the latter is the case, then we might ask why this need emerged at this particular time? Was it merely because populations were relocated to the Galilee at this time? 44 Had the local community grown and now required larger facilities? Was it a question of urban renewal and local shifts of population? Or was there some sort of destruction or upheaval that eliminated the older buildings? The last alternative appears improbable, as we have no evidence of persecution or destruction from the third century.45 Whatever the specic reasons for rebuilding, rabbinic literature fully corroborates the archaeological evidence that construction was being undertaken at this time. An interesting parallel to this Jewish phenomenon may be found in extant Samaritan traditions that speak of a major religious revival in the third century c.e. under the leadership of the legendary Baba Rabba. The Samaritan chronicles relate a series of reforms instituted under his aegis, including the reorganization of Samaritan leadership around an appointed council of seven elders (three priests and four laymen) and the establishment of a system of regional ocials in which one layman and one priest were assigned to each district.46 The cornerstone of this reformation was the synagogue (g. 15).47 Baba
42. Y Megillah 3, 1, 73d. 43. The only other comparable traditions come from third- and fourth-century Babylonia and involve R. isda, R. Papa, and R. Huna; see B Megillah 26b. 44. According to E. M. Meyers suggestion regarding the inux of Judaean immigrants into the Galilee following the two Jewish revolts (Galilean Regionalism, 99). 45. On third-century conditions in Palestine, see Lieberman, Palestine in the Third and Fourth Centuries, 32944 (= Texts and Studies, 11227); Avi-Yonah, Jews of Palestine, 35136. 46. Abu l-fath, in Samaritan Documents (pp. 14246); Kitab al-Tarikh (pp. 17786); Chronicle II, in Samaritan Chronicle (pp. 6576); Adler, Une Nouvelle Chronique, 8796. On these reforms generally, see Samaritan Chronicle (pp. 22838); Crown, Byzantine and Moslem Period, 5859; idem, Samaritans in the Byzantine Orbit, 11011; idem, Samaritan Religion, 3234, 4143; Kippenberg, Die Synagoge, 35160. On the reforms in Samaritan leadership, see Samaritan Chronicle (pp. 6674); Kitab al-Tarikh (pp. 18485 [see also p. 181]); Samaritan Documents (pp. 14345 and n. 170). On Samaritan synagogues generally, see Pummer, Samaritan Synagogues and Jewish Synagogues, 11860. 47. Samaritan Documents (pp. 142, 146); Kitab al-Tarikh (pp. 177, 183); Samaritan Chronicle (pp. 65, 71 72).

late roman pa le st i ne


[To view this image, refer to the print version of this title.]

15. Torah shrine from the mosaic oor of the Samaritan synagogue at Khirbet Samara.

Rabba allegedly reopened numerous synagogues that had been closed, rst by the emperor Commodus (180192 c.e.) and later by Alexander Severus (222235 c.e.).48 Moreover, he is said to have built a large synagogue on Mt. Gerizim, opposite the site of the former Samaritan temple, and next to a miqveh:
Later, the priest Baba Rabba built a ritual bath for purication on the boundary of the chosen place, Mount Gerizim Beth-El, the Mountain of Inheritance and Divine Presence. . . . He further built a Synagogue, adjoining the Chosen Place, Mount Gerizim Beth-El, so that the people could pray in it opposite this holy mountain.49

Baba Rabba also built eight additional synagogues in various Samaritan villages, seven of which are specically mentioned.50 Most revealing, however, is the following informa48. Adler, Une Nouvelle Chronique, 8587; Kitab al-Tarikh (pp. 16672). 49. Samaritan Chronicle (p. 71). According to Abu l-fath: On the periphery of the Holy Mountain Baba Rabba built a water pool for purication at prayer times, that is, before the rising of the sun and its setting. And he erected a prayer house for the people to pray in, opposite the Holy Mountain; see Kitab al-Tarikh (pp. 18283); see also Samaritan Documents (pp. 14546). 50. Samaritan Documents (p. 146); Kitab al-Tarikh (p. 183); Samaritan Chronicle (pp. 7172). These synagogues are said to have been built of stone (it is specically mentioned that no timber was used), and all had earthen oors. Chronicle II mentions the names of seven of the eight synagogues with earthen


historical development of the synagogue

tion: He built in them a place in which to read and to interpret, and to hear petitions in the southern part of the house of prayer, so that anyone with a personal problem could ask the ukama [i.e., the sage] about it and be given a sound answer. 51 Thus, we learn here of a school combined with a setting for the ukama to deal with issues brought to him by the people, a phenomenon not unfamiliar to the contemporary Jewish synagogue. Few scholars have reservations regarding the historical basis of Baba Rabbas reforms. However, a vigorously debated issue of late is that of dating, and the diversity of opinion in this regard has been fueled by the ambiguity and contradictory nature of the sources themselves. While the various chronicles clearly place Baba Rabbas activity in the mid third century, under the emperors Alexander Severus, Gordianus, and Philip, they also note the existence of Constantinople at the time as well as the fact that Baba Rabba was incarcerated and died there. This would, of course, bring us to the fourth century, for Constantinople was built in the 320s and dedicated by Constantine in 330. Moreover, several names mentioned in connection with Baba Rabba have been arguably identied with fourth-century gures. It is little wonder, then, that the dating of Baba Rabbas activity has been the subject of much controversy. Early in the last century, Montgomery advocated a fourth-century setting, more specically the era of Constantius (337361), and many have since followed his lead.52 In his edition of Chronicle II, J. M. Cohen has suggested that Baba Rabba lived around the turn of the fourth century and was publicly active for a twenty-year period, ca. 308328.53 Of late, however, a number of scholars, including Crown, Hall, and Stenhouse, have opted for the traditional Samaritan dating, assuming a mid third-century setting for these reformations.54 If this last suggestion is accepted, then Samaritan synagogue building would constitute an interesting chronological parallel to the appearance
oors, Abu l-fath the names of all eight. Magen, however, has suggested that these particular sites may, in fact, date to the thirteenth century, when these chronicles were composed (Samaritan Synagogues [Eng.], 22627). On the archaeological discoveries of Samaritan synagogues, see ibid., 193230; idem, Samaritan Synagogues (Heb.), 6690; idem, Samaritan Synagogues and Their Liturgy, 22964. The recently discovered Samaritan synagogues at el-Khirbe and Khirbet Samara have been tentatively dated to the fourth century c.e. See also Pummer, How to Tell a Samaritan Synagogue, 2435; Z. Safrai, Samaritan Synagogues, 84112. 51. Kitab al-Tarikh (p. 183). See also Samaritan Documents (p. 146). 52. Montgomery, Samaritans, 1014. See also Kippenberg, Die Synagoge, 351; Samaritan Documents (pp. 197201 nn. 192, 204); Broadie, Samaritan Philosophy, 2; Isser, Dositheans, 43, 95; MacDonald, Theology of the Samaritans, 26; Pummer, Samaritans, 4; Httenmeister and Reeg, Antiken Synagogen, II, 534; and most recently Magen, Samaritans in the Roman-Byzantine Period, 22122. 53. Samaritan Chronicle (pp. 22428). 54. Crown, Byzantine and Moslem Period, 56; idem, Samaritans in the Byzantine Orbit, 105108; idem, Samaritan Religion, 3234; Hall, Samaritan History, 5254; idem, Samaritan Religion, 214; Stenhouse, Fourth-Century Date, 31726; idem, Baba Rabba, 32732.

late roman pa le st i ne


of the mid third-century Jewish synagogues, and the two may even be related in some way.55 Returning to the question of how one might explain the resurgence in the building of Jewish synagogues in the latter part of the third century, given what we know of the period generally, our response can be only tentative. One approach would be to discount the literary and numismatic evidence, and the heretofore accepted scholarly opinion regarding a third-century crisis in Palestine. Either the period of anarchy was not as severe as once assumed or, more likely, dierent regions of the Empire may have been variously aected.56 Moreover, not all areas may have suered to the same degree.57 An engaging case has been made of late by Bar that not only was there no third century crisis in Palestine, but that the economy grew steadily from the second century c.e. into the Byzantine period.58 While Bar has undoubtedly brought a welcomed balance to the older picture, his analysis and conclusions may be somewhat overstated. Indeed, there is evidence of decline, and not all the data he cites are accurate. In short, it is doubtful if the picture was as rosy as he would have it, and one should be cautious in summarily dismissing the mid to late third-century sages negative comparisons of conditions in their own day with those that prevailed a generation or two earlier.59 Even if we discount part of their statements as hyperbole, it would be wholly gratuitous and arbitrary to disregard them altogether.60 Moreover, the numismatic as well as other literary evidence cannot be entirely ignored as if there were no instability whatsoever at the time. A second approach toward solving this conundrum is to accept both the historicity of political and economic diculties of the time and the archaeological evidence, but to assign each set of data to a dierent period. The nadir of economic and political conditions in the third quarter of the century is apparently what is reected in rabbinic sources. To55. A number of artistic similarities between Jewish and Samaritan synagogues of the third and fourth centuries have been noted by Amit ( Curtain, 57175), to which one might add the cluster of religious symbols appearing almost simultaneously in the fourth-century synagogues at ammat Tiberias (Jewish) and at el-Khirbe and Khirbet Samara (Samaritan). Crown has suggested a signicant Jewish inuence on Samaritan practice at this time: There are geographical, social and historical reasons for arguing that Jews and Samaritans lived in close symbiosis with a consequence that Rabbinic Judaism was a formative factor in Samaritanism despite Samaritan denials (Samaritan Religion, 35). On a possible Samaritan inuence on the rabbinic listings for names of Jerusalem about this time, see Shinan, Seventy Names of Jerusalem. 56. Av. Cameron, Later Roman Empire, 112; L. Levine, Palestine in the Third Century, 119. and esp. 13643. With regard to Sardis, see White, Social Origins, 317 n. 52. 57. Reservations as to the extent of the third century crisis have been stated in Watson, Aurelian and the Third Century, 120; and M. Grant, Collapse and Recovery of the Roman Empire, passim. 58. Bar, Was There a 3rd-c. Economic Crisis in Palestine? 4354. See also Brown, World of Late Antiquity, 4995. 59. Avi-Yonah, Jews of Palestine, 89.; Sperber, Roman PalestineLand, 1199. 60. See Alfldy, Crisis of the Third Century, 89111.


historical development of the synagogue

ward the end of this century, however, specically in its fourth quarter, during the reign of Diocletian (284311), the economic and political situation of the Empire ameliorated, and the archaeological material thus may reect primarily this period of stabilization. Such a chronological division is certainly possible, but it may be just a little too neat and compartmentalized. Some archaeological and literary evidence for synagogue building seems to point more toward the midand not latethird century, although, admittedly, precision with regard to dating is impossible, while certain literary traditions indicate a continued decline in the late third century, if not into the fourth as well. A third approach would accept the evidence of both types of sources, archaeological and literary, claiming that the building of synagogues was, at least in part, in direct response to the challenges of the third century. Having lost the Temple, Jerusalem, and much of the region of Judaea following the unsuccessful rebellions in the rst and second centuries, and now nding themselves in the throes of political and social instability, not to speak of a growing Christian presence and perhaps a reassertive paganism, many Jewish communities may have sought to reconrm their identity and demonstrate their cohesiveness by erecting communal buildings, at times of monumental proportions.61 Moreover, the breakdown of a political order could have had salubrious economic ramications, allowing people to avoid taxes and other payments and thus direct their monies toward local projects such as the building of synagogues. Renewed building activity in the face of political and economic distress is not an unknown phenomenon in the course of history and can be documented in other thirdcentury locales as well. The drive to build transcended all boundaries, touching upon Jew and non-Jew alike across the Empire. Emperors, urban aristocracy, and villagers from Egypt, Libya, Asia Minor, Numidia, and Gaul undertook building projects at this time.62 The third century witnessed the erection of walls and the refurbishing of cities such as Tiberias and Caesarea,63 and if the above Samaritan evidence is taken into account, then synagogues were also being restored and built anew in Samaria. Christians, too, were engaged at this time in building new churches. Porphyry refers to the great buildings of the Christians, which he says imitate the construction of temples, and Eusebius describes the extensive building activity of Christians throughout the Empire even before the large-scale persecutions of 303. According to him, Christians already began to build large halls to accommodate the masses of people who attended services, even though such ostentatious building might arouse the envy and hostility of their pagan neighbors.64 Lac61. Later on, Jews would express their need for self-identity through the massive use of Jewish symbols, a phenomenon that seems to make its rst appearance at this time. See L. Levine, History and Signicance of the Menorah, 13153. 62. MacMullen, Roman Governments Response, 11920. 63. Tiberias: B Bava Batra 7b. Caesarea: Holum et al., King Herods Dream, 10753; L. Levine, Roman Caesarea, passim. 64. Porphyry, Adversos Christianos, frag. 76; Eusebius, Ecclesiastical History 8, 1, 56.

late roman pa le st i ne


tantius reports that in 303, at the beginning of the persecution of Christians, Diocletian ordered the destruction of a large church on a hill adjacent to his palace in Nicopolis (Asia Minor).65 On the basis of these reports and other evidence, White asserts that the third century witnessed a spate of church building with larger and more formal structures than the domus ecclesiae that he calls aula ecclesiae.66 This last-noted alternative, i.e., synagogue building as an act of rearmation in the face of socio-economic stress and perhaps religious uneasiness, seems to be the most plausible explanation available. Thus, the third century, no less than other eras in the course of history, dees sweeping generalizations that only tend to obliterate cross-currents, obfuscate distinctions, and blur nuances.


In one very fundamental way, the synagogue of Late Antiquity exhibited a marked continuity from the pre-70 era; both literary sources and archaeological data indicate that it served as the central public framework for each Jewish community. Rabbinic literature, for its part, takes note of no other communal institution, and this is true of non-Jewish sources as well. The synagogue and its ocials, including the Patriarch as a supreme authority, are referred to by various fourth-century church fathers as well as by several edicts in the Theodosian Code; contrastingly, the rabbinic academy, along with the rabbis, rarely merit attention in these sources. The synagogues importance is persuasively conveyed by archaeological remains. Throughout late Roman Palestine, communities emphasized this fact by erecting the building in the very center of town and out of all physical proportion to the surrounding structures. This is strikingly evident, for example, at Khirbet Shema, Nevoraya, Qatzrin, and orvat Ammudim.67 Additionally, the synagogues prominence might be expressed topographically by placing it on an articial podium, as at Capernaum,68 high on a hill and overlooking the town, as at Khirbet Shema and Maon ( Judaea), or literally perched at the very peak of a mountain, towering over the village itself, as at Meiron (g. 16). An elaborate facade, as in most Galilean-type synagogues (e.g., Baram), or an imposing interior, as at ammat Tiberias, also gave expression to this prominence.69 A number of rabbinic sources emphasize the synagogues centrality. The Tosefta is especially intriguing in this regard:
65. Lactantius, De mortibus persecutorum 12, 45; see also Millar, Paul of Samosata, 1415. 66. White, Building Gods House, 12339. 67. For these and following sites, see E. Stern, NEAEHL, passim; and Z. Ilan, Ancient Synagogues, passim. 68. The Capernaum synagogue was especially prominent owing to the use of limestone instead of the local basalt. 69. On the variety of synagogue styles in the third and fourth centuries, see below, Chap. 9.


historical development of the synagogue

16. Suggested reconstruction of the Meiron synagogue at the crest of the mountain.

One should not behave lightheartedly in a synagogue. One should not enter them in the heat because of the heat, nor in the cold because of the cold, nor in the rain because of the rain. And one should not eat in them nor drink in them nor sleep in them nor stroll in them, nor just relax [lit., enjoy oneself ] in them, but [one should] read Scriptures and study laws [lit., Mishnah], and engage in midrash [i.e., exegetical commentary] in them. A public eulogy may be delivered therein. R. Judah [b. Ilai] said: The above relates to a standing synagogue, but if destroyed, they are to be left alone, and let grass grow there as a sign of sadness (or despair). 70

This source is often mistakenly quoted as an indication of what did not take place in the synagogue. In reality, however, it indicates that these were recurrent behavioral patterns to which the rabbis objected; whether or not the sages were successful in eecting change in this regard is unknown. As of the time of the above statement at least, these objectionable practices were still very much a part of the synagogue scene and inspired the above apodictic declaration. By focusing on what the rabbis wished to prohibit, we may gain a clearer idea of how, in fact, many of these synagogues actually functioned. Indeed, the institution referred to in the above Tosefta source was a community center. It was a place to gather and socialize (often involving food and drink), conduct business (following one interpretation of the term lightheartedly),71 and take shelter in inclement weather; it was a hostel and a place of relaxation; nally, it was a place for study and a platform for public eulogies. The sages were in agreement about the two last-mentioned functions; however, the community at large apparently recognized the legitimacy of the other activities as well. Thus, the sages made the above declaration in an attempt to preserve the reverence they deemed betting for such an institution. The Tosefta, in fact, describes an institution akin to the synagogue in Jerusalem per the Theodotos inscription; the Sardis and Halicarnassus synagogues per the edicts preserved by Josephus; and
70. T Megillah 2, 18 (p. 353). 71. B Megillah 28b.

late roman pa le st i ne


[To view this image, refer to the print version of this title.]

17. Remains of the main entrance of the Chorazim synagogue.

the Dura Europos, En Gedi, Capernaum, and ammat Tiberias (last stage) synagogues per the archaeological remains. Thus, the above-noted functions seem to have characterized the ancient synagogue throughout antiquity and across all geographical borders.72 A related Palestinian tradition is cited in the Bavli:
It has been taught: R. Ishmael b. Elazar says: The ammei ha-aretz die because of two transgressions: for calling the holy ark [simply] an ark [ ]and for calling the synagogue a bet am [lit., house of people, or a community center].73

This source is indeed extreme in its censure. Ostensibly, the two matters singled out as cardinal sins (that ought to incur death) do not appear to be so severe. While the sages objection to these terms certainly reects his negative attitude toward these practices, the source would seem to describe what appears to have been a widespread reality. The latter term (bet am) indicates that many people related to the synagogue as a community center and not primarily as a religious institution. However, in addition to its communal dimension, a standard feature of all later synagogues that rst made its appearance in third-century structures was the orientation toward Jerusalem.74 Almost all synagogues of the period faced this direction, although the expression of this orientation may have diered from building to building. With the exception of Arbel, Galilean-type synagogues positioned their elaborate and ornate facades toward Jerusalem and the Temple that once stood there (g. 17). The interior was oriented in this direction as well; most of these structures had three rows of columns, on the west, north, and east. Only the southern side of the building, facing Jerusalem, had
72. For somewhat similar roles of the pagan shrine in rural districts of Late Antiquity, see Libanius, Pro templis 30, 8. 73. B Shabbat 32a (see below). 74. See Landsberger, Sacred Direction, 181203; and below, Chap. 9. See also Wilkinson, Orientation, 1630 (to be used with caution). The uniqueness and sanctity of Jerusalem has also found expression in a number of other tannaitic traditions; see, for example, M Kelim 1, 69; M Ketubot 13, 11.


historical development of the synagogue

18. Isometric reconstruction of the Gush alav synagogue. Note the Torah shrine below, to the left of the entrance.

none, as this direction served as the halls focus.75 Other synagogues, at Khirbet Shema and Gush alav, had a bima against the southern wall (g. 18), and still others, as at Arbel, had a niche facing south. Synagogues in the southern part of the country, e.g., Eshtemoa and Susiya, thus faced north. On occasion, the orientation in a given building changed over time. So, for example, the synagogues at Merot, ammat Tiberias, and En Gedi at rst had entrances facing Jerusalem, but later on these were replaced by a bima and the entrances were relocated to another side of the building.76 Orienting oneself toward Jerusalem in prayer, clearly documented already in biblical literature, is emphatically articulated in third-century rabbinic sources.
Those who stand outside Israel must direct their hearts [i.e., face] toward the Land of Israel, as it is written: And they will pray toward their land [II Chr. 6:38]. And those standing in the Land of Israel direct their hearts toward Jerusalem and pray, as it is written: And they shall pray toward this city [ibid.]. Those standing in Jerusalem shall direct their hearts toward the Temple, as it is written: And they shall pray toward this House [ibid., 6:32]. Those standing in the Temple should direct their hearts toward the Holy of Holies and pray, as it is written: And they shall pray toward this place [I Kgs. 8:30]. Thus, those who stand in the north will face south, those who stand in the south will face north, those in the east will face west and those in the west will face east. Thus, all Israel will be praying to the same place.77
75. Foerster, Ancient Synagogues, 291. 76. On these sites, see above, note 67. 77. T Berakhot 3, 1516 (pp. 1516); SifreDeuteronomy 29, 26 (p. 47); Y Berakhot 4, 5, 8bc; B Berakhot 30a; Pesiqta Rabbah 33 (p. 149b), and elsewhere. See also M Berakhot 5, 5; Tanuma, Qedoshim, 10 (p. 39b); and the thorough discussion in Ehrlich, Non-Verbal Language, 6496. This last source empha-

late roman pa le st i ne


The emphasis on physical orientation toward Jerusalem undoubtedly reects profound changes transpiring within the Jewish community and its synagogues in the late Roman period. On one level, it was a powerful statement of religious and ethnic particularism. No longer was the main communal institution to be a neutral gathering place, as were the rst-century Judaean synagogue buildings. The synagogue was now meant to embody historical memories, a communal attachment to the present, and, perhaps, hopes (messianic and other) for the future.78 This physical expression of an attachment to Jerusalem and the Templedespite the fact that each buildings ornamentation and architectural style were almost exclusively borrowed, selected, and adapted from the material culture of the surrounding worldwas expressed rst and foremost via its orientation. However, the synagogues newly articulated orientation was also a function of a more profound change touching upon the very nature of the institution: The synagogues of Roman Palestine began to acquire a distinctly religious prole. While its manifold communal functions continued unabated, the synagogues main assembly hall gradually became something more than a large gathering place, as its liturgical role was now emphasized more than ever before. In some synagogues, this new emphasis found expression only in the outward orientation and in the interior arrangement of columns; in others, a permanent bima, niche, or aedicula was placed in or against the Jerusalem-oriented wall to emphasize the desired liturgical focus.79 Clearly, worship was becoming an increasingly important component of the synagogues activities, as it presumably had been for some time in the Diaspora. By the third century, this new emphasis was receiving an appropriate architectural expression. Indeed, there can be little question that the increasingly dominant religious dimension of the synagogue was the major factor in making the issue of orientation so pronounced in synagogue architecture of Late Antiquity. The Jews now began to imitate the widespread practice of a xed orientation that was customary in pagan and, later, Christian buildings. For the most part, both pagan temples and Christian churches faced east,80 as did the Tabernacle and both Jerusalem Temples for well over one thousand years. The Jews of late Roman Palestine, from the Galilee to southern Judaea (and in the Diaspora),
sizes Jerusalem and the Temple as the center (omphalos) of the world. See Alexander, Jerusalem as the Omphalos, 10419, and esp. 11316. 78. See Gafni, Land, Center and Diaspora, 5873. On the belief that the divine Shekhinah continues to reside in the Temple (as well as rival beliefs), see Ehrlich, Place of the Shekhina, 31529. 79. See Hachlili, Ancient Jewish Art and ArchaeologyIsrael, 16687. See also idem, Niche and the Ark, 4353. It is for this reason, perhaps, that the elders who had previously faced the congregation with their backs to Jerusalem (T Megillah 3, 21 [p. 360]) now also faced in the direction of Jerusalem (Differences in Customs, no. 36 [p. 156]). For the continued development of this emphasis in the subsequent Byzantine period, see below, Chap. 7. 80. Cf., however, the comments in Wilkinson, Orientation, 2629; Herbert, Orientation of Greek Temples, 3134.


historical development of the synagogue

altered this practice completely and dramatically. While possibly inuenced by the earlier Jewish tradition of praying toward Jerusalem,81 the communities also may have done this out of a demonstrative rejection of pagan practice or owing to their strongly felt religious and ethnic identication with the city. The seven-branched menorah as a decorative (symbolic?) element also rst appeared in late third-century synagogues. The large menorah incised on the northern lintel of the Khirbet Shema building is a case in point, as is the menorah inside a wreath in the center of the Nevoraya synagogue lintel. The possibility that an actual three-dimensional menorah existed in some synagogues is indicated by a number of rabbinic sources, the earliest of which is the Tosefta, which states that such an object was, in fact, donated to a synagogue; several note that the legendary emperor Antoninus made such a donation.82 While the latter traditions concerning Antoninus may be of questionable historical value, the example cited by these sources was undoubtedly drawn from the synagogue reality of the time and would thus attest to the presence of menorot in some synagogue buildings.83 The rabbis now began to claim that synagogue liturgy requiring ten male participants guaranteed Gods presence (= Shekhinah) there. By invoking the verse God stands in the congregation of the Lord (Ps. 82:1), the sages asserted: Wherever ten persons assemble in a synagogue, Gods presence is with them. 84 It is this enhanced religious dimension of the synagogue in the rabbis eyes that seems to have been behind their protest against using the synagogue for secular or non-worship purposes. Clearly, they did not approve of what they saw, and for that reason they articulated above-quoted prohibition. There were other types of protest as well. As noted above, the second-century sage R. Ishmael b. Elazar declared that the am ha-aretz deserves to die for calling the holy ark a chest (arana) and the synagogue a house of people (bet am).85 It is eminently clear that any secular reference to the synagogue or its appurtenances was anathema to this sage (and probably others as well; for example, those who preserved and transmitted this particular statement). Whether such admonitions were eective is yet another issue. Several inscriptions indicate that this rabbinic dictum was either unknown or ignored in some communities. The term arana, used to identify the Torah shrine, has been found at Naveh in the Hauran and at Dura Europos.86 The rabbinic desire that the synagogue evolve into a substantially dierent institution following the destruction of the Temple is reected in a series of taqqanot (enactments)
81. I Kgs. 8:44; II Chron. 6:34, 38; Dan. 6:11. See also M Berakhot 4, 5. 82. T Megillah 2, 14 (p. 352); Y Megillah 3, 2, 74a; B Arakhin 6b, as well as Chaps. 9 and 17. 83. On the Khirbet Shema and Nevoraya (Nabratein) synagogues, see NEAEHL, loc. cit. 84. Mekhilta of R. Ishmael, Baodesh, 11 (ed. Horowitz-Rabin, p. 243 = Lauterbach, II, p. 287); B Berakhot 6a; see also Y Berakhot 5, 1, 8d9a; PRK 5, 8 (p. 91); Seder Olam 15 (p. 64). On the Shekhinahs presence in rabbinic literature and in Matthew, see Sievers, Where Two or Three . . . , 17182. 85. B Shabbat 32a. On the term am ha-aretz, see below. 86. Kraeling, Excavations at Dura: Synagogue, 269; Naveh, Aramaic and Hebrew Inscriptions, 307.

late roman pa le st i ne


and comments that sought to enhance its religious and liturgical dimension. Immediately following the events of 70 c.e., R. Yoanan b. Zakkai issued a series of taqqanot aimed at transferring a number of Temple practices to this new setting.87 While some of these enactments were directed toward Jewish courts, and although the synagogue is never specically mentioned, it is the latter institution that was undoubtedly the primary beneciary. Blowing the shofar on Rosh Hashanah, using the lulav and ethrog during the seven-day Sukkot festival, and following the priestly practice of not wearing sandals when blessing the people all point to their implementation in a synagogue setting. How many synagogues responded to R. Yoanan b. Zakkais taqqanot is unknown. Some of his contemporaries might have viewed his innovations with alarm.88 The story of his outwitting the Sons of Bathyra regarding the blowing of the shofar on the Sabbath attests to the traditions intent to exemplify R. Yoanans cleverness.89 As it stands, however, the story not only demonstrates R. Yoanans success at handling the challenge to his authority, but also the opposition engendered by these measures. Some rabbinic traditions aimed at encouraging Jews to build synagogues in imitation of the Temple: One does not build the entrances to the synagogue except toward the east, as we nd with regard to the sanctuary [i.e., the Tabernacle], which was open toward the east, as is written: And those encamped before the Tabernacle toward the east, before the Tent of Meeting eastward [Num. 3:38]. One does not build them except on the highest point of the city, as it is written: at the height of . . . you will call [Prov. 1:21]. 90 Just as the Temple portals faced east and the Temple edice itself was located on the highest point along Jerusalems eastern ridge, towering above the biblical City of David, so one should build a synagogue entrance facing east at the highest point in a town. Whether intended to be descriptive or prescriptive, evidence shows that the guidelines delineated in this tradition were implemented only sporadically. While synagogues were often located on prominent sites in a city or village, more so in some than in others, the tradition that portals should face east found expression in only a handful of northern synagogues (Arbel, Maoz ayyim, orvat Sumaqa, and presumably Bet Shean [north] and Sepphoris) and in southern Judaean synagogues.91 The builders of most synagogues in Late Antiquity (i.e., the overwhelming majority of archaeological nds) were either unaware of this latter recommendation or ignored it.
87. M Rosh Hashanah 4, 14; M Sukkah 3, 12. On these enactments, see Alon, Jews in Their Land, I, 10718; L. Levine, Judaism from the Destruction of Jerusalem, 13336. 88. On the opposition to R. Yoanan in a variety of areas, see Alon, Jews, Judaism and the Classical World, 31443. 89. B Rosh Hashanah 29b. 90. T Megillah 3, 2223 (p. 360). See also Lieberman, TK, V, 12001201; Fine, This Holy Place, 4149. 91. The southern Judaean synagogues include Eshtemoa, Susiya, Maon, orvat Anim, and perhaps orvat Rimmon (rst stage); see Amit, Architectural Plans, 635; idem, Synagogues, passim; NEAEHL, passim.

200 h i s t o r i c a l d e v e l o p m e n t o f t h e s y n a g o g u e

Further links between the synagogue and Temple also nd expression in tannaitic literature. One tradition claims that a synagogue and its quorum of ten men would assure Gods attentiveness to Israel, and that this was similar in its eect to the recitation of the divine name in the Temple.92 Another is reected in a statement attributed to R. Judah. When prescribing what should not be done in a destroyed synagogue, this second-century sage invoked Lev. 26:31: and I will leave waste your sanctuaries. Just as a sanctuary (i.e., the Temple) retains its sanctity when in ruins, so, too, does a synagogue.93 Some time in the third or fourth century, the rabbis took additional steps to link these two institutions. Memories of the Temple cult were introduced into synagogue liturgy. Thus, it was claimed that the Amidah prayer was instituted, inter alia, to replace Temple sacrices (prayer was introduced instead of sacricesY Berakhot 4, 1, 7b; B Berakhot 26b), and numerous sources from the amoraic period make the claim that prayer, a substitute for sacrices, could serve as a means of atonement.94 About this time, the content of the Additional Amidah (Mussaf ) recited on Sabbaths and holidays had also changed. Whereas a selection from Deut. 32 (the Song of Moses) had reportedly been read earlier, the Mussaf Amidah now included recitation of the sacrices that used to be oered in the Temple.95 Finally, at some point in Late Antiquity, a supplementary Torah reading (maftir) from Numbers 29 was introduced, in which the sacrices of each holiday are enumerated.96 These last modications in synagogue liturgy take us one step further in demonstrating explicitly what always had been implicit when associating the synagogue with the Temple, and especially when transferring Temple practices to the synagogue setting. It was at this time that the synagogue itself began assuming a degree of sanctity. This idea was new to the Palestinian setting and was not always easily assimilated, even within rabbinic circles.97 An important source concerning the rabbinic view of synagogue sanctity is found in Mishnah Megillah, wherein a sequence of objects is placed in hierarchical order:

92. B Berakhot 6a. See also Y Berakhot 5, 1, 8d9a. 93. M Megillah 3, 3. See also Fine, Did the Synagogue Replace the Temple? 1826, 41. 94. B Berakhot 26a; SifreDeuteronomy 41 (p. 88); PRK 24, 19 (p. 377); Tanuma, Kora, 12; Tanuma, Ki Tavo, 1 (p. 23a). 95. B Rosh Hashanah 31a; Y Berakhot 4, 6, 8c. See also Elbogen, Jewish Liturgy, 98; J. Homan, Surprising History of Musaf Amidah, 4145; L. A. Homan, Canonization, 157. 96. Elbogen, Jewish Liturgy, 138. About this time (though precision here is impossible), the daily psalms once recited by the Levites in the Temple (M Tamid 7, 4) were now transferred to the synagogue setting (B Rosh Hashanah 31a; Tractate Soferim 18, 2 [pp. 31013]). See also Maharshen, Daily Psalm, 19899. 97. For a full discussion of the evolving sanctity of the synagogue in Late Antiquity, see Fine, This Holy Place, 35126. Cf., however, the remarks of A. Goldberg, Service of the Heart, 195211.

late roman pa le st i ne


When the people of the city sell their town square,98 they may only buy a synagogue with this money. If a synagogue [is sold], they may only buy a Torah chest. If a chest [is sold], they may only [buy] mantles for wrapping the scrolls. If the mantles [are sold], they may only buy [sacred] books [i.e., Prophets and Writings]. If the [sacred] books [are sold], they may only buy a Torah. However, if they sell a Torah, they may not buy [sacred] books; if [sacred] books [are sold], they may not buy mantles; if mantles [are sold], they may not buy a Torah chest; if a Torah chest [is sold], they may not buy a synagogue; if a synagogue [is sold], they may not buy a square. And so, too, with the remaining monies.99

The rst part of the above source lists places and objects in an ascending order of sacrality. Monies made from the sale of a less holy object must be used to acquire something more holy, but not vice versa. According to the Talmud, the sanctity of the square appears to be the opinion expressed by R. Menaem b. Yose.100 Thus, the synagogue is less holy than the objects it contains, objects that are all associated with the Torah scroll. It is clear from this list that the Torah scroll itself was considered the holiest object in the synagogue that bestowed degrees of sanctity on other objects, depending on how close to it they were in kind (i.e., other sacred books) or in physical proximity (e.g., mantles, chests, synagogue).101 This mishnah regarding the sanctity of a still-functioning synagogue stands in marked contrast to the following one, which oers a wide spectrum of opinions regarding this buildings status once it ceased to function as such:
A synagogue is sold only on the condition that, if they wish, they can return it [to its former status as a synagogue], says R. Meir. The sages say: If sold, it is sold forever [i.e., there can be no such condition], except that it may not then be used as a bathhouse, a tannery, a place of immersion, or a urinal. R. Judah says: They may sell it as a courtyard and the buyer may do with it as he pleases. 102

This mid second-century debate focuses on the option of reselling a synagogue building. It was unanimously agreed that such a transaction is legitimate; the only issue is whether there should be any restrictions. The more conservative R. Meir recommends a clause allowing its reclamation as a synagogue whenever desired; the sages stake out a more ex98. The public square acquired a modicum of sanctity since the Torah was occasionally read there: R. Yoanan said: This [mishnaic tradition is to be ascribed] to R. Menaem b. R. Yose, for R. Menaem b. R. Yose said: A town square has sanctity since they [sometimes] take out a Torah and read it publicly (Y Megillah 3, 1, 73d; see also M Taanit 2, 1). 99. M Megillah 3, 1. 100. See above, note 98. 101. Ironically, the special status of the Torah is also reected in the fast-day ceremony, for which the Torah scroll was taken into the public square. According to R. iyya bar Abba, this act was to demonstrate that the one precious item we have has been humiliated because of our sins (Y Taanit 2, 1, 65a). My thanks to David Levine for calling my attention to this source. 102. M Megillah 3, 2.


historical development of the synagogue

ible position, allowing a permanent sale except in cases in which its new use would discredit the buildings former status. R. Judah, the most liberal of all, recommends that the building be regarded as a courtyard, allowing the buyer to do what he wishes with it. Note the contrasting positions between these traditions and those of several third-century rabbis (cited above) who would not allow stones from one synagogue to be moved and reused to build another synagogue. How dramatically the sacred status of the synagogue developed among the rabbis within the span of a century! 103 What caused this profound transformation of the synagogue from a principally communal institution into one with an increasingly distinct religious stamp? Was there an idea, an object, or a practice that served as a catalyst for this change? No source addresses this issue head on and thus certainty is elusive. Nevertheless, we can suggest a number of likely causes that are reected in various rabbinic traditions. We have noted the ways in which the synagogue had begun to emulate the Temple, which was regarded as the sacred Jewish institution par excellence. Was it this increasing identication with the Temple that bestowed sanctity on the synagogue? 104 Alternatively, was the reason for this transition the fact that the Torah scroll, the holiest object in the synagogue, now to be found on an increasingly permanent basis in the main hall of the building? The appearance of bimot, apses, and niches in third- and fourthcentury buildings clearly indicates this development (g. 19), and the presence of scrolls undoubtedly played a role in according an enhanced sanctity to the synagogue building itself. There is no better testimony to this fact than the diatribe of Chrysostom in one of his homilies against the Jews:
But since there are some who consider the synagogue to be a holy place, we must say a few things to them as well. Why do you revere this place when you should disdain it, despise it and avoid it? The Law and the books of the prophets can be found there, you say. What of it? You say, Is it not the case that the books make the place holy? Certainly not! This is the reason I especially hate the synagogue and avoid it, that they have the prophets but do not believe in them, that they read these books but do not accept their testimonies. . . . Therefore stay away from their gatherings and from their synagogues and do not praise the synagogue on account of its books. Rather, hate it and avoid it for that very reason, for they have mangled the saints because they do not believe their words and they accuse them of extreme impiety.105

103. Y Megillah 3, 1, 73d. See also A. Goldberg, Die Heiligkeit des Ortes, 2631. On the superiority of a number of extra-Temple religious practices over those associated with the Temple in amoraic literature, see Bokser, Rabbinic Responses to Catastrophe, 3761. 104. See, for example, M Megillah 4, 3. 105. John Chrysostom, Homily 1, 5, trans. in Meeks and Wilken, Jews and Christians, 9496. See also E. M. Meyers, Torah Shrine, 30338. Amoraic literatures preference for aron and not tevah when referring to the Torah chest is indicative of a change in perspective. The term aron carries with it very clear associations with the biblical ark containing the tablets with the Ten Commandments that was placed in the holiest precinct of the Tabernacle and then of the Temple. Later on, the Karaites attacked the Rabban-

late roman pa le st i ne


[To view this image, refer to the print version of this title.]

19. Qatzrin synagogue, looking southwest. Note the bima along the southern wall.

Despite this poignant evidence from the late fourth century, it remains unclear whether the presence of scrolls was the pivotal factor in the evolving synagogue sanctity of late Roman Palestine.106
ites precisely because of this association; see Lieberman, Midreshei Teiman, 25. It is perhaps noteworthy that over a millennium earlier, it was the introduction into Jerusalem of the holy ark with the two stone tablets that David perceived as essential to granting the city its sacred status (II Samuel 6). Moreover, the sanctity of the Torah is reected in the dramatic account of Levi b. Sisi (early third century), who, upon the approach of Roman soldiers to his town, took a Torah scroll in hand, climbed onto a roof, and exclaimed: Master of the universe, if I have ignored one word in this Torah scroll, let them [the soldiers] enter; if not, have them leave (Y Taanit 3, 8, 66d). Compare this to the use of a Torah scroll by Joshua son of Sapphias, in rst-century Tiberias (Josephus, Life 13435), who accused Josephus of betraying the city. The process of introducing the Torah scrolls into the synagogue on a permanent basis was a long one. From the extant, though limited, evidence, it is rather clear that pre-70 synagogues did not have any bimot or aediculae, and the Torah would have been brought into the hall for reading and then removed. By the third and fourth centuries, a permanent installation was becoming more common, and by the end of Late Antiquity, it was almost a universal xture. See, for example, T Megillah 3, 21 (p. 360); B Sotah 39b; below, Chap. 9. 106. As argued in Fine, This Holy Place; see also Lightstone, Commerce of the Sacred, 104. On the sanc-

204 h i s t o r i c a l d e v e l o p m e n t o f t h e s y n a g o g u e

Several other possibilities also present themselves. We have noted above with regard to the synagogues orientation that, in rabbinic sources at least, prayer was becoming an ever more signicant element in Jewish liturgy of Late Antiquity. The synagogues orientation toward Jerusalem from the third century onward clearly reects an enhanced connection with the holy city that could plausibly be linked to the requirement to face Jerusalem during prayer. Thus, the growing role of congregational prayer in the synagogue liturgy may well have contributed to the buildings heightened sanctity.107 It is also possible that this newly emerging status of Palestinian synagogues was inuenced in some way by Diaspora models. We have seen with regard to the synagogues sanctity that the Diaspora preceded Palestine by centuries. Already before 70, many Diaspora synagogues were not only a place for Torah reading, as in its Palestinian counterpart, but also a setting for public prayer, and at times were considered sacred space.108 Several early examples of Diaspora synagogue architecture, at Delos and Ostia, may indicate a rst- or second-century c.e. orientation toward Jerusalem. Might such Diaspora models have helped shape, through channels as yet unknown to us, the later Palestinian one? On the basis of the sources available, however, it is well nigh impossible to ascertain which of the above was the pivotal factor in the synagogues transformation in status and role in late Roman Palestine. It is quite probable that the change was not due to any one of the above-mentioned factors, but rather to a combination of some, or perhaps all, of them. Moreover, the subsequent Byzantine era was to contribute further to a signicant increase in synagogue sanctity (see below, Chap. 7). Although we have noted the extent of rabbinic support for viewing the synagogue in a new religious light, particularly with regard to the Temple, an important caveat is in order; it appears that not all sages were comfortable with the idea that the synagogue and its practices should resemble those of the Temple. One tannaitic tradition often quoted in the Bavli oers the following words of caution: One should not make a house in the likeness of the Sanctuary; nor an exedra in the likeness of the Temple porch [ ,]nor a courtyard like the [ i.e., the Temple courtyard], nor a table like the [Temple] table, nor a menorah like the [Temple] menorah. Rather, one may make [a menorah] of ve, six, or eight [branches], but one of seven he shall not make. 109 Sages subscribing to such a
tity of the Torah scroll and its use as a magical object in medieval and modern Jewry, see Sabar, Torah and Magic, 14979 (Hebrew). 107. For the emphasis on public prayer, see, for example, B Berakhot 32b; Tanuma, Ki Tavo, 1 (p. 23a); Pirqei de Rabbi Eliezer 5 (= Seder Eliyahu Zuta 23 [p. 42]). On the comparison of a prayer leader with one who oers a sacrice, see Y Berakhot 4, 4, 8b; Genesis Rabbah 49, 23 (pp. 5067). Cf., however, Ginzberg, Commentary, III, 35051. See also Bokser, Rabbinic Responses, 4757; Fine, This Holy Place, 6267; Z. Safrai, Communal Functions, 200; Urman, House of Assembly and the House of Study, 23536. 108. See above, Chap. 4; as well as Goodman, Sacred Space, 116. 109. B Rosh Hashanah 24ab; B Avodah Zarah 43a; B Menaot 28b; Midrash Hagadol, Exodus 20:20 (pp. 41011); Blidstein, Prostration and Mosaics, 22.

late roman pa le st i ne


tradition clearly preferred to maintain a distinction between synagogue and Temple, an emphasis quite dierent fromthough not necessarily contradictory tothat reected in the above-noted sources, which clearly attempted to create physical and psychological ties between the two institutions. The many archaeological nds of seven-branched menorot from Late Antiquity, including a number of three-dimensional ones, indicate that many communities did not subscribe to (or perhaps did not know of ) these rabbinic restrictions.110 It is also possiblethough dicult to substantiatethat the above prohibition may reect the attitudes prevalent between 70 and 135, when the hopes of the more messianically inclined were still riding high regarding the imminent restoration of the Temple.111 The issue of imitating Temple practices gave rise to disagreements in other realms as well. Apparently, several second-century Galilean rabbis tried to conduct fast-day ceremonies as they were once carried out in the Temple:
It once happened in the days of R. alafta and R. anania b. Teradion that someone led services and completed the entire benediction, and no one responded Amen (this being the case in the Temple as well). [The azzan called out:] Blow [the shofar], priests, blow. [He recited:] He who answered our father Abraham on Mt. Moriah, He will answer you and will listen to the voice of your cries this day. [The azzan then said:] Blow, sons of Aaron, blow. [He then continued to recite:] He who answered our fathers in the Sea of Reeds, He will answer you and listen to the voice of your cries this day. When the report [of what had happened] came before the sages, they said: We would not have acted thusly except at the eastern gate on the Temple Mount. 112

The more expanded Tosefta version adds that R. alafta instituted this practice at Sepphoris and R. anania b. Teradion in Sikhni (or Sikhnin).113 It appears that these two sages were attempting to perpetuate Temple practices, although it is unclear just which component(s) of this ceremony constituted a continuation: omitting the response Amen, thereby according priests a central role in the ritual; reciting these particular prayers; or prostrating oneself in the prayer setting.114 The aversion of other sages to perpetuating

110. Hachlili, Ancient Jewish Art and ArchaeologyIsrael, 23656. See also Goodenough, Jewish Symbols, IV, 7198; Sperber, History of the Menorah, 13559; Negev, Chronology, 193210; L. Levine, History and Signicance of the Menorah, 13153. It may be signicant that the above-noted prohibition was preserved in the Bavli and not in the Yerushalmi. Might this indicate that the Palestinian amoraim did not object to such representations or that they saw no reason to do so in light of the popular practice? 111. This line of reasoning nds interesting corroboration on a half-dozen lamps from southern Judaea dating between 70 and 135 c.e. on which the menorah is displayed. Of these, ve have more than seven branches (Sussman, Ornamental Jewish Oil-Lamps, nos. 16). 112. M Taanit 2, 5. 113. T Taanit 1, 14 (pp. 32728) for this and other variations. See also Lieberman, TK, V, 1075. 114. On prostration at fast-day ceremonies, see Y Avodah Zarah 4, 43d; B Megillah 22a; and comments in Ginzberg, Commentary, III, 11922; Blidstein, Prostration and Mosaics, 3436.

206 h i s t o r i c a l d e v e l o p m e n t o f t h e s y n a g o g u e

one or more of these customs in all probability stemmed from the desire to clearly dierentiate between Temple and synagogue practice.115 Thus, from the various traditions cited above, it would seem that while there was general consensus within rabbinic circles over the synagogues sanctity, there was less agreement as to the institutions precise status, degree of sanctity, and extent of identication with the Temple. This issue engaged many Jewish communities as well, although it is not clear to what extent or intensity. Steps were gradually being taken in various locales throughout the country to steer the synagogue toward sanctication, and it is in the third century that we see the rst traces of this process as manifested in building orientation and the presence of sacred objects. The change was indeed a giant step from what the synagogue had been in the pre-70 era, yet still a far cry from what would evolve in subsequent centuries. In what appears to be a quintessentially elegant formulation reecting the ambiguity of the synagogues new role and the ambivalence in identifying it with the Temple, the third-century R. Samuel b. R. Isaac dened the synagogue as a , a diminished or small sanctuary.116 This formulation gives eloquent expression to the ambiguity. There was indeed something sacred about the synagogue; it was comparable to the .Yet, it was only a replica of sorts, whose status was not to be confused with that of the Jerusalem Temple itself. In reality, it was diminishedsomething less than the Temple, not quite as sacred, not quite as special.


We have explicit evidence for a large number of synagogues throughout the country in the late Roman era, especially the third and fourth centuries. As noted, archaeological remains of synagogues during this period have been discovered not only in the Galilee, but in Judaea and the Shephelah as well. Many cities and even villages undoubtedly had more than one synagogue, as did Alexandria, Rome, and Jerusalem before 70. Pre135 Betar reputedly had more than 480 such institutions, although this is clearly a highly inated gure; Tiberias had thirteen, Sepphoris and its environs possibly eighteen, and Lydda an undened (but large) number.117 Even towns and villages such as Gush alav
115. Blidstein (Prostration and Mosaics, 3536) has suggested that opposition to the practice of prostration may also stem from the fact that this practice had become widespread in Christian circles by this time. See also ibid., 22, for several other examples of rabbinic opposition to imitating Temple practices. 116. B Megillah 29a, citing Ezek. 11:16. This is the correct reading following the Munich and other MSS (e.g., British LibraryHarley 5508 (400); Columbia X893T 141; Gttingen 3), and not the printed Vilna edition. See R. Rabinowitz, Diqduqei Soferim, loc. cit. 117. Betar: Y Taanit 4, 8, 69a; B Gittin 58a. Tiberias: B Berakhot 8a. See also B agigah 15ab, as well as B Sotah 22a; and Miller, On the Number of Synagogues, 5558. In the eighth century, Tiberias re-

late roman pa le st i ne


and Baram had several such buildings. Some synagogues were named after the professions of their founders or leading members, such as the Tarsian synagogues in Lydda;118 others were named after a town or city, presumably the place of origin of those who founded the synagogue, as, for example, those of the Alexandrians, Cyrenians, Cilicians, and Asians in pre-70 Jerusalem and those of Gophna, Babylonia, Sidon, Tyre, and perhaps Cappadocia attested in Sepphoris.119 A synagogue may have been named after a section of a city, such as the Kifra and ammat synagogues in Tiberias.120 One Tiberian synagogue was named Boule, whether because boule members founded it and dominated its aairs, the building was located near the citys boule, or the building itself was (or had once been) the meeting place of the citys boule.121 Finally, a synagogue may have been named because of its antiquity (the Old Synagogue or Sarugnaia, south of Tiberias), size (the Large or Great Synagogue of Sepphoris), or a historical event (the Synagogue of [the] Rebellion in Caesarea).122 In addition to the above, some individual anonymous synagogues are mentioned in connection with specic communities, such as Maon and Migdal near Tiberias, and Sikhnin, Caesarea, Lydda, Tivon, Bet Shean, En Gedi, and Bostra.123 We also read in rabbinic literature of synagogues that catered to Greek-speaking communities, presumportedly boasted at least thirty synagogues, which were destroyed in the major earthquake of 749 c.e.; see Avissar, Sefer Teveria, 92. On the dating of this earthquake, see Tsafrir and Foerster, Dating of the Earthquake, 23135. Sepphoris: Y Kilaim 9, 4, 32b; Y Ketubot 12, 3, 35a; Ecclesiastes Rabbah 7:11; but cf. reservations of Miller, in On the Number of Synagogues, 5963. Lydda: Y Sheqalim 5, 6, 49b; Y Peah 8, 21b. 118. Leviticus Rabbah 26:3 (pp. 83031); B Nazir 53a. 119. Jerusalem: T Megillah 2, 17 (pp. 35253)Alexandrians; Y Megillah 3, 1, 73dCyrenians; B Megillah 26aCilicians; Acts 6:9Asians, Cilicians, Cyrenians, and Alexandrians. Sepphoris: Y Berakhot 3, 6a; Y Nazir 7, 1, 56aGophna; Y Sanhedrin 10, 1, 28a; Y Shabbat 6, 8a; Y Berakhot 5, 1, 9a; Genesis Rabbah 33, 3 (p. 305); PRK 25 (p. 381); see also Y Yoma 7, 5, 44b; Y Megillah 4, 5, 75b; and, for medieval evidence, Goitein, Mediterranean Society, II, 167Babylonia; Roth-Gerson, Greek Inscriptions, 10510 Sidon, Tyre; Y Sheviit 9, 5, 39aCappadocia. 120. Kifra: Y Megillah 1, 1, 70a; Pesiqta Rabbati, Supplement B (p. 196b). ammat Tiberias: Y Sotah 1, 4, 16d. 121. Y Sheqalim 7, 3, 50c; Y Taanit 1, 2, 64a. Cf. S. Klein, Galilee, 99. 122. : Y Kilaim 9, 32c; Leviticus Rabbah 22:4 (p. 511). : PRK 18, 5 (p. 297). : Y Bikkurim 3, 3, 65d; Y Sanhedrin 1, 1, 18a; Y Nazir 7, 1, 56a; Midrash on Samuel 6 (p. 34b); Numbers Rabbah 12, 3. A similar phenomenon may be noted in Bursa (Turkey), where a synagogue was named Gerush (lit., exile), after the Spanish exiles who founded it. As noted in Chaps. 4 and 8, the community in Rome used names for synagogues in each of the categories listed and also after leading historical gures. 123. Maon: B Yevamot 64b; B Shabbat 139a; B Zevaim 118b; see also Y Megillah 3, 2, 74a; B ullin 97a. Migdal: Y Megillah 3, 1, 73d. Sikhnin: Y Megillah 4, 5, 75b. Caesarea: B Yevamot 65b. Lydda: Y Sheqalim 5, 6, 49b; Y Peah 8, 9, 21b; Leviticus Rabbah 35, 12 (pp. 83031). Tivon: T Megillah 2, 5 (p. 349); Y Megillah 4, 1, 74c. Bet Shean: Y Megillah 3, 4, 74a. En Gedi: Lewis, Documents from the Bar-Kokhba Period, 8485. Bostra: B Shabbat 29b.


historical development of the synagogue

ably located in Palestine, such as the one in Caesarea, where members did not know enough Hebrew to recite even the most basic of prayers, the Shema.124 A number of synagogues in Roman Palestine were organized around particular social constituencies, a practice to which the rabbis objected. Undoubtedly, Jewish-Christian communities had their own places of worship, although we know next to nothing about them.125 Rabbinic traditions exhibit particular hostility to ammei ha-aretz and their synagogues. This term had borne a series of meanings since biblical times, often strikingly dierent from one period to another.126 While any consensus regarding the meaning of the term in the second and third centuries is illusive, it appears to have referred to a large segment of non-rabbinic Jewish society, at whom the rabbis took umbrage on the grounds that their religious observance was unacceptable and their study of the Torah inadequate. The sages identied certain synagogues with these Jews and referred to these places in pejorative terms. In one tradition, the rst-century R. Dosa b. Hyrcanus equates sitting in their synagogues with behavior such as sleeping in the morning, drinking wine in the afternoon, and engaging in childrens talk takes one out of this world. 127 As noted above, the second-century source attributed to R. Ishmael b. Elazar is even more disdainful, claiming that the ammei ha-aretz deserve to die because they call the holy ark arana and the synagogue bet am.128 It is hard to explain the reason for such hostility to what appears to be mildly deviant behavior by rabbinic norms;129 death is not a light curse. Obviously, there is much more to R. Ishmaels attitude toward am ha-aretz than meets the eye. Unfortunately, however, there is little more to go on, although we have already noted that several communities (in Dura Europos and Naveh) did, in fact, refer to their ark as an arana or bet arana. Thus, R. Ishmaels objections were not at all theoretical, and his contempt was similar to that felt at times by other intellectual and religious elites toward the unschooled and unso124. T Megillah 3, 13 (p. 356); Y Sotah 7, 1, 21b. On Greek culture in Caesarea, see Geiger, Voices Reciting the Shma in Greek, 2736. 125. See Visotzky, Fathers of the World, 12949; Saldarini, Gospel of Matthew, 2627; A. Baumgarten, Literary Evidence for Jewish Christianity, 50. On Jewish Christianity in rabbinic literature generally, see the important remarks of Visotzky, op. cit., 12949. With regard to the seven synagogues on Mt. Zion in the time of Aelia Capitolina, as reported by the Bordeaux Pilgrim (333 c.e.), which presumably belonged to Jewish Christians, see Taylor, Christians and Holy Places, 21020; cf. Mimouni, La synagogue judo-chrtienne, 21534. 126. See Oppenheimer, Am Ha-aretz, passim; L. Levine, Rabbinic Class, 11217, and the literature cited therein; Viviano, Study as Worship, 4243. 127. M Avot 3, 10. The reading in a number of MSS (Kaufman, Parma, and the Genizah fragments) amounts to basically the same thing. See my Rabbinic Class, 114. 128. B Shabbat 32a; Midrash Hagadol, Genesis 25:24 (p. 439). 129. The rabbis also urged one another never to sit with these people (B Berakhot 47a) and not to associate with them in any manner (B Pesaim 49b).

late roman pa le st i ne


phisticated masses. Such an attitude was widespread among the intelligentsia of antiquity and held true for pagans and Christians no less than for Jews.130 We have no way of assessing how widespread the synagogues of ammei ha-aretz were. It may be that they were quite well known and were perhaps even more reective of what was generally taking place than those synagogues following rabbinic dicta. Here and there in rabbinic writings we nd other expressions of the sages unhappiness with, and objections to, various synagogue practices. For instance, they would exclude laborers from leading services and from participating in the priestly blessing (although the reasons for doing so are unclear), and they objected to inappropriately dressed prayer leaders and the inclusion of certain additions to the service that they considered sectarian.131 The steps that the rabbis advocated against a range of sectarians are well known. The introduction into the daily prayer of references to Jewish Christians and other heresies considered dangerous was a signicant move.132 Rabbinic involvement in implementing this innovation is explicitly documented;133 what is unclear is whether they were taking the lead in initiating something entirely new or merely perpetuatingand further developingan already existing breach.134 Clearly, the rabbis could beand often werequite judgmental of their surroundings. They would record what they liked and either criticize or ignore what displeased them. The reality of ancient synagogues, their forms, practices, and organization, is a subject far wider than rabbinic sources permit us to see. While this corpus is of enormous value with regard to the subject at hand, it is also restricted in scope and tendentious in outlook, touching only on limitedalbeit importantaspects of the topic. It behooves us to be aware of these limitations.
130. See MacMullen, Paganism, 8 and nn. 3336. 131. Leading services: T Berakhot 2, 9 (pp. 78). Priestly blessing: B Berakhot 16a. See the explanation of Rashi, loc. cit. The rabbis often looked askance at such workers (see Ayali, Workers and Artisans, 95 97), and tensions appear to have been mutual; see Genesis Rabbah 65, 15 (p. 728); and the comments of S. Klein, Galilee, 1078. See also S. Krauss, Synagogale Altertmer, 19496. Inappropriate dress: M Megillah 4, 8. Sectarian prayer additions: M Berakhot 5, 3; M Megillah 4, 89. 132. Regarding references to Christians specically in the synagogues, see John 9:22; 12:42; 16:2; Epiphanius, Panarion 29, 9, 2. 133. B Berakhot 28b29a. See Kimelman, Birkat Ha-Minim, 22644; Katz, Issues in the Separation of Judaism and Christianity, 6376. See also Kalmin, Christians and Heretics, 15569. 134. See Lieberman, TK, I, 5354; Flusser, Jewish Religion, 2324.



yzantine Palestine (fourth to seventh centuries) not only witnessed the continued evolution and development of the synagogue, but also provided a rich context in which to understand additional dimensions of this institution. Synagogue remains from this period exist in far greater numbers and in more geographical areas than do those from earlier periods. Indeed, this period constitutes a peak in synagogue development, which is reected in its architectural, artistic, and epigraphic remains. The synagogue building acquired an ever more distinctive religious character, and synagogue liturgy continued to expand, becoming more elaborate and varied. These changes were not merely a continuation of the past. A new element entered Byzantine Palestine at this time that not only aected the climate of Jewish life generally, but also had a signicant inuence on the synagogue, its appearance, status, and modus operandi. This new element was Christianity, whose dramatically expanding presence in Palestine from the fourth century onward impacted profoundly on Jewish life.1 At worst, it was a disruptive, threatening, and at times destructive force, while at best, it was a stimulating and fructifying one. By the latter part of the fourth century, and with increasing frequency in the fth and sixth centuries, Jews, Judaism, and the synagogue were being assaulted on a number of fronts throughout the Empire. Church legislation attacked Jews and Jewish practices,
1. See L. Levine, Between Rome and Byzantium, 748.

byzant i ne pa le s t i ne


Imperial edicts became more restrictive, and Christian preachers fulminated against a religion and people they regarded as anachronistic, loathsome, and rejected by God. On occasion, mobs were incited against the Jews, resulting in damage, loss of lives, and either the desecration or outright destruction of synagogues.2 Imperial legislation was invoked on a regular basis to try and prevent such destruction.3 In some cases, Christians appropriated synagogues and converted them into churches, for example, in Stobi, Gerasa, and Apamea according to archaeological remains, and in Callinicum, Ravenna, and elsewhere in Italy, Mauretania, Spain, Gaul, Syria, and Minorca per our literary sources.4 Nevertheless, despite this litany of anti-Jewish words and deeds, we have become well aware that sermons were often not heeded nor legislation always enforced. Despite ocial restrictions, the Jews of Byzantine Palestine continued to build synagogues (e.g., Merot, Capernaum, Bet Alpha, southern Judaea), repair those already standing (e.g., Maoz ayyim, ammat Tiberias, ammat Gader, En Gedi), and entirely rebuild and refurbish others after a period of abandonment and disrepair (Nevoraya). In many instances, it was at this time (i.e., the sixth and seventh centuries) that a synagogue building reached its greatest dimensions (e.g., ammat Tiberias, En Gedi, Nevoraya, orvat Rimmon, ammat Gader). In several regions (e.g., the Golan), many synagogues were erected where few, if any, seem to have existed beforehand.
2. Baron, Social and Religious History of the Jews, II, 2079; III, 418; Parkes, Conict of the Church and the Synagogue, 16395, 26369; Simon, Verus Israel, 20233; Avi-Yonah, Jews of Palestine, 20831; H. H. Ben-Sasson, Trial and Achievement, 312. 3. The earliest such legislation was in evidence in 393 c.e. (Cod. Theod. 16, 8, 9 [Linder, Jews in Roman Imperial Legislation, 190]) and continued in 397 (16, 8, 12 [19798]), 412 (16, 8, 20 [26265]), 420 (16, 8, 21 [28386]), and 423 (16, 8, 25 [28789]). See also the law from 438 c.e. (Linder, Jews in Roman Imperial Legislation, 323.). That such legislation was often ineective is clear not only from the need to reissue protective laws, but also from the numerous cases of synagogue destruction. Incidents are reported to have occurred in Rome, Constantinople, Edessa, Sicily, Sardinia, North Africa, Alexandria, Antioch, and northern Italy. See Parkes, Conict of the Church and the Synagogue, 187, 21214, 23638, 25051. In Palestine as well it is reported that the early fth century witnessed the destruction of synagogues and temples as Barsauma and some forty other monks went on a rampage that supposedly lasted several years (ibid., 230). See also Simon, Verus Israel, 225; Avi-Yonah, Jews of Palestine, 218; and below, Chap. 8. On the destruction of pagan shrines by zealous Christians, see Trombley, Hellenic Religion and Christianization, I, 123., 207., 342.; Liebeschuetz, Antioch, 23739. On the other hand, Imperial restrictions on the building of synagogues were often not implemented, as was the case later on, in the Middle Ages, under Islamic rule (Assis, Synagogues in Medieval Spain, 8). 4. Juster, Les Juifs, I, 464 n. 3; Parkes, Conict of the Church and the Synagogue, 16668, 187, 2047, 22529, 244, 263; Simon, Verus Israel, 22529; Avi-Yonah, Jews of Palestine, 21820; Bradbury, Severus of Minorca, 30, 2 (p. 123) and 130 n. 25. On the conversion of temples into churches, see Trombley, Hellenic Religion and Christianization, I, 108., 123.; II, 377.; R. Cormack, Temple as the Cathedral, 7588. There were also instances where attempts were made to protect Jewish rights and synagogues, as was the case with Gregory and the Terracinan Jewish community; see Katz, Pope Gregory, 12022. For a recent overview, see T. Braun, Jews in the Late Roman Empire, 14271.


historical development of the synagogue

On the periphery of Judaea (to the east and west, and especially in the southern region), a dozen or so synagogues have been discovered to date.5 Given Hadrians prohibition of Jewish settlement around Jerusalem,6 the extensive destruction in Judaea in the aftermath of the Bar-Kokhba revolt, and the strong Galilean focus in rabbinic literature, the number of Judaean synagogues found to date is indeed surprising. However, the most unexpected discoveries have been in the Golan, where the bulk of the twenty-ve synagogues exposed to date ourished in the Byzantine era.7 As noted above on several occasions, the overwhelming majority of synagogues excavated in Palestine date to the Byzantine period. Most were built at that time, but even those constructed earlierwith the exception of the handful of pre-70 structurescontinued to function throughout most, if not all, of this period.8 The very existence of Byzantine synagogue remains at well over one hundred sites has been one of two main catalysts in reassessing Jewish life in this historical period. Since the beginning of Jewish historiography in the nineteenth century, this era was considered the dawn of the Dark Ages for the Jews of Palestine. Jewish life was then viewed as being in steady decline after the destruction of the Temple in the rst century. Crises in subsequent centuries served only to exacerbate this trend: the failure of the Bar-Kokhba revolt and Hadrians persecutions in the second century; the Empire-wide anarchy and instability of the third; the rise of Christianity and the beginning of anti-Jewish legislation in the fourth; and, nally, the disappearance of the Patriarchate in the fth.9 Even the edited form of the Yerushalmi, which appears both partial and hasty, especially when compared to its more polished Babylonian counterpart, has usually been interpreted as an indication of the sudden closure of this enterprise in the face of impending disaster.10 Therefore, to nd that Jews throughout the Byzantine era were building synagogues everywhere, often on a grand and imposing scale, requires a major reevaluation of this period.11

5. Z. Ilan, Ancient Synagogues, 254321, which also includes a number of dubious identications, and Amit, Synagogues, 6588. 6. Eusebius, Ecclesiastical History 4, 6, 4; Justin, Dialogue with Trypho 16; idem, First Apology 47; Schrer, History, I, 55355. 7. Z. Ilan, Ancient Synagogues, 61113; Maoz, Art and Architecture of the Synagogues of the Golan, 98115; idem, Golan, 53845; Hachlili, Late Antique Jewish Art from the Golan, 183212. Cf., however, the assumption of an earlier, late Roman date for some synagogue sites by Urman (Golan, 80116; Gregg and Urman, Jews, Pagans and Christians, 30510). 8. The Second Temple synagogues that never survived the revolt of 66 are excluded. 9. See, for example, Graetz, History of the Jews, II, 559.; Baron, Social and Religious History of the Jews, II, 17275, 20914; Avi-Yonah, Jews of Palestine, 275. 10. Ginzberg, On Jewish Law and Lore, 2429. 11. A caveat is in place here. It must be remembered that it is usually the latest stratum of archaeological remains that is best preserved, often obliterating, partially or entirely, earlier strata. Thus, impressive remains of Byzantine synagogues do not mean, ipso facto, that similar buildings did not already exist in

byzant i ne pa le s t i ne


[To view this image, refer to the print version of this title.]

20. Aerial view of the Capernaum synagogue, looking northeast. The courtyard lies to the east.

The synagogue at Capernaum (g. 20) is the most striking example of the revised perception of this period as one also characterized by relative stability, and at times even remarkable prosperity, for the Jews. This monumental and ornate building was completed in the plan as we know it today only in the latter part of the fth (or perhaps sixth) century, i.e., well into the Byzantine era.12 The buildings prominence was enhanced by the articially raised podium on which it stood, dwarng the nearby Church of St. Peter, which was also built at this time.13 A second catalyst in re-evaluating this era is based on documents from the Cairo Genizah, some of which were formerly dated to the early Middle Ages but are now acknowledged to have been written in the Byzantine era. Moreover, several new literary genres once thought to have been the products of later centuriesalso rst appeared in Late Antiquity: synagogue poetrythe piyyutoriginated in Palestine between the fourth
the earlier, Roman, era. Moreover, as we will discuss below (Chap. 9), the use of spolia (i.e., material from earlier structures) later on may attest to similar structures beforehand. 12. The earlier, fth-century date has been suggested by Corbo, Cafarnao, 11369, the later sixth century one by Magness, Question of the Synagogue, 1826. See also Chap. 9. 13. Corbo, House of St. Peter, 1618. See Brent, Die Christianisierung, 1528.


historical development of the synagogue

and sixth centuries; the earliest aggadic midrashim were edited then; the Hekhalot mystical traditions began crystallizing at this time; a number of apocalyptic works were composed in the early seventh century; and many targumic traditions took shape from the fourth century onward.14 Thus, synagogue nds, together with Genizah material, have led to a serious reassessment of the Byzantine era. This period is now accorded a far dierent appraisal than that which was common earlier; it was not a post-Classical era of decline, but one that generated new forms and institutions, as well as new cultural and spiritual foci, while, at the same time, sustaining many traditions and institutions from the past.15 Many Palestinian cities, though perhaps less lavishly ornamented than their precursors, reached an apogee of physical growth precisely during Byzantine rule; the total number of settlements (villages, towns, and cities) increased signicantly and, in general, the eastern Mediterranean world ourished.16 The earlier picture of a cultural wasteland characterized by political decline and economic decay should now be replaced by a very dierentand more nuancedunderstanding of this era.


The ourishing Jewish life attested by synagogue nds reects a rich diversity throughout the country. Beginning with the material remains, the vast majority of scholars today have eschewed the rigid classication of synagogues that once linked typology with chronology.17 Whereas it was once assumed that synagogues of the Byzantine period were built only on a basilical plan, following the contemporary church model, we know today not only of a far richer variety of such basilical types than before,18 but also of other
14. Piyyut: Heinemann and Petuchowski, Literature of the Synagogue, 2078; Y. Dan, City in EretzIsrael, 3240; Mirsky, Piyyut, 7781; Yahalom, Poetry and Society, 6484. Midrash: Strack and Stemberger, Introduction to Talmud and Midrash, 300. Hekhalot: Stemberger, Non-Rabbinic Literature, 3036; Elior, Temple and Chariot, 21276. Apocalyptic literature: Ibn Shmuel, Midreshei Geulah; Baron, Social and Religious History of the Jews, III, 1617. Targum: Hayward, Date of Targum Pseudo-Jonathan, 730; Shinan, Dating Targum Pseudo-Jonathan, 10916; idem, Embroidered Targum, 1115, 19398. 15. See the companion volumes by Av. Cameron, Later Roman Empire and Mediterranean World, as well as Browns World of Late Antiquity, 22., 96., 137. For a critique of Gibbons account of the Byzantine era (Decline and Fall of the Roman Empire) in light of the current understanding of Late Antiquity, see Vryonis, Jr., Hellas Revisited, 92118. 16. On Caesarea, see, for example, Holum et al., King Herods Dream, 16299; Tsafrir, Some Notes on Settlement and Demography, 26983; and Av. Cameron, Mediterranean World, 15286; Downey, History of Antioch, 317. 17. For a fuller discussion of the matter, see below, Chap. 9. 18. This is what Avi-Yonah (Ancient Synagogues, 41) calls a bewildering variety. See also Foerster, Ancient Synagogues, 4041; idem, Dating Synagogues with a Basilical Plan, 8892; Tsafrir, Archaeology and Art, 28599.

byzant i ne pa le s t i ne


synagogue types that were being either built or refurbished at this time. Merot, Nevoraya, Chorazim, and Capernaum were constructed between the fourth and sixth centuries following a Galilean model; broadhouse-type synagogues were being built in the third to fth centuries (Susiya, Eshtemoa, Maon, orvat Anim, orvat Rimmon, and perhaps ammat Tiberias, Yaa, and Kefar ananiah as well); and a variety of buildings, all more or less under Galilean inuence, began appearing in the Byzantine Golan.19 A striking example of this diversity among synagogues is evident in the Bet Shean area. To date, we know of ve contemporaneous synagogue buildings that functioned in the sixth century in this region.20 To date, no other urban setting boasts such a concentration of remains having not only geographical but also chronological propinquity. The synagogues referred to are Bet Shean A, just north of the city wall; Bet Shean B, near the southwestern city gate; Bet Alpha, to the west; Maoz ayyim, to the east; and Reov, to the south. Although these buildings functioned at the same time, they are, in fact, remarkably dierent from one another in a variety of ways. Let us compare three aspects of these buildings: architectural plan, art, and inscriptions (i.e., language and culture). Architectural Plan. Three of these buildings (Bet Shean A, Bet Alpha, and Maoz ayyim in its later stages) are apsidal basilicas; Reov is a basilica-type building, but with a raised bima at its southern end anked by two side rooms; Bet Shean B is a kind of chapel or prayer rooma simple, almost square room, possibly with a niche (of which little remains) in the direction of Jerusalem. Reov and Bet Alpha have a narthex in the north; Bet Shean A has auxiliary rooms on a number of sides, however no narthex has been recovered; and Maoz ayyim has a side entrance and no narthex. The most stunning variation, and indeed a striking exception to the norm at Bet Shean and elsewhere, is the orientation of Bet Shean A. This building faces northwest; i.e., its apse points in this direction while its entrances face southeast. Various theories have been suggested to explain this apparent deviation, ranging from its being a Samaritan synagogue 21 or some kind of sectarian building, based on the congregations aversion to ori-

19. Maoz, Art and Architecture of the Synagogues of the Golan, 98115; and above, note 7. 20. Chiat, Handbook of Synagogue Architecture, 12144; idem, Synagogue and Church Architecture, 624; Z. Ilan, Ancient Synagogues, 169.; and, for specic sites, Httenmeister and Reeg, Antiken Synagogen; NEAEHL. See also Roth-Gerson, Greek Inscriptions, nos. 49. For a recent presentation of the Reov ndings, see Vitto, Interior Decoration, 29397. 21. Whether this building is a Jewish or a Samaritan synagogue has been a subject of controversy for years. The excavator Zori (Ancient Synagogue at Beth-Shean, 73) posits a Jewish identication of this building, while Httenmeister and Reeg (Antiken Synagogen, II, 57475) and Foerster (Ancient Synagogues of the Galilee, 313; cf. also NEAEHL, I, 234) argue for a Samaritan identication; a noncommittal position is espoused by Naveh (On Stone and Mosaic, 7677). See also Chiat, Handbook of Synagogue Architecture, 13132; Z. Ilan, Ancient Synagogues, 18081. The Samaritan identication stems primarily from a Samaritan inscription, i.e., one written in the palaeo-Hebraic script used exclusively by the


historical development of the synagogue

enting the synagogue southward, toward Tel Bet Shean and its pagan temples and later Byzantine church. However, it is also possible that too much has been made of this apparent deviation. Then, as now, there may have been all sorts of extenuating circumstances (e.g., topography or the particular layout of the plot of land) that induced the founders to so position the building. Whatever the reason, this building shares many characteristics with other known synagogues, and even its northwestern orientation has been duplicated by the recently discovered Sepphoris synagogue, which also faces in this direction.22 Moreover, other synagogues display a variety of deviations from the generally accepted Jerusalem orientation; thus, the perceived uniqueness of this synagogues orientation has become somewhat attenuated over the years.23 Art. The artistic representation in the Bet Shean synagogues is about as broad as one could imagine, ranging from strictly conservative to strikingly liberal. At the former end of the spectrum stands the Reov building, in which the only decorations are of a geometSamaritans at this time, which was discovered in a side room of the building. However, since this room may be an addition to the original structure, the signicance of the Samaritan inscription remains unclear. Jacoby (Responses, 13031) has added two further considerations in favor of this being a Samaritan synagogue: the absence of animal gures in the panel depicting the Torah shrine (especially when the decorations are compared to those in the nearby Bet Alpha synagogue, which was executed by the same artisans) and the absence of the lulav and ethrog. Both, she claims, reect Samaritan beliefs. The rst claim, however, is questionable, as panels depicting religious symbols in many Jewish synagogues (e.g., ammat Tiberias) are also devoid of gural representation. Thus, other than indicating congregational preference, the absence of gures in this particular context proves very little. However, her second point, the absence of the lulav and ethrog, is more persuasive. Samaritans interpret the biblical reference to the four species (Lev. 23:3944) as parts of the sukkah and not four separate items (which was the Jews interpretation). Thus, their absence as independent items in Samaritan art is not surprising, and we may well have here a distinctively Samaritan feature. Still and all, a Samaritan identication is inconclusive, and the latest excavations at el-Khirbe and Khirbet Samara by Magen (Samaritan Synagogues [Eng.], 193230) have even further beclouded the issue. The newly discovered mosaic oors from these two Samaritan synagogues are quite dierent from the Bet Shean A panel. For example, at Bet Shean two menorot ank the ark, a parokhet (curtain) covers a portal and not an ark, and the parokhet is not gathered to one side, while a symmetrical display of a shofar and incense shovel is found alongside the menorot. None of these features is duplicated in these Samaritan buildings, but they are characteristic of many Jewish synagogues. This Bet Shean synagogue will thus be included in the discussion, albeit with some reservations. In any case, neither its inclusion nor its exclusion will aect the overall picture of diversity, sinceorientation asidethis building does not represent an extreme example of Jewish architectural, artistic, or cultural tendencies among the Bet Shean synagogues. The unusual orientation of this building is irrelevant to the present discussion. The Samaritans, too, seem to have been equally as concerned with their synagogues orientation. The northwesterly direction presents problems in either case. 22. Z. Weiss and Netzer, Promise and Redemption, 1213. 23. See below, Chap. 9. See also Naveh, On Stone and Mosaic, no. 26; Z. Ilan, Ancient Synagogues, 180 82; NEAEHL, loc. cit.; Chiat, Handbook of Synagogue Architecture, 12832.

byzant i ne pa le s t i ne


21. Mosaic oor of the prayer room in Bet Shean B. Note the menorah in the center.

ric nature, except for one plaster fragment depicting the facade of a building (the Ark of the Covenant?) and a menorah.24 In the Bet Shean A synagogue, the geometric patterns on several panels are complemented by one containing a facade (probably referring to the Temple) anked by pairs of menorot, shofarot, and incense shovels. The Maoz ayyim synagogue also features geometric patterns, several representations of birds, and a few religious symbols. The mosaic oor in the prayer room of the Bet Shean B synagogue, with its inhabited scrolls, features gural representations of animals along with an elaborate oral motif (g. 21). This prayer chapel and a large room nearby appear to have shared a common courtyard, leading many to assume that these rooms were part of either a large synagogue complex or a wealthy individuals home that contained a prayer room. If the latter, this setting might have been similar to the one in Stobi,25 where Claudius Tiberius Polycharmos designated part of his house as a synagogue for the local Jewish community. If one accepts this interpretation, then these two sites would exemplify what rabbinic literature refers to as the synagogue of an individual. 26 The mosaic oor of the nearby large room in Bet Shean B is most unusual. One of its panels features scenes from Homers Odyssey, while a second depicts the god of the Nile together with Nilotic motifs (i.e., a series of animals and sh) and a symbolic representation of Alexandria with its customary Nilome-

24. See Vitto, Interior Decoration, 29396. 25. See below, Chap. 8. 26. Y Megillah 3, 4, 74a.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

22. Central panel in the Bet Alpha mosaic oor depicting the four seasons, zodiac signs, and Helios in the center.

ter. Between these two panels is a third one, containing a Greek dedicatory inscription naming one Leontis as the benefactor or owner of the building. Sharing the inclination of Bet Shean B and the nearby Leontis mosaic toward expansive artistic representation is the Bet Alpha synagogue, rich in its diversity, Jewish content, and pagan motifs. Its well-known mosaic oor contains three panels surrounded by geometric borders. One of the panels includes many of the same religious symbols appearing in Bet Shean A, although here birds ank the shrine (as do a lion and ox beside two inscriptions at the northern end of the pavement). The central panel depicts Helios, the zodiac signs, and the four seasons (g. 22), while a third panel illustrates the biblical narrative of the Binding of Isaac (Gen. 22). Although the same artisans, Marianos and his son anina, laid the mosaic oors at both Bet Alpha and Bet Shean A, the style and content at each location are strikingly dierent. Clearly, various types of pattern books were in circulation, leading to very dierent oor designs.

byzant i ne pa le s t i ne


[To view this image, refer to the print version of this title.]

23. Greek (upper) and Aramaic (lower) inscriptions from the Bet Alpha synagogue.

Thus, the oors of these ve contemporaneous Bet Shean synagogues exhibit a rich range of strictly aniconic patterns on the one hand to elaborate representations of Jewish and non-Jewish gural motifs on the other. Inscriptions. The linguistic and cultural ambience of the communities reected in inscriptions also varies widely.27 Not surprisingly, Reov is the most conservatively oriented synagogue, with almost exclusively Hebrew and Aramaic inscriptions. The inscription from Maoz ayyim contains only one Hebrew word, Shalom; Bet Shean A has a number of Greek inscriptions and a palaeo-Hebrew (Samaritan?) one, while Bet Shean B and Bet Alpha have Greek and Aramaic inscriptions (g. 23). The Aramaic inscription from Bet Alpha preserves the date of construction, i.e., the time of Justin, probably referring to Justin I (518527 c.e.). The Greek inscription at this site notes the names of the artisans Marianos and anina.28 The dierences between these communities cultural proclivities are no less striking than their dierences in languages. Not only did Reov avoid using Greek, but the contents of its inscriptions are quintessentially Jewish. As noted, fragmentary inscriptions found on the columns and walls appear to have been blessings for various occasions,29 and the monumental twenty-nine-line, 365-word mosaic inscription in the narthex is entirely halakhic in content (g. 24). A similar inscription (as yet unpublished) appears on one of the synagogues columns, making these two sui generis among synagogue epigraphical evidence anywhere. This stands in contrast with the Homeric and Alexandrian scenes on the Leontis oor or with the Helios and zodiac depictions from Bet Alpha. Moreover, the Bet Alpha mosaic presents an intriguing balance between Jewish motifs and those drawn from the surrounding culture. The zodiac panel was placed between two others bearing
27. For these inscriptions, see Naveh, On Stone and Mosaic; Roth-Gerson, Greek Inscriptions. 28. Roth-Gerson, Greek Inscriptions, nos. 4, 5. 29. Z. Ilan, Ancient Synagogues, 18687.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

24. Halakhic inscription from the Reov synagogue.

a distinctly Jewish stampone with religious symbols and the other with the Binding of Isaac narrative. Taken together with the Alexandrian and Homeric motifs of the Leontis mosaic, the Bet Shean B complex is in some ways the most hellenized of all these sites. In fact, an Aramaic inscription in the prayer room itself contains a number of linguistic mistakes that, according to Kutscher, most likely reect a highly acculturated stratum of Palestinian Jewry.30 There was signicant diversity among ancient synagogues outside of Bet Shean as well. This would have been especially true of synagogues in urban centers, which might have diered considerably from each other in their architectural style, language, and use of motifs, given their varying constituencies. Archaeological evidence from other urban centers, however, is sparse, and thus any comparisons with Bet Shean are almost impossible to draw. The remains of three buildings have been excavated in Tiberias: the famous Severus synagogue excavated in 196163, another one excavated by Slousch in 1921, and a third excavated by Berman and On in 1989. While the latter two are either poorly documented or poorly preserved, the few nds there seem to indicate that these buildings differed in plan, decoration, and other features.31 Similarly with regard to Sepphoris. Aside
30. Kutscher, Language, 5760. 31. NEAEHL, II, 57477; Z. Ilan, Ancient Synagogues, 13943, 14647.

byzant i ne pa le s t i ne


[To view this image, refer to the print version of this title.]

25. Mosaic oor in the Naaran synagogue.

from the recently discovered synagogue, with its mosaic oor rich in artistic motifs and inscriptions, we know of only one fth-century Greek inscription that was originally in a synagogue, probably located in another part of the city. It speaks of leaders from several communities in Phoenicia who settled in the city.32 Finally, only one synagogue has been excavated in Caesarea, although a number of inscriptions, as well as rabbinic sources relating to local synagogues, may point to others there as well.33 Even in the Jericho area there seem to have been striking dierences between neighboring synagogues. One local sixth-century synagogue had demonstrably aniconic decorations, featuring geometric patterns and a stylized ark. Yet, several kilometers away and at about the same time, Naaran boasted a zodiac design, a representation of the biblical Daniel, and an assortment of animal depictions (g. 25). Both synagogues, however, contained only Aramaic or Hebrew inscriptions; Greek was not in evidence. The Golan synagogues also reect a variety of attitudes toward gural art, as do Galilean-type buildings.34
32. Roth-Gerson, Greek Inscriptions, no. 24. 33. Ibid., nos. 2529; L. Levine, Synagogue Ocials; idem, Roman Caesarea, 4245. 34. Maoz, Art and Architecture of the Synagogues of the Golan, 10912; idem, Jewish Art in the Golan; Hachlili, Late Antique Jewish Art from the Golan, 183212.


historical development of the synagogue

Outside the main urban centers of Byzantine Palestine, one is struck by the distinct regional dierences throughout the country. The structures that dominate the Galilee feature monumental facades facing Jerusalem, relief carvings, agstone pavements, two or three rows of columns, few inscriptions or Jewish motifs, and occasionally a permanent bima or two. The Golan synagogues are built of basalt, internally oriented to the south or west, characterized by a richly decorated single entrance (Dikke excepted), a limited repertoire of motifs, and a score of inscriptions, mostly in Aramaic and Hebrew.35 The synagogues further south, in the Jordan Valley, Hebron hills, and Shephelah, generally follow a Byzantine basilical plan but likewise exhibit a good deal of variation.36 Some buildings had an apse, niche, or bima, or a combination thereof. The use of gural representation, as well as of Greek, Aramaic, and Hebrew, also varied considerably (g. 26). By and large, it can be said that gural representation and the use of Greek were ubiquitous in hellenized urban settings and much less common in more remote rural regions. Thus, the less accessible Upper Galilee was considerably more conservative in such matters than the Lower Galilee, with its cities and major roads,37 and southern Judaea, including En Gedi, was less receptive to gural representation than were communities in the coastal region. Diversity among synagogues, however, was not only external, but also penetrated beyond the facades, plans, art, and language of the building; synagogue liturgy likewise varied from place to place at this time. As noted in Dierences in Customs, a composition listing the variant religious practices of Palestine and Babylonia in Late Antiquity: As for the residents of Eretz Israel, the section of the Torah read in one area is not read in another. 38 And, in fact, we know of a number of dierent customs for reading the Torah and haftarah at that time. The prayer service also seems to have varied in dierent locales, as was surely the case with regard to the piyyut or the types of targumim and sermons being oered.39 Having established the rich variety among synagogues in Byzantine Palestine, two caveats are in order. The rst is that this variety existed in an institution that was common to Jews everywhere. There may have been dierent architectural models, but the building served as the public space of each and every Jewish community. The art in these edices may have been selected according to local criteria, but the use of depictions with Jewish content was universal. From one synagogue to the next, the Torah may have been read at a dierent pace or a dierent haftarah may have been recited, but the centrality of these forms of worship was shared by all.
35. Kohl and Watzinger, Antike Synagogen, passim; Hachlili, Late Antique Jewish Art from the Golan, 183212. 36. Kloner, Ancient Synagogues: An Archeological Survey, 1118; Foerster, Ancient Synagogues, 3842. 37. E. M. Meyers, Galilean Regionalism, 93101. 38. No. 48 (pp.188 and 169). 39. Shinan, Sermons, Targums and Reading from Scriptures, 97.; see below, Chap. 16.

26. Plan and mosaics of the Bet Alpha synagogue.


historical development of the synagogue

The second caveat relates to the diversity discussed above. Then, as in earlier times, decentralization and local autonomy characterized Jewish life. To the best of our knowledge, no centralized institution or political-religious oce determined how communities would function on a local level; the result is a rich tapestry of behavioral and communal preferences. We shall return to these two points below.


The abundance of Jewish art from the synagogues of the Byzantine period tells us much about the institution and its adaptation in dierent communal settings. This art reects both the unity and diversity among synagogues and communities. The media of artistic expression utilized in these synagogues are varied and include stone moldings, mosaic oors, and frescoes. The forms of expression range from simple geometric and oral shapes to human and even mythological gures.40 The gural representations themselves are diverse; at times they consist only of birds (En Gedi, Maoz ayyim), an assortment of animals (Maon-Nirim, Gaza, Bet Shean B, Yaa, Chorazim), and, not infrequently, human representations. A common pattern in this last category is that of the four seasons, zodiac signs, and a representation of the sun god on a series of mosaic oors (ammat Tiberias, Sepphoris, Bet Alpha, useifa, Naaran, and Susiya).41 Zodiac depictions have been identied, albeit arguably, on stone fragments from Merot and Baram. Other human depictions appear in various biblical scenes (see below). The ubiquity of such gural representations in synagogue art, as well as in the necropolis of Bet Shearim, has indeed revolutionized our understanding of Jewish attitudes toward this type of artistic expression.42 Before the deluge of archaeological discoveries, beginning with Bet Alpha (192829), Dura Europos (1932), and Bet Shearim (from 1936 onward), it was assumed that Jews did not engage in gural art, or at least did so only on rare occasion. The apparent prohibition of the second commandment, together with a number of Josephus accounts regarding Jewish aversion to gural representations in the late Second Temple period, particularly in Jerusalem,43 essentially determined the matter. However, with the steady accumulation of archaeological material over the past several generations, and the increased scholarly attention, particularly following the publication of Goodenoughs monumental work on Jewish symbols in the 1950s and 1960s, this phe40. For this and what follows, see Hachlili, Ancient Jewish Art and ArchaeologyIsrael, 200224. See also the detailed note of Baron, Social and Religious History of the Jews (1937 ed.), 5153 n. 15. 41. Hachlili, Ancient Jewish Art and ArchaeologyIsrael, 3019; idem, Zodiac, 6176; Goodenough, Jewish Symbols, VIII, 167218. At Sepphoris, the sunand not the god Heliosis depicted above the chariot. 42. See L. Levine, Finds from Beth-Shearim, 27781. 43. See, for example, Antiquities 17, 14963; 18, 5559, 262.

byzant i ne pa le s t i ne


[To view this image, refer to the print version of this title.]

27. Eagle depicted on a Yehud coin (fourth century b.c.e.).

nomenon has been completely reevaluated, rendering a far more nuanced and complex picture than heretofore imagined.44 Today, we can safely conclude that Jews in the biblical and early Second Temple periods did, in fact, make use of a variety of gural representations. Examples from the biblical period include the cherubs over the holy ark, the cherubs and animal gures used by Solomon in his Temple and palace decorations, and the twelve oxen supporting the large basin in the Temple courtyard.45 In additon, the bronze serpent attributed to Moses, the golden calves in the northern sanctuaries of Dan and Beth-el, and innumerable gurines and seal engravings (e.g., of lions, horses, gazelles, cocks, snakes, and monkeys) found at Israelite sites and dating primarily to the eighth and seventh centuries b.c.e. all point to a generally permissive attitude toward this art form in the First Temple era.46 Even as late as the Persian and early Hellenistic eras, the Yehud coins minted in Jerusalem feature a wide variety of gural representations, including owls, eagles, a winged leaping animal, a Persian king, a divine gure sitting on a winged wheel, a warrior, a governor, a high priest, and depictions of Ptolemy, Berenice, and Athena (g. 27).47 At the turn of the second century b.c.e., the Tobiad Hyrcanus used a variety of carved animal reliefs when building his estate east of the Jordan River.48
44. Avigad, Beth Shearim, III, 27587. See also Schubert, Jewish Pictorial Traditions, 14759. 45. Exod. 25:20; 36:35; I Kgs. 7:23. 46. Num. 21:9; II Kgs. 18:4; I Kgs. 12:2633; A. Mazar, Archaeology of the Land of the Bible, 5017; Isserlin, Israelite Art, 4450. 47. Meshorer, Ancient Jewish Coinage, I, 1334; idem, Treasury, 1121; Barag, Coin of Bagoas, 9799; idem, Silver Coin, 414. It is interesting to note that in the Persian period, although Jewish society was largely isolationistat least as far as our literary sources indicateJudaean coins were heavily inuenced by foreign motifs, including what appear as rather daring gural representations. In contrast, the Hasmonean era witnessed a great deal of Hellenistic inuence, especially within the ruling elite, although its coins and other forms of artistic representation were strictly aniconic. 48. NEAEHL, II, 64748.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

28. Coin of John Hyrcanus. Obverse: palaeo-Hebrew inscription. Reverse: double cornucopias with a pomegranate.

However, beginning in the later Hellenistic (i.e., Hasmonean) period, the pendulum swung sharply in the opposite direction. Strict avoidance of animal and human depictions became the norm in Jewish society for some three hundred years or so, commencing with the rise of the Hasmoneans and lasting until the aftermath of the Bar-Kokhba revolt (ca. 150 b.c.e.150 c.e.).49 The reasons for this radical about-face with respect to gural representation are not entirely clear (g. 28), and a variety of suggestions has been put forth: (1) a traumatic reaction to the 167 b.c.e. desecration of the Temple by Antiochus IV, which included coercive pagan worship; (2) a more stringent position in this regard reecting the attitude of the Sadducees, who were generally in control of aairs in Jewish Palestine in the late Second Temple period; (3) a strict policy advocated by the Pharisees; (4) the direct result of Hasmonean policy, which aimed, in part, at cultivating unique Jewish modes of expression, including art; and (5) a Jewish reaction to hellenization and the threat of foreign inuences.50 Thus, the reintroduction of gural representation beginning in the late second and
49. This gural avoidance is evident rst and foremost in the coins of the era; see Meshorer, Ancient Jewish Coinage, I, 3598; II, 5165; idem, Treasury, 2357. See also Avigad, Discovering Jerusalem, 14450; Kon, Jewish Art, 6064; Hachlili, Ancient Jewish Art and ArchaeologyIsrael, 65119. 50. For these various opinions, see Avigad, Beth Shearim, III, 27778; M. Smith, Goodenoughs Jewish Symbols, 60; Goodenough, Jewish Symbols, IV, 67; Simon, Verus Israel, 23. See also Avi-Yonah, Jewish Art and Architecture, 25063; idem, Oriental Art in Roman Palestine, 1327; B. Cohen, Art in Jewish Law, 167. Nevertheless, here, as elsewhere, not all the evidence corresponds with this avoidance, especially in the Herodian era. A few depictions of sh and birds appear among the archaeological remains from the Jewish Quarter, while Herod (on one occasion) and some of his descendants (Philip and Agrippa I) used gures on some of their coins, as did Antipas in his palace in Tiberias. See Avigad, Discovering Jerusalem, 150, 169; Meshorer, Ancient Jewish Coinage, II, 29, 4446, 6061; Josephus, Life 6566. According to Josephus (War 5, 181), the fountains in Herods Jerusalem palace were in the shape of bronze animals. Finally, an enigmatic statement in the Tosefta (Avodah Zarah 5, 2 [p. 468]) claims that stamps or seals bearing imagesnot humanwere known in Jerusalem.

byzant i ne pa le s t i ne


third centuries c.e., and rst evidenced on Galilean urban coins and at Bet Shearim, is far from exceptional in the wider perspective of Jewish history. It reects yet another shift of the pendulum with regard to Jewish attitudes toward gural representation owing to internal needs and in response to the wider cultural, social, and political contexts in which the Jews found themselves (g. 29). Here, too, a number of suggestions (not necessarily mutually exclusive) have been oered to explain the reintroduction of gural art: (1) the ever increasing hellenization of the Jewish population; (2) the Jews increasing minority status, which made the need for social, economic, and cultural accommodation to the outside world ever more pressing; (3) the decline of paganism and the acceptance of the view that images were not a threat; (4) the disappearance of a cluster of religious phenomena (e.g., the apocalyptic-messianic mode, the striving for political independence, widespread purity concerns) that characterized the Jewish society of the late Second Temple period (from the Hasmonean era onward); and (5) the liberal stance in this regard adopted by Patriarchal circles.51 Whatever the reason(s), the exigencies of the time were ultimately the determining factors in this last-noted shift. While we do not know why and how this happened within Jewish society generally, as reected in the many opinions cited above, we do have some evidence of this shift within rabbinic, and more specically Patriarchal, circles. Flexibility among the sages often derived from a creative use of hermeneutics, which itself was driven by historical necessity and, at times, ideological convictions.52 This was certainly the case with the Patriarchate, the leading communal oce among Jews from the second to fth centuries c.e., and was dramatically expressed in a story regarding Rabban Gamaliel II (ourished ca. 90120 c.e.):
Proklos, the son of Philosophos, asked [a question of ] Rabban Gamaliel while the latter was bathing in the bath of Aphrodite in Acre, saying to him: It is written in your Torah, and there shall not cleave to you any of the devoted [i.e., forbidden] thing [Deut. 13:18]. Why do you thus bathe in the bath of Aphrodite? He answered: One may not give an answer

51. See Simon, Verus Israel, 2327; Urbach, Rabbinical Laws of Idolatry, 154., 236.; J. Baumgarten, Art in the Synagogue, 198, 201, 206, and my forthcoming book, Visual Judaism. Urbach has built his oft-quoted theory on this very assumption, namely, that the fear of idolatry lurked behind the earlier aniconic posture, and that paganisms decline in the second and third centuries led to a more liberal rabbinic legislation vis--vis gural art. Such an assumption, however, is problematic. There is no evidence that at any time in the Second Temple period idolatry was perceived as a threat to the Jews, except, of course, during the three years of Antiochus persecutions (167164 b.c.e.). This view is attested in both Second Temple non-rabbinic and later rabbinic sources; see Judith 8:18; Josephus, War 2, 19598; Song of Songs Rabbah 7, 13 (ed. Dunesky, p. 160; Vilna edition, 7, 8); B Yoma 69b (= B Sanhedrin 64a). Thus, Jewish avoidance of gural art ought to be attributed to other causes, no less than their subsequent re-adoption of this form of representation. See also Hachlili, Ancient Jewish Art and ArchaeologyIsrael, 28587, as well as my Visual Judaism. 52. See below, Chap. 13.

[To view this image, refer to the print version of this title.]

29. Mosaic oor in the ammat Tiberias synagogue.

byzant i ne pa le s t i ne


in the bath. And when he came out he said: (1) I did not come into her [Aphrodites] area [lit., boundary], she came into mine. People do not say, Let us make a bath for Aphrodite, but rather, Let us make a [statue of ] Aphrodite as an adornment for the bath; (2) moreover, even if they would oer you much money, you would not enter your place of worship [lit., idolatry] naked and suering pollution and urinate in front of her [i.e., Aphrodite]. But she stands at the edge of the gutter, and everyone urinates in front of her; (3) the verse only refers to their gods; that which they treat as a god is forbidden and that which they do not treat as a god is permitted. 53

The response attributed to Rabban Gamaliel is as fascinating as it is far reaching. He is quoted as oering three reasons for frequenting a pagan-ornamented bathhouse. The rst deals with the denition of the buildings function: Was it built to serve as a pagan sanctuary or a bath? Was the statue of Aphrodite inherent to the buildings function, or was it merely an ornamentation? Rabban Gamaliels answer was that the statue was meant to be purely decorative, as the building itself was intended to fulll a secular purpose. Secondly, the nature of a facility should also be judged by what people actually do there. When one walks around naked and urinates with no regard to the presence of a statue of a deity, the statue is clearly of no real consequence for those in attendance. Thus, the bathhouse was not to be regarded as something sacred or specically pagan. The third claim, however, is the most far reaching. One should view a place or an object as idolatrous only if it is so regarded by the pagans themselves; if it is not considered idolatrous, but only decorative, it ought to allow for a more permissive attitude and behavior. This last response is the most revolutionary precisely because it is cast as a general principle. Nothing, not even a statue, is inherently forbidden; everything depends on its function and on the intention of those who placed it there.54 The transition from a rigid aniconic posture in the late Second Temple period to a less restrictive stance in Late Antiquity was not always easy or smooth. Dierences of opinion might often be sharp and even bitter, and this was true within rabbinic circles as well.55 However, it is only for Byzantine Palestine that our diverse sources furnish us with
53. M Avodah Zarah 3, 4. For a dierent dating of this story, to the time of Rabban Gamaliel III (early third century), see Wasserstein, Rabban Gamliel and Proclus, 25767. See also the story recorded in T Moed Qatan 2, 15 (p. 372), recounting how Rabban Gamaliel sat on a bench of gentiles (perhaps ordinarily used for commercial purposes) on Shabbat, much to the chagrin of his Jewish hosts. 54. The same Rabban Gamaliel had various representations of the moon in his upper chamber, which he would use when cross-examining witnesses testifying to a new moon (M Rosh Hashanah 2, 8). An analogous situation cropped up in the mid third century regarding the use of the Bostra nymphaeum by Jews for drinking water. R. Yoanan overruled his friend and colleague Resh Laqish and permitted such usage, despite the presence of statues and the fact that the site was occasionally used for pagan rituals. See Y Sheviit 8, 11, 38bc; B Avodah Zarah 58b59a; and Blidstein, R. Yohanan, Idolatry and Public Privilege, 15461. 55. See J. Baumgarten, Art in the Synagogue, 198204; L. Levine, Judaism and Hellenism, 10610; and esp. below, Chap. 13.


historical development of the synagogue

a detailed and nuanced picture of the varied attitudes toward gural art within Jewish society. The use of animal and human gures had become, on the one hand, quite acceptable in many communities, although some were clearly more daring than others. On the other hand, our data also point to individual communities or entire regions that tended to be more conservative, studiously avoiding such representations. Predictably, many of the latter communities tended to be in rural districts or relatively isolated regions, such as the Upper Galilee and southern Judaea, although conservative communities may have also been found in or near large urban centers, as, for example, the Reov synagogue in the Bet Shean area.


A survey of synagogue art in Late Antiquity reveals a seemingly endless variety of patterns, designs, and motifs of both Jewish and non-Jewish content. Jews borrowed heavily from the Byzantine world, and many motifs and patterns appear in both Christian and Jewish frameworks. This commonality encompasses all kinds of representations, from geometric and oral patterns to depictions of birds, sh, animals, and humans.56 The motifs include round medallions formed by vine tendrils emerging from an amphora, baskets of bread and fruit, cornucopias, clusters of grapes, owers, oxen, eagles, and lions. Except for Dura Europos, where a sui generis plethora of biblical scenes appear, there is a surprising overlap between scenes appearing in ancient synagogues and churches: the Aqedah, Daniel in the lions den, Noahs ark, and even King David. The exceptional panel in this regard is Helios and the zodiac, an eminently pagan theme that appears in a half dozen Palestinian synagogues but never in churches. The similarity between certain synagogues and churches is at times so striking that it has often been assumed that they used the same artisans or pattern books. In comparing the synagogues of Gaza and Maon and the nearby Shellal church, it becomes clear that the Jews substituted the menorah for the cross when adapting and judaizing a well-known Byzantine mosaic pattern for their own use.57 Whether all these mosaics were produced in the same workshop is a moot issue, although Rutgers has suggested the use of common workshops for Jews and Christians in Rome.58 Similarly, the Susiya synagogue may have
56. Ovadiah, Art of the Ancient Synagogues, 30118; B. Narkiss, Pagan, Christian and Jewish Elements, 18386; Hachlili, Ancient Jewish Art and ArchaeologyIsrael, 310., 366. 57. Avi-Yonah, Art in Ancient Palestine, 38995; idem, Mosaic Floor of Maon Synagogue, 7785; Ovadiah, Mosaic Workshop, 36772; as well as the comments of Hachlili, On the Gaza School of Mosaicists, 4658; eadem, Ancient Jewish Art and ArchaeologyIsrael, 31016, 390; and N. Stone, Notes on the Shellal Mosaic, 20714. On the situation in Rome, see Rutgers, Jews in Late Ancient Rome, 70 76, 83. 58. Ibid., 5099.

byzant i ne pa le s t i ne


[To view this image, refer to the print version of this title.]

30. Aqedah scene in the Bet Alpha mosaic.

adopted a pattern of the Temple facade, which was also being used in the sixth-century Madeba church.59 This common reservoir of motifs came from a limited menu of Jewish symbols and depictions, together with a much wider range of motifs hailing from the Byzantine world. This was the setting for a series of creative syntheses in synagogue art as Jewish communities throughout the country forged their own combinations from the options available. An important facet of synagogue art in the Byzantine period is the greatly increased use of Jewish motifs and symbols as compared with earlier times. Interestingly, biblical scenes or gures were only sparingly used.60 The Aqedah (Binding of Isaac) scene at Bet Alpha is the most elaborate example of biblical art (g. 30); others include the gures of Daniel (Naaran, Susiya), David (Gaza and possibly Merot; g. 31), and Aaron, as well as fragmentary representations of the Aqedah and possibly the visit of the three angels to Abraham and Sarah (Sepphoris), and what is arguably a depiction of the symbols for the twelve tribes (Yaa). Regarding Jewish symbols, the menorah is far and away the most
59. Foerster, Allegorical and Symbolic Motifs, 54647. 60. No less remarkable is the fact that there are no traces of biblical gures or scenes in Jewish funerary art in either Roman-Byzantine Palestine or the Diaspora.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

31. Representation of David from the Gaza synagogue.

common one, often appearing with a shofar, lulav, and ethrog. A frequent pattern in synagogue mosaics features a Torah shrine anked by pairs of menorot, shofarot, lulavim, ethrogim, and incense shovels.61 These clusters of Jewish symbols frequently accompanied the zodiac signs, although they often appeared independently as well (e.g., Bet Shean A; g. 32). The meaning and signicance of the above-mentioned cluster of symbols have long been debated (g. 33). One popular theory maintains that they were primarily intended to recall the Jerusalem Temple (or possibly the Wilderness Tabernacle),62 with the facade representing that of the Temple and the menorah, shofar, lulav, ethrog, and incense shovel symbolizing accoutrements once used in that setting. If this interpretation is granted,
61. Hachlili, Ancient Jewish Art and ArchaeologyIsrael, 234300; B. Narkiss, Representational Art, 366. 62. B. Narkiss, Scheme of the Sanctuary, 13; Khnel, Jewish Symbolism, 14749; Grossberg, Reactions, 6465. See also Branham, Vicarious Sacrality, 31945; Z. Safrai, From Synagogue to Little Temple, 2328. For other possible explanations of the incense shovel, see M. Narkiss, Origins of the Spice Box, 32.

byzant i ne pa le s t i ne


[To view this image, refer to the print version of this title.]

32. Panel from the Bet Shean A synagogue exhibiting a series of Jewish symbols.

then the clear implication is that remembering the Temple was of paramount importance in many Byzantine Jewish communities. The appearance of this motif in the synagogue could then be viewed as triggering a memory of that institution and hopes for its restoration, or perhaps as reecting a desire that the synagogue be considered a continuation of the Temple in terms of sanctity and religious signicance. A second approach regards these religious symbols within the context of the synagogue itself.63 The facade is thus interpreted as a representation of the Torah shrine, while the other symbols represent the various objects found in the synagogue setting or used in the synagogue service. By Late Antiquity, the shofar and lulav had become integral parts of synagogue worship. Each of the above interpretations, however, has its weaknesses. To interpret the facade as a depiction of the Temples exterior is problematic. It rarely resembles either what we know of the Temple facade (via Josephus or Mishnah Middot) or, in fact, of the Temple facade depicted in second- and third-century Jewish art (e.g., on the Bar-Kokhba coins or the Dura frescoes). Given this fact, it has been further suggested that the synagogue depictions refer to an inner portal of the Temple, but this, too, is unconvincing. The incense shovel and menorah (though not two menorot) are certainly appropriate for the
63. Dothan, Hammath Tiberias, 3339; Goodenough, Jewish Symbols, IV, 11136; XII, 8386; Hachlili, Ancient Jewish Art and ArchaeologyIsrael, 27280. Amit, Reactions and Comments, 6566.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

33. Cluster of Jewish symbols from the ammat Tiberias synagogue.

Temple context, but one wonders why the shofar, lulav, and ethrog were so emphasized, since they appear to have been quite peripheral to ocial Temple worship. Moreover, with the Temple in mind, one could readily suggest other items that would have been even more closely identied with its ritual: the altars, showbread table, priestly garments, etc. On the other hand, the symbols depicted seem to t a synagogue context rather nicely, the one problematic element being the incense shovel. It patently does not apply to the synagogue or its liturgy, at least on the basis of what we know to date.64 A third approach, intriguing though not without its own problems, interprets a number of these symbols as referring to the major holiday season during the month of Tishri, when three major Jewish festivals occur in rapid succession: Rosh Hashanah (the New Year), Yom Kippur (the Day of Atonement), and Sukkot.65 Thus, some of the above-noted symbols can easily be associated with one of these holidays: shofarRosh Hashanah; in64. It seems, however, that at least some Jews used incense in tombs, as reported by the sixth-century Christian pilgrim Antoninus Martyr regarding the Cave of Machpelah in Hebron; see Tobler and Molinier, Itinera Hierosolymitana, 374; Antoninus Martyr, in PPTS (p. 24). 65. First suggested by Braslawski (Symbols, 11518) and later espoused by Wirgin (Menorah as a Symbol of Judaism, 14142); Eitan (Menora as a Symbol, 49); and Fine (This Holy Place, 121).

byzant i ne pa le s t i ne


cense shovelYom Kippur;66 lulav and ethrogSukkot. The last-mentioned holiday was considered the most popular and important of the pilgrimage festivals in the late Second Temple period, while Rosh Hashanah and Yom Kippur became known in rabbinic tradition as the Days of Awe (or, in modern parlance, the High Holidays). The latter two holy days were greatly developed in scope and content by the rabbis following the destruction of the Second Temple and came to represent a wide range of ideological themes and religious values. Indeed, it may not be coincidental that the High Holidays were more exclusively synagogue-centered than other festivals. In this vein, it is interesting that John Chrysostom chose to inveigh against these holidays and symbols in one of his homilies against the Jews:
What is this sickness? The festivals of the wretched and miserable Jews which follow one after another in successionTrumpets, Booths, the Fastsare about to take place. And many who belong to us and say that they believe in our teaching, attend their festivals, and even share in their celebrations and join their fasts. It is this evil practice I now wish to drive from the church.67

Elsewhere, Chrysostom rails at those Christians who are drawn to the sound of the shofar (Rosh Hashanah), dancing (!) and fasting (Yom Kippur), and building booths (Sukkot).68 Thus, it is not inconceivable that synagogues featured these particular symbols, which were associated with the holidays of the month of Tishri and recognized for their centrality by non-Jews as well. It is these very symbols that are often depicted on fragments of gold glass found in the Jewish catacombs of Rome.69 Nevertheless, it must be admitted that the weakness of this theory lies in the fact that it relates to only three of the symbols appearing in these panels, and the least prominent ones at that. The predominant menorah and Torah shrine are not addressed (g. 34). Still another interpretation of this cluster of Jewish symbols is more inclusive in nature. Rather than viewing all of these symbols as a reection of one particular institution (the Temple or the synagogue) or time framework (Tishri), we may indeed be dealing with several simultaneous allusions. Thus, these symbols may actually point to both the Temple and synagogue at one and the same time by amalgamating their representative symbols. Alternatively, one might suggest that each community chose the symbol most meaningful to it, either the Temple or the Torah shrine. The common denominator, i.e., the facade or holy objects, could be interpreted in accordance with each communitys preference.
66. On the importance of the incense shovel for the Yom Kippur ceremony, see Lauterbach, Rabbinic Essays, 5183. 67. Adv. Iud. 1, 1, 844 (also in Meeks and Wilken, Jews and Christians, 86). 68. See Adv. Iud. 1, 5, 851; 1, 2, 846; 7, 1, 915; and Wilken, John Chrysostom and the Jews, 75. 69. Goodenough, Jewish Symbols, II, 10819; Leon, Jews of Ancient Rome, 21824; Noy, JIWE, II, nos. 58897; Rutgers, Jews in Late Ancient Rome, 8185.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

34. Torah shrine anked by menorot and other symbols from the Bet Alpha synagogue.

Finally, a variation of the last-mentioned inclusive approach might view these symbols as representing two basic concepts in Judaism, the Temple and the Torah. Certain symbols are clearly associated with the Temple setting; others would seem to indicate the sanctity of the Torah shrine. Since these two dimensions were often associated with each other in Jewish tradition, starting with placing the two tablets of stone bearing the Ten Commandments together with Moses Torah (Deut. 31:9, 26) in the Wilderness Tabernacle and later in the First Temple, this combination may have found expression here as well. Whatever their specic signicance and meaning, the widespread use of these motifs in a synagogue context certainly gave vivid expression to the institutions religious dimension as well as to the shared symbolic language of Jewish communities everywhere.


The Byzantine period witnessed signicant strides in the transformation of the synagogue building into a distinctively religious institution. What had begun to emerge in

byzant i ne pa le s t i ne


35. Reconstruction of the Khirbet Shema synagogue.

late Roman Palestine was now more fully realized. The Jerusalem orientation of synagogue buildings was further emphasized by the almost universal practice of placing a platform, niche, or apse for the Torah shrine against the Jerusalem-oriented wall.70 In the Galilean-type synagogue, such a podium was located between the entrance portals facing Jerusalem, either on one or both sides of the main entrance (e.g., Gush alav, Nevoraya, and Merot). In other types of buildings, the platform was situated in the center of the wall facing Jerusalem (as at Reov and Khirbet Shema [g. 35]), and in Susiya there seem to have been two such podiums. The emphasis on a Jerusalem orientation was even greater in synagogues adopting a Christian basilical plan that incorporated a niche or apse along the wall facing Jerusalem, with the entrance, atrium, and narthex on the opposite side. Architecturally, these plans guided the synagogue participant to face in a specic direction, and if the prayer leader, Torah reading, or preaching were likewise positioned in or near the apse (or niche), then this focus would have become even more pronounced.71 The appearance of a bima or apse as a xture in most synagogues was not simply an architectural addition. It signied that the Torah shrine was now accorded a permanent and central status within the hall. It will be remembered that Second Temple synagogues had no such arrangement; the Torah scroll(s) was kept elsewhere and introduced into the assembly hall at an appointed time. This feature began to change in the latter part of the Roman period, and by Late Antiquity the presence of a Torah shrine had become the norm as well as an important component in determining the religious ambience of the synagogues main hall; all those present faced the Torah ark and the Jerusalem-oriented wall. This is in contradistinction to the earlier custom, whereby all faced the center of the hall or the elders faced the congregation.72
70. See Fine, This Holy Place, 10511. For further discussion of synagogue orientation, see below, Chap. 9. 71. Hachlili, Niche and Ark, 353. 72. See above, Chap. 3 as well as T Megillah 3, 21 (p. 360); Dierences in Customs, no. 36 (p. 156).


historical development of the synagogue

[To view this image, refer to the print version of this title.]

36. Greek (upper) and Aramaic (lower) inscriptions from the ammat Tiberias synagogue.

The artistic dimension, with its rich repertoire of Jewish symbols and biblical representations, is another indication of the increased religious prole of the Late Antique synagogue. In addition, literary sources 73 and a considerable number of inscriptions from this period refer to the synagogue as a holy place ( or hagios topos) and to the community as a holy congregation ( ) or a holy avurah (or association .) References to a holy place appear in synagogues throughout the country, in ammat Tiberias (twice; g. 36), Naaran (four times), Kefar ananiah, Ashkelon, and Gaza, and to the most holy [place] in Gaza and Gerasa.74 Inscriptions from Bet Shean, Jericho, and Susiya mention a holy congregation or community.75 The term the language of the holy house ( ) occurs frequently in Targum PseudoJonathan and may refer to the language used in the synagogue setting,76 while in the same targum of Lev. 26:1, it is the synagogue that almost assuredly is referred to as a sanctuary.77 Thus, we are on safe ground in assuming that in the course of the Byzantine period, the synagogue came to be widely viewed as a holy place,78 a status already articulated in fourth-century Byzantine Imperial legislation. Valentinian I (ca. 370 c.e.) refers to the synagogue as a religionum loca when prohibiting soldiers from seizing quarters there.79
73. For example, Ecclesiastes Rabbah 8, 10. See also Y Bikkurim 3, 3, 65d, where a Caesarean synagogue is likened to Gods Temple. 74. Holy place: Naveh, On Stone and Mosaic, nos. 16, 26, 60, 64, 65; Roth-Gerson, Greek Inscriptions, nos. 3, 17, 23. The most holy place: ibid., nos. 10, 21. 75. Naveh, On Stone and Mosaic, nos. 46, 69, 84. 76. Shinan, Aramaic Targum as a Mirror, 24849. 77. See comments in Blidstein, Prostration and Mosaics, 3739; and below, Chap. 13. On the issue of synagogue sanctity, see Schubert, Jewish Pictorial Traditions, 16170; and esp., for a detailed treatment, Fine, This Holy Place. 78. In the early third century, the term holy also appears on coins from Sepphoris with regard to the city and its council (Meshorer, Sepphoris and Rome, 16869). 79. Cod. Theod. 7, 8, 2 (Linder, Jews in Roman Imperial Legislation, 16163).

byzant i ne pa le s t i ne


Such a status is likewise assumed in other edicts issued over the next half-century that aimed at protecting synagogues from violence.80 The tendency to associate synagogue and Temple, rst noted in Palestine in the postTemple era, gained momentum in the Byzantine period. The appearance of plaques containing lists of the twenty-four priestly coursesin Caesarea, Ashkelon, Reov, perhaps Kissum (near Gaza), Nazareth, and Yemenmay serve as evidence of the associations being forged at this time between the synagogue and the memory of the Temple.81 In contemporary piyyutim, the subject of the twenty-four priestly courses became a frequent motif.82 One Byzantine paytan, Hadutha (or Hadutaha), wrote piyyutim for each of the twenty-four priestly courses, and other synagogue poets were themselves priests and seem to have been inclined to focus on Temple-related matters as well.83 We noted in the previous chapter that during the third and fourth centuries Templerelated matters were being introduced into various parts of the synagogue liturgy, for example, the Mussaf Amidah and the maftir reading for holidays, both of which focus on the Temple sacrices.84 It was also about this time that the Avodah service recalling the Temple ritual on Yom Kippur was added to that days synagogue liturgy, as was the recitation of a psalm for each day of the week, a custom rst documented for priests in the Temple.85 In truth, this latter practice is rst explicitly mentioned as an element of synagogue worship only in a source from the seventh or eighth century, but it almost surely originated before then.86 The addition of the psalm not only served to expand the syna80. See above, note 3. 81. Caesarea: Avi-Yonah, Caesarea Inscription, 4657. Ashkelon: Sukenik, Three Ancient Jewish Inscriptions, 1617. Reov: Z. Ilan, Ancient Synagogues, 186. Kissum: Z. Ilan, Broken Slab, 22526; Nazareth: H. Eshel, Fragmentary Hebrew Inscription? 15961. Yemen: Degan, Inscription of Twenty-Four Priestly Courses, 3023. See also Naveh, On Stone and Mosaic, nos. 51, 52, 56, 106. However, these plaques may possibly reect the increasing prominence of priests in Jewish society of Late Antiquity; see Irshai, Role of the Priesthood, 7585. 82. S. Klein, Galilee, 17792; and the numerous articles on the subject by Fleischer, e.g., Regarding the [Priestly] Courses, 14261; Piyyutim of Yannai, 17684; see also idem, Additional Data, 4760. 83. Zulay, On the History of the Piyyut, 11120; Fleischer, Hadutha, 7196. Some paytanim were priests, e.g., R. Pinas biribi Hakohen (Fleischer, Early Paytanim of Tiberias, 370); R. Simeon bar Megas (Yahalom, Liturgical Poems of imon bar Megas, 11). In addition, piyyutim written for the Ninth of Av often focus on the Temple priests, their roles, and the Temple cult; see Fleischer, Qaliric Compositions, 140. On the connection between hekhalot mysticism and Temple traditions, see Elior, From Earthly Temple, 21767. 84. See above, Chap. 6. 85. On the Avodah service, see B Yoma 36b and 56b. Although these traditions are from fourthcentury Babylonia, there is little doubt that such a practice was current in Palestine as well, as attested by the scores of piyyutim that highlight the Avodah service; see Elbogen, Studien, 49.; Goldschmidt, Mahzor, I, xviiixxv; L. A. Homan, Canonization, 107. On the psalms associated with specic days of the week, see Avot de R. Nathan, A, 1 (p. 3a). 86. Tractate Soferim 18, 2 (pp. 31013).

240 h i s t o r i c a l d e v e l o p m e n t o f t h e s y n a g o g u e

gogue liturgy by adopting older patterns of Temple worship but, in turn, reinforced the link between these two institutions. Moreover, we recall the claim ascribed to R. Joshua b. Levi (alternatively: the sages) that the Amidah prayer was introduced as a substitute for sacrices ( 78.) Temple terminology seems to have permeated other aspects of synagogue life as well. In a revealing account of Joseph the Comes collecting taxes for the Patriarch in Cilicia, Epiphanius refers to these monies as tithes and rstfruits. If indeed accurateand it would be dicult to imagine a reason for inventing such termsthen Temple-related expressions for obligatory donations were being invoked by the fourth-century Patriarchate.88 Two very dierent sources from Late Antiquity make this connection between the synagogue and Temple eminently clear. The rst is a halakhic work from the end of the Byzantine period, referred to today as the Book of Court Cases ( ) or the Literature of Court Cases (:)
And thus said the sages: One must not enter the Temple Mount with his sta and shoes. And if, owing to our sins, the Temple Mount is no longer available to us, a lesser sanctuary is and we must behave in [it] in a spirit of holiness and fear, as is written: You must fear My sanctuary [Lev. 19:30]. Therefore, our ancestors have determined that in all synagogue courtyards there should be basins of fresh water for sanctifying [i.e., washing] hands and feet. 89

Even more explicit is a fragment of a midrash on Deuteronomy found in the Cairo Genizah:
As long as the Temple existed, the daily oerings and sacrices would atone for the sins of Israel. Nowadays, the synagogues of Israel replace the Temple, and as long as Israel prays in them, they, in eect, replace the daily oerings and sacrices; and when prayers are recited [therein] at the proper times and [the Jews] direct their hearts [to God through their prayers], they gain merit and will see the rebuilding of the Temple and the sacricing of the daily oering and [other] sacrices, as it is written: And I will bring them to My holy mountain, and I will rejoice in My house of prayer; their sacrices and oerings are welcome on My altar, for My house will be called a house of prayer for all peoples [Isa. 56:7].90

In the former source, the synagogue is acknowledged as inferior to the Temple, yet was still assigned a sucient degree of holiness to warrant a vessel being placed there for the washing of hands and feet. Purication, of course, was crucial for Temple visitors, and the purpose of this custom in the synagogue was undoubtedly to introduce a modicum of such purity and sanctity. In the latter source, however, the synagogue is accorded an
87. B Berakhot 26b. Y Berakhot 4, 1, 7b, ascribes this tradition to the sages ( .) Cf. however, the comments of Blidstein, Prayer, 243 n. 26. 88. Panarion 30, 11, 2. 89. Margoliot, Palestinian Halakhah, 13132. 90. Ginzberg, Geniza Studies, I, 15253.

byzant i ne pa le s t i ne


even higher status than before, as it, along with prayer, replaced the Temple and sacrices. Moreover, it claims that proper observance of prayer will lead to the reestablishment of the Temple, a situation already foreseen by the prophet Isaiah. The addition of an eternal light ( ) to the synagogue at some point during these centuries further strengthened its association with the Temple. Although lighting xtures for the synagogue are noted in tannaitic sources, there is no indication that these were intended for anything more than basic illumination within the building.91 In several late midrashim, however, the presence of an eternal light is explicitly attested. Clearly, the association of synagogue and Temple had become so accepted during this period that, at some point, a biblically ordained appurtenance intended for the latter was introduced into the former. Another tradition explains the commandment to have lights in synagogues in a very simple and straightforward manner: Synagogues and academies are like the Temple. 92 A striking example of the perpetuation of the Temples memory within the synagogue is the mosaic oor in the fth-century synagogue at Sepphoris, which portrays a series of Tabernacle-Temple oerings.93 Each is depicted graphically and is accompanied by the Hebrew term, e.g., the daily sacrice ( ,)the oil ( )and the meal ( )oerings. Various items used in Tabernacle-Temple ceremonies are likewise depicted, such as trumpets ( ,)the showbread table, and the basket of rstfruits. No other synagogue, with the possible exception of Dura Europos, makes as clear a statement regarding Temple worship in a synagogue setting. Some of the reasons for the synagogues evolution into an institution with a pronounced religious character were suggested in the previous chapter. We noted reasons such as the association of the synagogue with the Temple, the beginnings of a permanent presence of the Torah scrolls in the main hall, the increasing importance of public prayer, and the possible inuence of Diaspora models, some of which, as we have seen, had already achieved a holy status. This process continued into the Byzantine era with a heightened impetus, and several additional factors appear to have played a signicant role at the time. The rst was the remarkable development in the concept of holiness throughout many parts of Late Antique society. Holiness as a religious category characterizing places, people, and objects was becoming an ever greater concern in a wide variety of religious circles. For instance, the holy man had become a well-recognized phenomenon in Late Antiquity.94 Browns pathbreaking studies of the Christian holy man or saint have been followed by that of Fowden with regard to pagan holy gures and by Kirsch-

91. M Pesaim 4, 4. See also Y Pesaim 4, 9, 31b; B Pesaim 53b. 92. Midrash Hagadol, Numbers 8:1 (p. 119). 93. Z. Weiss and Netzer, Promise and Redemption, 2025. 94. On the holy person in the Hellenistic and Roman worlds, see G. Anderson, Sage, Saint and Sophist. Regarding holiness in paganism, see Bowersock, Hellenism, 1528.


historical development of the synagogue

ners comparison of pagan, Christian, and Jewish models.95 It would be dicult to isolate the Jews or the synagogue from these developments; some sort of inuenceat the very least, the proverbial Zeitgeistought to be posited. This was certainly a factor in the practice of placing amulets containing personal prayers on synagogue premises.96 However, a second and more immediate impetus was the dramatic development and growing presence of Christianity throughout much of Palestine at this time.


There is no doubt that the era of Constantine ushered in a period of profound change in the way Christians related ideologically and liturgically to Palestine, Jerusalem, and other holy sites, as well as to the increasingly widespread phenomenon of pilgrimage to the Holy Land.97 There is no gainsaying the dramatic eect of Constantines massive infusion of money and resources into recasting Jerusalem as a Christian city, rst and foremost by building a series of magnicent churches, sometimes referred to as pilgrimchurches, in order to commemorate events in the life of Jesus. This shift from a spiritual emphasis to one that included the temporal as well is what J. Z. Smith has termed a move from the utopian to the locative mode.98 Constantine created, for the rst time, a Christian Holy Land, laid palimpsest-like over the old, and interacting with it in complex ways, having for its central foci a series of Imperial-dynastic churches. 99
95. Brown, Society and the Holy, 10365; idem, Saint as Exemplar, 125; Fowden, Pagan Holy Man, 3359; Kirschner, Vocation of Holiness, 10524. See also Valantasis, Spiritual Guides, 133; Markus, End of Ancient Christianity, 199211; Phillips, Sociology, 275264. Regarding rabbinic tradition specically, see L. Levine, Rabbinic Class, 1059, and literature cited therein, to which might now be added Diamond, Hunger Artists and Householders, 2947. 96. See, for example, Naveh and Shaked, Amulets and Magic Bowls, 40122; idem, Magic Spells and Formulae, 43., for which other examples are now known (e.g., Baram). 97. See Markus, Christianity in the Roman World, 87.; MacMullen, Christianizing the Roman Empire, 4351. The revolution in Christian life that began unfolding in Palestine in the fourth century could not have been foreseen. Whatever his motivations, Constantine initiated a process whereby Jerusalem was transformed into a focal point of Christian life throughout the Empire. Until then, Christianity had by and large shunned earthly, physical, Jerusalem in favor of a heavenly, spiritual, one (John 4:1926). For Paul, places could not be holy: The God who made the world and everything in it, being the Lord of heaven and earth, does not dwell in shrines made by men (Acts 17:24). Christians believed, moreover, that if there was a real temple, it was within the believers themselves: The temple of God is holy; you are that temple (I Cor. 3:17; see also II Cor. 6:16). After taking a generally inhospitable posture (ocially, at least) prior to the fourth century, Christianity had now become receptive to the idea of a locus sanctus. See Barnes, Constantine and Eusebius, 22021; Hunt, Holy Land Pilgrimage, 3235; Holum, Hadrian and St. Helena, 7177; Wilken, Land Called Holy, xiv, 83. See also Nibley, Christian Envy, 97123. 98. Hunt, Holy Land Pilgrimage, 627, 190202; J. Z. Smith, Map Is Not Territory, 1012, 14751, 3089. 99. J. Z. Smith, To Take Place, 79. Markus (End of Ancient Christianity, 13955; How on Earth Could Places Become Holy? 25771) has suggested that the sacred and holy in Christianity owe much to the

byzant i ne pa le s t i ne


The second component of this revolution concerned pilgrimage. Scholars have disputed whether this phenomenon was a seminal and dramatic transformation, new and revolutionary in the fourth century, or the continuation of a process that had originated a century or two earlier, only now with heightened intensity and scope owing to Imperial and church support.100 However, while pilgrimage existed in the second and third centuries, all evidence points to the fact that it was of a dierent kind. Early pilgrimage involved limited distances that pilgrims traveled to reach their destinations and was often restricted to a religious elite whose goals seem to have been cognitive and exegeticali.e., they visited biblical sites for the historical memories with which they were associated.101 What distinguished the new wave of Christian pilgrims were not only the dramatically increased numbers, but also the spiritual and devotional experiences evoked. It is this ontological dimension that was most often primary, as pilgrims from the fourth century onward sought traces of God in these holy places. The temporal and spatial dimensions of Christian Palestine were fast becoming inextricably intertwined with the sanctity of Jesus words and deeds.102 A third element in the Byzantine Christian revolutionand one that relates to the above twois the emergence of a vigorous monastic movement in Palestine beginning in the fourth century. Pilgrims provided a constant source of recruits for the burgeoning monasteries of Jerusalem, Bethlehem, and the Judaean Desert, and one of the monks functions was to provide pilgrims with food, water, lodgings, as well as to introduce them to the holy places, which might include prayer and scriptural readings, in addition to explanations of the sites themselves. Indeed, the development of a vigorous monastic community in Palestine seems to have been closely connected with pilgrimage.103
veneration of martyrs and their tombs and relics, and that this can best account for the growing sanctity of places then appearing in the Byzantine Christian world. 100. Some scholars maintain the newness of this phenomenon; see, for example, Holum, Hadrian and St. Helena, 6681; Sivan, Pilgrimage, Monasticism, 5463; Taylor, Christians and the Holy Places, 30618, 32931; Wilken, Land Called Holy, 82100; Markus, How on Earth Could Places Become Holy? 26162. Walkers study (Holy City, Holy Places? 3641, 31115) of the dierences of opinion between Eusebius and Cyril of Jerusalem regarding the sanctity and centrality of Jerusalem and other holy places dramatizes the transformation that was taking place in Christian thought in the fourth century. See MacCormack, Loca Sancta, 1214. Other scholars maintain continuity with the pre-Constantinian era; see Windisch, Die ltesten christlichen Palstinapilger, 14558; Hunt, Holy Land Pilgrimage, 25; Wilkinson, Jewish Holy Places, 4153. Common to these scholars is the assumption that Jewish veneration of tombs and holy places had spawned similar traditions in early Christianity, which then, almost organically, attained a more developed expression in the Byzantine period. See also Lane Fox, Pagans and Christians, 476; MacCormack, Loca Sancta, 20; Chadwick, History and Thought, 38. 101. MacCormack, Loca Sancta, 2021; Limor, Holy Land Travels, 48. 102. Walker, Holy City, Holy Places? 3738; Holum, Hadrian and St. Helena, 69; Wilken, Land Called Holy, 8384; Taylor, Christians and the Holy Places, 310.; Limor, Holy Land Travels, 1112. 103. Sivan, Pilgrimage, Monasticism, 55: neither the scholarly interest displayed by many of the

244 h i s t o r i c a l d e v e l o p m e n t o f t h e s y n a g o g u e

Thus, from the fourth century on, the concept of holiness became associated with myriad Christian sites, particularly in Jerusalem and its environs. This association beneted greatly from Imperial initiatives, an emerging Jerusalem-centered ideology (per Cyril of Jerusalem), and increased pilgrimage. Moreover, this newly discovered sanctity was institutionalized in the developing Christian liturgy, which ritualized these associations. This process rst took root in Jerusalem and was often brought into local churches throughout the Empire by pilgrims returning from the Holy Land.104 In light of our discussion regarding the association of synagogue with Temple, it should be noted that sanctity was often ascribed to churches through their identication with the Jerusalem Temple. Eusebius refers to the Church of the Holy Sepulchre as a New Jerusalem and a temple (), and he similarly applied the latter designation to the church built in Tyre (a temple of God).105 It is reported that in the sixth century, upon completion of his magnicent Hagia Sophia edice in Constantinople, Justinian exclaimed: Solomon, I have conquered you! 106 Moreover, much of the symbolism associated with the Temple and Temple Mount in Jewish tradition was now being transferred to Golgotha and the Church of the Holy Sepulchre. This New Jerusalem built over against the one so famous of old 107 was where Adam was created and Isaac was bound as a sacrice, where Egeria saw King Solomons ring and the horn from which the kings of Judah were anointed. The dedication of the Church of the Holy Sepulchre coincided (intentionally?) with the biblical date for the completion of the First Temple by Solomon, and the church too was identied as the omphalos, or navel, of the world.108
early visitors nor the piety that imbued all pilgrims is sucient in itself to explain the vast expansion of Christian topography in early Byzantine Palestine. Nor are Imperial-sponsored projects, like those of Constantine, enough to account for the numerous localities shown to pilgrims toward the end of the 4th century. The renewal of biblical traditions and their association with specic contemporary localities appears to have been largely the unique result of the mutual interests and combined eorts of monks and pilgrims in the 4th century. Otherwise it would be dicult to explain the spread of monasticism in the area, the active involvement of monks and priests in hosting and guiding pilgrims, and the response of pilgrims to the wide-scale promotion of the Holy Land. See also Hunt, Holy Land Pilgrimage, 50106; Hirschfeld, Holy Sites, 11230. 104. Hunt, Holy Land Pilgrimage, 10727. See also Baumstark, Abendlndische Palstinapilger, 8083. See, however, the important reservations of Bradshaw (Inuence of Jerusalem), that the city may have been as much an importer of liturgical traditions as an exporter. For this and other aspects of sanctity in Byzantine Christian life (e.g., community, liturgy, time), see Markus, End of Ancient Christianity, passim. 105. Eusebius, Vita Constan. 3, 33, 36, 45; idem, Eccles. Hist. 10, 4, 13, 25, 69. Regarding the use of naos for other churches, see idem, Vita Constan. 3, 45. 106. Referred to in Wilken, Land Called Holy, 93. See also Procopius, Buildings, 1. A Syriac hymn notes that the cathedral of Edessa was compared to the Wilderness Tabernacle built by Bezalel (Mango, Art of the Byzantine Empire, 57). 107. Eusebius, Vita Constan. 3, 33. 108. Wilken, Land Called Holy, 9397; Khnel, Jewish Symbolism, 15051; see also Alexander, Jerusalem as the Omphalos, 10419.

byzant i ne pa le s t i ne


Given this monumental change in the status of Palestine for Christians, which included the Galilee as well as Judaea, we might well ask what eect this had on the local Jewish community. It is dicult to imagine that the Jews would or could have been impervious to such developments. Here, however, our sources fail us. Very little is recorded in rabbinic literature about Christianity in general, and what there is stems primarily from the second and third centuries. A number of sources have been cited by scholars which seem to indicate an awareness of fourth-century Christian claims and a rabbinic attempt to counter them.109 However, these sources are only vaguely suggestive (perhaps owing in part to medieval censorship), each bearing a greater or lesser degree of probability. On the other hand, a number of Byzantine Aramaic piyyutim contain fairly explicit barbs at a number of Jesus-related traditions.110 We have suggested elsewhere that the increased use of Jewish symbols in the Byzantine period resulted, in large part, from this Christian ascendance, which moved the Jews to reassert and reestablish their self-identity.111 Moreover, it may well be that the synagogue assumed an increased spiritual and religious role for the Jews as a result of Christianitys emphasis on the sanctity of the land in general and of specic sites in particular, not to speak of the many hundreds of churches now being built throughout the country. The fact that churches were also being referred to as holy or compared to a temple () generally, and to the Jerusalem Temple in particular, may have motivated Jews to make similar assertions regarding synagogue sanctity. In this case, the synagogue would have provided a setting for the Jewish community to express whatever disappointment and de109. See J. J. Schwartz, Encaenia, 26581; Yahalom, Angels Do Not Understand Aramaic, 4244; and the rich trove of material collected and analyzed by Visotzky in his Fathers of the World as well as his Anti-Christian Polemic, 83100. See also the material marshaled (at times, somewhat forced) by AviYonah, Jews of Palestine, 16674. One opportunity for at least some Jews to witness rsthand the power and grandeur of Christian Jerusalem was on the Ninth of Av, when they were allowed to visit the city and mourn the destruction of the Temple, a custom reported in Itinerarium Burdigalense (p. 22). 110. Lately, this inuence of Byzantine Christianity on rabbinic Judaism generally has been emphasized by J. Neusner in a series of publications ( Judaism and Its Social Metaphors, 21204; Judaism in the Matrix of Christianity, 67137; City of God in Judaism, 24178). Neusner, however, seems to have posited too sharp a distinction between pre- and post-Constantinian developments. It is hard to see, for example, how one can condently assume that a midrashic pericope (e.g., from Genesis or Leviticus Rabbah) reects a Byzantine setting. Perhaps such traditions originated earlier, as many of the midrashic texts indeed claim. Such a realization is already reected in the study of Christian liturgy, where former assumptions regarding post-Constantine origins were abandoned in favor of the realization that such practices arose, in fact, much earlier (Bradshaw, Ten Principles, 67). Too radical a redactionalist approach (i.e., following solely the assumed time of compilation to date a tradition) has serious drawbacks, as does the assumption that assigned statements are unquestionably accurate. 111. See L. Levine, History and Signicance of the Menorah, 14553. We are speaking of increased usage; all of the themes that became prominent in the Byzantine period (sanctity of the synagogue, representation of Temple-related motifs, use of religious symbolism, etc.) had already surfaced earlier in one form or another. S. Schwartz seems to overemphasize the extent of Christianitys impact on the Byzantine synagogue, not to speak of the community and its religious life; see his Imperialism, 179202.

246 h i s t o r i c a l d e v e l o p m e n t o f t h e s y n a g o g u e

spair it felt, on the one hand, and its longings and hopes, on the other. What they were powerless to realize in the political realm, Jews might have hoped to achieve within the connes of their synagogues, albeit in an associative and symbolic vein.112 The piyyut is clearly a primary source for charting such sentiments. In a sense, the emphasis of most synagogue buildings in this period (Galilean-type and Golan buildings excluded) on a modest exterior and relatively richly decorated interior may oer concrete expression of external political realities and internal communal needs. Galilean-type (mostly in the Upper Galilee) and Golan synagogues, with their impressive facades, were generally located in more isolated areas. Synagogue structures found in the area of the Sea of Galilee, Bet Shean, the coastal and Judaean regionsareas with a far greater Christian presencewere clearly fashioned like the Christian basilica. This type of building provided an architectural model for many contemporary synagogues, which, inter alia, t the social, political, and religious contexts now unfolding.113 It would be interesting to understandorigins asidewhat the Jews of Late Antiquity might have meant when they used the terms sacred and holy with regard to their synagogues. Were the buildings holy because God was present, an idea already found in several statements in rabbinic literature from the second and third centuries but appearing with far greater frequency in the fourth, in both Palestine and Babylonia? 114 Alternatively, did holiness derive from the holy congregation, from sacred objects or functions (i.e., the presence of Torah scrolls and the holding of prayers), or perhaps from the synagogues increasing association with the Temple? 115 A denitive answer to this question is elusive, and it may well be that there is, in fact, no single answer. Various communities might have used the same term but with markedly dierent or overlapping connotations. Moreover, it is conceivableand even quite likelythat even dierent members of the same community did not relate to these terms in the same way.
112. The polemical aspect of Jewish art from the period has been highlighted of late, primarily on the basis of the Sepphoris mosaic; see Khnel, Synagogue Floor Mosaic in Sepphoris, 3143; H. Kessler, Sepphoris Mosaic and Christian Art, 6472; Z. Weiss, Sepphoris Synagogue. 113. On the Byzantine synagogues indebtedness to contemporary Christian models, see Tsafrir, Byzantine Setting, 14757. 114. Second and third centuries: Y Berakhot 5, 1, 8d9a; B Berakhot 6a. See also Urbach, Sages, 5563. Fourth century: B Berakhot 6a; Y Berakhot 5, 1, 9a; 9, 1, 13a; B Megillah 29a; PRK 5, 8 (p. 90) and parallels; Deuteronomy Rabbah 7, 2; Midrash on Psalms 84, 4. PRK 28, 8 (pp. 43132) is instructive in this regard: R. Judan [said] in the name of R. Isaac: Whenever the Jews [lit., Israel] close themselves up in synagogues and academies, the Holy One, Blessed be He, also closes Himself up with them. . . . R. aggai [said] in the name of R. Isaac: Whenever the Jews [lit., Israel] gather in synagogues and academies, the Holy One, Blessed be He, gathers his Shekhina (= Divine Presence) there with them. See Ehrlich, Place of the Shekhina, 32029. 115. See Y Berakhot 5, 1, 9a; B Berakhot 6a, where Gods presence is directly linked to the congregations constituting an assembly of the Lord ( .) See also above, Chap. 6. On the dierent conception of sanctity in Judaism and Christianity, see J. Dan, On Sanctity, 1130.

byzant i ne pa le s t i ne


Nevertheless, given the developments of the Byzantine era, we would suggest that the sociopolitical context at the time was as signicant a factor in the Jewish usage of holy as any internal spiritual or religious aspirations. With the Byzantine synagogue being associated with and related to Jerusalem and its Temple,116 it may well be that the concepts of holy and sacred were now being utilized to enhance and deepen this connection in light of Christian usage. Moreover, the fact that some churches at this time were identied as replicas or substitutes for the Jerusalem Temple is, as noted, of signicance.117 It presented a challenge to the Jews and forced them to assert their claim that the synagogue was the legitimate continuation of Temple practice and presence.118 The midrash fragment quoted above focuses squarely on this issue, claiming that the synagogue and prayer had now replaced the Temple and its sacrices.119 Churches at this time were also being referred to in inscriptions as holy places, and this, too, is probably not unrelated to contemporary synagogue usage.120 Thus, the association of the synagogue with the Temple, Jerusalem, and holiness, a process that commenced well before Constantine (and especially documented in rabbinic sources), gained impetus in the Byzantine era, undoubtedly having been spurred by Christian practice and polemics.121
116. B. Narkiss, Image of Jerusalem in Art, 1120. 117. Cf. above, note 105; as well as Socrates, Eccles. Hist. 1, 17. 118. It is dicult to ascertain whether the presence of water installations in many synagogues (as in churches) stemmed from a desire to imitate the Temple. Such installations were also found in the Hellenistic-Roman world, in temples, private homes, and other buildings. However, it is impossible to know if their use was interpreted in terms of holiness or as temple-related. The above-cited text from Margoliots Palestinian Halakhah indeed points to the fact that this connection was made, at least in this tradition. 119. See above, note 90. An interesting synagogue practice unique to Palestine reects vividly the awe and reverence for the holy ark and perhaps Jerusalem. Among the series of customs that distinguished Palestinian from Babylonian Jewry and date to Late Antiquity is the following: Those [elders] in the east (i.e., Babylonia) face the congregation [when sitting in the synagogue], and their backs are toward the [Torah] ark; those living in Eretz-Israel face the [Torah] ark (Dierences in Customs, no. 36 [ed. Margalioth, p. 156; ed. Lewin, pp. 7576]). Palestinian elders faced the ark and thus also Jerusalem and the site of the Temple; the Babylonians did not. This practice also diers from that recorded in the T Megillah 3:21 (p. 360), where it is stated that the elders (here: Palestinians) sat with their backs toward the holy (ark? Jerusalem? both?) while facing the congregation. Babylonian Jewry thus continued this older Palestinian practice, while in Byzantine Palestine a new custom evolved that reected a more egalitarian stance (both the elders and the congregation sat facing the ark together) as well as greater reverence vis--vis the holiest object in the room. This renewed interest, if not fascination, with Jerusalem and the Temple on the part of Christians and Jews in the fourth century may throw additional light on Julians revolutionary plan to rebuild the Jewish Temple. 120. See, for example, the references to holy place in regard to the church on Mt. Nebo and twenty or so other sites in Byzantine Palestine in Piccirillo, Mt. Nebo, 36, 51. 121. It should be noted that many Christian communities began to view their own local churches as


historical development of the synagogue

Thus, the synagogue of Byzantine Palestine came to ll many more needs of the Jewish community than ever before, particularly in the religious and spiritual realms, and this development cannot be viewed as a solely internal matter, divorced from the Byzantine historical setting. Many features of the synagogue, both physical and liturgical, are to be linked to patterns, models, and stimuli from the surrounding world.122

In reviewing the Byzantine synagogue, we have noted several major trends. One is the continued development of the institution throughout Palestine, evident in the increasing number of synagogue buildings, with their elaborate iconography and rich epigraphic remains.123 Imperial legislation prohibiting the building and repair of synagogues was clearly being honored in the breach. However, lurking behind the facade of growth and expansion was a disconcerting reality. With the rise of Christianity, forces were being unleashed and restrictions introduced that threatened to limit the construction and functioning of synagogues, and even destroy them.124 The most violent example of this latter threat were the activities of the fth-century monk Barsauma, who, with his band of forty
holy places. Despite the above-noted insistence of church leaders following New Testament guidelines, i.e., that God is omnipresent and that buildings per se have no inherent sanctity, attitudes in many communities were often strikingly dierent. As Crowfoot (Early Churches, 7) has commented, many church buildings in the eastern Mediterranean largely resembled the church of Tyre, which Eusebius describes as a replica of the Jerusalem Temple. According to Markus: new ways of speech were making their appearance in media which reect more directly the instinctive habits of imagination: we have inscriptions that speak of the house of God, the hall of Christ, and the like; and visual imagery represented the saint in his shrine just as age-old representations showed the dead in his tomb. Before long locus sanctus narratives would come to adorn churches (How on Earth Could Places Become Holy? 264). 122. See below, Chap. 18, as well as Stemberger, Jewish-Christian Contacts, 13146. 123. The abundance of archaeological data for the Byzantine synagogue raises the intriguing methodological question whether this institution actually emerged and ourished in this period or was the latest stage of a phenomenon that existed in earlier periods as well (except that we have many fewer material remains for earlier periods). It is clear from our presentation that, given its role as the primary Jewish public institution, we have assumed the synagogues centrality throughout the Hellenistic-Byzantine eras. The fact that we have so much evidence for Late Antiquity may be best explained as a well-documented phenomenon in archaeology, namely, that the latest strata are those largely preserved, much as we argued in the previous chapter with respect to the absence of rst- and second-century synagogues. Here, too, I dier from S. Schwartzs presentation in his Imperialism, as regards both methodology and, of course, conclusions. 124. See Rubin, Christianity in Byzantine Palestine, 1078; Peters, Jerusalem, 15861. Perhaps the destruction layer in the Reov synagogue was a result of this wave of attacks. See NEAEHL, IV, 1272. See also the decrees of 393 and 397 granting protection to synagogues (Linder, Jews in Roman Imperial Legislation, nos. 21, 25). On the destruction of synagogues generally or their conversion into churches, see below, Chaps. 8 and 9.

byzant i ne pa le s t i ne


or so fanatic monks, reputedly destroyed synagogues and pagan temples throughout the region. Indeed, these two trendsImperial legislation and locally inspired destruction may not be unrelated. Perhaps it was the laxity, ineectiveness, and perceived corruption of the Imperial bureaucracy in enforcing its decrees limiting non-Christian practice that inuenced some elements within the church, from bishops to monks, to seize the initiative.125 These contradictory forces coexisted for generations. It was only the Moslem conquest, with its far-reaching political, social, and economic consequences, that eectively began to constrict local Jewish life, one of the indications of which was the precipitous decline in the number of synagogues throughout the country.
125. See also Fowden, Bishops and Temples, 5378; MacMullen, Christianizing the Roman Empire, 86101; Rubin, Christianity in Byzantine Palestine, 10711.



he evidence for the Diaspora synagogue in Late Antiquity invites comparison with the proverbial cup of water only partly full. On the one hand, we have material remains of thirteen buildings (excluding Delossee Chap. 4) as well as hundreds of inscriptions relating to the synagogue or its ofcials.1 In addition, literary sources note scores of synagogues throughout the Persian and Roman-Byzantine worlds, although in most cases nothing substantive is conveyed about the institution.2 On the other hand, given the existence of an extensive and far-ung Diaspora, this evidence appears woefully fragmentary. There can be little doubt that what we have is but a small sample of this Diaspora institution in Late Antiquity. Archaeological remains of synagogues derive from all parts of the Empire, from Dura Europos (Syria) in the east to Elche (Spain) in the west. Between these geographical extremes, synagogue remains have been found at Gerasa in Provincia Arabia (Jordan), Apamea in Syria, Sardis and Priene in Asia Minor, Aegina in Greece, Stobi in Macedonia, Saranda in Albania, Plovdiv (ancient Philippopolis) in Bulgaria, Ostia and Bova

1. For recent surveys of Diaspora synagogue remains from Late Antiquity, see Feldman, Diaspora Synagogues, 4866; Rutgers, Diaspora Synagogues, 6795; idem, Hidden Heritage of Diaspora Judaism, 97123; and esp. Hachlili, Ancient Jewish Art and ArchaeologyDiaspora, passim. 2. See the listing of some sixty-six places in Feldman, Diaspora Synagogues, 49. Rutgers suggests at least 150 synagogues but does not enumerate (Diaspora Synagogues, 67).

diaspora synagogues


37. Map of the excavated Diaspora synagogues of Late Antiquity.

Marina in Italy, and ammam Lif (Naro) in North Africa (g. 37).3 Inscriptions from these synagogues alone number over 200. Taken together with inscriptions found elsewhere (e.g., in Asia Minor), and especially those discovered in the catacombs of Rome and Venosa that mention the titles of synagogue ocials,4 the total number of synagogue3. Excavations at Saranda were carried out in September 2003 under the auspices of the Institute of Archaeology of the Albanian Academy of Sciences and the Institute of Archaeology of Hebrew University of Jerusalem. Two mosaic pavements were found, one with a menorah anked by an ethrog and shofar, the other depicting camels, trees, and the facade of a structure (either a Temple or a Torah shrine). The synagogue dates to the fth or sixth century c.e. and was found beneath a church. Excluded from this list are nine sites whose identication as synagogues remains uncertain. These include Athens; Miletus, Pergamum, and Mopsuestia in Asia Minor; Palmyra in Syria; Carthage and Leptis Magna in North Africa; Chersonesus in the Crimea; and Qana in Yemen. On these sites, see Stavroulakis and DeVinney, Jewish Sites, 4647; Goodenough, Jewish Symbols, II, 78; von Gerkan, Eine Synagoge in Milet, 17781; Kraabel, Diaspora Synagogue, 48889; Avi-Yonah, Mosaics of Mopsuestia, 18690; Foerster, Survey of Ancient Diaspora Synagogues, 165; Duval, Art palochrtien, 41315; Lund, Synagogue at Carthage? 24562 (see also Rives, Religion and Authority in Roman Carthage, 221 23); Foerster, Fifth-Century Synagogue in Leptis Magna, 5358; MacLennan, In Search of the Jewish Diaspora, 4451; Edwards, Jews and Christians at Ancient Chersonesus, 16670 (see also E. Eshel, Hebrew Grato, 28999). Information regarding Qana was communicated by A. V. Sadov. See also the comments in Hachlili, Ancient Jewish Art and ArchaeologyDiaspora, 4952, 20916. 4. The number here could uctuate between several score to well over one hundred, depending on whether one interprets certain oces as being connected to a synagogue, for example, those of presbyter and, more notably, archon, in Rome (this title alone is mentioned in almost fty inscriptions). See Frey, CIJ, I, lxviii.; Leon, Jews of Ancient Rome, 167.; Noy, JIWE, I, 32829. Given the fact that the


historical development of the synagogue

related inscriptions from the Diaspora is well over three hundred.5 The buildings themselves have been discovered over the course of the past century. The earliest report, from the end of the nineteenth century, relates to the ammam Lif synagogue. At the beginning of the twentieth century, synagogues were identied (sometimes arguably, at rst) at Priene, Aegina, and Elche. Between 1929 and 1934, four synagogues were discovered, at Gerasa (1929), Stobi (1931), Dura Europos (1932), and Apamea (1934). Three decades later the synagogues at Ostia (1961) and Sardis (1962) were excavated, in the 1980s, those at Bova Marina and Plovdiv came to light, and in 2003 the synagogue at Saranda was uncovered. In contradistinction to the situation in the pre-70 period, for which literary sources (primarily Philo, Josephus, and Acts) are of inestimable value in any discussion of the Diaspora synagogue, the situation in Late Antiquity leaves much to be desired. There are no literary sources that oer any serious discussion or description of these synagogues or on how they functioned. Jewish sources are restricted almost exclusively to rabbinic literature, which addresses the situation in Babylonia only and even then in a limited fashion. Regarding non-Jewish sources, the synagogue is indeed mentioned both in Roman Imperial sources and by church fathers, and many of these references are of immense value. However, the Roman sources are few in number, and the Christian material, with but few exceptions, is polemical in nature, frequently reporting only a synagogues destruction.

We begin with a discussion of the buildings that have been recovered to date, moving geographically from east to west.

Dura Europos
Discovered in 1932, in the course of the extensive excavations carried out by Yale University and the Department of Antiquities of Syria, the synagogue at Dura Europos is the most complete and important one yet recovered.6 The rst of the synagogues two
synagogue was the basic communal institution in Rome, it can be assumed that all ocials named were associated with it in one way or another (see below). 5. In addition to the aforementioned collections, mention should be made of other inscriptions listed by Lifshitz (Donateurs et fondateurs), several from Greece (Corinth, Athens), three from Hungary, at least a dozen more from western Europe, and those known to us from literary sources. An example in the last category would be the inscriptions from the synagogue in Edessa; this building was converted by the bishop Rabbula (died 435 or 436) into the church of St. Stephen; see J. B. Segal, Edessa, 182. 6. The bibliography on this synagogue is understandably extensive. The two basic works are Kraeling, Excavations at Dura: Synagogue; and Goodenough, Jewish Symbols, IXXI. See also White, Building Gods House, 9397; idem, Social Origins, 27293; Hachlili, Ancient Jewish Art and ArchaeologyDiaspora,

diaspora synagogues


38. Plan of the Dura Europos synagogue.

stages dates from the late second or early third century; the second stage was built in 244245 and destroyed in 256. The entire complex in both its stages is clearly identiable (g. 38). Toward the middle of the third century, the synagogue community expanded signicantly and additional facilities were required. A second building was therefore acquired and integrated into the original structure, which itself underwent extensive remodeling. Undoubtedly, the full range of synagogue functions was carried out in the new complex, although it is impossible to designate which room was assigned what function. In each stage, the focus of the building was the sanctuary (house of assembly), where an aedicula serving as a Torah shrine was located in its western wall. Benches lined the four walls of the room, and there may have been some sort of bima or table in the center. The pice de rsistance of this building is its astounding display of Jewish art (g. 39). In its later stage, the synagogue walls were covered from oor to ceiling with frescoes
96197, 42432; Roth-Gerson, Jews of Syria, 84104, 27377. For a listing of the major authors on this subject, see Gutmann, Early Synagogue and Jewish Catacomb Art, 133842.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

39. Western wall of the Dura Europos synagogue with a Torah niche in its center.

depicting scenes from the Bible. Just above the aedicula on the western wall are Templeassociated representations, i.e., a menorah, the Temple facade, and an Aqedah scene (from the time of Chronicles at least, Mt. Moriah of Genesis 22 has been identied with the Temple Mount). Above these depictions is a series of biblical scenes illustrating the bestowal of blessingsJacob to his sons and grandsonsas well as one of a seated royal gure (David? the Messiah?) holding court. David, dressed as Orpheus and enchanting animals with his music, is also depicted. Flanking this upper panel are four large gures, one of whichon the upper righthand sideis clearly identied as Moses. The identication of the others is unclear, and they have been given many dierent interpretations, including that of Goodenough, who views all of them as Moses.7 The remainder of the wall space in the synagogue illustrates an array of what must have been between fty and sixty scenes taken from the biblical narrative. The nding of Moses, the exodus from Egypt, and Ezekiels vision of the dry bones are the most extensive single representations preserved (gs. 4041), but there are also several cycles of se7. The upper right-hand portrait is Moses at the burning bush and is so identied by an inscription. Based on this identication, Goodenough ( Jewish Symbols, IX, 11023) opines that all the other gures are of Moses as well; the upper left-hand one is Moses at Sinai (another opinion: Joshua); the bottom right-hand one, Moses reading the Law, per Deuteronomy (other opinions: Josiah, Ezra, Samuel, Jeremiah); and the bottom left-hand one, Moses before his death; see Midrash on Proverbs 14 [p. 118] (other opinions: Jacob, Joshua, Isaiah). See also Kraeling, Excavations at Dura: Synagogue, 22739; Weitzmann and Kessler, Frescoes, 12732, 17073.

diaspora synagogues


[To view this image, refer to the print version of this title.]

40. Panel on the western wall of the Dura Europos synagogue depicting Moses and the Children of Israel crossing the Sea of Reeds.

quential events narrated in the Bible, as, for example, the loss of the ark at Even ha-Ezer and its sojourn in Philistine territory, a series of vignettes from the life of Elijah, and a number of depictions relating to the Wilderness Tabernacle and Solomons Temple. It has long been debated whether there is an overall pattern to these scenes.8 Are they arbitrary or is there a comprehensive theme that informs the varied depictions? Barring this, is there any logic in the arrangement of the three horizontal registers in the room i.e., does each represent a dierent idea, and, if so, howif at alldo they interrelate? Despite scores of attempts to answer these questions, no consensus has been reached; assuming an overall plan may be positing a far more ambitious agenda than the Duran Jews had ever dreamed of. Alternatively, as Kraeling and Bickerman noted decades ago, what is represented here might well be a kind of Heilsgeschichte that drew exclusively from the biblical narrative. According to this approach, each individual panel or set of panels has its own meaning and signicance and does not presume one overall theme.9 Whatever the case, the implications of the Dura synagogue representations vis--vis Jewish art are enormous. This is the earliest evidence of sophisticated Jewish art, and it appears enigmatically, almost ex nihilo, in such a developed, detailed, and complex fashion. Studies on the paintings themselves abound, and the latter have also sparked renewed
8. L. Levine, Synagogue at Dura-Europos, 17277; Gutmann, Programmatic Painting, 13754; idem, Early Synagogue and Jewish Catacomb Art, 132228; Hachlili, Ancient Jewish Art and ArchaeologyDiaspora, 18082. 9. Kraeling, Excavations at Dura: Synagogue, 35658; Bickerman, Symbolism in the Dura Synagogue, 12751. Carrying this idea even further, Wharton suggests that the Dura narratives are a pastiche and should be viewed as postmodernist (deconstructive, circumstantial, local and multicultural). See Wharton, Good and Bad Images, 125; as well as idem, Reguring, 3851. Regarding other aspects of the Dura frescoes, see below, Chap. 17.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

41. Panel on the northern wall of the Dura Europos synagogue depicting Ezekiel and the vision of the dry bones.

interest in ancient Jewish art generally. The Dura synagogue constitutes the most impressive example to date of this art as well as of midrashic (and not necessarily rabbinic) traditions of the Bible that most probably did not originate there. Lying on the fringes of the eastern Empire, the Dura community was too small and peripheral, and its history too short, to have created such a rich iconographic-midrashic tradition. Most likely, these motifs were found in other Diaspora synagogues. If there were any doubts before its discovery as to whether a developed Jewish art existed at this time, then Dura surely has put them to rest. To date, however, nothing even remotely comparable has been recovered elsewhere. Thus, while the euphoria over the rst revelations of Dura and the expectations of nding other such synagogues has waned somewhat in the more than seventy years that have passed since the sites discovery, these nds clearly indicate that a wider Jewish artistic tradition must have existed.10 The uniqueness of the Dura synagogue also lies in the fact that its immediate urban context has also been extensively excavated and explored.11 This important dimension makes eminently clear the degree to which the builders of this synagogue adapted local
10. See Weitzmann and Kessler, Frescoes, 14350. 11. Perkins, Art of Dura, 3369; Gates, Dura-Europos, 16681.

diaspora synagogues


architectural and artistic models in order to integrate this edice into the urban architectural and artistic landscape of Dura. The Torah shrine, for instance, was a close approximation of the aediculae in local pagan temples, with the distinction, of course, that in the Jewish context, the aedicula was intended to house a scroll (or scrolls) and not an idol. Linguistically as well, the Jews of Dura merged into the cultural milieu of their surroundings. Some of the nineteen Greek and twenty-two Aramaic inscriptions found are dedicatory; others identify gures and scenes depicted in the frescoes. The ten Iranian inscriptions are enigmatic grati, possibly documenting a series of visits by outsiders to the synagogue. No other site known to date is more illustrative of the adaptability of a Diaspora community to its social, religious, and cultural environment, while at the same time preserving its distinctive character, than that of Dura.12

Excavated in 1929, the synagogue in Gerasa, Jordan, is situated on a mound west of the centrally located Temple of Artemis, at the highest point of the city.13 Remains of the synagogues mosaic oor were found about fteen centimeters beneath the oor of the church that replaced it. East of the synagogue building are the remains of an atrium surrounded by a colonnade on its northern, eastern, and southern sides. It has been conjectured, on the basis of dierent-sized tesserae found in the center of the atrium, that a basin for ablutions may have once stood here, although no actual remains were discovered. From this atrium, one approached the synagogue building by ascending a series of steps to the west. The oor of the vestibule was completely covered with a mosaic pavement containing the familiar cluster of Jewish symbols (menorah, shofar, incense shovel, lulav, and ethrog), a Greek inscription surrounded by several animals (probably lions), and a scene from the Noah story. The last includes a depiction of animals exiting the ark; to its left are the heads of two young men, Noahs sons Shem and Yaphet (whose names appear in Greek), as well as a dove carrying an olive branch in its beak. The Greek inscription anking the menorah reads: To the most holy place, Amen, Selah. Peace upon the congregation (g. 42). Three doors led into the buildings interior, which was arranged in a basilical plan. Two rows of columns separated the nave from the side aisles, and at its western end, in the direction of Jerusalem, there appears to have been some sort of niche. The oor of the nave was paved with marble slabs, and the two side aisles were covered with mosaics. Although fragmentary, the mosaic remains indicate a high level of execution. The west12. This issue will be addressed at greater length in my forthcoming Visual Judaism. On the polemical nature of these synagogue frescoes, see Weitzmann and Kessler, Frescoes, 17883 (vs. Christianity); and Elsner, Cultural Resistance, 269304 (vs. paganism). 13. Crowfoot and Hamilton, Discovery of a Synagogue at Jerash, 21119; Kraeling, Gerasa, 23639, 31823; Goodenough, Jewish Symbols, I, 25960; Roth-Gerson, Greek Inscriptions, no. 10; Naveh, On Stone and Mosaic, no. 50; Crowfoot, Churches at Jerash, 1620.


historical development of the synagogue

[To view this image, refer to the print version of this title.]

42. Depiction from the Gerasa synagogue of animals marching to Noahs ark. Note also the menorah and the Greek inscription.

ern part of the pavement was removed and destroyed when the building was converted into a church; only geometric designs and a guilloche border have been preserved. A veline Hebrew inscription was found in the northern aisle: Peace on Israel, Amen, Amen, Selah. Pinas son of Baruch, Yose son of Samuel, and Judan son of izqiyah. The synagogue dates from the fourth or fth century and was replaced by a church in 530531 c.e., at the beginning of Justinians reign, as we learn from an inscription found there.

The building at Apamea, situated on the cardo maximus in the very heart of the city, approximately one hundred meters south of the main intersection, was discovered during excavations carried out in the mid 1930s (g. 43).14 The local Jewish community was evidently of sucient stature and prominence to have been able to acquire such a central location. The building was constructed in the late fourth century but appears to have existed for only a number of decades before it was destroyed and converted into a church in the early fth century. The main hall of the synagogue measured about 15.50 by 9 meters; it had a square niche in its southern wall for the Torah scrolls as well as a lavish mosaic oor with impressive carpet-like geometric patterns, as at Sardis (g. 44; see below), as well as a menorah and twenty dedicatory inscriptions that often included the names of the donors and their other family members. Unique to these inscriptions was a listing of the number of feet of mosaic oor contributed by each person: Ilasios150 feet, Sapricia150 feet, Euthalis Scholastikos
14. Brenk, Die Umwandlung der Synagoge von Apamea, 125; Hachlili, Ancient Jewish Art and ArchaeologyDiaspora, 3234, 198204, 402; Sukenik, Mosaic Inscriptions, 54151; Frey, CIJ, II, nos. 803 18; Lifshitz, Donateurs et fondateurs, nos. 3856; and most recently, Roth-Gerson, Jews of Syria, 5483, 26265. See also Mayence, La quatrime campagne de fouille Apame, 199204. For the nave inscription mentioning women, see Brooten, Women Leaders, 15859.

diaspora synagogues


43. Plan of the Apamea synagogue with the niche to the south. Remains of several Byzantine churches with apses facing east were discovered in the synagogues ruins.

44. Mosaics from the Apamea synagogue.

140 feet, Alexandra100 feet, Ambrosia50 feet, Domitilla (or Domina, Domnina) 100 feet, Eupithis100 feet, Diogenis100 feet, Basilidas100 feet, Thaumasis (along with his wife Hesychios and mother-in-law Eustathia)100 feet, Hierios (and his wife Urania)100 feet, Colonis75 feet, Theodoros (and his wife Hesychios)35 feet. Among those mentioned, either as donors or honorees, were the head (gerousiarch) of

260 h i s t o r i c a l d e v e l o p m e n t o f t h e s y n a g o g u e

the Antioch council, several archisynagogoi, a presbyter, a azzan, and a deacon. Of special interest are several inscriptions that note the date of donation (i.e., the year 703 of the Seleucid era, or 391 c.e.), as well as the participation of community leaders in these endeavors:
In the time of the most illustrious archisynagogoi Eusebios, Nemeos, and Phineos, and the gerousiarch Theodoros, and the most illustrious presbyters Isakios, Saulos, and the others, Ilasios, archisynagogos of the Antiochans, made the entrance of mosaic, 150 feet, in the year 703, the seventh day of the month of Audynaios [= January 7, 391]. Blessings upon all! Ilasios [son] of Isakios, archisynagogos of the Antiochans, for the welfare of Photion his wife, and of [their] children, and for the welfare of Eustathia his mother-in-law, and in memory of Isakios, Edesios, and Hesychios [his] ancestors, made the mosaic of the entrance. Peace and mercy upon all your holy congregation! 15

The above inscriptions reect very dierent types of dedications. The rst records the gift of Ilasios, archisynagogue of the Antiochan synagogue in Apamea, who contributed a sizable section of the oor in honor of his colleaguesthree archisynagogues, a gerousiarch, and at least two presbyters. In contrast, the second inscription speaks of this same Iliasos honoring his family. The former inscription is most unusual and is sui generis among the inscriptions from this synagogue. Thirteen other inscriptions from this synagogue note that donations were made by, or in honor of, a family. Nine of the inscriptions were donated by women, and in another three, wives are mentioned as benefactors along with their husbands. All the inscriptions are in Greek, as are most of the donors names; many are transliterated from Hebrew: Nemeos, Phineas, Isakios, Saulos, and Hesychios. Both Iliasos inscriptions were placed in the entranceway or portico of the synagogue. Another inscription found there states that an anonymous donation was made in the time of Nemeos, the azzan and diakonos of the synagogue, and refers to the synagogue building as a naos.16

Discovered in a major city of western Asia Minor in 1962, the Sardis synagogue has deservedly attracted a great deal of scholarly attention over the past generation, particularly through the articles of Kraabel and Seager, and more recently Bonz.17 The Sardis structure is by far the most monumental of all ancient synagogues. Its impressiveness
15. Following Sukeniks transcription and translation in Mosaic Inscriptions, 54445, with several emendations of Roth-Gerson, Jews of Syria, 54, 57. 16. See Roth-Gerson, Jews of Syria, 59. 17. For Kraabels and Seagers contributions to this subject, see above, Chap. 1, note 13; see also Bonz, Jewish Community of Ancient Sardis, 34359; idem, Diering Approaches to Religious Benefaction, 13954. See also Trebilco, Jewish Communities, 3754; White, Building Gods House, 98101; idem, Social Origins, 31023; Hachlili, Ancient Jewish Art and ArchaeologyDiaspora, 5863, 21831, 41012.

diaspora synagogues


45. Sardis synagogue as part of the municipal palaestra complex.

stems from its prominent location, large dimensions, and rich remains. Located on the main street, at an important intersection of the city, the synagogue was housed in what formerly had been a wing of the citys palaestra, or gymnasium (g. 45). Outside its southern wall, facing the main street, the building was fronted by a row of shops, some of which were owned by Jews; a side entrance connected these shops directly with the synagogues atrium.18 No other extant ancient synagogue can match that of Sardis in sheer physical size. The building was some 80 meters long; in its last stage it was divided into two parts, an almost 60-meter-long sanctuary and a 20-meter-long atrium. Compare this to the largest Palestinian synagogues known to dateCapernaum (24 meters), Meiron (27 meters), and Gaza (ca. 30 meters). The palaestra itself was completed some time in the second century c.e. and served as
18. Crawford, Byzantine Shops, 1718.


historical development of the synagogue

46. Four building stages of the Sardis building. The nal two stages were occupied by the synagogue.

the citys immense gymnasium and bath complex (g. 46).19 The complexs southeastern wing, like its northern counterpart, was divided into a series of rooms, which appear to have once functioned as apodyteria (dressing rooms) or exercise rooms, each opening onto the palaestra area. When this wing was converted into a civic basilica, its inner partitions were removed and entrances sealed, and a new entrance was made from the street to the east. The building had a forecourt in the east and an exedra with niches for statues of deities or emperors in the west. It seems unlikely that the building was used by the Jewish community at this juncture; rather, it probably served some public civic function for a period of time. In its next stage, dated to the last half of the third century, the building had clearly been taken over by the local Jewish community of Sardis.20 The wall separating the main hall

19. The history presented here of the synagogue building is based on Seager and Kraabel, Synagogue and the Jewish Community, 17173. 20. Adopting Seagers dating, Bonz (Jewish Community of Ancient Sardis, 34359; Diering Approaches to Religious Benefaction, 13954) oers an explanation of the local Jewish communitys rise to

diaspora synagogues


and atrium was removed to create an 80-meter-long hall. At its western end, three-tiered semicircular benches were built and the exedra was transformed into an apse for seating. The synagogue revealed in the excavations and partially reconstructed on site was built in the fourth century. All stratied archaeological evidence points to a mid century date: the atrium was completed around 360380 c.e.; the main hall, with its piers and elaborate mosaic oor, in the second quarter of the fourth century; and the wall inscriptions, along with the plaques and marble revetment on which they appear, from the mid fourth to early fth centuries. Those attending the synagogue would have entered the atrium from either one of the streets to the east or south. The atrium was large and attractive, with porticoes surrounding an open courtyard lavishly decorated with a mosaic pavement of multicolored geometric patterns. A chancel screen or balustrade stood between the columns that supported the roof. An impressive marble basin for washing and perhaps drinking was located in the center of the atrium. A reference to a fountain of the Jews in a municipal inscription may refer to this basin.21 Three portalsa large central door anked by two smaller onesled from the courtyard into the main sanctuary. Immediately inside the sanctuary, two aediculae on masonry platforms anked the main entrance, at least one of whichmost likely the southern onewas of a better quality and probably held the Torah ark and scrolls (g. 47). The function of the second aedicula remains unknown; additional scrolls or possibly a menorah might have been kept there. Alternatively, it may have served as a seat for an elder or some other ocial, although this seems unlikely given the fact that there were benches in the apse along the opposite (western) wall. As was the case in a number of other places in the synagogue, the stones used for these aediculae originally came from pagan buildings in the city. Pillars divided the central nave and two side aisles of the main hall. As there were no traces of a balcony or stone benches, the congregationwhich by some estimates might have numbered up to one thousand peopleprobably sat on mats or wooden benches, and some might have stood. The oor was lavishly decorated with geometric patterns and divided into seven bays, while the lower parts of the walls were decorated with marble wall panels or revetments (skoutlosis) and the upper parts with panels of brightly colored marble inlay. Toward the western end of the hall stood a massive stone table, labeled the eagle table because of the two large Roman eagles engraved in relief on each of its two supporting stones (g. 48). The table was anked by two pairs of freestanding lions sitting
prominence in the course of the third century, which allowed for the construction of the synagogue. For a suggestion that the synagogue building dates only to the fourth century, see Botermann, Die Synagoge von Sardis, 10321. 21. Seager and Kraabel, Synagogue and the Jewish Community, 169.

264 h i s t o r i c a l d e v e l o p m e n t o f t h e s y n a g o g u e

[To view this image, refer to the print version of this title.]

47. Reconstructed aedicula along the eastern wall of the Sardis synagogue, near the entrances.

back to back. Both the eagles and lions also appear in secondary use, the latter perhaps even dating back to the citys Lydian period, i.e., the sixth to fth centuries b.c.e. According to the excavators estimates, the semicircular benches at the western end of the sanctuary could have accommodated about seventy people (gs. 4950), a number similar to that of the leadership in the large Alexandrian synagogue, discussed above (Chap. 4).22 Directly in front of the apse is a nely executed mosaic oor featuring vine tendrils emerging from a vase or basin, similar, perhaps, to the one located in the atrium outside. The names of donors were incorporated into this mosaic. A stone parapet, perhaps a kind of chancel screen, separated the mosaic and the apse from the main hall. The excavations also yielded a rich harvest of epigraphical material; a total of 85 inscriptions (or fragments thereof ) were found, 79 in Greek and six in Hebrew.23 Of the thirty or so donors listed, only two names are Hebrew derivatives. The Sardis donors are identied either by profession, not uncommon in Jewish epigraphy, or by public oce, which was far less common. Among the synagogue members were provincial and city o22. T Sukkah 4, 6 (p. 273). 23. Kroll, Greek Inscriptions, 1127; Cross, Hebrew Inscriptions, 319.

diaspora synagogues


[To view this image, refer to the print version of this title.]

48. Eagle table from the main hall of the Sardis synagogue.

cials, procurators, city councillors, comes, and members of the decurionate. Altogether, this synagogue boasted well-placed and inuential members of the city, which would indeed account for the buildings central location, size, and impressiveness, as well as the ability of the Jewish community to maintain such an imposing facility for centuries.24 Of particular interest regarding the religious functioning of the synagogue is an inscription found in the very center of the mosaic oor mentioning one Samoe, hiereus [priest] and sophodidaskalos, the latter title referring to a wise teacher or a teacher of wisdom. Given the prominent position of this inscription, it is clear that Samoe at one time occupied a central religious role within the community. It has also been suggested that this Samoe may have preached or taught from the very spot where the inscription was found.25 Finally, synagogue inscriptions tell us something about the nature of the Judaism understood and practiced in this synagogue. One inscription refers to the Torah shrine as a nomophylakion (i.e., that which protects the Law); another bears the cryptic sentence Having found, having broken,26 read! observe! These inscriptions were carefully executed and may once have been prominently displayed in the synagogue hall near the Torah shrines. Eleven inscriptions mention the Greek term pronoia (divine providence), and the
24. For a somewhat dierent reconstruction of the communitys history, see Bonz, Jewish Community of Ancient Sardis, 10622. 25. Hanfmann and Bloom, Samoe, 10*14*. 26. The term may refer to the breaking open of a text (i.e., to decipher its meaning) or to breaking a seal in order to open a scroll; see Kraabel, Diaspora Jews and Judaism, 289.

266 h i s t o r i c a l d e v e l o p m e n t o f t h e s y n a g o g u e

49. Drawing of the western section of the Sardis synagogue, with the eagle table, two pairs of lions, and an apse with benches along the western wall.

appropriation of this cosmopolitan Greek philosophical-religious concept apparently reects a signicant degree of acculturation among the Jews.27 In addition, remains of some nineteen menorot were found, some incised in stone, brick, metal, or pottery. The most impressive is an ornate stone menorah that also bears the name of its donor, Socrates.28 It is little wonder that such synagogue nds have thrown into question many of the negative assumptions prevalent in former generations concerning late ancient Diaspora Jewish life. They have demonstrated, beyond all doubt, that at least some communities had achieved a high degree of recognition and status within their individual cities; in the case of Sardis, this privileged position continued for several centuries after the Christianization of the Empire, right up the Persian destruction of the city in 616.29

The synagogue at Priene, located on the Ionian coast of Asia Minor between Ephesus and Miletus, was identied as such only decades after it was excavated.30 At rst referred to as a house-church, it appears that the building originally functioned in the Hellenistic period as a private residence and was transformed into a synagogue some time in the second or third century.31 The main hall was an irregular rectangle, measuring 10.20 meters (east to west) by 12.5913.70 meters (north to south); two rows of stone slabs served as stylobates and were laid on an east-west axis; a niche 1.35 by 1.37 meters
27. Kraabel, Pronoia at Sardis, 7596; Rajak, Gifts of God at Sardis, 23236; and, more generally, Harrison, Benefaction Ideology, 10916. 28. Seager and Kraabel, Synagogue and the Jewish Community, 176. 29. Crawford, Multiculturalism at Sardis, 3847. See, however, Goodman (Jews and Judaism in the Mediterranean Diaspora, 20824), who suggests that, at least until the fth century, the synagogue was used by non-Jews who worshipped the God of Israel. 30. Wiegand and Schrader, Priene, 480; Goodenough, Jewish Symbols, II, 77; Kraabel, Diaspora Synagogue, 48991; White, Building Gods House, 6768; idem, Social Origins, 32532; Hachlili, Ancient Jewish Art and ArchaeologyDiaspora, 5658. 31. White, Social Origins, 32830.

diaspora synagogues


[To view this image, refer to the print version of this title.]

50. Reconstruction of the western section of the Sardis synagogue.

was cut into its eastern wall, thus indicating that this was a broadhouse-type synagogue (see Chap. 9) similar to the one at Dura Europos. The hall had benches along its northern wall and a small forecourt. A series of rooms, undoubtedly serving a wide variety of congregational purposes, surrounded the hall. The identication of this building as a synagogue became conclusive with the discovery of a number of smaller items bearing Jewish symbols.32 A relief, which appears to originally have been axed to the wall, was found on the oor in front of the niche; it depicted the usual symbols and featured a menorah anked by peacocks. Nearby, a pillar incised with a menorah was found on the oor. Presumably, the carving was never completed, as only three branches of the menorah are represented. Another relief, showing a menorah with a lulav and shofar on one side and an ethrog on the other, was found in a church next to the theater and may have originated in this synagogue. Lastly, a large basin, almost a meter in diameter, was found in the synagogue building.

32. Wiegand and Schrader, Priene, 480; Foerster, Survey of Ancient Diaspora Synagogues, 165; White, Social Origins, 332.


historical development of the synagogue

51. Mosaic oor of the Aegina synagogue.

The synagogue of Aegina, an island in the Aegean Sea near Piraeus, the port of Athens, was located close to the local harbor.33 Traces of ancillary rooms were found at the northern end of the building, which perhaps had a portico at its western end. The main hall measured 13.50 by 7.60 meters, and there was an apse on its eastern end measuring 5.50 meters in diameter. The oor of the nave had a mosaic pavement with geometric designs (g. 51). Dated to the fourth century, the synagogue appears to have been built over an earlier structure of unknown identity but identical in plan. The following two Greek inscriptions were found at the western end of the mosaic oor.
I, Theodoros, the archisynagogos who served for four years, built the synagogue from its foundations. Revenues [contributed] amounted to 85 gold pieces and oerings to God [i.e., from the synagogue treasury] [amounted] to 105 gold pieces. Theodoros the younger being in charge, the mosaic work has been done out of synagogue revenues. Blessings upon all who enter.34

33. Mazur, Studies on Jewry in Greece, 2533; Sukenik, Ancient Synagogues, 4445; Goodenough, Jewish Symbols, II, 7576; Foerster, Survey of Ancient Diaspora Synagogues, 16667; White, Social Origins, 35659; Hachlili, Ancient Jewish Art and ArchaeologyDiaspora, passim. 34. Frey, CIJ, I, nos. 72223; Lifshitz, Donateurs et fondateurs, nos. 12; Sukenik, Ancient Synagogues, 44; Wischnitzer, Architecture of the European Synagogue, 45; Goodenough, Jewish Symbols, II, 7576; White, Social Origins, 35659.

diaspora synagogues


[To view this image, refer to the print version of this title.]

52. Menorah from the mosaic oor in the Plovdiv (Philippopolis) synagogue.

Discovered in Bulgaria in 1981, this synagogue was located in the center of ancient Philippopolis (now Plovdiv), a city founded by Philip II of Macedonia.35 It was located on a cardo, not far from the forum and close to other major public buildings (a large bath complex and a basilica). The synagogue hall itself is part of a larger complex which, although poorly preserved, clearly included at some point a number of rooms and courtyards. The hall was oriented southward toward Jerusalem and there were entrances in the north, where an atrium was located. Although it appears to have been almost square, the hall had a basilical plan with two rows of columns. Measuring 13.5 meters (north to south) by 14.2 meters (east to west), its nave was 9 meters wide and the two aisles each measured 2.6 meters wide. The total area of the entire synagogue complex is estimated to have been 600 square meters. Remains of the buildings mosaic oor indicate a tripartite division, with each panel measuring 3.0 by 3.8 meters. Geometric designs predominate alongside an array of ivy leaves. A large, highly ornate seven-branched menorah with a circular base adorned the central panel (g. 52). The mosaic oor contained two inscriptions in Greek; the better-preserved one was found in a side panel: From his [resources], and according to design [or Providence],36 Cosmianus, otherwise known as Joseph, made this decoration. Blessing to all. Dated on palaeographic grounds to the third century, the inscription notes a wealthy benefactors gift to the synagogue. The synagogue had presumably suered, as did much of the city during the Gothic incursions of the mid third century, and the renovation recorded in
35. Danov and Kesjakova, Unique Finding, 21026; and, for somewhat more detail, Danov, Neues aus der Geschichte von Philippopolis, 10723; Koranda, Menorah-Darstellungen, 21839; Hachlili, Ancient Jewish Art and ArchaeologyDiaspora, passim. 36. See Danov and Kesjakova, Unique Finding, 212.


historical development of the synagogue

this inscription possibly came in their aftermath. It is unusual to see a Greek (or Latin) name together with a Hebrew (or Semitic) one, although the same phenomenon occurs in nearby Stobi as well. Nevertheless, each of these names is attested elsewhere for Jews.37 An early third-century inscription from Intercisa (Hungary) mentions one Cosmius, a customs ocial who likewise contributed to a synagogue.38 Finally, the formula Blessing to . . . appears, for example, in other Jewish inscriptions from Italy, Syracuse, Aegina, and Bet Shearim.39 The second inscription, which is fragmentary and very poorly preserved (made the arrangement and the decoration), was found at the foot of the mosaic menorah, on either side of its base, and is dated to the fourth century. By the fth century, the synagogue was no longer in use.

Located in Macedonia, about 160 kilometers north of Salonika, the ancient town of Stobi was excavated extensively between the years 1924 and 1934.40 Amid a cluster of buildings along one of the main streets between the Roman bridge and a church, a structure was identied as a synagogue on the basis of a monumental Greek inscription found therein.41 However, excavations soon made it clear that the building functioned as a church from some time in the fth or sixth century. The column on which the synagogue inscription appears was clearly in secondary usage. For decades since the excavation of this building in 1931, the only evidence for a synagogue was this very impressive and informative inscription, arguably the most important one found to date in any Diaspora synagogue setting (g. 53):
The year 311 [?]. Claudius Tiberius Polycharmos, also named Achyrios, father [pater] of the synagogue at Stobi, having lived my whole life according to Judaism, have, in fulllment of a vow, [given] the buildings to the holy place, and the triclinium, together with the tetrastoon, with my own means, without in the least touching the sacred [funds]. But the ownership and disposition of all the upper chambers shall be retained by me, Claudius Tiberius Polycharmos, and my heirs for life. Whoever seeks in any way to alter any of these dispositions of mine shall pay the Patriarch 250,000 denarii. For thus have I resolved. But the repair of the roof tiles of the upper chambers shall be carried out by me and my heirs.42
37. On the name Joseph in Jewish inscriptions, see Noy, JIWE, I, nos. 70, 79. 38. See Scheiber, Jewish Inscriptions, 29. 39. Noy, JIWE, 203. 40. Kitzinger, Survey of the Early Christian Town of Stobi, 81161; Marmorstein, Synagogue of Claudius Tiberius Polycharmus, 37384; Kraabel, Diaspora Synagogue, 49497; White, Social Origins, 34352; Hachlili, Ancient Jewish Art and ArchaeologyDiaspora, 6367, 23133, 410. See also Moe, Cross and Menorah, 14857. 41. See, for example, Wischnitzer, Architecture of the European Synagogue, 79. 42. Translation generally follows White, Social Origins, 35256. See also Kitzinger, Survey of the

diaspora synagogues


[To view this image, refer to the print version of this title.]

53. Stobi synagogue inscription engraved on a column.

This inscription is of enormous historical value. In the rst place, it mentions the name of the donor and his oce or title. The combination of Latin and Greek names would seem to indicate his position of prominence within the Jewish community and perhaps within the town as well. The title pater appears elsewhere in Jewish inscriptions, but its precise meaning remains elusive. Like the feminine mater, it clearly derives from the larger Greco-Roman world and may have been essentially honoric, although some have suggested that it may have referred to a member of the gerousia.43 Secondly, the inscription tells us something about the synagogue building itself. The main hall is referred to as a holy place, clearly indicating its distinct religious prole. In addition, Polycharmos built a series of other unspecied structures ( . . . ), perhaps referring to one or more buildings adjoining the synagogue. As we will see below, this may apply in particular to the building directly south of it, where there appears to have been a passage between the two (see below). In addition to the general designation,
Early Christian Town of Stobi, 14142; Hengel, Die Synagogeninschrift von Stobi; Feldman, Diaspora Synagogues, 62; Habas (Rubin), Dedication of Polycharmos from Stobi, 4178. 43. Noy, JIWE, I, 7778; and below, Chap. 11.


historical development of the synagogue

the inscription also mentions a triclinium, or dining room, which served, at the very least, as a place for communal meals and possibly as a general meeting place for other communal purposes as well. The term tetrastoon, a four-sided stoa, seems to refer to a courtyard or atrium, perhaps in some way connected to the triclinium. Finally, the building (or buildings) described in this inscription clearly had a second story, the upper chambers of which were reserved for use by Polycharmos family. Thus, it appears that Polycharmos was now ceding part of his home to the community at large. The inscription discusses details governing the arrangement between these two parties. One of the stipulations of this agreement requires an oender to pay an enormous sum of money to the Patriarcha reference, in all probability, to the third century Patriarch of Palestine (see below). If so understood, and this has been the most widely accepted interpretation to date (as against viewing the patriarch as a local ocial, noted in several laws in the Theodosian Code), the inscription constitutes an extremely important piece of evidence regarding the status and prestige of this oce toward the end of the third century, even in such a distant Diaspora community.44 The association of the Patriarch with the Stobi synagogue raises interesting questions regarding the relationship of this oce to the ancient synagogue generally; we shall return to this issue below (Chap. 12). In the early 1970s, excavations at the site were renewed after a forty-year hiatus, and it soon became clear that a synagogue building referred to in the inscription had once existed on this very site.45 In fact, it has been claimed that several stages of an earlier structure are discernible and include remains of an early second-century pavement, coins of Marcus Aurelius, a menorah-grato, and frescoes bearing the name Polycharmos in tabula ansata frames. A bronze votive plaque with the name Posidonia was found, and another such plaque bearing the name Eustathius, as well as a menorah, were discovered in a sewage canal. In any case, it would seem that Polycharmos houseand subsequently the synagogueincluded the aforementioned building south of the later church (referred to as the House of Psalms). What is not fully clear, however, is which stage should be associated with the abovementioned monumental inscription. Is it referring to a late second- or, as is usually assumed, a later third-century setting and structure? 46 This, in turn, is related to a date appearing in the inscription that, unfortunately, is partly obliterated. Over the years, scholarly consensus has focused on one of two alternatives, each dependent upon a different era as a basis for calculation: 163 c.e. or 280 c.e. Following Hengels detailed study in 1966, the latter date has been generally preferred.47 Before the third century, the Patri44. See generally, L. Levine, Status of the Patriarch, 132, esp. 13. Cf. also Goodman, Jews and Judaism in the Mediterranean Diaspora, 221. 45. Mano-Zissi, Stratigraphic Problems, 2089. 46. See White, Social Origins, 34356. 47. Hengel, Die Synagogeninschrift von Stobi, 11048. For a long period, the year 163 c.e. was widely preferred; see, for example, Frey, CIJ, I, 5047; Alon, Jews in Their Land, I, 25152; II, 672; Baron,

diaspora synagogues


54. Plan of the Stobi synagogue, with the bima facing east.

arch, to the best of our knowledge, had no such status, and the enormous sum of money recorded would have been utterly inconceivable to demand. The outlines of the interior plan of the third- to fourth-century synagogue are detectable although, as yet, far from clear (g. 54). The hall measured 7.9 by 13.3 meters; in it was the base of what was probably a bima along the eastern wall and the remains of a bench along the southern wall. The walls themselves appear to have been frescoed with geometric and oral motifs and perhaps given decorative stucco moldings. A mosaic oor pavement also has geometric motifs. In addition, per the synagogue inscription, other large communal areas included a triclinium and a tetrastoon. Some time in the later fourth or possibly fth century, the synagogue was converted into a church.

The synagogue of Ostia was excavated in 196162 under the direction of Squarciapino. Only a series of short preliminary reports in Italian and English appeared thereafter, and a nal report of the excavations has never been published.48 Nevertheless, over the last decade the Ostia synagogue has been the topic of much research and debate. White, in his work on ancient synagogues and churches, has studied this building extensively, and in a seminal article that appeared in 1997, he presented an alternative thesis to
Jewish Community, I, 84; III, 28 n. 24. More recently, Poehlman (Polycharmos Inscription, 23546) and Habas (Rubin) (Dedication of Polycharmos from Stobi, 4178) have again argued for a secondcentury date. See also White, Social Origins, 34648. 48. Squarciapino, Synagogue of Ostia, 1926; idem, Synagogue at Ostia, 194203; Fortis, Jews and Synagogues, 12128; Kraabel, Diaspora Synagogue, 497500; Noy, JIWE, I, no. 13; Fine and Della Pergola, Synagogue of Ostia, 4257; Binder, Into the Temple Courts, 32236. See also Wischnitzer, Architecture of the European Synagogue, 57.


historical development of the synagogue

that outlined by Squarciapino.49 At about the same time, the synagogue project at Lund University in Sweden, entitled The Ancient Synagogue: Birthplace of Two World Religions, under the direction of Olsson, adopted the Ostia synagogue as its joint research project. The brunt of this task was assigned to Runesson, who, having gained access to the original reports and documentation of the excavator, challenged Whites reconstruction, largely supporting Squarciapinos original conclusion.50 This then led to a series of intensive interchanges between White and Runesson, culminating in Runessons elaborate architectural presentation in a volume of studies on the Ostia synagogue and community.51 In addition, White has now undertaken a multi-year excavation of the Ostia synagogue and its environs; the results of this endeavor, called The Ostia SynagogueArea Masonry Analysis Project (OSMAP), will inevitably bring to light new material that will require revised explanations and reconstructions. The synagogue at Ostia is located outside the city wall, near the Porta Marina, i.e., close to the harbor that served Rome as well. Adjacent to the building on the north was the Via Severiana, an important coastal artery built around the turn of the third century. The synagogue remains visible today date to the fourth and fth centuries c.e., although there clearly were earlier stages, as far back as the rst and second centuries c.e. Since the major dierences in interpretation revolve around the rst stages of the buildings history, we will commence our description of the site with the later structure for which there is wide consensus (g. 55). The synagogue building measured 36.60 by 23.50 meters. The entrance from Via Severiana was anked by two small columns and led into a long vestibule (A), where a marble wellhead and well were found left of the entrance. On the right were a series of doors leading into the synagogues main rooms, with the rst three entrances leading into the sanctuary. This tripartite entranceway had a large central door anked by two smaller ones, reminiscent of the monumental portals in Sardis and Gerasa, as well as in many synagogues of Byzantine Galilee. This area contained several small rooms (B and C) partitioned by thin walls. A room to the right reportedly contained a large basin, perhaps serving as a ritual bath (miqveh), however it is not certain that it was in use at this stage. The central doorway gave entrance to the synagogues main hall (D, 15 by 12.50 meters)52 through two pairs of columns, a propylaeum of sorts, rising to a height of 4.70
49. White, Building Gods House, 6970; idem, Social Origins, 37994; idem, Synagogue and Society, 2358. 50. Runesson, Oldest Original Synagogue Building in the Diaspora: A Response to L. Michael White, 40933. 51. White, Reading the Ostia Synagogue: A Reply to Runesson, 43564; Runesson, Monumental Synagogue, 171220; idem, Synagogue at Ancient Ostia, 2999. For a comparison of these approaches, see Mitternacht, Current Views, 53344. 52. Following Runesson, Synagogue at Ancient Ostia, 52. White lists 14.31 by 12.50 meters (Synagogue and Society, 30).

diaspora synagogues


55. Plan of the Ostia synagogue.

meters (C). A raised podium or bima was located along the far wall, i.e., along the slightly curved wall at the western end of the main hall; just south of the propylaeum, on a podium, was a large apsidal aedicula in which the Torah scrolls were undoubtedly housed. The symbols carved on two corbels (at the end of the architraves extending from the apse) include a menorah, shofar, lulav, and ethrog. There is some dispute as to whether the aedicula is to be dated to the general fourth-century renovation or, perhaps, to a later stage (i.e., late fourthearly fth centuries). A small fragment of a stone lion was found on the oor of the main hall, although its original location is unknown. The oor of the hall was decorated in part with opus sectile marble slabs in secondary use (some of which bore inscriptions, presumably from other buildings) and in part with a black and white mosaic oor on which a Solomons knot was depicted, while two columns were found in the debris and may have stood in the middle of the hall. No benches dating to this period were found here. South of the main hall was a kitchen (G) in which an oven, a marble-topped table, and ve lamps decorated with menorot were found. Several oors were uncoveredan earlier mosaic one and a later earthen one; the dating of each remains uncertain. West of the kitchen were small (F) and large (E) rooms with mosaic oors. Benches lined the southern and western walls of Room E, and it has been suggested that it may have served as a


historical development of the synagogue

place of study, a social hall, a hostel, and almost certainly as a triclinium, but there is no rm evidence to support any of these identications.53 Regarding the earlier phases, there is sharp disagreement between Squarciapino, generally followed by Runesson on the one hand, and White on the other. Squarciapino had proposed from the outset that the building was erected as a synagogue some time in the mid to later rst century c.e. It originally boasted a large main hall (D) with benches on three sides, a podium (bima? ) along part of the slightly curved western wall, and a fourcolumned entranceway; south of this main hall was a triclinium, also having benches on three sides (Room G and the southern part of Room B), and to the east, near the entrance, was a well and cistern. The original stage of the synagogue building was followed by an intermediary phase dating to some time in the second, and possibly early third, centuries. Evidence of this phase include fragmentary remains of painted walls and mosaic paving, the partitioning of several areas in Rooms B and C, the addition of a basin in Room B (a miqveh?), and the removal of the benches from Room G.54 However, the most important nd here is an inscription that speaks of the existence of a holy ark. Found on a reused marble stone on the fourth-century vestibule oor (A), it consists of a rst line in Latin followed by four in Greek, and reads as follows:
For the safety of the Emperor [pro salute Aug(usti)]. Mindius Faustus with his family built and made [it] from his own gifts, and set up the ark [ ] for the holy law.55

The name Mindius Faustus was added to the last two lines of the inscription, and it seems quite evident that he restored the ark or possibly an entire aedicula housing the ark that had existed beforehand.56 If, as is generally assumed, the wording of the opening lines most likely reects a late second- or early third-century usage, then the addition of Faustus in the last two lines should be dated to the later third century. With the building of a monumental aedicula in the fourth century, the older ark and its accompanying dedicatory inscription became redundant and the latter was reused in the vestibule renovation (g. 56).57 White, in contrast, oers an entirely dierent reconstruction of the buildings history. In the rst place, he claims that the initial structure dating to the later rst or early sec53. Runesson suggests that a building to the immediate west of the synagogue (K) may have been associated with it, serving perhaps as a place of residence for one of its ocials (Synagogue at Ancient Ostia, 6263). 54. See Runesson,