Synagogues in the Hellenistic and Roman Periods: Archaeological Finds, New Methods, New Theories [1 ed.] 9783666522154, 9783525522158


120 100 28MB

English Pages [433] Year 2020

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Synagogues in the Hellenistic and Roman Periods: Archaeological Finds, New Methods, New Theories [1 ed.]
 9783666522154, 9783525522158

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Lutz Doering / Andrew R. Krause (eds.)

Synagogues in the Hellenistic and Roman Periods Archaeological Finds, New Methods, New Theories

Ioudaioi Schriften des Institutum Judaicum Delitzschianum Edited by Lutz Doering Advisory Board Martina Böhm, Bernadette J. Brooten, Matthias Henze, William Horbury, Sarah +,Pearce, Ishay Rosen-Zvi, Adiel Schremer, Daniel Stökl Ben Ezra, Jan Willem van Henten, Gregory E. Sterling

Volume 11

Lutz Doering /Andrew R. Krause (eds.)

Synagogues in the Hellenistic and Roman Periods Archaeological Finds, New Methods, New Theories ȱ ȱ in co-operation with Hermut Löhr

ȱ

Vandenhoeck & Ruprecht

Bibliographic information published by the Deutsche Nationalbibliothek: The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data available online: https://dnb.de. © 2020, Vandenhoeck & Ruprecht GmbH & Co. KG, Theaterstraße 13, D-37073 Göttingen All rights reserved. No part of this work may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without prior written permission from the publisher.

Vandenhoeck & Ruprecht Verlage | www.vandenhoeck-ruprecht-verlage.com ISBN 978-3-666-52215-4

Table of Contents

Preface............................................................................................................................. 7 Lutz Doering and Andrew R. Krause  Introduction: Synagogues in the Hellenistic and Roman Periods ......................... 9 I. Advances in the Archaeology of Synagogues from the Hellenistic and Roman Periods Zeev Weiss The Synagogue in an Age of Transition, from the Second Temple Period to Roman Times: Recent Developments in Research ............................................ 25 Uzi Leibner The Dating of the “Galilean”-Type Synagogues: Khirbet Wadi amam as a Case-Study ............................................................................................................ 43 Mechael Osband and Benjamin Arubas The Discovery of a Roman Period Synagogue in the Golan at Majduliyya ............................................................................................................... 71 Monika Trümper The Synagogue in Delos Revisited ............................................................................ 81 II. Interpreting Material Remains and Literary Sources Lutz Doering The Synagogue at Magdala: Between Localized Practice and Reference to the Temple........................................................................................... 127 Judith H. Newman Contextualizing the Magdala Synagogue Stone in its Place: An Exercise in Liturgical Imagination ................................................................... 155 Andrew R. Krause The Rhetoric of Synagogue Space: Theoretical Issues in the Study of Jewish Institutions in Literary Sources .................................................................. 175 Jordan J. Ryan The Contributions of Historical and Archaeological Study of Early Synagogues to Historical Jesus Research................................................. 189

6

Table of Contents

III. Theorizing Practice in Ancient Synagogues Jutta Leonhardt-Balzer What Were They Doing in Second Temple Synagogues? Philo and the προσευχή ............................................................................................ 215 Hermut Löhr In Search of the Petichah: Some Thoughts on the Torah, the Prophets, and the Scriptures in the Synagogues and Beyond ............................................... 239 Ruth Langer Rabbis, Nonrabbis, and Synagogues in Roman Palestine: Theory and Reality .................................................................................................... 253 Clemens Leonhard The Origins of Torah Reading as a Ritual and its Social Context ...................... 277 IV. Legal, Political, and Cultural Contexts of Ancient Synagogues Kimberley Czajkowski “Synagogues” in Ptolemaic and Early Roman Egypt ........................................... 295 Benedikt Eckhardt Synagogues as Associations in the Roman Empire .............................................. 313 Markus Öhler Synagogues in Inscriptions from Asia Minor: The Iulia Severa Inscription Reconsidered.............................................................................................................. 339 Katrin Kogman-Appel Dress Codes in the Synagogue of Dura Europos? ................................................ 369

Contributors to this Volume ................................................................................... 401

Index of Ancient Sources ......................................................................................... 403 Index of Modern Authors ........................................................................................ 420 Index of Subjects ....................................................................................................... 429

Preface

The origins of the present volume lie in the international conference, “Synagogues in the Hellenistic-Roman Period: New Finds—New Theories—New Methods”, which took place at Westfälische Wilhelms-Universität Münster, 13– 15 June 2017. This conference was part of the research programme in the project, “EXC 212 C2-24 Integration and Diversification in the Judaism of Palestine during the Hellenistic-Early Roman Period (300 BCE–135 CE)”, headed by one of the editors, Lutz Doering, within the DFG-funded Münster Cluster of Excellence Religion and Politics during the years 2014–2018. The other editor, Andrew Krause, was employed as a postdoctoral researcher by the Cluster of Excellence between 2016 and 2018. One area of enquiry of the project was the relationship between the Jewish institutions of the Jerusalem Temple and the local synagogues, as well as the role and function of early synagogues. Alongside chapters and articles authored by the editors individually and published elsewhere, this volume presents some of the results of this collaborative enquiry. The 2017 conference was organized in co-operation with Hermut Löhr, who led the Cluster of Excellence project, “EXC 212 A2-10 The Jewish Nomos between Normativity and Identity using the Example of Alexandria in the 1st–3rd Centuries A.D.”, until 2017. We thank the board of directors of the Cluster of Excellence Religion and Politics and the International Office of WWU Münster for the generous funding of the conference. While not all the papers given at the conference were eventually submitted for publication in this volume, we invited a further chapter by Jordan Ryan on the contribution of synagogue studies to Historical Jesus research, in order to round out the volume with an aspect that was not present at the conference. We are grateful to Hermut Löhr for his co-operation not only in organizing the conference but also in preparing the present volume following his move to Bonn. We would also like to thank both Elisabeth Hernitscheck at Vandenhoeck & Ruprecht for help with planning the volume and, more recently, her successor Izaak de Hulster for seeing it through the publication process, as well as Renate Rehkopf for advice relating to the production of the volume. We thank Laura von Bartenwerffer and John Dik (Münster) for help with proofreading and Franziska Prokopetz, Yannick Golchert (Münster), and Sebastian Rogowsky-Schmidt (Bonn) for assistance in preparing the indices. The Editors

Introduction: Synagogues in the Hellenistic and Roman Periods

Lutz Doering and Andrew R. Krause Research into ancient synagogues has made significant advances over the past thirty years.1 New sites have been discovered, excavated, and studied, which has led to a better view of the architecture and art of ancient synagogues, particularly those in the Land of Israel. One focus has been on synagogues from the Middle to Late Roman and Byzantine periods.2 Spectacular finds of mosaics have allowed new perspectives in the research of synagogue art,3 while the debate about the interpretation of individual scenes continues, revealing differences in method and emphasis.4 In addition to the burgeoning field of synagogues in the Middle to Late Roman and Byzantine periods, however, there has also been a considerable increase in the number of excavated sites from the late Second Temple for which the identification of a building as synagogue period has been suggested. It is the earlier synagogues from the Hellenistic and Roman periods5 on which the present volume focuses predominantly, both in the Land of Israel and the Jewish Diaspora, both from the Second Temple period and from the period between 70 and 300 CE. Several decades ago, an identification as a “Second Temple synagogue” was considered only for buildings at Gamla, Masada, and the Herodium (Levine  1

For a brief history of research until 2005 see Levine 2005: 9–17. E.g., “Synagogue II” at Khirbet Wadi amam (Leibner 2018; and Leibner in this volume), uqoq (Magness et al. 2018a), and orvat Kur (Zangenberg et. al 2013). 3 E.g., mosaic floors from Khirbet Wadi amam (Miller and Leibner 2018), uqoq (Magness et al. 2018b), and orvat Kur (Zangenberg 2017). 4 See particularly the debate on the Elephant mosaic panel from uqoq: Britt and Boustan 2017 and the contributions in JRA 31 (2018) by K. M. D. Dunbabin (506–508), J. Balty (509–512), R. Talgam (513– 523), B. D. Gordon and Z. Weiss (524–541), as well as A. Erlich (542–558). 5 In distinguishing between the Hellenistic and Roman periods, we follow widely accepted convention amongst archaeologists and historians, although the precise starting points of these periods may vary for the different areas of the Mediterranean as well as between scholarly disciplines. From the point of view of Levantine archaeology, it is useful to subdivide the Roman period into the Early, Middle, and Late Roman periods, as suggested by Leibner 2009: Early Roman, ca. 50 BCE–135 CE; Middle Roman, ca. 135– 250 CE; Late Roman, 250–350 CE; followed by the Early Byzantine period, ca. 350–450 CE. These distinctions do not exclude the possibility of viewing the Hellenistic and Early Roman periods historically as belonging closely together in the sense of a “long Hellenistic Age”, as suggested by Chaniotis 2018 (e.g., 3). 2

10

Lutz Doering and Andrew Krause

1996: 428–429), and by some for the so-called “mini-sinagoga” at Magdala (Corbo 1976). Of these, only the building in Gamla was widely seen as a purposebuilt synagogue, dated to either the middle of the first century BCE or, as the final excavation report prefers, not earlier than the turn of the era.6 In contrast, the buildings at Masada and Herodium had previously been used for other purposes and were converted into synagogues only by the rebels of the First Revolt or, at the Herodium, according to some scholars, only in the Bar Kokhba Revolt.7 Other scholars raised doubts even about the use of the Masada building as synagogue in the time of the First Revolt (e.g., Flesher 1995). Moreover, Ehud Netzer (1987) made a compelling case for the Magdala “mini-sinagoga” to have been a fountain house right from the beginning. Thus, the evidence for pre-70 CE synagogues in Palestine remained very slim at the time. Textual scholars, particularly from North America, nurtured further scepticism about the Second Temple synagogue. Thus, Howard C. Kee, in a series of articles, expressed doubts that, in the first century CE, “synagogue” denoted a building—rather than an “assembly”— of Jews; in order to maintain this, he had to date the Theodotus inscription from the city of David (CIJ II 1404 = CIIP I 9), which appeared to attest to the existence of such a building, to a much later period (Kee 1990; 1999). Kee found some defenders but also several detractors, who battled the issue out for the better part of the 1990s, with somewhat better arguments on the side of Kee’s critics.8 The archaeological evidence has expanded dramatically in recent years.9 Apart from Gamla, Masada, and probably Herodium (both in secondary use), buildings identified as Second Temple synagogues have been excavated at Qiryat Sefer (Khirbet Badd ‘Isa; Magen, Tzionit, and Sirkis 2004: late first century BCE to first century CE), Modi‘in (Khirbet Umm el-‘Umdan; On and Weksler-Bdolah 2005: a Hasmonaean-period synagogue, followed by one from the Herodian period), and in the northern part of Magdala, in an area acquired by the Legionaries of Christ and excavated by the Israel Antiquities Authority (Avshalom-Gorni and Najar 2013: two pre-70 CE stages). All of these buildings feature a central room surrounded by one or more rows of benches on three or four sides, and rows of columns along the sides that once supported a roof. There is no indication of a specific orientation of the buildings, for example, to Jerusalem, as is the case in later synagogues. Netzer (2003: 282) had suggested that two further features would be essential for identifying a building as Second Temple synagogue, namely, the existence of a niche, as in the Gamla synagogue, and the connection  6

Yavor 2010: 60–61. The earlier date was suggested by S. Gutmann: Levine 2014: 135–136. For the debate see Runesson, Binder, and Olsson 2008: 35. 8 For a brief review of scholarship on this issue see Catto 2007: 2–5 and passim, as well as Ryan 2017: 102–103, which is, however, not free from inaccuracies; see, e.g., the placing of the work of Claußen (2002) with those who deny the existence of synagogue buildings, against which see only Claußen 2002: 129 (Kee’s theory is “[v]öllig abwegig”), though Claußen still deemed evidence for pre-70 CE synagogue buildings slim and hence suggested private houses as additional—and perhaps more frequent—meeting places for Jewish communities. 9 The evidence is in part summarized in Hachlili 2013: 23–39; and now Ryan 2017: 61–67. 7

Introduction

11

with a ritual bath or miqweh, as evidenced by many of the Second Temple finds. However, Netzer’s claim was disputed by David Amit (2007: 27–29). In fact, there is no consistent evidence for a niche or any other form of Torah shrine inside the central room of these buildings, although adjacent rooms—as are evident, for example, in Gamla, Masada, Qiryat Sefer, and Magdala—may well have housed scrolls. In addition, not each of these buildings was directly connected with a miqweh, and in some cases, including Magdala, it may be questioned whether the miqwa’ot are specifically connected with the synagogue. It has therefore been argued that miqwa’ot were often placed close to synagogues as public places for general purposes of purification, not specifically for the visitors of a synagogue.10 While many synagogues feature miqwa’ot in their vicinity, the absence of a miqweh is not a hard and fast criterion against the identification of a building as a synagogue. In addition to the sites mentioned above, there are other buildings for which the identification as a synagogue has been proposed. One is a structure next to the Hasmonaean winter palace at Jericho, which Netzer (1999; 2004: 11–17) suggested was a synagogue. However, several scholars have raised doubts about this identification (Ma‘oz 1999; Claußen 2002: 185–186; Levine 2005: 72–74), while Anders Runesson (2014: 270–271) allows for its interpretation as an “association synagogue” (a hypothesized synagogue type which will be discussed below).11 Moreover, a first-century CE date is considered for the basalt floor—and the building to which it originally belonged—underneath the limestone synagogue at Capernaum (Binder 1999: 188–193; Levine 2005: 71). In contrast, claims about finds of Second Temple synagogues in Chorazin and Northern Jerusalem (Khirbet er-Ras) remain doubtful (Levine 2005: 72). Thus, archaeological evidence for Second Temple synagogues from Palestine—both Judaea and the Galilee/Golan—is stronger today than previous estimations had allowed. To this, we may add that the Theodotus inscription—for both archaeological and epigraphical reasons—is now firmly dated to the period from the first century BCE to the first century CE (Kloppenborg Verbin 2000; Cotton et al. 2010: 54) and thus attests to a Second Temple synagogue in a Greek speaking milieu in Jerusalem (cf. Acts 6:9). Lidia Matassa in her posthumously published PhD thesis (Matassa 2018), who has doubted the existence of firstcentury CE synagogues except perhaps for Gamla, seems to have paid too little attention to some of the finds mentioned above. However, she rightly pointed to some of the methodological problems surrounding the identification of early  10

See Haber 2008: 169–170. Haber thought that persons handling the scrolls inside the synagogue might wash their hands, as suggested by the hand basin in the Gamla synagogue. Adler 2008 suggested that men with seminal emissions immersed before handling scrolls. Adler now believes that miqwa’ot were so common that they appeared also close to synagogues (in a paper in the Hellenistic Judaism section at the 2019 SBL Annual Meeting in San Diego). 11 See also Runesson, Binder, and Olsson 2008: 40–42; and Hachlili 2013: 42, who lists the Jericho building among Second Temple synagogues.

12

Lutz Doering and Andrew Krause

synagogues. It is indeed not easy to distinguish between an early synagogue and other public, communal buildings in Jewish settlements of Palestine. Or, to put it differently: A public, communal building in an area with a Jewish population in the Land of Israel in the late Second Temple period, with benches and an open space in the middle, should be considered a synagogue unless some other function can be established.12 But what about assembly rooms of smaller groups? Allowing for this distinction, Runesson (2001a: 395–400, 478–482; 2014: 267–270) has suggested that synagogues usually functioned as “public synagogues” in places where Jews were in charge of the town or city administration (that is, in the Land of Israel), whereas synagogues run by specific Jewish groups in the homeland (e.g., the Essenes) and also those in the Diaspora were “semi-public” or “association synagogues”. Apart from the central role of communal reading and interpretation of the Torah (and, according to Luke-Acts, also the Prophets), the latter form would have included activities typical for the association, inter alia, communal dining. While there is evidence for dining facilities in some early Diaspora synagogues (e.g., Ostia, as per Runesson’s dating, see below), one might, for the homeland, point to the room L77 at Qumran (Levine 2005: 65; Runesson 2014: 273), which is the larger of the two dining rooms at Qumran (see Magness 2002: 122–126). However, it is unclear whether the reading and studying of scripture would have taken place in this room as well, and there are other rooms at Khirbet Qumran, such as L4, which seems to feature a low bench along three sides and may have been a small assembly or council room.13 Runesson’s distinction between “public” and “semi-public” or “association” synagogues, for which he assumes different origins—the former deriving from Torah assemblies in the city-gate, the latter emerging with the rise of voluntary associations, in the Diaspora allegedly by way of a transformation of Jewish temples—has some explanatory potential (see Ryan 2017: 31–33), though it remains debated in current scholarship. In the homeland, there are further sites for which the identification of a building as a Second Temple synagogue has been suggested, some of which are, however, contested. One such site is orvat Etri in the Judaean Shephelah, where a room featuring no benches and a row of three columns in the middle has been identified as a synagogue by Boaz Zissu and Amir Ganor (2004; 2009: 101–110); others, however, have compared this room with halls in domestic dwellings of extended families, such as the one at orvat Burna in the vicinity of Lod (Amit,  12 For a different type of public building in a Jewish first-century context, without benches and unable to accommodate a large group in a single space, see the “basilica” (L2100) at Gamla: Syon and Yavor 2005: 52–59. The excavators tentatively suggest that the northern part of the aisle “could have accommodated judiciary and public functions under the auspices of the town council, while in the side chambers transactions and meetings could be held” (59). 13 As suggested by de Vaux; see Magness 2002: 51. Rapuano 2001 compares this room with the Jericho synagogue argued for by Netzer. Hirschfeld 2004: 101, however, reconstructs L4 as serving the storage of clay jars.

Introduction

13

Torgü, and Gendelman 2008: 102–103). Another debated issue is the identification of synagogues in small rural settlements or farmsteads. One of these is a building in Khirbet Diab north of Jerusalem recently excavated by Benjamin HarEven (2016), another one a room in a Jewish farmstead on Tel Rekhesh not far from Mt. Tabor in Lower Galilee, excavated by Mordechai Aviam (Aviam et al. 2019). Whether there were “small” rural synagogues catering for a household or a farmstead remains to be discussed,14 and the current volume will offer a contribution to this debate. Moreover, despite the increasing evidence for Second Temple synagogues it is unclear how widespread in fact synagogue buildings were in the period. Another issue becoming more prominent in recent research is the question of continued use of synagogues—potentially following restoration—after 70 CE, as well as the founding of synagogues between the revolts. Here, the current research is moving away from the alleged “clear watershed” of 70 CE (Levine 2005: 175)15 to the question of both continuities and changes in the period after the Temple destruction. The archaeological evidence for such synagogues is still slim and partially debated, and hence more research is called for. Thus, it is unclear whether the tentative synagogue building at Khirbet e-uwani, excavated by Har-Even (2012), operated before or after 70 CE. The building is also unusual in that the benches, preserved on three sides, were hewn into the rock, and there is no evidence of columns, the absence of which the excavator explains with the modest size of the building. Moreover, the newly discovered synagogue at Majduliyya in the Golan may have operated before and after the First Revolt; it is discussed in the present volume. Additionally, a building at Khirbet Qana in the Lower Galilee has been interpreted as a “possible synagogue”, with the current estimation being that it was built after 70 CE when refugees from the south arrived in the Galilee (McCollough 2015: 141–142). While Levine (2005: 182) still attested “meager evidence, at best” for synagogues from the period between 70 and the third century, such synagogues are now becoming more firmly attested. This may allow new insights into the transformation of the synagogue from a Second Temple/Early Roman building to a post-Temple, Middle to Late Roman period institution. A case in point is the early phase of the synagogue at Nabratein, which—unlike the early Roman synagogues—featured a fixed Torah shrine and an orientation towards Jerusalem; according to Eric Meyers (2010), it dates from the second century CE. More generally, a crucial issue in this respect is the dating of the “Galilean”type synagogues. While Heinrich Kohl and Carl Watzinger (1916) classically dated these synagogues to the second or third century CE, a time range recently confirmed for the synagogues in the Meiron area by the team headed by Eric  14 15

CE.

See also Claußen’s suggestion of synagogue communities in private houses, above n. 8. But see Schwartz and Weiss 2012 for a more mixed general picture regarding the “watershed” of 70

14

Lutz Doering and Andrew Krause

Meyers (see Meyers 2001; Strange 2001), Jodi Magness, in a series of studies, has proposed revised dates for these synagogues from the fourth to the early sixth centuries.16 More recently, this issue has been addressed afresh with the excavation of Khirbet Wadi amam, a site apparently abandoned in the late fourth century CE (Leibner 2018). The results of this excavation seem to suggest that this synagogue type began to develop in the Middle Roman to Early Byzantine periods. In addition, the discovery of an Early Roman “public building” underneath the synagogue at Khirbet Wadi amam by Uzi Leibner and his team raises the possibility of continuity in location between a potential synagogue from the Early Roman period and one erected in the Middle Roman period. Observations of this kind might shed light on the situation of other places with a potential sequence of synagogue buildings, such as Capernaum or, perhaps, Kefar Shikhin, where a synagogue, “probably not earlier than the second century” reused “pieces of an earlier private villa or public building” (Strange 2015: 105). A further controversial topic of synagogue research over the past decades has been the earliest archaeological evidence for synagogues in the Jewish Diaspora. There is ongoing debate about the date from which on two of the oldest synagogues in the Jewish Diaspora operated as such: the synagogue at Delos and the synagogue at Ostia. For the latter, two contrasting views are defended, one, viewing the origins of the Ostia synagogue in the first century CE, as the earliest purpose-built synagogue extant from the Diaspora (Runesson 2001b), the other one, suggesting that a multi-use insula complex at Ostia was converted into a synagogue only from the second or third century CE onwards (White 1997); White’s research since 2002 suggests that the complex was not constructed before the third century, went through several phases of renovation lasting into (at least) the sixth century, was probably not converted into a synagogue prior to the midforth century, with the latest phase of renovation (which included the Torah shrine with menorah reliefs) taking place after 475. The forthcoming publication of this research will surely advance the discussion. Debate has also been raging on the synagogue at Delos. Thus, the late Lidia Matassa criticized scholars arguing for the identification of building GD 80 as a synagogue (Matassa 2007; 2018), most notably Monika Trümper (2004), who allowed for a function of the building as synagogue well before 88 BCE. Trümper offers a rejoinder in this volume. Finally, several issues regarding the function, status, and role of synagogues in the Second Temple period and shortly thereafter continue to be debated in scholarship. What do we know about the function of, and activities in, synagogues, both in the homeland and in the Diaspora when it is realized that the presentation of synagogues by authors such as Flavius Josephus (see Krause 2017) or Philo answers specific rhetorical purposes? What was going on in Second Temple synagogues, and might some of what was going on in them be labelled “worship” (thus van der Horst 2002, contra McKay 1994)? How central was prayer—and  16

Magness 1997; 2001a; 2001b; 2001c; 2005; 2007; 2009; 2010.

Introduction

15

what kind of prayer: communal or private—in early synagogues or proseuchai, literally, “prayer houses”? Some have suggested that prayer was rather not central in usual (“public”) synagogues in Palestine and more so in the Diaspora and perhaps Judaean sects (Levine 2005: 162–169), while others have deemed it central throughout (Binder 1999: 404–415). What do we know about practices of reading and interpreting scriptures in early synagogues, and how do these connect with later evidence? Were the rabbis taking control of synagogues right after 70 CE, or do we see changing attitudes towards synagogues within the developing rabbinic corpus? How can we responsibly correlate material evidence and literary sources? What is the relation between the Jerusalem Temple and synagogues, and were synagogues at the time really “extensions” of the Jerusalem Temple (as suggested by Binder 1999)? These and similar questions have been raised in previous scholarship (see, e.g., Levine 2005: 169–173; or Runesson, Binder, and Olsson 2008: 7–13, distinguishing between spatial, liturgical, social, and institutional aspects), but in view of the emergence of new finds—apart from new proposed synagogue sites, also the discovery of the decorated stone table in the first-century CE synagogue at Magdala17—the deployment of new methods, and the development of new theories the present volume will take up several of these aspects. The essays to follow, which predominantly originated in a conference at Westfälische Wilhelms-Universität Münster in June 2017, are attempts to push research on the earliest synagogues forward through the discussion of new material finds, the deployment of emergent methods and theories, and/or the timely reassessment of previous work in the field. Those new finds and definitional issues surveyed above dominate these essays. The recent, rapid growth of possible Hellenistic-Roman period synagogues from the Land of Israel excavated over the past decade has expanded our data set for understanding of how synagogues were conceived, constructed, and placed within specific geographic and community contexts during this period. Likewise, the introduction of new critical theory and methods to the sub-field of ancient Judaism has further nuanced the picture of what a synagogue was and how people acted within them over this same period, both in terms of reading specific sources and as we seek to create a composite picture based on the combination of these sources and the relevant material cultures. Moreover, the increased participation by and dialogue with classicists have allowed the field to contextualize the architectural, institutional, and ritual elements that we find in these early synagogue exemplars and the corresponding sources, as well as to incorporate inscriptions at the confluence of the material and literary streams. As with other fields of inquiry pertaining to religion, this combination of new data and new ways of processing the data have led to further problematization of normative reconstructions of Jewish assemblies extrapolated from the archaeological record of later periods and from rabbinic  17

See, e.g., Avshalom-Gorni and Najar 2013; Aviam 2013; Binder 2014; Bauckham 2015; Aviam and Bauckham 2018; Kershner 2015; Fine 2017; Hachlili 2017.

16

Lutz Doering and Andrew Krause

literature. This increased variegation has led to a toppling of many theories that had been dominant, though also allowing for an enrichment of the field. In the first section, “Advances in the Archaeology of Synagogues from the Hellenistic and Roman Periods,” Zeev Weiss commences with a methodological assessment of recent archaeological identifications of synagogues and studies of criteria in such identifications. Weiss voices caution as to the identification of the structures at Khirbet Diab and orvat Etri as synagogues and discusses the function of the Magdala stone as well as its contribution in reconstructing early synagogue liturgy. Following Weiss’ more programmatic treatment, Uzi Leibner reassesses the date of the “Galilean-type” synagogue through the lens of the synagogue from Khirbet Wadi amam. Leibner argues that both the stratigraphic dating method and the art-historical method of comparing the architectural decoration indicate that this synagogue, belonging to the “Galilean type” and being abandoned in the late fourth/early fifth century CE, originated in the third century CE. Hence, this evidence raises questions regarding the suggestion that the so-called “Galilean-type” synagogue emerged only in the Byzantine period. Continuing the focus upon specific synagogue finds, Mechael Osband and Benjamin Arubas analyse the discovery of a potential synagogue in the rural Golan settlement of Majduliyya. Osband and Arubas contend that this structure may be identified confidently as one of the few Roman period synagogues in the Golan Heights, and that it was used both before the First Jewish Revolt and beyond 67 CE, which is even rarer—in fact, down to the late third century CE. In the final essay of this section, we move from the analysis of new finds to the reassessment of older, well known finds, as Monika Trümper comprehensively defends her previous identification of the Delos synagogue. Taking the recent criticisms by Lidia Matassa as her starting point, Trümper argues that building GD 80 from Delos may still confidently be classed as a synagogue structure, potentially from its beginnings. She concludes by challenging other archaeologists and synagogue scholars to take up the areas of further research that she previously identified in her 2004 article. In the next section, “Interpreting Material Remains and Literary Sources”, we move to emergent theoretical and methodological studies, and new uses of established theories and methods in the study of the earliest synagogues and synagogue sources. Lutz Doering reviews the current evidence for first-century, late Second Temple synagogues and places the newly discovered synagogue at Magdala in this context. In doing so, he critically reviews previous proposals for the interpretation of the decoration and function of the Magdala stone table. Doering argues that, rather than providing a model of the Jerusalem Temple in the synagogue at Magdala, or bringing visitors of the synagogue into the Temple courts, the stone, through references to the Temple in its decoration, provides a connection between the local activity of Torah reading and the central institution of the Temple. Judith Newman, too, revisits the important Magdala synagogue stone and applies her mastery of ritual studies and religious experience in a study of the

Introduction

17

stone, as she moves beyond merely linking the various relief images to the Jerusalem Temple and Wilderness Tabernacle. Distinguishing between space and place (following Yi-Fu Tuan) and using the lens of cultural memory construction as well as ritualized reading of scripture, she argues that the synagogue-as-place was infused with memories of divine presence—in the Temple, in the tabernacle, or in the dispersion—within the textual community gathering in the synagogue on Sabbaths for instruction in scriptures. Subsequently, Andrew Krause assesses several forms of critical spatial theory in the reading of both literary texts (e.g., the works of Josephus) and inscriptions (e.g., the Theodotus inscription). Krause contends that rhetorical constructions of space in texts must be handled critically if they are to be incorporated into larger reconstructions of synagogue development, and critical spatial theory provides many of the tools necessary for such responsible readings. Similarly, Jordan Ryan advocates careful methodological consideration of history and the historicity of synagogue accounts in the New Testament Gospels and Historical Jesus research. According to Ryan, Collingwood’s evidentialist treatment of historical analysis that seeks corroboration in sources should take the place of more positivistic explorations of “what really happened”. Ryan’s essay also attempts to bridge the methodological and ideological divides involved in combining New Testament studies and the study of ancient Judaism, which are too often ignored. Moving from overarching theoretical concerns in the interpretation of material finds and literary sources to the more concrete—though equally fraught— issues of the constituent, day-to-day practices that were situated in the earliest synagogues, the third section, “Theorizing Practice in Ancient Synagogues”, addresses various sources and activities related to ancient Jewish institutions. Jutta Leonhardt-Balzer commences this section with a detailed study of Philo’s various clues regarding practices in first-century CE synagogues, often labelled proseuchai, including also the assemblies of the Essenes and Therapeutae. According to Leonhardt-Balzer, Philo presents the teaching of Torah as the primary activity of these gatherings, with special interest in the organization and discipline therein, while omitting several activities that arise in other sources. The next essay in this section transitions from the careful reading of a specific corpus of sources to the tracing of specific practices over time, as Clemens Leonhard tracks the development of scripture reading as a ritual practice in the synagogue. Leonhard applies Roman reading practices and literacy rates to this question, and he argues that in the early rabbinic period, Torah reading was more akin to elitist Roman reading and study groups than to the highly ritualized Torah reading that emerged only in Talmudic times. Hermut Löhr aptly follows with an analysis of the connection between synagogue readings, the bi- or tripartite division of the Jewish scriptures, and their exegesis. Drawing on sources from the Second Temple period and including early Christian texts, Löhr problematizes simple developmental models and direct attribution to synagogue practices for the socalled petichah. In the final essay of this section, Ruth Langer assesses the role of

18

Lutz Doering and Andrew Krause

the rabbinic movement and the introduction of the standardized synagogue liturgies of the rabbinic literature in the synagogue in the Tannaitic and early Amoraic periods. Langer challenges Lee Levine’s readings of several of the appropriate sources, as she contends that we must go even further than Levine in questioning the influence of rabbinic practices in the synagogues of the Roman period. Thus, all four of these essays push against simple, normative presentations of synagogue practices during this early period of synagogue development. The final section of this collection, “Legal, Political, and Cultural Contexts of Ancient Synagogues”, continues to address everyday matters in the synagogue, though moves to legal and ideological issues and practices pertinent to the study of the synagogue. Kimberley Czajkowski begins this section by surveying the Alexandrian Jewish community’s self-expression in both the city and the surrounding countryside. Czajkowski elucidates the applicability and meaning of the term “synagogue” in a context in which the term is given such short shrift. This is followed by two essays relating to the applicability of associational language and concepts in the study of synagogues, both of which focus on inscriptional data. In the first of these essays, Benedikt Eckhardt asks the fundamental question of what it means to portray synagogues as Roman collegia or thiasoi, in terms of both community practice and legal status. Eckhardt nuances the discussion by arguing that while synagogues before 70 CE certainly cohere with the general characteristics and were even treated as associations by some, we must still recognize the key difference in that Jewish associations were semi-public institutions which were not open to everyone and provided the context for legally practicing Judaism. Eckhardt ends with an invaluable appendix that analyses the evidence in a region-by-region survey. In the second such essay, Markus Öhler narrows the scope to query what makes a particular inscription Jewish. Öhler critiques the criteria used by earlier scholars and seeks to set out more workable principles for such identification, and he ends with the famous Iulia Severa inscription as a case study. In an art-historical coda to this section and the collection as a whole, Katrin Kogman-Appel studies the attire depicted in the iconic murals of the Dura Europos synagogue in order to explicate the “dress code” and its place in the visual storytelling of these paintings. Kogman-Appel argues that the deployment of Roman and Persian dress codes in the murals, rather than being used to demarcate self-identity and otherness, show the ability of the Durene Jews to be multilingual in a broader cultural sense. This collection thus seeks to survey the current state of the study of Hellenistic and Roman period synagogues, to move the conversation forward in this specialized sub-field, and to bring the various disciplines needed to clarify the meaning, role, and use of early synagogues into a meaningful conversation. Further, the volume suggests that there is still room for increased engagement with data relating to early synagogues in the areas of archaeology, biblical studies, and the historiography of ancient Judaism alike. It is thus hoped that this volume may

Introduction

19

both provide a point of entry for those new to the field and stimulate further conversation among the specialists in our burgeoning area of study.

Bibliography Adler, Y. 2008. “‫ הממצא הארכאולוגי וזיקתו להלכה קדומה‬:‫”בתי כנסת עתיקים ומקוואות טהרה‬. Cathedra 128: 51–72. Amit, D. 2007. “‫”בתי כנסת כפריים מימי בית שני‬, Michmanim 20: 21–32. Amit, D., Torgü, T., and Gendelman, P. 2008. “"orvat Burnat—a Jewish Village in the Lod Shephelah during the Hellenistic and Roman Periods”. Qadmoniot 136: 96–107 (in Hebrew). Aviam, M. 2013. “The Decorated Stone from the Synagogue at Migdal: A Holistic Interpretation and a Glimpse into the Life of Galilean Jews at the Time of Jesus”. NovT 55: 205– 220. Aviam, M. and Bauckham, R. 2018. “The Synagogue Stone”. In: Bauckham, R. (ed.), Magdala of Galilee: A Jewish City in the Hellenistic and Roman Period. Waco, TX: Baylor University Press, 135–159. Aviam, M. et al. 2019. “A 1st–2nd Century CE Assembly Room (Synagogue?) in a Jewish Estate at Tel Rekhesh, Lower Galilee”. Tel Aviv 46: 128–142. Avshalom-Gorni, D. and Najar, A. 2013. “Migdal”. Hadashot Arkheologiyot 125. http://www.hadashot-esi.org.il/Report_Detail_Eng.aspx?id=2304&mag_id=120 (accessed 27 February 2020). Bauckham, R. 2015. “Further Thoughts on the Migdal Synagogue Stone”. NovT 57: 113– 135. Binder, D. D. 1999. Into the Temple Courts: The Place of the Synagogues in the Second Temple Period. SBLDS 169. Atlanta, GA: Society of Biblical Literature. Binder, D. D. 2014. “The Mystery of the Magdala Stone”. In: Warner, D. A. and Binder, D. D. (eds.), A City Set on a Hill: Essays in Honor of J. F. Strange. Mountain Home, AR: BorderStonePress, 17–48. Britt, K. and Boustan, R. 2017. The Elephant Mosaic Panel in the Synagogue at Huqoq: Official Publication and Initial Interpretations. JRA Supplementary Series 106. Portsmouth, RI: Journal of Roman Archaeology. Catto, S. K. 2007. Reconstructing the First-Century Synagogue. LNTS 363. London: T&T Clark. Chaniotis, A. 2018. Age of Conquests: The Greek World from Alexander to Hadrian (336 BC – AD 138). London: Profile Books. Claußen, C. 2002. Versammlung, Gemeinde, Synagoge: Das hellenistisch-jüdische Umfeld der frühchristlichen Gemeinden. SUNT 27. Göttingen: Vandenhoeck & Ruprecht. Corbo, V. C. 1976. “La città romana di Magdala”. In: Studia Hierosolymitana in onore del P. B. Bagatti, Vol. I. SBFCMa 22/1. Jerusalem: Franciscan Printing Press, 355–378. Cotton, H. et al. (eds.) 2010. Corpus Inscriptionum Iudaeae/Palaestinae, Vol. 1: Jerusalem, Part 1: 1–704. Berlin: De Gruyter. Fine, S. 2017. “From Synagogue Furnishing to Media Event: The Magdala Ashlar”. Ars Judaica 13: 27–38. Flesher, P. V. M. 1995. “Palestinian Synagogues before 70 C.E.: A Review of the Evidence”. In: Urman, D. and idem (eds.), Ancient Synagogues: Historical Analysis and Archaeological Discovery, Vol. I. StPB 47/1. Leiden: Brill, 27–39. Haber, S. 2008. “Common Judaism, Common Synagogue? Purity, Holiness, and Sacred Space at the Turn of the Common Era.” In: eadem, “They Shall Purify Themselves”:

20

Lutz Doering and Andrew Krause

Essays on Purity in Early Judaism. Ed. by A. Reinhartz. EJL 24. Atlanta, GA: Society of Biblical Literature, 161–179. Hachlili, R. 2013. Ancient Synagogues—Archaeology and Art: New Discoveries and Current Research. HdO I 105. Leiden: Brill. Hachlili, R. 2017. “The Migdal Stone and its Ornamentation”. Revue Biblique. 124: 245–272. Har-Even, B. 2012. “ ‫ביזנטית והתקופה‬-‫ התקופה הרומית‬,‫ יישוב מימי הבית השני‬:‫חורבת א־טוואני‬ ‫”המוסלמית הקדומה‬. In: ‫ דברי הכנס השביעי‬:‫הְספר והמדבר בארץ ישראל‬, 15–29. Available from: http://www.atarsusya.co.il/pages/articles.asp (accessed 27 February 2020). Har-Even, B. 2016. “A Second Temple Synagogue at Khirbet Diab in Western Benjamin”. Qadmoniot 151, 49–53 (in Hebrew). Hirschfeld, Y. 2004. Qumran in Context: Reassessing the Archaeological Evidence. Peabody, MA: Hendrickson. Kee, H. C. 1990. “The Transformation of the Synagogue after 70 C.E.”. NTS 36: 1–24. Kee, H. C. 1999. “Defining the First-Century CE Synagogue: Problems and Progress”. In: idem and Cohick, L. (eds.), Evolution of the Synagogue: Problems and Progress. Harrisburg, PA: Trinity Press International, 7–26. Kershner, I. (2015). “A Carved Stone Block Found in Israel Upends Assumptions About Ancient Judaism”. The New York Times, 9 December 2015: page A4. Online version available from https://www.nytimes.com/2015/12/09/world/middleeast/magdala-stoneisrael-judaism.html (accessed 27 February 2020). Kloppenborg Verbin, J. S. 2000. “Dating Theodotos”. JJS 51: 243–280. Kohl, H. and Watzinger, C. 1916. Antike Synagogen in Galilaea. Leipzig: Heinrichs. Krause, A. R. 2017. Synagogues in the Works of Josephus: Rhetoric, Spatiality, and FirstCentury Jewish Institutions. AJEC 97. Leiden: Brill. Leibner, U. 2009. Settlement and History in Hellenistic, Roman, and Byzantine Galilee: An Archaeological Survey of the Eastern Galilee. TSAJ 127. Tübingen: Mohr Siebeck. Leibner, U. 2018. Khirbet Wadi amam: A Roman-period Village and Synagogue in the Lower Galilee. Qedem Reports 13. Jerusalem: Institute of Archaeology, Hebrew University. Levine, L. I. 1996. “The Nature and Origin of the Palestinian Synagogue Reconsidered”. JBL 115: 425–448. Levine, L. I. 2005. The Ancient Synagogue: The First Thousand Years. 2nd ed. New Haven, CT: Yale University Press. Levine, L. I. 2014. “The Synagogues of Galilee”. In: Fiensy, D. A. and Strange, J. R. (eds.), Galilee in the Late Second Temple and Mishnaic Periods, Vol 1: Life, Culture, and Society. Minneapolis, MN: Fortress, 129–150. Magen, Y., Tzionit, Y., and Sirkis, O. 2004. “Khirbet Badd ‘Isa—Qiryat Sefer”. In: Magen, Y. et al., The Land of Benjamin. Judea and Samaria Publications 3. Jerusalem: Israel Antiquities Authority, 179–241. Magness, J. 1997. “Synagogue Typology and Earthquake Chronology at Khirbet Shema‘, Israel”. JFA 24/1: 211–220. Magness, J. 2001a. “When Were the Galilean-Type Synagogues Built?”. Cathedra 101: 39– 70 (in Hebrew). Magness, J. 2001b. “The Question of the Synagogue: The Problem of Typology”. In: AveryPeck, A. J. and Neusner, J. (eds.), Judaism in Late Antiquity, Part III: Where We Stand: Issues and Debates in Ancient Judaism, Vol. 4: The Special Problem of the Synagogue. HdO I 55. Leiden: Brill, 1–48. Magness, J. 2001c. “A Response to Eric M. Meyers and James F. Strange“. In: Avery-Peck, A. J. and Neusner, J. (eds.), Judaism in Late Antiquity, Part III: Where We Stand: Issues

Introduction

21

and Debates in Ancient Judaism, Vol. 4: The Special Problem of the Synagogue. HdO I 55. Leiden: Brill, 79–91. Magness, J. 2002. The Archaeology of Qumran and the Dead Sea Scrolls. Grand Rapids, MI: Eerdmans. Magness, J. 2007. “Did Galilee Decline in the Fifth Century? The Synagogue at Chorazin Reconsidered”. In: Zangenberg, J., Attridge, H. W. and Martin, D. B. (eds.), Religion, Ethnicity, and Identity in Ancient Galilee: A Region in Transition. WUNT 210. Tübingen: Mohr Siebeck, 259–274. Magness, J. 2009. “Did Galilee Experience a Settlement Crisis in the Mid-Fourth Century?” In: Levine, L. I. and Schwartz, D. R. (eds.), Jewish Identities in Antiquity: Studies in Memory of M. Stern. TSAJ 130. Tübingen: Mohr Siebeck, 296–313. Magness, J. 2010. “The Ancient Synagogue at Nabratein”. BASOR 358: 61–68. Magness, J. et al. 2018a. “The Huqoq Excavation Project: 2014–2017 Interim Report”. BASOR 380: 61–131. Magness, J. et al. 2018b. “Bible Stories in the Huqoq Synagogue Mosaic Floor: Interim Report on the 2001–2017 Seasons”. Qadmoniot 156: 66–83. Ma‘oz, Z. U. 1999. “The Syn[a]gogue that Never Existed in the Hasmonean Palace at Jericho: Remarks Concerning an Article by E. Netzer, Y. Kalman and R. Loris [sc. Laureys]”. Qadmoniot 118: 120–121 (in Hebrew). Matassa, L. 2007. “Unravelling the Myth of the Synagogue on Delos”. BAIAS 25: 81–115. Matassa, L. D. 2018. Invention of the First-Century Synagogue. Ed. by J. M. Silverman and J. M. Watson. ANEM 22. Atlanta, GA: SBL Press. McCollough, C. T. 2015. “Khirbet Qana”, in: Fiensy, D. A. and Strange, J. R. (eds.), Galilee in the Late Second Temple and Mishnaic Periods, Vol. 2: The Archaeological Record from Cities, Towns, and Villages. Minneapolis, MN: Fortress Press, 127–145. McKay, H. A. 1994. Sabbath and Synagogue: The Question of Sabbath Worship in Ancient Judaism. RGRW 122. Leiden: Brill. Meyers, E. M. 2001. “The Dating of the Gush Halav Synagogue: A Response to Jodi Magness”. In: Avery-Peck, A. J. and Neusner, J. (eds.), Judaism in Late Antiquity, Part III: Where We Stand: Issues and Debates in Ancient Judaism, Vol. 4: The Special Problem of the Synagogue. HdO I 55. Leiden: Brill, 49–70. Meyers, E. M. 2010. “The Problem of the Scarcity of Synagogues from 70 to ca. 250 C.E.: The Case of Synagogue 1 at Nabratein (2nd–3rd Century C.E.)”. In: Weiss, Z. et al. (eds.), “Follow the Wise”: Studies in Jewish History and Culture in Honor of L. I. Levine. Winona Lake, ID: Eisenbrauns, 433–446. Miller, S. and Leibner, U. 2018. “The Synagogue Mosaic”. In: Leibner, U., Khirbet Wadi amam: A Roman-period Village and Synagogue in the Lower Galilee. Qedem Reports 13. Jerusalem: Institute of Archaeology, Hebrew University, 144–186. Netzer, E. 1987. “Did the Magdala Springhouse Serve as a Synagogue?”. In: Kasher, A., Oppenheimer, A. and Rappaport, U. (eds.), Synagogues in Antiquity. Jerusalem: Yad Izhak Ben-Zvi, 165–172 (in Hebrew). Netzer, E. 1999. “A Synagogue From the Hasmonean Period Recently Exposed in the Western Plain of Jericho”. IEJ 49: 219–221 (in collaboration with Y. Kalman and R. Laureys). Netzer, E. 2003. “The Synagogues from the Second Temple Period: According to Archaeological Finds and in Light of the Literary Sources”. In: Bottini, C. G., Di Segni, L. and Chrupcała, D. (eds.), One Land – Many Cultures: Archaeological Studies in Honour of S. Loffreda OFM. SBFCMa 41. Jerusalem: Franciscan Printing Press, 277–285. Netzer, E. 2004. “ ‫בתי הכנסת מימי הבית השני והשפעתם על התפתחות בתי הכנסת של תקופת‬ ‫”המשנה והתלמוד – סקירה אדריכלית‬. In: Eshel, Y. et al. (eds.), ‫ועשו לי מקדש – בתי הכנסת‬ ‫מימי קדם ועד ימינו‬. Ariel: College of Judea and Samaria, 9–24.

22

Lutz Doering and Andrew Krause

On, A. and Weksler-Bdolah, S. 2005. “Khirbet Um el-Umdan – a Jewish Village with a Second-Temple Synagogue in Modiin”. Qadmoniot 130: 107–116 (in Hebrew). Rapuano, Y., 2001. “The Hasmonean Period Synagogue’ at Jericho and the ‘Council Chamber’ Building at Qumran”. IEJ 51: 48–56. Runesson, A. 2001a. The Origins of the Synagogue: A Socio-Historical Study. ConBNT 37. Stockholm: Almqvist & Wiksell. Runesson, A. 2001b. “The Synagogue at Ancient Ostia: The Building and its History from the First to the Fifth Century”. In: Olsson, B., Mitternacht, D. and Brandt, O. (eds.), The Synagogue of Ancient Ostia and the Jews of Rome: Interdisciplinary Studies. Stockholm: Paul Åströms Förlag, 29–99. Runesson, A. 2014. “The Historical Jesus, the Gospels, and First-Century Jewish Society: The Importance of the Synagogue for Understanding the New Testament”. In: Warner, D. A. and Binder, D. D. (eds.), A City Set on a Hill: Essays in Honor of J. F. Strange. Mountain Home, AR: BorderStonePress, 265–297. Runesson, A., Binder, D. D., and Olsson, B. 2008. The Ancient Synagogue from its Origins to 200 C.E.: A Source Book. AJEC 72. Leiden: Brill. Ryan, J. J. 2017. The Role of the Synagogue in the Aims of Jesus. Minneapolis, MN: Fortress Press. Schwartz, D. R. and Weiss, Z. (eds.). 2012. Was 70 CE a Watershed in Jewish History? On Jews and Judaism before and after the Destruction of the Second Temple. AJEC 78. Leiden: Brill. Strange, J. F. 2001. “Synagogue Typology and Khirbet Shema‘: A Response to J. Magness”. In: Avery-Peck, A. J. and Neusner, J. (eds.), Judaism in Late Antiquity, Part III: Where We Stand: Issues and Debates in Ancient Judaism, Vol. 4: The Special Problem of the Synagogue. HdO I 55. Leiden: Brill, 71–78. Strange, J. R. 2015. “Kefar Shikhin”. In: Fiensy, D. A. and Strange, J. R. (eds.), Galilee in the Late Second Temple and Mishnaic Periods, Vol. 2: The Archaeological Record from Cities, Towns, and Villages. Minneapolis, MN: Fortress Press, 88–108. Syon, D. and Yavor, Z. 2005. “Gamla 1997–2000”. ‘Atiqot 50: 37–71. Van der Horst, P. W. 2002 [1999]. “Was the Synagogue a Place of Sabbath Worship Before 70 CE?”. In: idem, Japheth in the Tents of Shem: Studies on Jewish Hellenism in Antiquity. CBET 32. Leuven: Peeters, 55–82. White, L. M. 1997. “Synagogue and Society in Imperial Ostia: Archaeological and Epigraphical Evidence”. HTR 90: 23–58. Yavor, Z. 2010. “The Architecture and Stratigraphy of the Eastern and Western Quarters”. In: Syon, D. and idem (eds.), Gamla II: The Shmarya Gutmann Excavations 1976–1989. The Architecture. IAA Reports 44. Jerusalem: Israel Antiquities Authority, 13–112. Zangenberg, J. K. et al. 2013. “$orbat Kur 2011”. Hadashot Arkheologiyot 123. http://www.hadashot-esi.org.il/report_detail_eng.aspx?id=2230&mag_id=120 (accessed 13 March 2020). Zangenberg, J. K. 2017. “The Menorah on the Mosaic Floor from the Late Roman/Early Byzantine Synagogue at "orvat Kur”. IEJ 67: 110–126. Zissu, B. and Ganor, A. 2009. “Horvat Etri – A Jewish Village from the Second Temple Period and the Bar Kokhba Revolt in the Judean Foothills”. JJS 60, 90–136.

I. Advances in the Archaeology of Synagogues from the Hellenistic and Roman Periods



The Synagogue in an Age of Transition, from the Second Temple Period to Roman Times: Recent Developments in Research

Zeev Weiss

The synagogue was the central institution of the Jewish community in ancient Palestine and the Diaspora. The earliest excavated synagogue building dates to the end of the Second Temple period, but after the destruction of the Jerusalem Temple, and especially from the third century CE on, the synagogue underwent radical changes in both form and content. A key innovation in the synagogue structure that was maintained throughout history was the creation of a roofed and illuminated space designed to house a group of believers who actively participated in the ritual and, in some places, sat on benches built against the walls of the structure. Each community invested great efforts in building a spacious synagogue, embellishing it with architectural decorations, wall paintings, or mosaics. In the early stage its art was aniconic, but around the third century CE figurative art became prominent in Jewish circles, leaving its mark even in their sacred spaces.  The present study will focus on the early history of the synagogue—the communal building that operated in the late Second Temple period in the shadow of the Jerusalem Temple and developed in the following centuries into an independent institution in ancient Palestine and the Diaspora. The purpose of this article is to raise some methodological problems in current scholarship and to offer some preliminary thoughts about the changes that transpired in the building over the years. Therefore, it will not describe the archaeological finds systematically, but instead will use them to illustrate several issues pertaining to the synagogue in the late Second Temple period and Roman times. Synagogues were prominent in both urban and rural areas of first-century CE Palestine. To date, these early buildings were discovered in Judaea at Masada, Herodium, Modi‘in, and Qiryat Sefer, and in the Galilee at Gamla and, more recently, Magdala (Netzer 2003: 277–285; Levine 2005: 45–80). Jerusalem, too, boasted one such building that predated the destruction of the Temple, as evidenced by the Theodotus inscription uncovered in the city (Roth-Gerson 1987: 76–86; Kloppenborg 2000). The New Testament tells us that Jesus visited and brought his good news to synagogues in Capernaum, Nazareth, and virtually

26

Zeev Weiss

every settlement in the Galilee (e.g., Matt 4:23; 9:35; Mark 1:21–28, 39; Luke 4:16– 43), and Josephus testifies to the existence of a synagogue in Tiberias as well, claiming that it could accommodate a large crowd (Vita 277, 280). The interior space of the synagogues uncovered to date in the region, with their rectangular plans and dissimilar orientations, contained columns to support the roof and stepped benches for sitting along the walls, an arrangement that facilitated communal participation in the activities conducted inside the building (Fig. 1). Several synagogues also had annexed rooms, water installations, adjacent ritual baths, and other facilities of uncertain identification (Levine 2005: 74–80). The Galilean synagogues had yet another feature that was barely known at the other sites—the use of artistic decorations in the main halls. The synagogue at Gamla was graced with architectural features carved in relief (Yavor 2010: 42– 52), and at Magdala, besides the ornamented low stone table (to be discussed below), the building was decorated with frescoes and, at least on its eastern aisle and in one of its side rooms, was paved with a geometric mosaic (AvshalomGorni and Najar 2013).

Fig. 1. Plan of the Gamla synagogue (courtesy of the Gamla Excavations and the Israel Antiquities Authority).

Over the years, several suggestions have been raised as to which buildings in first-century CE Judaea or Galilee may have functioned as synagogues. The large dimensions of a building, the quality of its construction, the existence of a row of ashlar stones that may have served as one continuous bench, the location of the finds at a site, or its possible similarity to buildings known elsewhere in the region all helped scholars formulate their theories. Some argued that the basalt walls discovered beneath the late synagogue in Capernaum were to be identified as the remains of a synagogue, perhaps the one built by the Roman centurion mentioned in Luke 7:1–5 (Corbo 1982; Strange and Shanks 1983). The late Ehud Netzer identified one structure in the Hasmonean estate in Jericho as a synagogue

The Synagogue in an Age of Transition

27

(Netzer et al. 2004). Boaz Zissu and Amir Ganor maintain that a large hall at

orvat Etri was a synagogue that functioned between the two revolts (Zissu and Ganor 2009). Binyamin Har-Even identified a partially preserved building at Khirbet e-uwani as a synagogue (Har-Even 2012: 17–19). Douglas Edwards identified a large structure at Khirbet Qana as a synagogue he believed to have been constructed in the first century CE (Edwards 2002: 111–114). More recently, Binyamin Har-Even uncovered a small structure at Khirbet Diab (HarEven 2016), and Mordechai Aviam, at Tel Rekhesh in Naal Tabor (Aviam et al. 2019), each claiming the existence of a synagogue in the rural countryside. The scope of the present study does not allow for a comprehensive discussion of each and every building. Instead, it will focus on two structures—Khirbet Diab and orvat Etri—which will illustrate some of the problems and limitations encountered when attempting to identify Second Temple synagogues. After reviewing the caveats in identifying such buildings and rejecting the recent hypothesis that small synagogues, like the one in Khirbet Diab, attest to the existence of private synagogues in the late Second Temple period and between the two revolts against Rome, the discussion will focus on the rectangular stone base uncovered in the main hall at Magdala and will clarify its function, role, and contribution in reconstructing synagogue liturgy in the late Second Temple period and subsequent centuries.

Khirbet Diab A farmhouse dating to the late Second Temple period was uncovered in a fourdunam area at Khirbet Diab, north of Jerusalem, where an open courtyard containing several installations, including a ritual bath, was found (Har-Even 2016). Three structures were identified around the courtyard, however only the southern one, comprising five different-sized rooms, was fully excavated. The northeastern room (no. 125) was the entrance area through which one could approach the three western rooms (nos. 124, 130, and 131) or, alternatively, the southern room (no. 113), which is the focus of our discussion. Measuring 7.5 x 3.5 m, its walls were roughly built with small fieldstones while a single row of contiguous ashlars adjacent to the northern, western, and southern walls (height and width of each about 40 cm) suggests that they served as a continuous bench running along three sides of the room (Fig. 2). An 80 cm-wide mass of stones (no. 132) was found in the gap between the western wall (no. 105) and the row of ashlars to its east (no. 104); the excavator maintains that this mass was the foundation of another, wider, bench measuring twice the size of the others. A large, poorly preserved, base made of fieldstones was detected on the opposite side of the room, and although the findings are not sufficiently indicative, it has been suggested that it served either as a foundation

28

Zeev Weiss

for additional benches or even as a platform (bema) on the eastern side of the room (Fig. 3).

Fig. 2. View of the synagogue at Khirbet Diab, looking north (courtesy of Binyamin Har-Even).

Fig. 3. Khirbet Diab, suggested reconstruction of the synagogue (courtesy of Binyamin Har-Even).

The oblong layout of the southern room at Khirbet Diab, the ashlar benches arranged along three walls of the room, and the nearby ritual bath led the excavator to identify this structure as a synagogue. Maintaining that these features

The Synagogue in an Age of Transition

29

follow the pattern of the late Second Temple-period communal buildings known throughout the country, he further argues that one of the prominent elements of such synagogues—the columns—were not needed at Khirbet Diab since the room was not very wide and could have been roofed easily with wooden beams (Har-Even 2016: 53). The finds from Khirbet Diab provide important information about the hinterland of first-century CE Jerusalem. However, to what extent are such arguments, expressed also with regard to the building at Tel Rekhesh, convincing? Should they be added to the growing list of late Second Temple-period synagogues, and are we to assume that they even represent a new type of communal building, modest in size and simple in plan—as proposed by Har Even? It should be remembered that the oblong layout of the room at Khirbet Diab is not very different from that of the rooms in the western wing, whereas the miqweh in the courtyard apparently had no connection to it; only the benches, if they are to be identified as such, may support the proposed identification as a synagogue. The identification of the row of ashlars as benches at Khirbet Diab is indeed unequivocal and is important for assessing the nature of the room. Could the existence of these benches necessarily lead to the assumption that this room could only have been a synagogue and not another space with a different use? It should be noted that these so-called benches are not uniform in size or shape whereas the relationship between the ashlars, the height of the bedrock, and the reconstructed floor suggest that, if indeed benches, they might be a bit low and unsuitable for sitting. Even if they were benches, their location on three sides of the room does not necessarily indicate that they were the seating arrangement in a synagogue. Similar architectural elements were found along the walls of a room in the building excavated by the Mexican expedition in Area A at Magdala, and although two-stepped baths were found in the same house, no one has identified these rows of ashlars as benches in a synagogue and argued for the existence of another, smaller, synagogue in this town (De Luca and Lena 2015: 306–307). Furthermore, it seems rather odd that the hall identified as a synagogue at Khirbet Diab lies in close proximity to other rooms within the same compound that contained a millstone and a large cupmark (nos. 124 and 131) and that, theoretically, could have been used to shelter animals. Moreover, there are no parallels for such an arrangement in any other pre-70 CE synagogue. With no further support for the suggested identification, it could be argued that this room should be understood as part of the farmhouse compound that possibly functioned as living quarters for a family gathering or for hosting guests. Its construction was of a higher quality than that of the other rooms and was equipped with a ledge that may have been used for a variety of purposes. The thick western wall in the same room was probably intended to accommodate rectangular niches used as storage closets, similar to those found at Gamla (Yavor 2010: 63–67), Jerusalem (Geva 2007: 48–52), and other sites in the region (Hirschfeld 1995: 234–236).

30

Zeev Weiss

orvat Etri

orvat Etri, located in the Judaean Hills southwest of Jerusalem, presents further challenges in classifying and identifying spacious halls as synagogues (Zissu and Ganor 2004; 2009: 101–110). Measuring 13 x 7 m, the building constructed and used between the two revolts was entered from one of its long walls. An open courtyard with a stepped pool was located in front of the building, on its western side. It was a simple and modest structure with a plain plaster floor and no decorative elements or inscriptions. The remains of three square column bases found inside ran across the width of the hall, indicating that the columns themselves were meant to support the roof of the building (Fig. 4). It has been suggested that several architectural elements found elsewhere at the site also may have been used in this building (Zissu and Ganor 2009: 102).

Fig. 4. View of the synagogue at "orvat Etri, looking north (photo: Zeev Weiss).

The building at orvat Etri features none of the major elements typical of other contemporary synagogues. There are no benches inside the hall, and the three columns arranged in one row down its center actually blocked the open space found in all other synagogues where communal activities were held. Zissu and Ganor nevertheless claim that this was a synagogue! Aware of the meager evidence at hand, they rely on rabbinic literature to fill the vacuum and support their identification. Thus, for instance, they claim that the building was oriented

The Synagogue in an Age of Transition

31

to the northeast, toward Jerusalem, in compliance with the tannaitic halakhah of praying in this direction (Sifre Deut §29). Furthermore, based on a discussion in the Palestinian Talmud, the excavators argue that the furniture inside the building—the bema and benches—were made of wood (y. Meg. 3:1, 73d; Zissu and Ganor 2009: 108–109). Like other structures at the site, the building at orvat Etri was constructed of roughly cut stones. Architecturally, it not only lacks the unique elements characterizing synagogues at this time, but its layout—a long hall with a single row of columns in its center—resembles storerooms or other functional spaces in the region in Roman and Byzantine periods (e.g., Gamla [5.5 x 10.5 m]: Yavor 2010: 98–110; Sepphoris [9.4 x 6.2 m]: Netzer and Weiss 1995: 164–165; “Third Mile Estate” near Ashqelon [23 x 8 m]: Israel and Erickson-Gini 2013: 195–196). Nevertheless, the rationale for such an identification is methodological and relates to the use of rabbinic sources and the validity of testimonies in later compilations for the reconstruction of an earlier reality. How should one relate to a tannaitic or amoraic tradition in each and every case? Does it reflect an actual practice at the time of its creation? Is it indicative of a wider phenomenon regardless of the time or place of its creation? Or was it merely tailored to fit an academic discussion having no relevance to everyday life in the first and early second centuries CE? To what extent is the redactor’s involvement in placing the text in its appropriate context a reliable indicator of an earlier reality? The use of rabbinic literature is rife with methodological problems, and its relevance and applicability in reconstructing the realia of Jewish society in Roman and late antique Palestine have been questioned by various scholars in recent years.1 And yet, rabbinic sources can and should be employed when it is possible to contextualize their compatibility with the finds, but only after verifying the nature of the text, the time and context of the tradition, and its relevance to the actual find.2 This is definitely true with regard to orvat Etri, where none of the later sources provided by the excavators is relevant or is corroborated by evidence from the excavation site. The orientation of the building, with its long side facing northeast, emanated from the topography of the site; the location of the entrance was inferred from the adjacent courtyard. In principle, its rectangular layout, and especially the alignment of the columns, dictated the northwest–southeast orientation of the building and had nothing to do with facing Jerusalem. As mentioned, the arrangement of the interior resembles the layout of storerooms in the region and, I would argue, alludes to the use of the building for agricultural purposes and not as a synagogue.   1

Levine (2005: 179–182) and Schwartz (2001: 103–176) opine that the rabbis were marginal, had little charisma, few followers, and only minimal influence on Jewish society in the second and third centuries. 2 For a methodological discussion regarding the use of rabbinic sources in interpreting archaeological finds, see Weiss (2005: 226–228; 2017).

32

Zeev Weiss

Were There “Private Synagogues” in the Late Second Temple Period?  Synagogues constructed in the Second Temple period and between the Jewish revolts differ from each other but share architectural features. Their identification is not always evident since they lack religious components of the later synagogues, such as Jewish symbols, inscriptions, and interior furnishings (a bema, niche, and Torah ark) that by and large determined the nature of the building and the direction of prayer toward Jerusalem. A great deal of caution must be taken in analyzing archaeological finds; every detail must be examined and all options must be carefully studied before reaching misleading or poorly substantiated conclusions. Following the discovery of small rural sites on which one of the rooms was identified as a synagogue, similar to those discussed above, it has been suggested that while some synagogues existed in either cities or towns, others were located in isolated farmhouses.3 Such rooms served the landlord and his family, or whoever lived on the premises, for the purpose of conducting religious activities— reading and studying the scriptures, and listening to sermons—similar to those conducted in other synagogues on the Sabbath and holidays. According to this suggestion, the term “a synagogue of an individual” in rabbinic literature (y. Meg. 3:1, 73d; Ratner 1967: Megillah, 61) echoes the existence of such private institutions. Besides the methodological difficulty in using such sources for reconstructing an earlier reality, those discussing “a synagogue of an individual” do not refer at all to a private building owned by a single (extended) family, but rather to a communal synagogue whose construction was fully financed by one wealthy donor. “A synagogue of an individual” (‫ )בית כנסת של יחיד‬stands in contrast to “a synagogue of many” (‫)בית כנסת של רבים‬, where its members joined in a communal effort, each according to his ability, to help with the building expenses (y. Meg. 3:1, 73d; 3:4, 74a).4 The activities conducted in the late Second Temple-period synagogue varied and were intended for a wide audience, as indicated by several sources describing the institution at the time. Even if we were to assume that such rooms were designated for private gatherings of the landowner and his family, one ought to clarify what kind of religious activities may have been conducted there and who supervised them. It may be assumed, in principle, that the activities that took place at such isolated sites resembled those in other synagogues. If this theory is correct, it is probable that the landowner had several scrolls in his possession and that either he or one of his family members was literate enough to explain the Torah and, when  3

See, e.g., the opinion expressed by Aviam in Aviam et al. 2019: 141. See also b. Meg. 26a, which refers to a communal synagogue in Jerusalem that was sold to an individual. On the financing of synagogues by an individual or community, see Levine 2005: 386–388. 4

The Synagogue in an Age of Transition

33

needed, even deliver a sermon. Scrolls in the late second century CE were expensive and generally not found in every household, so it is doubtful that the landowner, who supposedly was not wealthy, could have owned a full set of scriptures for the Sabbath reading in his private synagogue. The archaeological finds at Khirbet Diab, orvat Etri, and perhaps also Tel Rekhesh suggest that such rooms could have been used for family gatherings either for joint meals or recreation, or merely for storage purposes, however there is no evidence to support the assumption that they were “private synagogues” designated for religious purposes. On the contrary, an analysis of the archaeological finds indicates, as demonstrated above, that such rooms could have had a variety of purposes, and not necessarily to house synagogue activity.

The Magdala Stone and Its Role in Reconstructing Synagogue Liturgy Situated at the northern end of the site, the first-century CE synagogue at Magdala was constructed on an east–west axis. It includes an entrance room, a rectangular hall to its east, and few small rooms adjacent to the building’s southwestern corner (Avshalom-Gorni and Najar 2013). The main hall was surrounded on all four sides by a continuous aisle and stone benches, and it also had six columns supporting the building’s roof. Its walls were decorated with frescoes and at least its eastern aisle was paved with a geometric mosaic containing a rosette inside a square. A rectangular stone base resembling a low table with four short legs (ca. 67 x 57 x 40 cm, the thick legs being 7 cm high) was found in the central area at some distance from its south-eastern corner. Its top and four sides are adorned with reliefs, including a seven-branched menorah resting on a square pedestal and flanked on both sides by an amphora, arcades, oil lamps, palm trees, rosettes, and other symbols (Fig. 5). Several attempts were made to explain these reliefs but, unfortunately, they have been over-interpreted and lack a methodological awareness of the anachronistic use of later artistic expressions for earlier phenomena (Aviam 2013; Binder 2014; Bauckham 2015; Schiffman 2017). This becomes evident not only in the examination of several rabbinic sources but also in the use of fourth- and fifthcentury artistic expressions when attempting to explain first-century depictions, without considering the format, style, and development of Jewish art from early Roman times through the Byzantine era.5 We do not intend to discuss the decorative program of the stone or even propose the meanings of certain features, but instead will proffer one general comment.  5

Aviam (2013: 212–217) argues that some of the depictions portrayed on the stone are related to the showbread table and its implements. He attempts to corroborate his identifications with artistic representations from Dura Europos, Sepphoris, and even Samaritan synagogues, failing to distinguish between time, space, and socio-religious groups.

34

Zeev Weiss

Fig. 5. The Magdala stone decorated with Jewish symbols on all sides. Note the four rounded elements in the corners of the stone’s upper surface and the raised edges on both narrow sides (used with the permission of the Israel Antiquities Authority; photo: Zeev Weiss).

Features such as the menorah or two amphorae clearly exhibit affinities to the Jerusalem Temple and its cult. The other images such as the rosette or palm trees, would simply reflect common artistic expressions in Judaea, and especially Jerusalem, in the Second Temple period (Peleg-Barkat 2016: 38–43). Identifying the depictions adorning the stone is important, but the prime interest in this stone—a prominent piece of furniture inside the prayer hall— concerns its functionality and role in reconstructing the liturgy in this synagogue in the Second Temple period and in those buildings constructed in the third and early fourth centuries CE. Several options have been raised thus far, but unfortunately none of them provides a satisfactory explanation that factors in the nature of the activity conducted in the synagogue at this time. Aviam argues that the four round elements in the corners of the stone’s upper surface “are the remains of four stone legs or four stone bases for wooden legs, on which a table (stone or wood) was placed and it is the table on which the Torah was read” (Aviam 2013: 216–217). Donald Binder offers four options but does not reach a decisive conclusion; it may have been a table for reading the scriptures (following Aviam), a

The Synagogue in an Age of Transition

35

base for a lampstand, a seat for the archisynagōgos, or a base for an offering vessel (Binder 2014: 41–43). More recently, Steven Fine has proclaimed that he is not “ready to interpret the ashlar as a reader’s table”, arguing, therefore, that the purpose of this stone is “still a mystery” (Fine 2017: 38). The rounded elements mentioned by Aviam are flat, relatively smooth except for the outer margins, and unequal in size, and there is no evidence that the conjectured “stones legs” were chiseled at a specific stage (Fig. 5). Rather, they seem to be sections of the relief left in the four corners of the stone that were molded into a round shape but not refined afterwards. It is equally difficult to assume that the rounded elements were used as “bases for wooden legs” that stabilized the table on which the Torah was read. The combination of wood and stone would be rather imperfect, unstable, and odd, whereas the suggested reconstruction seems to be influenced by the contemporary custom of placing the Torah scroll on a table with a sloped top, a lectern of sorts, for scriptural readings. What, then, was the purpose of the Magdala stone and how was it used during ceremonies conducted in the synagogue? Its modest shape and calculated dimensions attest to its use as a table on which Torah scrolls were placed. This is neither the “reading Torah table (sic)”, to use Aviam’s terminology, nor one of Binder’s earlier-mentioned suggestions (Aviam 2013: 216; Binder 2014: 41–43), but rather a low stand upon which the scroll was placed before the Torah reading, after its completion, and during the delivery of the sermon. Just as Ezra the scribe held the Torah in his hands, opened it, and lifted it for all to see (Neh 8:4–8)—a detail vividly and prominently depicted in one panel of the fresco in the third-century CE Dura Europos synagogue—in the Second Temple period a congregant stood and held the scroll in his hands and read the Scriptures to all those present in the synagogue (Kraeling 1956: 232–235).6 Similarly, upon the completion of the sacrifices in the Temple on the Day of Atonement (Yom Kippur), the Torah scroll was handed to the high priest who “stands and receives it, and reads standing” (m. Yoma 7:1). After completing the reading, the same mishnah states that “he rolls up the scroll, holds it close to his bosom, and says…”. A more detailed description of the ceremonies held in the synagogue appears in Luke 4:16–22: Jesus came on the Sabbath to the synagogue in his hometown of Nazareth. “And when he stood up to read, the scroll of the prophet Isaiah was handed to him. He opened the book and found the place where it was written…”. Upon completing the readings for that Sabbath, Jesus “rolled up the scroll, returned it to the attendant, and sat down” to deliver his preaching. T. Sukkah 4:6, too, indicates that whoever read the Torah in the synagogue at Alexandria approached the wooden bema, held the scroll, and read a section from it.7  6

Similarly, King Agrippa stood on the wooden platform in the Temple court, “received (the scriptures) standing and read standing, and the Sages praised him” (m. Soah 7:8). 7 Compare y. Sukkah 5:1, 55b, where it is indicated that “one stood and read the Torah”; see also Levine 2005: 91–96.

36

Zeev Weiss

Based on these sources, which are contemporaneous with, or at least presume a setting in, the Second Temple period, the exact use of the Magdala stone becomes clear. As elsewhere in the region at this time, members of the Magdala community sat in the synagogue facing the center, toward the decorated stone on which the scrolls, customarily kept in a side room, were brought into the communal hall at the appointed time of the Torah-reading ceremony. Examining the cross-section of the Magdala stone, it becomes evident that it is slightly concave, so that the raised edges on both narrow sides could support the scrolls and prevent them from falling off the table (Fig. 5). When it came time for the Torah reading, a congregant stood, walked toward the table, raised the scroll, and read the scriptures while still standing. At the end of the reading, according to Luke 4:20, Jesus rolled up the scroll and returned it to the attendant, but in Magdala, whoever read the scriptures presumably placed the scroll back on the table, possibly covering it with a cloth, and then went back to his seat to listen to the sermon.8 The finds from Qumran, dating from the Second Temple period, demonstrate that prior to the destruction of the Temple each scroll comprised a single book of the Bible, although a few of them contained two books of the Bible (Haran 1986). The height of each scroll was determined by the size of the letters, the number of rows per column, and its length. However, according to the data presented by Emanuel Tov in his study of the Dead Sea Scrolls, the height of the scrolls from 50 BCE onwards ranged from 25 to 46.5 cm (Tov 2004: 74–82). Assuming that the scrolls used in first-century CE Magdala were similar in size to those from Qumran, their height would fit the dimensions of the table perfectly (ca. 67 x 57 cm). Such a scroll could be easily lifted, held with both hands during the reading of the Scriptures, and returned to the table upon completion of the ceremony. The synagogue building at Magdala is not oriented toward Jerusalem, nor was the table inside the hall. Thus, the menorah facing southward is not indicative of anything, especially since it is not clear whether the table was discovered in its original position or whether it has been repositioned in antiquity.9 It is our understanding that whoever read the scriptures stood behind the long side of the table and, in keeping with its current location in the hall, faced westward so that most of those present could see him. The Magdala stone provides another important dimension that relates not only to the actual Torah reading in the synagogue in first-century Palestine, but also marks the early crystallization of synagogue liturgy in the Second Temple  8 Amit (2013: 173–175) also maintained that the table served the same purpose, but the scrolls rested on it for the entire Torah reading ceremony and the reader read while sitting on the floor, beside the table. 9 In contrast, see Aviam (2013: 208), who argues that the menorah was depicted on one of the short sides of the stone and that the location of the stone inside the building indicates that the direction of prayer was toward Jerusalem.

The Synagogue in an Age of Transition

37

period followed by major changes in subsequent centuries. The reading of Scriptures was one of the main activities conducted in the synagogue up to the Bar Kokhba revolt; afterwards it gave way to prayer, which eventually became the central component of synagogue liturgy (Levine 2005: 530–540). The orientation of prayer toward Jerusalem, derived from the tannaitic halakhah in Sifre Deut § 29 (ed. Finkelstein, 47), apparently influenced the design of many synagogues built in the third century CE and onwards in ancient Palestine and the Diaspora.10 To date, not a single synagogue constructed between 150 and 250 CE has been found in the region, however the buildings erected in the second half of the third century CE at ammat Tiberias (Weiss 2010: 329–338), ‘En Gedi (Porat, in press), and Khirbet amam (Leibner 2010: 230–235) are oriented toward Jerusalem but have no permanent ark.11 Judging by the description in several rabbinic sources, mainly tannaitic, it seems that the tevah (Hebrew for chest or ark) containing the Torah scrolls was brought into the prayer hall of these buildings for the Torah reading ceremony and was subsequently removed from the hall and returned to its place upon the completion. This practice seems to follow the custom in the Second Temple synagogues; however, the building was now oriented toward Jerusalem. When the tevah containing the scrolls was brought into the prayer hall, it was put down, so—as stated in t. Meg. 3:21—“it faces the people and its back is to the holy”, thus signifying its orientation toward Jerusalem (t. Meg. 3:21; Sapir 2007: 134–137). Only later, by the late third and mainly fourth centuries CE, with the introduction of a built bema (or platform) into the synagogue, did the Torah ark become a permanent fixture in the prayer hall, serving as a ritual focus and indicating conclusively that the direction of prayer was toward Jerusalem (Weiss 1990: 14–20; Levine 2005: 345–347). Synagogues constructed in the third century CE had an elevated bema added later on, in the fourth century CE, while those buildings constructed in the fourth century CE and onwards (e.g., Ma‘oz ayyim, Eshtemoa, and Sepphoris) had a permanent bema in the hall from the outset. But where were the scrolls placed in the Roman and late antique synagogues during the Torah reading, even after a permanent ark was set in the prayer hall? Some scholars maintain that several decorative elements in the center of some synagogue floors, or a change in the direction of a few stone pavers, allude to the possible existence of a wooden table or bema built in the middle of the prayer hall (Levine 2005: 343–344; Amit 2013: 179–181). It is quite possible that tables similar to the Magdala stone were in use concurrently in a few other late Second  10

Compare t. Ber. 3:15. For a discussion of the talmudic sources concerning the direction of prayer toward Jerusalem, its significance, and architectural expression within the building, see Weiss 1990; Ehrlich 1999: 76–89; Levine 2005: 326–330.  11 In contrast, see Amit (2013: 175) and Hachlili (2000), who argue that synagogues constructed after the destruction of the Jerusalem Temple already had a permanent focal point in their prayer halls—at first an aedicula (second century CE), then a niche (fourth century CE), and finally a Torah ark inside an apse (fifth–sixth centuries CE).

38

Zeev Weiss

Temple synagogues, or even in some Roman and late antique synagogues, however to date this piece of furniture has not been discovered at any site, except at one, orvat Kur, whose identification is questionable. A rectangular basalt table (ca. 70 x 68 x 35 cm), similar in size and shape to the one at Magdala, was found in the synagogue of orvat Kur; according to the excavator, it served as a “offering” or dining table (Zangenberg 2016). Aviam rejects the proposed identification and argues that, like the Magdala stone, this, too, served as the base of an elevated Torah reading table. Furthermore, he argues, in contrast to the reconstruction at Magdala, the rectangular depression at the top of the table here suggests that it was a monopodium (single-legged) table (Aviam 2016: 80*–82*). Here, too, the proposed reconstruction makes no sense for the same reasons mentioned above. Monopodia tables have an elegant leg for their support, but their stability does not require such a massive pedestal (Richter 1966: 112–113). Moreover, the rectangular depression is sloped and not deep enough to hold or stabilize the single leg, and therefore could not prevent the Torah reading table from falling. Needless to say, the basalt tabletop showed no indication that a leg was affixed to it, possibly with nails. The depictions adorning the orvat Kur stone—cultic vessels that may visually symbolize the lost Jerusalem Temple and its utensils—are entirely different from those appearing on the Magdala stone. However, its dimensions, shape, and concept resemble the low table from Magdala and imply that it was used, from the outset, for the same purpose. Furthermore, single books of the Bible were still in use in the talmudic period (late fourth–early fifth centuries CE; Haran 1986: 97–98), and so both the orvat Kur table and the Magdala stone would have served as a surface on which to rest the scrolls before the Torah reading and after its completion. Once the ceremony was over, the scrolls in the Magdala synagogue were taken to a side room, whereas in a late antique synagogue such as the one at orvat Kur the scrolls would be returned to the Torah ark, which was a permanent fixture inside the prayer hall.

Conclusions Excavations conducted throughout Israel in recent years have uncovered additional synagogues from the late Second Temple period and the years between the two revolts against Rome. Some archaeologists have identified other structures dated to the late Second Temple period as synagogues based on their dimensions, quality of construction, presence of columns, or existence of a bench along some of the walls. However, as explained above, the suggested identification is not always convincing. It is hoped that future finds, especially those dating to the interim period, will shed further light on the synagogue and its institutionalization in the late Second Temple period in the shadow of the Jerusalem Temple,

The Synagogue in an Age of Transition

39

after its destruction, and until the rise of Christianity in the early fourth century CE.

Figures Fig. 1. Plan of the Gamla synagogue (courtesy of the Gamla Excavations and the Israel Antiquities Authority). Fig. 2. View of the synagogue at Khirbet Diab, looking north (courtesy of Binyamin HarEven). Fig. 3. Khirbet Diab, suggested reconstruction of the synagogue (courtesy of Binyamin Har-Even). Fig. 4. View of the synagogue at "orvat Etri, looking north (photo: Zeev Weiss). Fig. 5. The Magdala stone decorated with Jewish symbols on all sides. Note the four rounded elements in the corners of the stone’s upper surface and the raised edges on both narrow sides (used with the permission of the Israel Antiquities Authority; photo: Zeev Weiss).

Bibliography Aviam, M. 2013. “The Decorated Stone from the Synagogue at Migdal: A Holistic Interpretation and a Glimpse in the Life of the Galilean Jews in the Time of Jesus”. NovT 55: 205– 220. Aviam, M. 2016. “Another Reading Table Base from a Galilean Synagogue: Some Comments on the Stone Table from "orvat Kur”. In: Patrich, J., Peleg-Barkat, O., and BenYosef, E. (eds.), Arise, Walk through the Land: Studies in the Archaeology and History of the Land of Israel in Memory of Y. Hirschfeld on the Tenth Anniversary of his Demise. Jerusalem: Israel Exploration Society, 79*–82*. Aviam, M. et al. 2019. “A 1st–2nd Century CE Assembly Room (Synagogue?) in a Jewish Estate at Tel Rekhesh, Lower Galilee”. Tel Aviv 46: 128–142. Amit, D. 2013. “ ‫השוני התפקודי בין בתי הכנסת של ימי הבית השני לאלה שלאחר החורבן‬ ‫”והשתקפותו באדריכלות‬, Jerusalem and Eretz-Israel 8–9 [FS A. Kloner, ed. B. Zissu]: 171– 184. Avshalom-Gorni, D. and Najar, A. 2013. “Migdal”. Hadashot Arkheologiyot 125. http://www.hadashot-esi.org.il/Report_Detail_Eng.aspx?id=2304&mag_id=120 (accessed 27 February 2020). Bauckham, R. 2015. “Further Thoughts on the Migdal Synagogue Stone”. NovT 57: 113– 135. Binder, D. D. 2014. “The Mystery of the Magdala Stone”. In: Warner, D. A. and Binder, D. D. (eds.), A City Set on a Hill: Essays in Honor of J. F. Strange. Mountain Home, AR: BorderStonePress, 17–47. Corbo, V. C. 1982. “Resti della sinagoga del primo secolo a Cafarnao”. Studia Hierosolymitana 3: 313–357. De Luca, S. and Lena, A. 2015. “Magdala/Taricheae”. In: Fiensy, D. A. and Strange, J. R. (eds.) Galilee in the Late Second Temple and Mishnaic Periods, Vol. 2: The Archaeological Record from Cities, Towns, and Villages. Minneapolis, MN: Fortress Press, 280–342.

40

Zeev Weiss

Edwards, D. R. 2002. “Hirbet Qana: From Jewish Village to Christian Pilgrim Site”. In: Humphrey, J. H. (ed.), The Roman and Byzantine Near East, III. Portsmouth, RI: Journal of Roman Archaeology, 101–132. Ehrlich, U. 1999. The Non-Verbal Language of Jewish Prayer. Jerusalem, Magnes (in Hebrew). Fine, S. 2017. “From Synagogue Furnishing to Media Event: The Magdala Ashlar”. Ars Judaica 13: 27–38. Geva, H. 2007. Jerusalem Quarter Excavations in the Old City of Jerusalem conducted by Nahman Avigad, 1969–1982, III: Area E and Other Studies. Jerusalem: Israel Exploration Society. Hachlili, R. 2000. “Torah Shrine and Ark in Ancient Synagogues: A Re-evaluation”. ZDPV 116: 147–183. Har-Even, B. 2012. “ ‫ביזנטית והתקופה‬-‫ התקופה הרומית‬,‫ יישוב מימי הבית השני‬:‫חורבת א־טוואני‬ ‫”המוסלמית הקדומה‬. In: ‫ דברי הכנס השביע‬:‫הְספר והמדבר בארץ ישראל‬, 15–29. Available from: http://www.atarsusya.co.il/pages/articles.asp (accessed 27 February 2020). Har-Even, B. 2016. “A Second Temple Synagogue at Khirbet Diab in Western Benjamin”. Qadmoniot 151: 49–53 (in Hebrew). Haran, M. 1986. “Torah and Bible Scrolls in the First Centuries of the Christian Era”. Shnaton 10: 93–106 (Hebrew). Hirschfeld, Y. 1995. The Palestinian Dwelling in the Roman-Byzantine Period. Jerusalem: Franciscan Printing and Israel Exploration Society. Israel, Y. and Erickson-Gini, T. 2013. “Remains from the Hellenistic through the Byzantine Periods at the ‘Third Mile Estate’, Ashqelon”. Atiqot 74: 167–222. Kloppenborg, J. S. 2000. “Dating Theodotos (CIJ II 1404)”. JJS 51: 243–280. Kraeling, C. H. 1956. The Excavations at Dura Europos: Final Report, VIII/1: The Synagogue. New Haven, CT: Yale University Press. Leibner, U. 2010. “Excavations at Khirbet Wadi Hamam (Lower Galilee): The Synagogue and the Settlement”. Journal of Roman Archaeology 23: 220–237. Levine, L. I. 2005. The Ancient Synagogue: The First Thousand Years, 2nd rev. ed. New Haven, CT: Yale University Press. Netzer, E. 2003. “The Synagogues from the Second Temple Period according to Archaeological Finds and in Light of the Literary Sources”. In: Bottini, C. G., Di Segni, L. and Chrupcała, D. (eds.), One Land – Many Cultures: Archaeological Studies in Honour of S. Loffreda OFM. SBFCMa 41. Jerusalem: Franciscan Printing Press, 277–285. Netzer, E. et al. 2004. “The Synagogue Complex”. In: idem, Hasmonean and Herodian Palaces at Jericho. Jerusalem: Israel Exploration Society and Institute of Archaeology, 159–192. Netzer, E. and Weiss, Z. 1995. “New Evidence for Late-Roman and Byzantine Sepphoris”. In: Humphrey, J. H. (ed.), The Roman and Byzantine Near East: Some Recent Archaeological Research. JRA Supplementary Series 14. Portsmouth, RI: Journal of Roman Archaeology, 162–176. Peleg-Barkat, O. 2016. “Interpreting the Uninterpreted: Art as a Means of Expressing Identity in Early Roman Judaea”. In: Leibner, U. and Hezser, C. (eds.), Jewish Art in Its Late Antique Context. TSAJ 163. Tübingen, Mohr Siebeck, 27–48. Porat, Y. (ed.) (in press), The Synagogue of Roman-Byzantine En-Gedi (Qedem). Jerusalem: Institute of Archaeology, Hebrew University. Ratner, B. 1967. Ahavat Zion v’Yerushalayim. 12 vols. Reprint. Jerusalem: no publisher (in Hebrew). Richter, G. M. A. 1966. The Furniture of the Greeks Etruscans and Romans. London: Phaidon.

The Synagogue in an Age of Transition

41

Roth-Gerson, L. 1987. The Greek Inscriptions from the Synagogues in Eretz Israel. Jerusalem: Yad Izhak Ben-Zvi (in Hebrew). Sapir, I. 2007. “The Ancient Synagogue in Israel in the Light of the Talmudic Sources”. PhD thesis, The Hebrew University of Jerusalem (in Hebrew). Schwartz, S. 2001. Imperialism and Jewish Society: 200 B.C.E to 640 C.E. Princeton, Princeton University Press. Schiffman, L. H. 2017. “The Magdala Stone”. Ami Magazine May 28, 2017: 162–164. Available from: http://lawrenceschiffman.com/wp-content/uploads/2017/ 06/magdalastone.pdf (accessed 27 February 2020). Strange, J. F. and Shanks, H. 1983. “Synagogue Where Jesus Preached Found at Capernaum”. Biblical Archaeology Review 9: 24–31. Tov, E. 2004. Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert. STDJ 54. Leiden: Brill. Weiss, Z. 1990. “The Location of the Sheliah Tzibbur during Prayer”. Cathedra 55: 8–21 (in Hebrew). Weiss, Z. 2005. The Sepphoris Synagogue: Deciphering an Ancient Message in Its Archaeological and Socio-Historical Contexts. Jerusalem: Israel Exploration Society. Weiss, Z. 2009. “Stratum II at Hammath Tiberias: Reconstructing Its Access, Internal Space, and Architecture”. In: Rodgers, Z. et al. (eds.), A Wandering Galilean: Essays in Honour of S. Freyne. Leiden: Brill, 321–342. Weiss, Z. 2017. “Is Rabbinic Literature a Valid Tool for Analyzing Archaeological Finds?”. In: Ferziger, A. (ed.), Studies in Judaism and Jewish Culture in Honor of Rabbi Professor D. Sperber. Ramat Gan: Bar-Ilan University Press, 419–435 (in Hebrew). Yavor, Z. 2010. “The Architecture and Stratigraphy of the Eastern and Western Quarters”. In: Syon, D. and idem (eds.), Gamla II: The Architecture, The Shmarya Gutmann Excavations 1976–1989. IAA Reports 44. Jerusalem: Israel Antiquities Authority, 13–112. Zangenberg, J. 2016. “A Basalt Stone Table from the Byzantine Synagogue at "orvat Kur, Galilee: Publication and Preliminary Interpretation”. In: Patrich, J., Peleg-Barkat, O., and Ben-Yosef, E. (eds.), Arise, Walk through the Land: Studies in the Archaeology and History of the Land of Israel in Memory of Y. Hirschfeld on the Tenth Anniversary of his Demise. Jerusalem: Israel Exploration Society, 61*–78*. Zissu, B. and Ganor, A. 2004. “The Public Building at Horvat Ethri – A Synagogue? From the Period Between the Jewish War and the Bar Kokhba Revolt”. In: Eshel, Y. et al. (eds.) And Let Them Make Me a Sanctuary: Synagogues from Ancient Times to the Present Day. Ariel: Judea and Samaria Academic College, 31–42 (in Hebrew). Zissu, B. and Ganor, A. 2009. “Horvat Etri – A Jewish Village from the Second Temple Period and the Bar Kokhba Revolt in the Judean Foothills”. JJS 60, 90–136.

The Dating of the “Galilean”-Type Synagogues: Khirbet Wadi amam as a Case-Study1

Uzi Leibner

Introduction The dating of the “Galilean”-type synagogues has been debated by scholars for nearly fifty years. The “Galilean”-type synagogues consist of some thirty monumental buildings located in the eastern Galilee, considered to be the earliest architecturally defined group of synagogues from the post-Second Temple period (Fig. 1). In the early twentieth century, the German scholars Heinrich Kohl and Carl Watzinger launched the study of these synagogues, pointing to their fairly uniform plan and characteristics: the buildings were erected on a north–south axis; two interior rows of pillars running parallel to the long walls of the building, and another running parallel to the rear wall, divided the space into a wide nave surrounded by three narrow aisles; the pillars supported a wooden construction covered by a tiled roof; some of the buildings had a second story above the aisles (galleries); the buildings were constructed of hewn stones and decorated primarily with carved architectural elements; the decorative emphasis was on the facade wall facing southward, toward Jerusalem; the facade contained ornate entranceways—usually a large central entrance flanked by two smaller ones, while small buildings had one entrance in the middle of the facade (Kohl and Watzinger 1916). Kohl and Watzinger’s research was based exclusively on a stylistic analysis of the architecture and decoration. They pointed to the similarity between these “Galilean”-type synagogues and various Roman-period structures in southern Syria. While the synagogues generally have no date-bearing inscriptions, many of the Syrian comparanda have inscriptions dating them primarily to the second and third centuries CE. Kohl and Watzinger’s conclusion that the “Galilean”-type synagogues were built in the same period and should thus be attributed to the thriving Galilean Jewish community of the second–third centuries CE, was the accepted view for over fifty years.

 1

A detailed presentation of the Kh. Wadi amam project has appeared in the final publication of the excavation (Leibner 2018).

44

Uzi Leibner

Fig. 1. Map of sites where “Galilean”-type synagogues (or components of such synagogues) have been identified (drawing by R. Sabar).

In the 1970s, the Franciscan Fathers Virgilio Corbo and Stanislao Loffreda excavated the synagogue of Capernaum, which is considered the best example of a “Galilean”-type synagogue, with the richest architectural decoration. Based on a careful stratigraphic excavation and finds sealed beneath the floors, the synagogue was dated to the late fourth century, and the completion of its courtyard to the late fifth century (Corbo 1972; 1975; Loffreda 1972; 1973; 1979; 1993). These conclusions raised an ongoing scholarly debate, as the building was perceived as belonging stylistically to the late second or third centuries, following Kohl and Watzinger (1916: 4–40, 184–203; Foerster 1972: 153–59; see also discussions on the date of the Corinthian capitals of Capernaum in Bloedhorn 1989; Rough 1989; Fischer 1990: 71). Thus, the Capernaum synagogue presented a sharp contradiction between two basic dating methods in archaeology—one stylistic and the other stratigraphic (see, e.g., Foerster 1971; Tsafrir 1995; Ma‘oz 1999). Excavation of the synagogues at Kh. Shema‘, Meiron, Gush alav and Nabratein in the Upper Galilee by a team headed by Eric Meyers, revealed that

The Dating of the “Galilean”-Type Synagogues

45

the first phase of these structures should be dated to the second (Nabratein) and third centuries, thus corroborating the view that these “Galilean”-type synagogues belong to the Roman-period building tradition (Meyers, Kraabel, and Strange 1976; Meyers, Strange and Meyers 1981; Meyers, Meyers, and Strange 1990; Meyers and Meyers 2009). These buildings stood for centuries and some of them underwent renovations or extensive modifications once or twice in the course of their existence. This excavation team almost entirely avoided comparative research of the architectural decoration, and their conclusions were based solely on stratigraphic considerations and small finds from sealed loci. In the last two decades, however, Jodi Magness has presented a renewed stratigraphic analysis of these excavations, claiming that all the buildings were misdated and were actually erected only in the Byzantine period: Kh. Shema‘, Meiron and Nabratein in the late fourth or early fifth centuries, Gush alav in the second half of the fifth century, and Capernaum in the first half of the sixth century (Magness 1997; 2001a; 2001b; 2009; 2010). Korazim is the only other “Galilean”-type synagogue that has been excavated and fully published, and its excavator dates the beginning of its construction to the early fourth century CE, and its completion to the late fourth century (Yeivin 2000: 31*, 106). In this case also, Magness proposes a later date and suggests the synagogue was constructed no earlier than the late fifth century (Magness 2007).2 In fact, there is no “Galilean”-type synagogue today for which there is a consensus that it was built in the Roman period (second–early fourth centuries CE). Some scholars believe that they were all built in the Byzantine period, primarily in the fifth–sixth centuries, perhaps in reaction to church construction (e.g., Schwartz 2001: 184–212; Magness 2001b: 90). This issue has far-reaching implications with regard to basic conceptions in the archaeological and historical research of these periods. If these synagogues were indeed built two or three hundred years after this architectural style prevailed in the region, then the basic principles of the art-historical method of dating on the basis of style and comparative research are thrown into question. In addition, their “late chronology” would leave us with no synagogue remains from the heyday of Galilean Jewry—the second–early fourth centuries—and date them precisely to an era that the sources portray as one of a declining Jewish population suffering from oppression under the Christian regime.

The Khirbet Wadi amam Excavations In an attempt to shed new light on this highly debated question, it was decided to conduct a case-study excavation of another “Galilean”-type synagogue, one that had never been excavated previously. The site chosen for excavation was  2

Several parameters of the decorative style of this synagogue, however, are different and indeed later than the “Galilean”-type synagogues mentioned above (Amir 2007).

46

Uzi Leibner

Khirbet Wadi amam, located in the eastern Lower Galilee above Naal Arbel (Wadi amam), 2 km west of the Sea of Galilee. The site (map ref. 2460/7480) is situated opposite Mount Arbel, on a steep slope at the base of the massive cliffs of Mount Nitai. Surrounded on three sides by the riverbeds of Naal Arbel and Naal Savyona, and by cliffs on the fourth, the site is naturally fortified. The settlement, whose ancient remains cover some thirty to forty dunams, was one of the largest villages in this region in the Roman period. A perennial spring immediately below the site, as well as the proximity of the ancient route that led from the Sea of Galilee through Naal Arbel to central Galilee, and from there to the Mediterranean coast, can explain the development of this large settlement. Despite its size, central location, and nearness to continuously settled sites such as Tiberias (6 km), the ancient name of this village has long disappeared. Kh. Wadi amam may be identified with Migdal aba‘aya (‫)מגדל צבעיא‬, a site mentioned several times in Palestinian rabbinic literature of the Amoraic period (third–fourth centuries CE). Literary sources indicate that Migdal aba‘aya should be located somewhere near Tiberias, and it is described as a prosperous settlement that was probably destroyed during the Jewish revolts against Rome, but is mentioned again in the Late Roman period . The ruin was first documented in the nineteenth century by the Survey of Western Palestine, but its surveyors noted here only “heaps of stones” (Conder and Kitchener 1881: 409). Joseph Braslavsky (1925: 140) was the first to draw attention to architectural elements at the site that seemed to belong to a “Galilean”-type synagogue. The remains were surveyed again in the 1970s and 1980s by Gideon Foerster (1983: 243) and Zvi Ilan (1991: 128). Recently, the site was surveyed as part of the Eastern Galilee Survey, in which hundreds of datable potsherds were collected from the surface (Leibner 2009: 205–210). In the same survey, the site was also sampled by a shovel-testing technique, whereby shallow probe pits were excavated in all parts of the village and all the soil was sifted. Altogether, close to 600 identifiable sherds were collected and all were attributed to the first centuries CE, none postdating ca. the late fourth century CE, leading us to suggest that the site was abandoned around that time (Leibner 2009: 71– 74). These findings pointed to the unique advantages of excavating here, as it was unlikely that the synagogue was constructed after the village had been abandoned; and in addition, it appeared that the village was not covered by later strata, a situation that at other sites had complicated the picture and resulted in interpretative debates. Six seasons of excavations were conducted at the site (2007–2012). As the synagogue was built by and for the ancient local community, it clearly could not be detached from its environment and had to be studied against the background of village life and history. This objective necessitated excavation of large portions of the surrounding village, an undertaking that corresponded well with our other research questions concerning life in rural Galilee, as well as our basic conception

The Dating of the “Galilean”-Type Synagogues

47

that only a holistic study of a site can provide faithful interpretations of its different components. This strategy proved to be crucial not only for the study of the synagogue itself, but also for an entire array of subjects concerning village life in Roman Galilee, such as periods of prosperity and decline, communal organization, means of livelihood, domestic architecture, household utensils and the ancient diet. The excavations extended over four areas in the village and unearthed in total some ten residential structures, two olive-oil presses, alleys, and a monumental synagogue (Fig. 2). The remains belong to two main and distinct strata: Stratum III dated to the first century BCE–early second century CE, and Stratum II dated to the third–fourth centuries CE.

Fig. 2. Khirbet Wadi "amam and Mount Nitai on a 5 m contoured topographical map. The synagogue is located in Area A (drawing by B. Arubas).

The Early Roman “Public Building” The finds indicate that the site was first settled in the Hasmonean period. Due to intensive later construction, only a few architectural remains could be clearly dated to this period. Nevertheless, Late Hellenistic pottery found in the fills beneath the Early Roman structures, as well as a dozen Seleucid and autonomous Phoenician coins from second-century BCE Akko-Ptolemais and Tyre (mainly the second half of that century), and some 90 Hasmonean coins, indicate that occupation here began around the early first century BCE.

48

Uzi Leibner



Fig. 3. Block plan of Stratum 3 remains (in black) on the terrace and its surroundings (Stratum 2 walls in light gray).

The settlement expanded and developed during the Early Roman period, reaching its peak in terms of size and prosperity, as is evident from the impressive domestic structures and rich finds from this period revealed in all the excavated areas. In the first half of the first century CE, a 17 m-wide terrace was leveled in the center of the village, overlooking the beautiful scenery of the Sea of Galilee and the Arbel cliffs. In order to prepare this steep area for construction, the builders erected a retaining wall to support a huge artificial fill, over 5 m deep in some places. Only a small part of the original superstructure that stood on this terrace has survived, as it was dismantled in the Late Roman period and replaced by a “Galilean”-type synagogue (Fig. 3). Nevertheless, what has survived enables us to conclude that the Early Roman structure that stood here was also a public building. Four rooms in the western section of this first-century building have survived (Rooms A6, A8, A5N, A5S); three of them were sealed behind the western wall of the “Galilean”-type synagogue when it was constructed, and the fourth (Room A6) was incorporated into it, although the floor was raised and some modifications were implemented. This latter room (extant dimensions 3.8 m × 2.5 m) was originally built of well-dressed basalt blocks, was surrounded by low

The Dating of the “Galilean”-Type Synagogues

49

benches along the walls, and had a plastered floor with a plastered element of unknown purpose (0.8 m × 0.6 m) in its center (Fig. 4).

Fig. 4. The side room of the Early Roman “public building” surrounded by low benches (photograph by G. Laron).

Its plan resembles that of the side room in the Second Temple-period synagogue at Gamla, which Shmarya Gutmann called “the study room” (Syon and Yavor 2010: 56). A side room surrounded by low benches was also discovered recently in the Second Temple-period synagogue at nearby Magdala (Avshalom-Gorni and Najar 2013). South of this side room at Kh. Wadi amam is another wellpreserved auxiliary room with a plastered floor, its walls coated with fine white plaster and preserved to a maximal height of over 2 m (Room A8). Negative impressions in the wall and floor plaster indicate that this room also had benches along its walls. It should be noted that the general plan of four small auxiliary rooms located along one wing of the building, is similar to that found in the Gamla synagogue. A drum of a basalt heart-shaped column was found in one of the rooms sealed behind the western wall of the “Galilean”-type synagogue, and most likely belonged to this Early Roman structure. Similar basalt heart-shaped columns are known from the Gamla synagogue. Other basalt architectural elements, such as column and pilaster bases and a pilaster or anta capital, also seem to have belonged to this structure. Building stones decorated with painted plaster in the

50

Uzi Leibner

secco technique, and many fragments of such plaster, were found in soundings excavated into the terrace, beneath the floor of the “Galilean”-type synagogue. The painted secco style is typical of the first century CE and similar examples were revealed on the walls of the synagogue at Magdala (Rozenberg 2018). The soundings also yielded fragments of stucco flutings of Doric columns that apparently decorated the Early Roman structure and were later buried when the terrace was leveled again before the “Galilean”-type synagogue was erected. Thus, the size of the terrace and its central location, the quality of the building and the special plan of the rooms in the western wing, and the style of the architectural decorations, indicate that this early, lavish structure was of a public nature. The most plausible interpretation of a public building in a Jewish village of this period is a synagogue; the resemblance between features of our building and the excavated synagogues of Gamla and Magdala strengthens this identification. It should be noted that remains of Second Temple-period “public buildings” were also found beneath the Late Roman synagogues at ammat Tiberias and Gush alav (Dothan 1983: 15–19; Meyers, Meyers, and Strange 1990: 62–64) and it is probable that these were synagogues as well. These examples, together with the synagogues discovered at sites such as Qiryat Sefer and Umm el-‘Umdan (Magen, Tzionit, and Sirkis 2004; Onn and Weksler-Bdolah 2008), indicate that the distribution of synagogues in rural areas during the Second Temple period was more extensive than previously thought. The “public building”, together with the entire, affluent Early Roman settlement at Kh. Wadi amam, came to an abrupt end in a massive destruction documented wherever our excavations were carried out. The ample numismatic finds, and mainly two coin hoards, enable us to pinpoint this destruction to the reign of Hadrian, ca. 125–135 CE (Leibner and Bijovsky 2013). This date, and the dramatic nature of the destruction, raise the possibility that it was connected to the Bar Kokhba Revolt, or perhaps to local unrest upon the arrival of a Roman legion to the Galilee a few years before that revolt.

The “Galilean”-Type Synagogue Following the second-century destruction, the site seems to have been abandoned for several decades, or at least suffered a serious decline, as is evident from the decrease in the quantity of coins dating to the latter part of that century. A wide range of evidence for habitation and the massive construction of new structures appears around the end of the second or early third century CE. Remains of Stratum II were uncovered in all the excavation areas in the village, indicating that Kh. Wadi amam flourished again in the third and early fourth centuries CE. The structures of Stratum III were not restored but rather systematically blocked and filled (in some cases to heights of 3–4 m), and new structures,

The Dating of the “Galilean”-Type Synagogues

51

alleys, and installations were built on top of them. While the new construction followed the lines of the re-used early terraces, the general layout differed from that of the previous stratum. A “Galilean”-type synagogue was now built on the wide, pre-existing terrace where the “public building” and an adjacent domestic structure had previously stood. This location was chosen due to the wide platform that was available here. If this Early Roman structure had indeed been a Second Temple-period synagogue, the choice for the location of the new synagogue may also be connected with the continuity and sanctity of sacred sites, a phenomenon well known at other synagogue sites and at ancient Near Eastern cultic places in general.3 On the northern and southern sides of the “Galilean”-type synagogue were alleys, and beyond them were large domestic insulae. The synagogue, alleys and insulae were all built according to a grid-like plan, and together with the datable finds within them, it is clear that the entire area was both conceived and constructed simultaneously. The chronological history of the “Galilean”-type synagogue is divided into two main phases (Synagogues I and II) spanning the Middle Roman to Early Byzantine periods. Based on the surviving pedestals, it seems that the floor level of Synagogue I was some 10–15 cm lower than that of the Early Roman “Public Structure”, attesting to deepening and leveling works that removed any trace of the former floor. While the main terrace was initially leveled in the early first century CE, the renewed construction works, such as the insertion of column foundations, disturbed its surface and caused some later material to penetrate beneath floor level. The walls of the western half of Synagogue I as well as four pedestals have survived in situ. Together with numerous architectural elements assigned to this phase that were found scattered around the building or embedded, in secondary use, in the Phase II structure, they enable us to reconstruct the plan of Synagogue I. It was a rectangular basilica with internal dimensions of 15.4 m × 12.8 m (17.2 m × 14.7 m external dimensions). Rows of columns lined three sides of the hall, dividing it into a nave surrounded by narrow aisles on the east, north and west. In comparison with other “Galilean”-type synagogues, which are usually elongated (with an average length to width ratio of ca. 3:2), this nearly square plan is odd, apparently due to reliance on the given shape of the Early Roman terrace. Also, as a result of the shape of the ancient terrace, the new synagogue and its façade did not face directly south, toward Jerusalem, but rather to the southeast (148°). In addition, the eastern aisle is wider than the western one by ca. 1.0 m. A possible explanation for this asymmetrical plan may be the architectural necessity to reduce the pressure on the vulnerable eastern wall by placing the eastern row of columns as far as possible from the wall.  3

See, for example, ammath Tiberias, where several synagogue structures were built one on top of the other between the third and eighth centuries (Dothan 1983; 2000).

52

Uzi Leibner

The current outer walls of the synagogue are composed of two distinct parts, distinguished by building techniques and materials. The western half of the frame, including the entire western wall (W2A05) and the western halves of the northern and southern walls (W1A01, W2A01), survived from Phase I. These walls, built of well-cut basalt blocks arranged in headers and stretchers, are bound together at the corners. The walls of the eastern half clearly differ from those of the western half in technique, materials and width, and belong to the restoration of Synagogue II (Fig. 5; see below).

Fig. 5. Aerial view of the synagogue area after excavation (photograph by I. Arbel).

Rich assemblages of finds, including hundreds of datable pottery sherds, were found along the western and northern walls and in their foundation trenches and facilitated the dating of Synagogue I. The construction of the western wall sealed behind it three of the auxiliary rooms of the “public building” of Stratum III. The foundation trench of this wall cut through the perpendicular walls of these rooms to bedrock, and through the early second-century destruction layer on the floors of the two southernmost rooms. The huge intentional fill that buried Room A8 when the “Galilean”-type synagogue was built contained a rich assemblage of finds that enabled us to date its sealing to the early third century CE. Room A6, the northernmost of the Early Roman auxiliary rooms, was integrated into the “Galilean”-type synagogue, with some alterations. The western synagogue wall cut through the eastern part of the room and its plaster floor. A threshold leading into the now-shortened room was incorporated into this wall and a floor made of hard-packed earth was leveled with the threshold. The new

The Dating of the “Galilean”-Type Synagogues

53

floor was ca. 0.4 m above the previous floor, burying the encircling benches and the stump of the central element of the earlier phase. The finds in the fill beneath this new floor included Middle Roman pottery and oil lamps, also indicative of a third-century date for its laying. The southern wall of this room (W2A16) was significantly thickened in order to support the impressive vault made of limestone ashlars that roofed the newly designed room. The care and attention invested in the construction of this room, and the fact that it could only be reached from within the synagogue, attest to its significant role during this phase. It may have been used to store the Holy Scriptures in the early stages, before the bema and presumably the ark that stood on it were added inside the synagogue hall, or it may have served as a treasury.4 The foundation trench of the northern synagogue wall cut through an Early Roman domestic structure that was buried partly beneath the synagogue and mostly beneath the northern alley (see above, Fig. 3). Rich assemblages of finds from the foundation trench and the intentional fills that buried the earlier structure date this undertaking to the early third century CE. A similar chronological picture was revealed further to the east, where the northeastern corner of the “Galilean”-type synagogue overlies another Early Roman domestic structure. This structure (Unit A12), preserved to a height of over 3 m, went out of use in the early second century CE, and the rich finds indicate that it was intentionally buried when the synagogue was erected in the early third century CE. In the foundation trench of the northeastern corner of the synagogue were mainly Middle Roman sherds with a few intermediate Middle–Late Roman sherds, the latter probably reflecting the repairs here in the late third century (Phase II, see below). Six large soundings were excavated beneath floor level in different parts of the “Galilean”-type synagogue, in order to determine the building’s date, to understand its construction techniques, and to reveal any earlier structures buried beneath it. In total, the soundings yielded 559 diagnostic potsherds and 17 coins. These finds indicate that the wide terrace on which the synagogue was built was originally leveled in the first half of the first century CE, as the vast majority of the pottery and all of the 17 coins clearly demonstrate. A few dozen Middle Roman sherds were retrieved in certain locations and attest to works that penetrated below the surface of the terrace when Phase I of the “Galilean”-type synagogue was constructed; for example, in the vicinity of the column foundations (Fig. 6).

 4

Similar auxiliary rooms were found in other Galilean synagogues, such as Kh. Shema‘ (Meyers, Kraabel, and Strange 1976: 58–61, 76–80), Korazim (Yeivin 2000: 15), and Meroth, where the room was similarly vaulted and a treasury was also found (Ilan and Damati 1987: 44, 66–68, 97); see, in general, Levine 2005: 316–319.

54

Uzi Leibner

Fig. 6. Representative sample of Middle Roman (Nos. 1–2), Early Roman (Nos. 3–9) and Late Hellenistic (Nos. 10–12) pottery from Sounding A4 beneath the synagogue floor in the southeastern corner of the hall.

A few Late Roman pottery sherds were found in the bedding of the mosaic floor and in a drainage channel added beneath the center of the building, indicating works performed during Phase II of the “Galilean”-type synagogue (see below). It should be emphasized that not a single vessel of types that began to appear in Galilean assemblages from the early decades of the fourth century on, was found. In addition, over 6,000 body sherds from beneath the floor were also collected and examined. All of these body sherds are of common Late Hellenistic and Roman-period fabric, while none are of the easily distinguished fabrics that began to prevail in the Galilee from the early decades of the fourth century. To summarize the stratigraphic evidence: the abundant datable finds collected from the soundings beneath the floor level and from the foundation trenches of the western and northern walls that have survived from Phase I of the “Galilean”type synagogue, all indicate that this structure was erected in the early third century CE. The architectural decoration of Synagogue I is characterized by the fine workmanship of the local, yellowish-white, hard Eocene limestone utilized in the construction of the columniation, the seating tiers and the elaborate façade, all displaying high-quality carving. Hundreds of fragments of these architectural elements were recovered during the excavations, and here we present only a few components that have decorative or chronological significance (for details of the architectural decoration of the synagogue, see Leibner 2018: 99–129). The columniation included nine pseudo-Tuscan columns resting on well-dressed

The Dating of the “Galilean”-Type Synagogues

55

pedestals, four of which have survived in situ from Phase 1 (Columns 4, 5, 8, 9). Based on the surviving components, the estimated height of the columns, from the bottom of the pedestal to the top of the capital, was ca. 3.9–4.0 m with a ca. 0.5–0.6 m diameter. Similar columns, but of smaller dimension, found among the debris outside the building, apparently indicate that the building had secondstory galleries. The pseudo-Tuscan (Roman-Doric) capitals consist of an echinus with a “convex-cavetto” molding separated by narrow fillets, and above it a square, straight abacus. Some of these capitals were apparently reused in Synagogue II for their original purpose. Others were found discarded outside the building, buried beneath the southern alley or embedded in the benches of Synagogue II, and were clearly not in use in the latter phase (Fig. 7).

Fig. 7. A pseudo-Tuscan capital embedded in the foundation of the eastern benches of Synagogue II (photograph by G. Laron, drawing by O. Reicher).

Pseudo-Tuscan capitals are common in “Galilean”-type synagogues, as at Bar‘am, Gush- alav, Meiron, orvat ‘Ammudim, Capernaum and Arbel (Kohl and Watzinger 1916: Figs. 6, 39, 49, 138, 148, 158, 169, 172, 178; Meyers, Meyers, and Strange 1990: Figs. 35–36). They are quite rare in civil and pagan buildings in the region, and were found in a few structures in Lebanon and Syria (especially in the Hauran), such as the great sanctuary at Qala‘at Faqra (Krencker and Zschietzschmann 1938: 43, Fig. 63), the theater at Bora (Brünnow and von Domaszewski 1909: Figs. 966, 956–957, 963 [pilasters]), and the “Praetorium” at Musmiye (Hill 1975: Fig. 1; see also Foerster 1972: 117–118; Ma‘oz 2017: 29–30). The Syrian parallels are dated mainly to the second–early third centuries CE, and no later examples are known. None of the benches of Synagogue I survived in situ, although numerous bench blocks of yellowish-white limestone with a molded cyma reversa profile and a grooved fillet were found in the debris inside the synagogue, buried beneath the later southern alley, and discarded around the building. These seem to represent an entire set of limestone benches of Synagogue I, some of which were apparently reused as building stones in Synagogue II. These molded benches

56

Uzi Leibner

resemble benches in Roman theaters in our region, generally dated to the second–third centuries CE, and may thus also serve as a chronological indicator.5

Fig. 8. A reconstruction of the entire lintel and photograph of the lintel fragment (drawing by N. Arad and B. Arubas, photograph by G. Laron).

Numerous finely hewn ashlars made of yellowish-white limestone were also found discarded around the building (mainly in front of the southern facade), buried beneath the surface of the southern alley, and re-used in Synagogue II. In all likelihood, these ashlars originated in the facade of Synagogue I, while the rest of the framing walls of the synagogue were of basalt, as evident from their surviving remains. A lintel of yellowish-white limestone, ca. 2.1 m wide and decorated with a relief of a pair of eagles flanking a wreath, most likely adorned the single doorway in the center of the facade. Only half of the lintel has survived. The accuracy of the anatomic details in which the raptor is depicted points to a highly skilled and perceptive artist (Fig. 8). While not comparable in artisanship to eagle carvings from Roman urban centers in the region, this lintel was clearly  5

For example, Tiberias theater: Atrash 2012: 82; Beth She’an theater: Mazor 2015: 573, Fig. 9.118; Gadara theater: Weber 2002: Pl. 10; Gerasa theaters: Kraeling 1938: Pls. V:a; XXXIV:b; Samaria theater: Crowfoot, Kenyon, and Sukenik 1942: Pl. LVII:2, Fig. 25; Philippolis theater: Brünnow and von Domaszewski 1909: Fig. 1065; Bora theater: eidem: Fig. 967; Caesarea circus: Porath 2013: 45, Fig. 3.16.

The Dating of the “Galilean”-Type Synagogues

57

influenced by Roman-period art and bears similarities, for example, to the eagle on an altar from Legio, and those on a lintel of the Ba‘alshamin temple in Palmyra, both probably dating to the second century CE (Avi-Yonah 1946: Pl. XXVI.6; Collart and Vicari 1969: Pl. 97.1). The closest parallels to our relief from other “Galilean”-type synagogues are the solitary eagles with outstretched wings on the soffit of the lintel of the Gush alav synagogue, and on the lintel of the main entrance to the Capernaum synagogue, both of which are likely from the third century (Meyers, Meyers, and Strange 1990: 84–85, Fig. 25, Photo 40; Kohl and Watzinger 1916: 12, Figs. 17–18.).6 The lintel rested on molded doorjambs adorned with pseudo-Tuscan pilasters, in accordance with the style of the free-standing columns inside the building.7 Corner pseudo-Tuscan pilasters adorned the corners of the facade, and flat pilasters apparently decorated the front of the facade, in line with the rows of columns inside, as was common in monumental columned structures of the Roman period (cf. Capernaum: Kohl and Watzinger 1916: 7, Figs. 5–6, 39, 74, Tafel III; Korazim: Yeivin 2000: 40, Figs. 87–88, Plan 10). These pseudo-Tuscan pilasters also provide chronological evidence: similar ones were found in the Bora theater dated to the second half of the second century CE (Brünnow and von Domaszewski 1909: Figs. 956–957, 963), while pilasters identical to ours adorn the facades of Catacombs 11, 14 and 20 and the mausoleum near Catacomb 11 in Beth She‘arim, all dated approximately to the late second–first half of the third centuries CE (Maisler 1941: Pl. IV; Avigad 1976: 43, 86, Figs. 20, 36, 37.8, Pls. XXXI, XVII:2–3). Fragments of a modillion cornice of yellowish-white limestone were found in the southern alley and it clearly decorated the synagogue’s facade (Fig. 9). The cornice displays a high level of stonework artisanship, and bears the patterns and moldings of the typical “Syrian sequence” (Weigand 1914: 86–87), which was common on cornices in the region during the second–third centuries CE (Turnheim 1996). An almost-identical cornice was found at the Sumaqa synagogue (Turnheim 1999: 236–239, Pls. 10–13),8 and similar examples are seen at Capernaum (Kohl and Watzinger 1916: 18, Figs. 32–34). The cornice was apparently  6 The lintel from Gush alav apparently originated in the first phase of the synagogue, which was dated by the excavators to ca. the mid-third century (Meyers, Meyers and Strange 1990: 10–12). At Capernaum, the notion that many of the architectural elements of the Byzantine synagogue were reused and originated in earlier structures, has become widely acknowledged (see, e.g., Ma‘oz 1999; De Luca 2013: 175). Based on style and artisanship, Arubas and Talgam divided these elements into three chronologically distinct groups. The lintel of the main entrance was assigned by them to the earliest group, apparently originating in a second- or third-century building (Arubas and Talgam 2014: 248). 7 Doorjambs with pseudo-Tuscan pilasters adorned the synagogues of . ‘Ammudim and Sumaqa; in the latter they were preserved in situ in the side portals of the facade; see Kohl and Watzinger 1916: 74, Fig. 142 ( ‘Ammudim); Turnheim 1999: Pls. 1, 3–4, 7 (Sumaqa). 8 The initial phase of the Sumaqa synagogue was dated between the end of the second and the midthird centuries CE; see Dar 1999: 19; Turnheim 1999: 248.

58

Uzi Leibner

positioned above the lintel of the main entrance, as seen in various buildings in the Land of Israel and southern Syria.9

Fig. 9. Fragments of a modillion cornice found in front of the synagogue facade (photograph by G. Laron, drawing by M. Edelcopp).

Numerous fragments of limestone cyma recta cornices, belonging to three distinct types, were found outside the synagogue, concentrated mainly in the southern alley. Based on parallels that survived in situ in similar structures, one type was probably located between the first and second stories, another apparently adorned the base of the pediment, and the third, of which only a few fragments were found, may have adorned a window (for parallels, see Kohl and Watzinger 1916: 92–97, Figs. 177, 180–186 [Bar‘am]; idem.: 83–84, Figs. 163–166 [Meiron]; Brünnow and von Domaszewski 1909: 4–9, Figs. 871–878; 15–16, Figs. 888–897; 62–63, Figs. 945–946, 957 [Bora]; Butler 1903: 65–67, 70, 100–101, 121, 346, 402  9 See, e.g., the central entrance to the postscaenium of the Beth She’an theater: Atrash 2015: 318, Fig. 7.22; the Tychaeon at es-Sanamen: Kohl and Watzinger 1916: 151, Fig. 287; the temples at Bziza, Hössn Soleiman, Niha and Majdal ‘Anjar: Krencker and Zschietzschmann 1938 (Tafelband): Plates 4, 33, 45–46, 55, 76, respectively; the temple of Jupiter at Damascus: Watzinger and Wulzinger 1921: 8, Figs. 8–9; the temple of Ba‘alshamin at Palmyra: Lyttelton 1974: Pl. 151; a tomb at Yarhai: idem.: Pl. 161.

The Dating of the “Galilean”-Type Synagogues

59

[the Syrian sites of Babutta, Burdj Bakirha, Benabil, Ruwea, ‘Atil and Dmer, respectively]). Taken together, the various limestone architectural elements enable us to offer a fairly certain reconstruction of the facade of Synagogue I (Fig. 10). It was an impressive and elegant facade, but relatively restrained in comparison with the facades of the nearby synagogues at Capernaum (Kohl and Watzinger 1916: Tafel III) and Korazim (Yeivin 2000: 57, Plan 10).

Fig. 10. Suggested reconstruction of the facade of Synagogue I (drawing by M. Edelcopp).

In summary, the architectural decoration of Synagogue I at Kh. Wadi amam displays many similarities with other “Galilean”-type synagogues, such as those at Bar‘am, Gush alav, Arbel, Meiron and especially those at . ‘Ammudim and Sumaqa. The comparative stylistic analysis of the architectural elements points to a date between the late second and the mid-third centuries CE for the construction of Synagogue I. This dating relies mainly on the pseudo-Tuscan columns and pilasters, the style of the theater-like benches, the modillion cornice with the “Syrian sequence”, and the carved-eagles lintel of the main entrance. It should be emphasized that this dating is independent from that reached by the stratigraphic analysis and the study of the associated small finds. Nevertheless, similar chronological conclusions were reached. Synagogue I came to an end in the late third century CE, when its eastern side collapsed. Consequently, Structure A11, at the foot of the synagogue’s eastern

60

Uzi Leibner

wall, was buried beneath a huge pile of debris containing building stones, architectural elements and roof tiles. A hoard of 37 coins, mainly silver tetradrachmas and dinarii, was probably hidden in the eastern synagogue wall and was discovered in the debris. The latest coins in the hoard are of Gallienus and provide a terminus post quem for the collapse of not before—and apparently not long after—264–265 CE. A late third-century date for the collapse is also supported by the assemblages of finds buried beneath the debris. The synagogue was reconstructed shortly thereafter (Synagogue II). The new phase is assigned to ca. 300 CE, mainly on the basis of stratified finds from the foundation trenches of the newly constructed walls, and those sealed beneath the mosaic floor. The restoration was based on the surviving remains, with the same dimensions and general plan, although probably without second-story galleries. The eastern half of the building frame, including the entire eastern wall (W2A15a) and the eastern halves of the northern and southern walls (W1A01, W2A02/W2A04), were reconstructed. The technique, building materials and width of these new walls clearly differ from those of the western half of the synagogue that survived from Phase I (see above). They are built of a combination of basalt blocks arranged in headers on the outside, and light gray concrete mixed with small fieldstones cast against a wooden mold on the inside. This casting technique (known by the Arabic term debesh) is typical of the Late Roman period and onward. The relationship between the two halves is clearly illustrated by a seam in the middle of the northern wall, where the later eastern half overlaps the stump of the earlier western half. The prominent feature of Synagogue II is the splendid mosaic floor (Miller and Leibner 2018). However, in this phase the synagogue lost its monumental architectural splendor, and displays a clear tendency to a more introverted style and utilitarian building technique, making secondary use of building materials and architectural elements. A whole array of new basalt benches, rather crudely carved, lined the western, northern and eastern walls. As the benches abut the two distinct halves of the building, they were evidently built during the reconstruction of the later eastern half. Two limestone pseudo-Tuscan capitals of Phase I are embedded in these benches, one in the western bench, and the other in the eastern bench.  The columniation consisted of nine columns, the pedestals or bases of seven still embedded in the floor. Four of these are finely carved, limestone pedestals that survived in situ from Phase I (Nos. 4, 5, 8, 9), while the other three are simple square blocks, or of poor workmanship (Nos. 2, 3, 6). The columns consist of a mixture of fine limestone shafts and column drums from Phase I, together with cruder basalt ones that apparently replaced those that could not be reused. Similarly, capitals of both limestone and basalt were used together in this phase. A large, north–south drainage channel was also added beneath the center of the building in this phase, perhaps an attempt to relieve the enormous pressure on the eastern terrace wall that caused the collapse of Phase I.

The Dating of the “Galilean”-Type Synagogues

61

Fig. 11. Plan of the Synagogue II mosaic, with preserved fragments indicated in gray (by B. Arubas and M. Edelcopp).

The mosaic was laid on a ca. 10 cm-thick bedding of pebbles in mortar, and beneath it was a ca. 10 cm layer of smoothed, beaten earth. The mosaic abuts the southern wall and the benches along the northern wall, and its design and remains relate to the various column bases and pedestals mentioned above. However, while it is level with the bases of the inferior group, leaving them completely visible, it covers ca. 15 cm of the lower molded part of the well-dressed pedestals, confirming that this was not the floor on which they originally stood. Thus, the stratigraphic relationships attest that the mosaic floor was laid together with the

62

Uzi Leibner

inferior group of columns and the construction of the basalt benches, and all these components clearly belong to the second phase of the synagogue. On the basis of the finds in the drainage channel, in the mosaic foundations and in the foundation trenches of the new walls, the construction of Synagogue II should be dated to the end of the third–beginning of the fourth centuries CE. The mosaic was already severely damaged in antiquity; however, the thirty-orso fragments that have survived enable us to reconstruct the floor’s decorative pattern (Fig. 11). The identification and interpretation of parts of the mosaic were recently aided by the current excavation of the synagogue at uqoq, where a mosaic with close similarities to ours is being exposed (Magness 2013; Britt 2014; Britt and Boustan 2017). The nave of the Kh. Wadi amam synagogue was apparently covered with a large mosaic carpet, perhaps divided into three panels, and was surrounded by a wide band containing inscriptions and geometric and floral decorations. Faded remains of two concentric circles, divided into voussoirshaped panels, have survived in its center, and a few Hebrew letters were discerned near the outer circle. This pattern, together with a parallel recently revealed at uqoq, seems to indicate that this was a zodiac, although little else can be said. Segments of a few figurative scenes have survived in the eastern and western aisles, facing the nave. Additional fragments of bands and border frames were uncovered in all three aisles. Altogether, the extant fragments indicate that the mosaic in the aisles was divided into twelve panels, each occupying the space between a pair of columns. Panel 4 contains a construction scene with various craftsmen shown working on a monumental building. Based on a similar panel at uqoq, which is much better preserved, this scene should be interpreted as the construction of the Tower of Babel (Fig. 12). Panel 11 depicts a battle between a giant and a group of soldiers. The interpretation of this panel as Samson smiting the Philistines with the jaw of a donkey (Judges 15), is based on Byzantine illuminated bibles that contain close parallels with a similar composition, but post-date our mosaic by at least five hundred years.10 Interestingly, two mosaic panels at the uqoq synagogue depict other stories of the Samson cycle (Britt, 2014: 348–349). The popularity of this marginal biblical figure in synagogue art in the region is indeed intriguing, and is viewed by scholars as reflecting messianic expectations (Grey 2013). Panel 12 depicts a chariot that is tilted on an angle as if it is about to flip over, pulled by three horses that are kneeling in an unnatural position. Behind the chariot is a huge fish swimming above an armed soldier who lies supine on the sea floor. This scene clearly presents Pharaoh’s army drowning in the Red Sea, and a depiction of this narrative in one of the uqoq synagogue panels reinforces this identification.  10 For example, a ninth-century copy of the Homilies of Gregory of Nazianzus (Paris gr. 510; Der Nersessian 1962: Fig. 14); and an eleventh-century Octateuch (Vat. 747; Weitzmann and Bernabò 1999: Fig. 1514).

The Dating of the “Galilean”-Type Synagogues

63

Fig. 12. Panel 4: The Tower of Babel (photograph by G. Laron).

Thus, it appears that the mosaic floor plan in the aisles contained a chain of twelve panels depicting biblical scenes. This kind of mosaic layout, with depictions of various biblical themes, was until recently unknown from ancient synagogues of Palestine. The exposure of such mosaics in the neighboring synagogues of Kh. Wadi amam and uqoq, and the reinterpretation of the scant mosaic remains in the synagogue of Meroth (Talgam 1987) and perhaps also of

. ‘Ammudim (Levine 1982) as sharing this layout, seem to point to a group of synagogue mosaics with a compositional scheme that is different from that known at other synagogues, perhaps due to chronological or geographical factors (Weiss 2016). Poorly executed renovations of Synagogue II, designated Sub-Phase IIb, were carried out in the late fourth century and should likely be attributed to damage caused by the severe earthquake of 363 CE. The repairs include the replacement of large segments of the mosaic with a simple plaster floor, and the addition of a low, flimsy bench and a stone bema against the southern wall, directly on top of the mosaic floor. The debris inside the synagogue and the finds sealed beneath it provided important information concerning the last stage of the building and its final collapse, which was apparently caused by another earthquake. The latest finds on the floor beneath the debris date to the fourth–early fifth centuries. They were probably washed in from above after the abandonment of the building, but

64

Uzi Leibner

before its collapse, and sealed inside shortly thereafter. Thus, the building probably collapsed in an earthquake in the early years of the fifth century (the earthquake of 419 CE?; see Amiran, Arieh, and Turcotte 1994: 226). Within the village, the latest finds in most of the Stratum II structures included coins from the Constantinian dynasty, glass vessels, and local pottery and lamps characteristic of mid-fourth-century assemblages. These, and the nature of the collapse, seem to attest to the earthquake of 363 CE as the cause of their abandonment. Evidence of later activity was documented only in a restricted area in the center of the site: in the synagogue itself (Phase IIb) and in an oil press adjacent to the synagogue on the north. The recovered materials here included Early Byzantine pottery, as well as minimi (tiny bronze coins) dating to the late fourth century CE. No later stratified remains were found. These data indicate that occupation of the site ceased at the end of the fourth or beginning of the fifth century CE.

Conclusions The chronological data revealed in the excavation of the Kh. Wadi amam synagogue are of special significance in light of the above-mentioned debate, as both phases of the structure clearly date to the Roman period, before the rise of imperial Christianity.11 Both the stratigraphic dating method based on the study of the development of the structure and the small finds from sealed contexts, and the art-historical method of studying the architectural decoration in comparison with other buildings in the region, reached, independently, the same date for Synagogue I around the first half of the third century. As noted, the long and heated scholarly debate over the date of the “Galilean”-type synagogues was based mainly on an alleged discrepancy between the results achieved by these two dating methods (excluding the complicated case of Capernaum with its massive use of spolia in the Byzantine period-synagogue, see Ma‘oz 1999). Thus, the results of our research seem to have brought the study of these synagogues to full circle, returning us to the chronological and stylistic conclusions reached by Kohl and Watzinger over a century ago, and disproving the claim by some current scholars that this building tradition emerged only in the Byzantine period. The abandonment of the village and synagogue at Kh. Wadi amam around the late fourth–early fifth centuries CE, and the absence of later Byzantine occupation at the site, facilitated the task of dating the structure and provided an  11 It should be noted that recent discoveries at Shikhin and Kh. Qana seem to indicate that the synagogues there also date to the Roman period, perhaps even predating that of Kh. Wadi amam (Strange 2015: 103–105; McCollough 2015: 141–142, respectively). It is not clear, however, if these structures belong architecturally to the “Galilean”-type group, and we must await the completion of their excavation and publication before we can place them in the larger framework.

The Dating of the “Galilean”-Type Synagogues

65

important test case, relatively free of factors that could have complicated the picture. Indeed, it seems that the major disputes concerning the date of other “Galilean”-type synagogues, such as those of Gush alav and Nabratein, were due to the continuous occupation of those sites deep into the Byzantine period, and interpretative debates concerning renovation phases and Byzantine-period finds in structures that were thought to have been built centuries earlier. In this respect, it should be emphasized that the Kh. Wadi amam case demonstrates that even a relatively short-lived synagogue, which stood for about a century and a half, underwent two considerable renovations during its existence.

Figures Fig. 1. Map of sites where “Galilean”-type synagogues (or components of such synagogues) have been identified (drawing by R. Sabar). Fig. 2. Khirbet Wadi "amam and Mount Nitai on a 5 m contoured topographical map. The synagogue is located in Area A (drawing by B. Arubas). Fig. 3. Block plan of Stratum 3 remains (in black) on the terrace and its surroundings (Stratum 2 walls in light gray). Fig. 4. The side room of the Early Roman “public building” surrounded by low benches (photograph by G. Laron). Fig. 5. Aerial view of the synagogue area after excavation (photograph by I. Arbel). Fig. 6. Representative sample of Middle Roman (Nos. 1–2), Early Roman (Nos. 3–9) and Late Hellenistic (Nos. 10–12) pottery from Sounding A4 beneath the synagogue floor in the southeastern corner of the hall. Fig. 7. A pseudo-Tuscan capital embedded in the foundation of the eastern benches of Synagogue II (photograph by G. Laron, drawing by O. Reicher). Fig. 8. A reconstruction of the entire lintel and photograph of the lintel fragment (drawing by N. Arad and B. Arubas, photograph by G. Laron). Fig. 9. Fragments of a modillion cornice found in front of the synagogue facade (photograph by G. Laron, drawing by M. Edelcopp). Fig. 10. Suggested reconstruction of the facade of Synagogue I (drawing by M. Edelcopp). Fig. 11. Plan of the Synagogue II mosaic, with preserved fragments indicated in gray (by B. Arubas and M. Edelcopp). Fig. 12. Panel 4: The Tower of Babel (photograph by G. Laron).

Bibliography Amir, R. 2007. “Style as a Means of Periodization: The Relative Chronology of Synagogues in the Golan”. Cathedra 124: 29–50 (in Hebrew). Amiran, D. H. K., Arieh, E., and Turcotte, T. 1994. “Earthquakes in Israel and Adjacent Areas: Macroseismic Observations since 100 B.C.E.” Israel Exploration Journal 44: 260– 305. Arubas, B. and Talgam, R. 2014. “Jews, Christians and ‘Minim’: Who Really Built and Used the Synagogue at Capernaum – A Stirring Appraisal”. In: Bottini, G. C., Chrupcała, L.

66

Uzi Leibner

D., and Patrich J. (eds.), Knowledge and Wisdom: Archaeological and Historical Essays in Honor of L. Di Segni. SBFCMa 54. Milan: Edizioni Terra Santa, 237–274. Atrash, W. 2012. “The Roman Theater at Tiberias”. Qadmoniot 144: 79–83 (in Hebrew). Atrash, W. 2015. “Architectural Analysis and Proposed Reconstruction”. In: Mazor, G. and idem: Bet She’an III, Nysa-Scythopolis: The Southern and Severan Theaters, Part 2: The Architecture. IAA Report 58/2. Jerusalem: Israel Antiquities Authority, 273–348. Avigad, N. 1976. Beth She‘arim: Report on the Excavations during 1953–1958, Vol. III: Catacombs 12–23. Jerusalem: Rutgers University Press. Avi-Yonah, M. 1946. “Newly Discovered Latin and Greek Inscriptions”. Quarterly of the Department of Antiquities of Palestine 12: 84–102. Avshalom-Gorni, D. and Najar, A. 2013. “Migdal”. Hadashot Arkheologiyot 125. http://www.hadashot-esi.org.il/Report_Detail_Eng.aspx?id=2304&mag_id=120 (accessed 27 February 2020). Bloedhorn, H. 1989. “The Capitals of the Synagogue of Capernaum: Their Chronological and Stylistic Classification with Regard to the Development of Capitals in the Decapolis and Palestine.” In: Hachlili, R. (ed.), Ancient Synagogues in Israel, Third–Seventh Century C.E. BARIS 499. Oxford, BAR, 49–54. Braslavsky, J. 1925. “Remains”. Journal of the Jewish Palestine Exploration Society 1, b: 139– 142 (in Hebrew). Britt, K. 2014. “The Mosaic Panels”. In: Magness, J., Kisilevitz, S., Britt, K., Grey, M. J., and Spigel, C.: “"uqoq (Lower Galilee) and Its Synagogue Mosaics: Preliminary Report on the Excavations of 2011–2013”. JRA 27: 348–355. Britt, K. and Boustan, R. 2017. The Elephant Mosaic Panel in the Synagogue at Huqoq: Official Publication and Initial Interpretations. JRA Supplementary Series 106. Portsmouth, RI.: Journal of Roman Archaeology. Brünnow, R. E. and von Domaszewski, A. 1909. Die Provincia Arabia: Auf Grund zweier in den Jahren 1897 und 1898 unternommenen Reisen und der Berichte früherer Reisender. Strasbourg: K. J. Truebner. Butler, H. C. 1903. Architecture and Other Arts: Architecture, Sculpture, Mosaic, and WallPainting in Northern Central Syria and the Djebel Hauran. New York : The Century Co. Collart, P. and Vicari, J. 1969. Le sanctuaire de Baalshamin à Palmyre, Vol. II. Rome: Institut suisse de Rome. Conder, C. R. and Kitchener, H. H. 1881. The Survey of Western Palestine, Memoirs of the Topography, Orography, Hydrography and Archaeology, Vol. I, Sheets I–VI, Galilee. London: Palestine Exploration Fund. Corbo, V. C. 1972. Cafarnao. Jerusalem: Franciscan Printing Press. Corbo, V. C. 1975. Cafarnao: Gli edifici della città. Jerusalem: Franciscan Printing Press. Crowfoot, J. W., Kenyon, K. M., and Sukenik, E. L. 1942. The Buildings at Samaria. London: Palestine Exploration Fund. Dar, S. 1999. Sumaqa: A Roman and Byzantine Jewish Village on Mount Carmel, Israel. BARIS 815. Oxford: BAR. De Luca, S. 2013. “Capernaum”. In: Master, D. E. (ed.), The Oxford Encyclopedia of the Bible and Archaeology, Vol. 1. New York: Oxford University Press, 168–180. Der Nersessian, S. 1962. “The Illustrations of the Homilies of Gregory of Nazianzus: Paris gr. 510: A Study of the Connection between Text and Images”. DOP 16: 195–228. Dothan, M. 1983. ammath Tiberias, I: Early Synagogues and the Hellenistic and Roman Remains. Jerusalem: Israel Exploration Society. Dothan, M. 2000. ammath Tiberias, II: The Late Synagogue (completed and edited by B. L. Johnson). Jerusalem: Israel Exploration Society.

The Dating of the “Galilean”-Type Synagogues

67

Fischer, M. L. 1990. Das korinthische Kapitell im Alten Israel in der hellenistischen und römischen Periode: Studien zur Geschichte der Baudekoration im Nahen Osten. Mainz: Philipp von Zabern. Foerster, G. 1971. “Notes on Recent Excavations at Capernaum (Review Article)”. IEJ 21: 207–211. Foerster, G. 1972. “Galilean Synagogues and Their Relation to Hellenistic and Roman Art and Architecture”, PhD thesis, The Hebrew University of Jerusalem (in Hebrew).  Foerster, G. 1983. “The Synagogues of the Galilee”. In: Shmueli, A., Sofer A., and Kliot, N. (eds.), The Lands of the Galilee. Haifa: Haifa University and Ministry of Defense, 231– 256 (in Hebrew). Grey, M. J. 2013. “‘The Redeemer to Arise from the House of Dan’: Samson, Apocalypticism, and Messianic Hopes in Late Antique Galilee”. JSJ 44: 553–589. Hill, S. 1975. “The ‘Praetorium’ at Musmiye”. DOP 29: 347–349. Ilan, Z. 1991. Ancient Synagogues in Israel. Tel Aviv: Ministry of Defense (in Hebrew). Ilan, Z. and Damati, E. 1987. Meroth: The Ancient Jewish Village. Tel Aviv: Society for the Protection of Nature in Israel (in Hebrew). Kohl, H. and Watzinger, C. 1916. Antike Synagogen in Galilaea. Leipzig: Heinrichs. Kraeling, C. H. (ed.) 1938. Gerasa, City of the Decapolis: An Account Embodying the Record of a Joint Excavation Conducted by Yale University and the British School of Archaeology in Jerusalem (1928–1930), and Yale University and the American Schools of Oriental Research (1930–1931, 1933–1934). New Haven, CT: American Schools of Oriental Research. Krencker, D. and Zschietzschmann, W. 1938. Römische Tempel in Syrien: Nach Aufnahmen und Untersuchungen von Mitgliedern der deutschen Baalbekexpedition 1901–1904, Text und Tafelband. Berlin: De Gruyter. Leibner, U. 2009. Settlement and History in Hellenistic, Roman, and Byzantine Galilee: An Archaeological Survey of the Eastern Galilee. TSAJ 127. Tübingen, Mohr Siebeck. Leibner, U. 2018. Khirbet Wadi amam: A Roman-Period Village and Synagogue in the Lower Galilee. Qedem Reports 13. Jerusalem: The Institute of Archaeology, Hebrew University. Leibner, U. and Bijovsky, G. 2013. “Two Hoards from Khirbat Wadi "amam and the Scope of the Bar Kokhba Revolt”. Israel Numismatic Research 8, 109–134. Levine, L. I. 1982. “Excavations at the Synagogue of Horvat ‘Ammudim”. IEJ 32: 1–12. Levine, L. I. 2005. The Ancient Synagogue: The First Thousand Years. 2nd rev. ed. New Haven, CT: Yale University Press. Loffreda, S. 1972. “The Synagogue of Capernaum: Archaeological Evidence for its Late Chronology”. LASBF 22: 5–29. Loffreda, S. 1973. “The Late Chronology of the Synagogue of Capernaum”. IEJ 23: 37–42. Loffreda, S. 1979. “Potsherds from a Sealed Level of the Synagogue at Capernaum”. LASBF 29: 215–220. Loffreda, S. 1993. “Capernaum”. NEAEHL 1: 291–295. Lyttelton, M. 1974. Baroque Architecture in Classical Antiquity. London: Thames and Hudson. Magen, Y., Tzionit, Y., and Sirkis, O. 2004. “Khirbet Badd ‘Isa—Qiryat Sefer”. In: Magen, Y. et al., The Land of Benjamin. Judea and Samaria Publications 3. Jerusalem: Israel Antiquities Authority, 179–241. Magness, J. 1997. “Synagogue Typology and Earthquake Chronology at Khirbet Shema‘, Israel”. JFA 24/1: 211–220. Magness, J. 2001a. “The Question of the Synagogue: The Problem of Typology”. In: AveryPeck, A. J. and Neusner, J. (eds.), Judaism in Late Antiquity, Part III: Where We Stand:

68

Uzi Leibner

Issues and Debates in Ancient Judaism, Vol. 4: The Special Problem of the Synagogue. Leiden: Brill, 1–48. Magness, J. 2001b. “A Response to E. M. Meyers and J. F. Strange”. In: Avery-Peck, A. J. and Neusner, J. (eds.), Judaism in Late Antiquity, Part III: Where We Stand: Issues and Debates in Ancient Judaism, Vol. 4: The Special Problem of the Synagogue. Leiden: Brill, 79–91. Magness, J. 2007. “Did Galilee Decline in the Fifth Century? The Synagogue at Chorazin Reconsidered”. In: Zangenberg, J., Attridge, H. W. and Martin, D. B. (eds.), Religion, Ethnicity, and Identity in Ancient Galilee: A Region in Transition. WUNT 210. Tübingen: Mohr Siebeck, 259–274. Magness, J. 2009. “Did Galilee Experience a Settlement Crisis in the Mid-Fourth Century?” In: Levine, L. I. and Schwartz, D. R. (eds.), Jewish Identities in Antiquity: Studies in Memory of M. Stern. TSAJ 130. Tübingen: Mohr Siebeck, 296–313. Magness, J. 2010. “The Ancient Synagogue at Nabratein”. BASOR 358: 61–68. Magness, J. 2013. “Samson in the Synagogue”. BAR 39: 32–39, 66–67. Maisler, B. 1941. “The Fourth Campaign at Beth Shearim, 1940 (Preliminary Report)”. BJPES 9: 5–20 (in Hebrew). Ma‘oz, Z. U. 1999. “The Synagogue at Capernaum: A Radical Solution”. In: Humphrey, J. H. (ed.), The Roman and Byzantine Near East: Some Recent Archaeological Research, Vol. 2. JRA Supplementary Series 31. Portsmouth, RI: Journal of Roman Archaeology, 137– 148. Ma‘oz, Z. U. 2017. On Ancient Synagogues in Galilee and Golan. Qazrin, Archaostyle. Mazor, G. 2015. “The Architectural Elements”. In: idem and Atrash, W., Bet She’an III, Nysa-Scythopolis: The Southern and Severan Theaters, Part 2: The Architecture. IAA Report 58/2. Jerusalem: Israel Antiquities Authority, 371–613. McCollough, T. C. 2015. “Khirbet Qana”. In: Fiensy, D. A. and Strange, J. R. (eds.), Galilee in the Late Second Temple and Mishnaic Periods, Vol. 2: The Archaeological Records from Cities, Towns and Villages. Minneapolis, MN: Fortress Press, 127–145. Meyers, E. M., Kraabel, A. T., and Strange, J. F. 1976. Ancient Synagogue Excavations at Khirbet Shema‘, Upper Galilee, Israel 1970–1972. Durham, NC: Duke University Press. Meyers, E. M. and Meyers, C. L. 2009. Excavations at Ancient Nabratein: Synagogue and Environs. Winona Lake, IL: Eisenbrauns. Meyers, E. M., Meyers, C. L., and Strange, J. F. 1990. Excavations at the Ancient Synagogue of Gush alav. Meiron Excavation Project 5. Winona Lake, IL: Eisenbrauns. Meyers, E. M., Strange, J. F. and Meyers, C. L. 1981. Excavations at Ancient Meiron, Upper Galilee, Israel 1971–72, 1974–75, 1977. Cambridge, MA: American Schools of Oriental Research.  Miller, S. and Leibner, U. 2018. “The Synagogue Mosaic”. In: Leibner, U., Khirbet Wadi amam: A Roman-period Village and Synagogue in the Lower Galilee. Qedem Reports 13. Jerusalem: Institute of Archaeology, Hebrew University, 144–186. Onn, A. and Weksler-Bdolah, S. 2008. “Khirbet Umm el-‘Umdan”. NEAEHL 5 [supplementary vol.]: 2061–2063. Porath, Y. 2013. Caesarea Maritima, Vol. I: Herod’s Circus and Related Buildings. Part 1: Architecture and Stratigraphy, IAA Report 53. Jerusalem: Israel Antiquities Authority. Rough, R. H. 1989. “A New Look at the Corinthian Capitals at Capernaum”. LASBF 9: 119– 128. Rozenberg, S. 2018. “Wall Painting, Stucco and Opus Sectile Fragments”. In: Leibner, U., Khirbet Wadi amam: A Roman-period Village and Synagogue in the Lower Galilee. Qedem Reports 13. Jerusalem, Institute of Archaeology, Hebrew University, 593–604. Schwartz, S. 2001. Imperialism and Jewish Society, 200 B.C.E. to 640 C.E. Princeton: Princeton University Press.

The Dating of the “Galilean”-Type Synagogues

69

Strange, J. R. 2015. “Kefar Shikhin”. In: Fiensy, D. A. and Strange, J. R. (eds.), Galilee in the Late Second Temple and Mishnaic Periods, Vol. 2: The Archaeological Records from Cities, Towns and Villages. Minneapolis, MN: Fortress Press, 88–108. Syon, D. and Yavor, Z. 2010. Gamla II: The Architecture. IAA Report 44. Jerusalem: Israel Antiquities Authority. Talgam, R. 1987. “Remarks on the Mosaics of the Synagogue and Bet-Midrash”. In: Ilan, Z. and Damati, E. (eds.), Meroth: The Ancient Jewish Village, Tel Aviv: Society for the Protection of Nature in Israel, 149–153 (in Hebrew). Tsafrir, Y. 1995. “The Synagogues of Capernaum and Meroth and the Dating of the Galilean Synagogue”. In: Humphrey, J. H. (ed.), The Roman and Byzantine Near East: Some Recent Archaeological Research. JRA Supplementary Series 14. Portsmouth, RI: Journal of Roman Archaeology, 1–21. Turnheim, Y. 1996. “Formation and Transformation of the Entablature in Northern Eretz Israel and the ‘Golan’ in the Roman and Byzantine Periods”, ZDPV 112: 122–138. Turnheim, Y. 1999. “The Design and the Architectural Ornaments of the Synagogue at "orvat Sumaqa”. In: Dar, S., Sumaqa: A Roman and Byzantine Jewish Village on Mount Carmel, Israel. BARIS 815. Oxford: BAR, 233–262. Watzinger, C. and Wulzinger, K. 1921. Damaskus: Die antike Stadt. Berlin: De Gruyter. Weber, T. M. 2002. Gadara Decapolitana: Untersuchungen zur Topographie, Geschichte, Architektur und der Bildenden Kunst einer “Polis Hellenis” in Ostjordanland. Gadara – Umm Qēs 1. ADPV 30. Wiesbaden: Harrassowitz. Weigand, E. 1914. “Baalbek und Rom. Die römische Reichskunst in ihrer Entwicklung und Differenzierung”. JdI 29: 35–91. Weiss, Z. 2016. “Decorating the Sacred Realm: Biblical Depictions in Synagogues and Churches of Ancient Palestine”. In: Leibner, U. and Hezser, C. (eds.), Jewish Art in Its Late Antique Context. TSAJ 163. Tübingen: Mohr Siebeck, 121–137. Weitzmann, K. and Bernabò, M. 1999. The Byzantine Octateuchs. Princeton: Princeton University Press. Yeivin, Z. 2000. The Synagogue at Korazim: The 1962–1964, 1980–1987 Excavations. IAA Report 10. Jerusalem: Israel Antiquities Authority (in Hebrew).

The Discovery of a Roman Period Synagogue in the Golan at Majduliyya

Mechael Osband and Benjamin Arubas

Roman-Byzantine Synagogues in the Golan Archaeological evidence for approximately thirty ancient synagogues has been discovered in the Golan (Meir and Meir 2015; The Bornblum Eretz Israel Synagogues Website, Ancient Synagogues Interactive Map, 2020). The first evidence for synagogues was documented in the late 19th century (Oliphant 1885: 82–93; 1886: 73–81; Schumacher 1888).1 Limited excavations by Kohl and Watzinger took place in two of these synagogues, at Dikkeh and Umm el-Qanaatir (Kohl and Watzinger 1916: 112–134). Additional synagogues have been discovered in surveys since 1967. Six of these synagogues—En Nashut, Dabiyye, Qarin, Kanaf, Deir Aziz, and Umm el Qanaatir—were excavated and dated to the Byzantine period (Ma‘oz 1993: 538–539; Ma‘oz and Ben David 2008; Dray et al. 2017). At sites that were not excavated, architectural elements determined by scholars to have been belonging to synagogues were dated by their style primarily to the Byzantine period (Ma‘oz 1993: 538–545; 1995; 2007).2 Ben David, based mainly on his survey results from village sites in the western central Golan, notes that monumental structures like synagogues and churches are found almost exclusively at sites that include a Byzantine period occupation. In contrast, these monumental buildings are noticeably absent at Roman period sites that did not continue into the Byzantine period (Ben David 2007). Only the synagogue at Gamla that was destroyed along with the site in 67 CE during the First Jewish Revolt has been dated to the Early Roman period (Yavor 2010: 60–61). No synagogues were attributed to the second to third centuries CE.3

 1

For additional descriptions of synagogues in the Golan before 1967 see Meir and Meir 2015: 21–22. Amir suggests the possibility of an earlier date for some of these synagogues (Amir 2011). 3 Urman critiques the conclusions presented by Ma‘oz and suggests that the Jewish presence in the Golan continued in the Roman period after the First Jewish Revolt (Urman 1995: 381–385; 607–617). Archaeological evidence for settlement continuity was found in Ben David’s survey of the western central Golan (Ben David 2005). 2

72

Mechael Osband and Benjamin Arubas

Fig. 1. Map with location of Majduliyya and sites mentioned in this article. Photo from Google Earth (accessed 17 September 2020).

The Site of Majduliyya The ancient village at Majduliyya4 (ca. 8.5 acres) is located in the Golan (Fig. 1), south of the Samekh Stream (Israel Grid: 2720/7522) on the border of two administrative districts in antiquity, the Hippos district to the south and the Gaulanitis district to the north (Ben David 2010). The site was first mentioned by Schumacher who notes that it was only brought to his attention on aside and concluded that it does not present anything of interest (Schumacher 1888: 134). Epstein identified the site as Byzantine with no notable remains (Epstein and Gutman 1972: 281). The only notable finds mentioned in past surveys is an olive press that can still be seen at the western part of the site (Ben David 1998: 28).  

According to Schumacher (1888: 134, 289), elders of a nearby village told him the name of the site is el-Mejdulîyeh (‫)امل ل ّة‬.

A Roman Period Synagogue at Majduliyya

73

Ben Ephraim (2003: 46, 52, 57, 63, 67) dated the site mainly to the Roman and Byzantine periods based on pottery collected in survey. A survey at the site in 2010 by our research team and shovel test pits in 20125 along with a review of available previous survey material (in total over four hundred identifiable potsherds) found that the site was occupied mainly in the Roman period and that it was later reoccupied in the Islamic periods (mainly Abbasid and Mamluk), which point to a marked settlement gap in the Byzantine period.6 During the survey, building remains of a residential nature built in basalt masonry of hewn and field stones were found throughout the site. Remains of olive presses were discovered along the boundaries of the site. These finds are consistent with the finds from surveys at other Roman period rural sites in the Golan (Ben David 2005). In order to address research questions concerning rural Golan in the Roman period, it was important to select a site for excavation that dates to the Roman period and specifically did not continue into the Byzantine era, and with little modern disturbance, typical in size, and located in the area of interest (i.e., the central and southern Golan or the Gaulanitis and Sussita districts, respectively, as they were known in antiquity). Based on the survey evidence briefly described above, Majduliyya was found to be a suitable example of a typical Roman period settlement in this region that could help attain the overall goal of characterizing a Roman period village as well as relating to the different topics presented above.7

The Synagogue of Majduliyya Following repeated visits in different seasons a peculiar row of stones that differed from those common in the village was discerned in area full of natural over growth and thorns on the flattened eastern part of the site in a field of many basalt stones. These peculiar stones were finely dressed on their top and front. During the surveys, a few of the basalt stones in the field after being turned over were found to be the remains of broken columns, including a base, of fine workmanship. In addition, a few small fragments of roof tiles were also found in this area. All of these finds indicated the presence here of a building of a public nature. None of these finds were noted in the previous surveys.  5 The survey and shovel test pits were conducted under the Israel Antiquities Authority Licenses S213/2010 and G-4/201, respectively. 6 A few potsherds of common cooking ware may belong to the Late Byzantine period but could also belong to the Early Islamic period. To date, following the study of thousands of potsherds from excavations in different areas of the village along with additional surveys, no evidence for a Byzantine period settlement has been found. 7 It is important to note that the site was not selected because of possible synagogue remains but rather due to broader questions concerning rural Golan in the Roman period. The synagogue was only identified during excavation and besides the scant remains described here preceding the excavation was not recognizable as such from the survey.

74

Mechael Osband and Benjamin Arubas

Fig. 2. Aerial view of the synagogue, looking southeast (A. Wiegmann).

For the reasons stated above, the site was selected for excavation with the overall goal of characterizing a Roman period village in the Golan.8 We decided to begin the excavation in this area because of the few interesting remains that were found. The building at the time of this writing has only been partially excavated (Fig. 2). The structure including architectural element, was built exclusively from basalt. The row of well fitted stones (mentioned above) seen in the surface survey was found to be part of a stone bench. It is attached to a two-faced wall that formed the eastern exterior wall of a public building of rectangular shape, measuring 13 m x 23 m in a north-south direction. It is important to note that there are no entrances on the southern wall and no features marking a southern orientation. In general, the preservation is from one to three courses and in some places the outer walls were completely robbed. The stones of the external walls on the western and southern sides are made of high quality ashlars. The building had benches along all four walls, which were partially preserved. On the northern eastern and southern walls, the benches apparently ran continuously along the entire wall. On the western wall two entrances were found on the northern and southern ends, and based on symmetry we suggest that there was also a central  8

The site has been excavated under Israel Antiquities Authority Licenses G-70/2014, G-72/2015, G20/2016, G-69/2017, and G-53/2018. The first season was with the Archaeology Institute at Bar-Ilan University and since 2015 and continuing through today with the Zinman Institute of Archaeology, the University of Haifa. Funding for the excavation has been provided by the Shamir Research Institute and the Israel Science Foundation research grant 722/17 (“An Archaeological Study of the Roman Period Territorium of Hippos: Urban and Rural Settlement Relationships”), along with M. Eisenberg at the University of Haifa.

A Roman Period Synagogue at Majduliyya

75

entrance. However, this part of the wall is completely missing. Presumably the bench on the western side was built in segments between the entrances. The building contains a central lower floor in the middle and two upper floors on the eastern and western sides. The lower middle floor runs along the entire building and ends in double benches on the northern and southern sides (Fig. 3).

Fig. 3. Northeast corner of the synagogue with remains of benches, looking northeast (A. Wiegmann).

Fig. 4. Column fragments, base, and Doric capital discovered in surveys (M. Osband).

Mason marks were found on many of the benches on the southern side. The lower floor is separated from the raised aisles on the eastern and western sides by two parallel rows of benches, which included plinths that supported columns, and extended to the northern and southern outer benches. Column fragments were found in the excavation and also scattered outside the building in some cases in secondary use in a stone fence or in the village (Fig. 4). The capitals

76

Mechael Osband and Benjamin Arubas

discovered are in a Doric style (Fig. 4). Remains of a badly preserved colorful mosaic were discovered in different excavation areas throughout the building (Fig. 5). Roof tiles were found scattered throughout the excavation both inside and outside the building.

Fig. 5.

Southwest corner of the synagogue with mosaic remains, looking north (A. Wiegmann).

The building was clearly a public structure used for gathering. The probes under the floor show that the building dates to the Roman period, but did not provide conclusive dating within this period. Differences in building techniques between the northern and eastern walls from the southern and western walls suggests the existence of at least two phases. The best parallels for benched structures in this region are ancient synagogues of the Early Roman period (ca. late first century BCE to early second century CE). The Early Roman synagogues found at the northern sites of Gamla (Yavor 2010: 40–61) and Magdala (Avshalom-Gorni and Najar 2013), both of which ceased to exist in 67 CE, are of particular importance. The building in Majduliyya has similarities to these two structures. The capitals and overall masonry has strong similarities to Gamla (Peleg-Barkat 2010). The higher aisles and lower nave have similarities to Gamla and Magdala. At Magdala there is also a mosaic. While there are many similarities, there are some important differences. For example, the lower floor of the central hall extends the length of the entire struc-

A Roman Period Synagogue at Majduliyya

77

ture unlike Magdala where the outer benches and upper floor completely surround the lower central area. Auxiliary rooms of a later stage in Majduliyya show continued use of the original building into the third century. The scant remains of the colored mosaic design may suggest a later date than that of Magdala, most probably to the third century CE. In addition, many fragments of roof tiles were found at Majduliyya, and these are completely absent from the synagogues in Gamla and Magdala. Majduliyya, unlike Gamla and the synagogue at Magdala, continued beyond 67 CE until the late third century, as is evidenced by the pottery and coins found in the area of the synagogue and throughout the site. Some of the differences noted here are best explained by the continued use of the Majduliyya synagogue beyond 67 CE.

Conclusion The site of Majduliyya was chosen for excavation as a representative Roman period rural settlement in the Golan. Public monumental structures had not been previously noted at similar rural sites from this time. It was therefore surprising to find an ancient synagogue not identified in previous surveys. The remains from Majduliyya raise the possibility that additional synagogues may in fact be found at other Roman period sites but only intensive surveys accompanied with excavation will provide definitive answers.9

Postscript Since this manuscript was submitted, two more seasons have been conducted in the synagogue, and additional architectural and stratigraphic evidence have been revealed. For these newer seasons see our preliminary report, Osband and Arubas forthcoming.

Figures Fig. 1. Map with location of Majduliyya and sites mentioned in this article. Photo from Google Earth (accessed 17 September 2020). Fig. 2. Aerial view of the synagogue, looking southeast (A. Wiegmann). Fig. 3. Northeast corner of the synagogue with remains of benches, looking northeast (A. Wiegmann). Fig. 4. Column fragments and Doric capital discovered in surveys (M. Osband).

 9

The possibility of additional Roman period synagogues in this region are currently being studied by the authors along with C. Ben David.

78

Mechael Osband and Benjamin Arubas

Fig. 5. Southwest corner of the synagogue with mosaic remains, looking north (A. Wiegmann).

Bibliography Amir, R. 2011. “Style as a Chronological Indicator: On the Relative Dating of the Golan Synagogues”. In: Jews in Byzantium 14: 337–370. Avshalom-Gorni, D. and Najar, A. 2013. “Migdal”. Hadashot Arkheologiyot 125. http://www.hadashot-esi.org.il/Report_Detail_Eng.aspx?id=2304&mag_id=120 (accessed 27 February 2020). Ben David, C. 1998. “Oil Presses and Oil Production in the Golan in the Mishnaic and Talmudic Period.” ‘Atiqot 34: 1–62 (in Hebrew; English summary, 5*–6*). Ben David, C. 2005. The Jewish Settlement on the Golan in the Roman and Byzantine Period. Qazrin: Golan Studies Institute, Qazrin Museum (in Hebrew). Ben David, C. 2007. “A Ceramic Survey and the Dating of the Monumental Synagogues in the Golan”. Cathedra 124: 13–27 (in Hebrew). Ben David, C. 2010. “The Preservation of Roman and Byzantine Place Names from the Golan Heights”. Semitica et Classica 3: 265–271. Dray, Y., Gonen, I., and Ben David, C. 2017. “The Synagogue of Umm El Qanatir – Preliminary Report”. IEJ 67: 209–231. Ben Ephraim, Y. 2003. “The Boundary between the Provinces of Palaestina and Phoenicia in the Second and Third Centuries in the Central Golan Heights”. M.A. thesis, Bar Ilan University, Ramat Gan (in Hebrew). Epstein, C. and Gutman, S. 1972. “The Golan”. In: Kochavi, M. (ed.), Judaea, Samaria and the Golan: Archaeological Survey 1967–1968. Jerusalem: Carta, 244–298 (in Hebrew). Kohl, H. and Watzinger, C. 1916. Antike Synagogen in Galilaea. Leipzig: Heinrichs. Ma‘oz, Z. U. 1995. “Ancient Synagogues in the Golan: Art and Architecture”. Thesis for the Degree of Doctor of Philosophy, Golan Archeological Museum (in Hebrew). Ma‘oz, Z. U. 2007. “Chronological Elements in the Art of the Golan Synagogues”. Cathedra 124: 51–64 (in Hebrew). Meir, D. and Meir, E. 2015. Ancient Synagogues of The Golan. Jerusalem: Yad Izhak BenZvi. Oliphant, L. 1885. “Explorations North-East of Lake Tiberias, and in Jaulan”. PEQ 17(2): 82–93. Oliphant, L. 1886. “New Discoveries”. PEQ 18(2): 73–81. Osband, M. and Arubas, B. forthcoming. “The Excavation of a Roman-Period Village and Synagogue at Majduliyya”. IEJ. Peleg-Barkat, O. 2010. “Architectural Decoration”. In: Syon, D. and Yavor, Z. (eds.), Gamla II: The Shmarya Gutmann Excavations 1976–1989. The Architecture. IAA Reports 44. Jerusalem: Israel Antiquities Authority, 159–174. Schumacher, G. 1888. The Jaulân: Surveyed for the German Society for the Exploration of the Holy Land. London: Richard Bentley and Son. The Bornblum Eretz Israel Synagogues Website, Ancient Synagogues Interactive Map 2018. Available at: http://synagogues.kinneret.ac.il/excavated-synagogues/synagogues-interactive-map/ (accessed 14 November 2018). Urman, D. 1995. “Public Structures and Jewish Communities in the Golan Heights”. In: Urman, D. and Flesher, P. V. M. (eds), Ancient Synagogues: Historical Analysis and Archaeological Discovery. 2 vols. Leiden: Brill, [2.]373–617.

A Roman Period Synagogue at Majduliyya

79

Yavor, Z. 2010. “The Synagogue and Miqveh Complex”. In: Syon, D. and Yavor, Z. (eds.), Gamla II: The Shmarya Gutmann Excavations 1976–1989. The Architecture. IAA Reports 44. Jerusalem: Israel Antiquities Authority, 40–61.

The Synagogue in Delos Revisited

Monika Trümper In 1912 and 1913 André Plassart excavated a building on the eastern shore of the island of Delos that he identified as a synagogue, based on its plan and furnishings, as well as inscriptions found here and in the vicinity. Because of the integration of inscribed bases from the nearby gymnasium, Plassart suggested a construction date after 88 BCE. This discovery was ground-breaking because few ancient synagogues were known at this time, let alone from the first century BCE. Meanwhile, the knowledge of and state of research on early synagogues has changed significantly, as vividly demonstrated by this conference. While research on the synagogue building in Delos has profited from increased interest in this topic, the identification of this building has been repeatedly challenged since 1913 and remains debated until today. My own contribution to this debate has been an article published in 2004.1 The aim of this paper is to present the current state of research and offer suggestions for future research that may provide new evidence to advance the debate and possibly even solve the contested identification once and for all. After a brief introduction to the site and building, the research debate is analysed with a focus on contributions published since 2004, followed by concluding remarks for future perspectives.

Introduction Delos is a small island in the Cyclades, with a surface area of about 360 ha. The island had several harbours on the west and east coast, and its terrain is hilly, including a mountain (Kynthos) of 113 m height. Delos was particularly renowned for its Panhellenic sanctuary of Apollo, which developed from the seventh century BCE onwards. From the sixth century onwards, the sanctuary was repeatedly under control of Athens, until the city became independent in 314 BCE. The city reached its maximum extension after 167/6 BCE when the Romans  I would like to thank the conference organizers for inviting me to this very stimulating and interesting conference and for their hospitality. I am also very grateful to the other participants at the conference for comments and discussions. 1 Trümper 2004; cf. also Trümper 2011: 61–63.

82

Monika Trümper

declared it a free port and gave control back to the Athenians. While the island was sacked twice by troops of Mithridates and pirates in 88 and 69 BCE, respectively, its successive decline from the mid-first century BCE onwards is commonly explained with the emergence of rival ports in Italy and thus with a loss of economic significance. The settlement of the Roman Imperial and Byzantine periods is usually identified as small and insignificant in ancient and modern literature.

Research until 2004 Since 1873 the École française d’Athènes has excavated 24 ha and thus about a quarter of the Late Hellenistic city. Among the revealed remains is the synagogue building, which is denominated number 80 in the French guidebook of Delos, hereafter referred to as GD 80.2 The building occupies a lot of 28.30 m x 30.70 m and consists of three parts (Fig. 1):  two large rooms A and B that were found decorated with marble benches and a throne in A; a large lime kiln was installed in room A at an unknown point after the building’s abandonment;  area C that includes a marble stylobate running north-south, benches in the northwest corner, and an entrance in the south;  and the D-complex with several small rooms, one of which gave access to a large water reservoir. The original extension of the building in the east is unknown as the sea has risen for about 2.5 m since antiquity and destroyed and submerged ancient structures on the coast. Plassart excavated the building in just two weeks in 1912 and 1913, and immediately published two brief—albeit identical—reports in 1913 and 1914. The later report includes a schematic plan of the building.3 In 1962, Philippe Bruneau resumed the study of GD 80, cleaning out its water reservoir and providing the first excellent state plan of the building as well as several plans and sections of different parts of the building. Bruneau confirmed Plassart’s identification of GD 80 as a synagogue and its construction date. Furthermore, he emphasized that finds suggest use of the building until the late second or early third century CE.4 The chance find of two inscribed stelae 92.50 m to the north of GD 80 in  2 Bruneau and Ducat 2005: 254–255, no. 80. In the following, this standard handbook will be cited as GD followed by the number of the monument (e.g., GD 2 = Bruneau and Ducat 2005, monument no. 2). For reasons of language consistency monuments and quarters in Delos will be designated with English names. 3 Plassart 1913; 1914, pl. I preceding p. 523. 4 Bruneau 1970: 480–493, pls. B–H.

The Synagogue in Delos Revisited

83

1979, which include references to the Israelites in Delos (Samaritans), did not entail further fieldwork, but a prompt publication by Bruneau.5

Fig. 1.

Delos, Synagogue building GD 80: state plan; © EFA: Philippe Bruneau; cf. Bruneau 1970, pl. B.

Between 2000 and 2003, I examined GD 80 with the aim of reconstructing its history and function, and provided a detailed discussion and reconstruction of  5

Bruneau 1982.

84

Monika Trümper

the building history, with five different phases (Fig. 2).6 Since the third phase includes the reuse of inscribed bases from the nearby gymnasium, which was presumably abandoned after 88 BCE, phases 1 and 2 can be dated before 88 BCE, and phases 3–5 after 88 BCE. But when exactly the respective modifications were made before or after 88 BCE, could not be determined. The most distinctive trait of the building, the large undivided hall A/B, was maintained from phase 1 to phase 4, making possible or even suggesting continuity in function. Depending on what is considered conclusive for a potential use of GD 80 as a synagogue, the building would have been used as a synagogue either from the very beginning or only in phase 5: the benches and throne that were often recognized as conclusive for use as a synagogue can only be safely attributed to phase 5, but some inscriptions found here have been dated to the early first century BCE. More crucially, the overall design of phases 1–4 is not at all compatible with a private house or a meeting place of a pagan cultic association in the local context. Therefore, I also argued that GD 80 could have been a synagogue, right from its beginning and built well before 88 BCE.

Fig. 2. Synagogue building GD 80: reconstruction of phases 1, 3, 4, 5; Trümper 2004: 559, fig. 34; 563, fig. 37; 566, fig. 38; 568, fig. 39.

 6

Trümper 2004.

The Synagogue in Delos Revisited

85

However, the optimistic evaluation of GD 80 by three experts on Delian architecture and archaeology has been countered by critical or even pessimistic voices. Among early critics were Belle D. Mazur and Hershel Shanks, who in 1935 and 1979, respectively, claimed that GD 80 was never used as a synagogue.7 Others suggested compromises proposing that the building had not been built as a synagogue but only used as such in a later phase. For example, based on fieldwork in 1985, Michael White provided the first phase plan of the building and argued that it had been built as a private house in the second century BCE and was only later transformed into a synagogue, albeit already before 88 BCE.8 After a visit to the site, Donald Binder resumed the reconstruction proposed by Belle D. Mazur in 1935: the building would have been constructed as a cultic hall by a pagan association in the second BCE, including a large peristyle courtyard; after 88 BCE, Jews or rather Samaritans would have renovated and transformed the building into a synagogue.9 Despite significant attention and fieldwork of several scholars in the period between 1913 and 2004, the building GD 80 has not been comprehensively studied and published, including all of its finds. This picture has not changed since 2004 and may be the major reason why the debate about the function of GD 80 has continued since then.

Research since 2004 The following brief overview of literature shows how scholars have assessed the building GD 80 from 2004 until today. The first volume of the Inscriptiones Judaicae Orientis includes a detailed discussion of the epigraphic and archaeological evidence pertaining to Jewish presence in Delos.10 In assessing the archaeological evidence and the history of research, the authors pay particular attention to the various reconstructions of the building’s history proposed in scholarship until 1999. They provide the most exhaustive currently available discussion of the inscriptions found in and close to GD 80, which will frequently be referred to below. Although they are sceptical about the “Jewishness”11 of several of these inscriptions, they have chosen to include all of them in their corpus. They argue that the Samaritans certainly had their own building in Delos, at the findspot of their inscriptions. In contrast, they conclude that GD 80 currently cannot be safely identified as the “synagogue” of the Jewish community of Delos. GD 80 may have been “the building of a pagan  7

Mazur 1935: 15–24; Shanks 1979: 28–29. White 1987; 1996: 64–67; 1997: 332– 342. 9 Binder 1999: 299–316. 10 IJO I: 210–242; the plan shown on p. 213 is the state plan provided by Bruneau 1970, pl. B. The last entry in the bibliography on p. 211 is Binder 1999: 299–316. 11 For this term, see IJO I: 224. 8

86

Monika Trümper

cultic society under Jewish influence” or even an “association of Judaizers. Alternatively, the building might have housed a Jewish association that borrowed from or assimilated to pagan practice, as suggested by the form of the dedications to Theos Hypsistos.”12 In the second edition of his monumental study on the ancient synagogue from 2005, Lee Levine maintains that GD 80 served as a synagogue. While he refers to various theories regarding the original date and function of GD 80, he does not take sides but reprints the phase plan provided by Michael White in 1987. His assessment refers exclusively to the last stage of GD 80.13 In an article from 2005, which discusses the relationship between Jewish religious architecture in Palestine and the Diaspora, Inge Nielsen identifies GD 80 as a synagogue, used as such from the beginning. She recognizes local buildings of religious voluntary associations as models and argues that the Delian synagogue had a more inclusive function than its Palestinian equivalents: it would have served for social as well as religious activities, which would be reflected in its denomination as προσευχή/prayer house, as opposed to συναγωγή/assembly house.14 Stephen Catto in his monograph on the first-century synagogue from 2007 provides a good summary of the debate.15 He supports Binder’s theory that GD 80 was first constructed as a public building, and not as a private house. Following Binder, he argues that it can currently not be determined whether GD 80 was originally built as a temple or guild house by a pagan group, or as a προσευχή by Jews or Samaritans. The source book published by Anders Runesson, Donald Binder and Birger Olsson in 2008 includes a detailed entry for Delos, discussing both textual and archaeological sources. The authors argue that GD 80 was constructed as a public building before 88 BCE, either for a non-Jewish/non-Samaritan or a Jewish/ Samaritan association, and definitely used as a synagogue by either Samaritans or Jews after 88 BCE.16 In his book on ancient synagogue seating capacities from 2012, Chad Spigel occasionally mentions GD 80, referring to its potential use as a synagogue with a

 12

IJO I: 219. Levine 2005: 107–113; reprinting the phase plan provided by White 1987: 157, Fig. 2; 1996: 65, Fig. 10, which, according to Trümper 2004, is not correct. 14 Nielsen 2005: 84–85, 97. She reprints (2005: 78, Fig. 12; 79, Fig. 14; 87, Fig. 15) the plan with different wall systems and the reconstructions of various phases from Trümper 2004: 520, Fig. 3; 559, Fig. 34; 562, Fig. 36. 15 Catto 2007: 61–67. The book is based on his dissertation, submitted in 2005. He does not cite Trümper 2004. 16 Runnesson, Binder and Olsson 2008: 123–134, no. 3.1.1.4. They reprint (132, Fig. 17) the plan with different wall systems from Trümper 2004: 520, Fig. 3. 13

The Synagogue in Delos Revisited

87

question mark, but using the measures of the benches and the throne for his calculations.17 Lidia Matassa discusses GD 80 in two articles published in 2006 and 2007 as well as in a chapter of her PhD dissertation that examines the invention of the First Century Synagogue from an archaeological perspective and was submitted in 2010.18 She visited Delos in October 2003, but does not specify for how long she stayed and how intensively she could examine the remains of GD 80 and other Delian buildings, and whether she could study material in the storerooms.19 Reassessing the literary, epigraphic and archaeological evidence, Matassa aims at “unravelling the myth of the synagogue on Delos” and concludes that it is impossible to identify GD 80 or any other building on Delos as a synagogue based on the available evidence.20 Since she provides the most detailed discussion of GD 80 published after 2004, her arguments require more attention here.21 Following her line of arguments, first written sources with a focus on inscriptions and then the archaeological evidence will be examined. While Matassa’s arguments play an important role in the following, other recent literature will also be taken into account.

Epigraphic Evidence Eight inscribed monuments have been related with the building GD 80, of which five were found in the building itself, one in a nearby house to the northwest of GD 80, and two 92.50 m to the north of GD 80. Issues that have been contested for a long time and have been brought up again by Matassa include the translation and meaning of the word προσευχή, mentioned in two of the inscriptions;  17

Spigel 2012: 29, 39, 55, 69, tab. 3.3. He does not cite Trümper 2004 or Fraisse and Moretti 2007. He attributes the throne to the first century BCE without any further discussion. According to Fraisse and Moretti 2007: 73–74, this throne would have been made “quelques décennies” after its model in the theatre of Athens, which is dated to the period of Lykurgos. The theatre of Delos was built between the late fourth century BCE and the third quarter of the third century BCE; thus, the throne was made sometime in the third century BCE. It can only be safely ascribed to the fifth phase of GD 80, cf. Trümper 2004: 539 n. 59. 18 Matassa 2006; 2007; 2010: 51–107. Matassa 2006 is a briefer summary of what is provided in more detail in Matassa 2007. The texts in Matassa 2007 and 2010 are largely identical. While the dissertation is available online, in the following, the article from 2007 will be referred to, because it was published first and appeared in a peer-reviewed journal. – After completion of the present chapter, Matassa’s PhD thesis (2010) was posthumously published (Matassa 2018), with only very light editorial work in the chapter on Delos. 19 Matassa 2007: 97; 2010: iv, xi, 55 n. 166–167, 81. 20 Matassa 2007: esp. 112. 21 Matassa’s contributions are fervently argued and, at times, polemical. Unfortunately, her argumentation includes a significant number of mistakes, such as wrong citations in the bibliography, misquotes, wrong translations, insufficient credits given to earlier scholars, and factual mistakes regarding Delian archaeology. Some of these will be pointed out in the following, but this is not the place for an in-depth study of all of her arguments and errors.

88

Monika Trümper

the identity of Theos Hypsistos/Hypsistos, referred to in four inscriptions; and the format and function of the monuments that carry these inscriptions.22 The discussion begins with the two inscriptions that were found 92.50 m north of GD 80, but were indisputably issued by a Jewish group. They were carved on marble stelae decorated with wreaths and record that Israelites, who make contributions to Mt. Garizim, honoured two euergetai from Crete and possibly another region.23 Matassa acknowledges that these stelae are clearly linked to a community of Israelites, but is cautious regarding their significance for the presence of Jewish groups on Delos. She argues, “it is also possible that the dedications were made by Samaritan visitors and traders to the island on behalf of their religious communities at home,”24 and might “relate to a non-resident donor or group of Samaritans.”25 The names of the persons honoured would not be specifically identifiable as Jewish or Samaritan. The first is a stela of white marble (H. 0.70 m, W. 0.53–0.56 m, D. 0.06–0.09 m) that is badly preserved at the bottom, and the original upper part of which is entirely missing (Fig. 3).26 The stela was reused to record, over an earlier erased inscription, that the Israelites honoured Menippos from Herakleia for constructing and dedicating something from his own funds ἐπὶ προσευχῇ. Translations and interpretations of this inscription vary significantly: [Ὁι ἐν Δήλῳ] Ἰσραηλῖται οἱ ἀπαρχόμενοι εἰς ἱερòν ἅγιον Ἀργαριζεὶν ἐτίμησαν vac. Μένιππον Ἀρτεμιδώρου Ἡράκλειον αὐτὸν καὶ τοὺς ἐγγόνους αὐτοῦ κατασκευάσαντα καὶ ἀναθέντα ἐκ τῶν ἰδίων ἐπὶ προσευχῇ τοῦ θε[οῦ] ΤΟΝ [- - - - - - - - - - - - - - ] ΟΛΟΝΚΑΙΤΟ [- - - - - καὶ ἐστεφάνωσαν] χρυσῷ στε[φά-]

 22 Trümper 2004 was focused on the architecture of GD 80, but the monuments were briefly discussed, esp. 570–571 n. 122–123; 585 n. 170. – The literary evidence pertaining to Jews in Delos has frequently been discussed; since no new arguments based on material evidence from Delos can be brought forward here, it is not discussed in the following; see IJO I: 211; Catto 2007: 61; Runesson, Binder, and Olsson 2008: 123–125, no. 93. 23 The donors come from Knossos in Crete and Herakleion, which was a popular city name in antiquity; so far, cities in Crete and Caria have been proposed, cf. Collar 2013: 250. Matassa 2007: 109 criticizes Trümper 2004: 571 for obfuscating the argument by incorrect statements regarding the findspot of these two inscriptions and ID 2329, discussed further below. On the same page, Trümper 2004: 571, the find spots of all three inscriptions are, however, fully and correctly listed in the footnote 123. 24 Matassa 2007: 90. – For the inscriptions, Matassa refers to the Inscriptions de Délos (ID) and the Corpus Inscriptionum Iudaicarum (CIJ), but not to the most recent edition in the Inscriptiones Judaicae Orientis (IJO), the first volume of which includes the inscriptions from Delos and appeared in 2004. Reference to the extensive discussions in IJO would have most superseded most of her arguments and prevented her mistakes in dating and translations. 25 Matassa 2007: 92. 26 Bruneau 1982: 471–475, no. 2; IJO I Ach66. – The following abbreviations are used for bases: H.: height; W.: width; D.: depth; Dm. diameter.

The Synagogue in Delos Revisited

89

νῳ καì [- - - - - - - - - - - - - - - - - ] ΚΑ - Τ - -] [Les] Israélites [de Délos] qui versent contribution au sacré et saint Garizim ont honoré Ménippos, fils d’Artémidoros, d’Héraclée, lui-même ainsi que ses descendants, pour avoir établi et dédié à ses frais, en ex-voto [à Dieu], le [ ] et le [et l'ont couronné] d'une couronne d’or et [ ].27 The Israelites [on Delos?] who make offerings to the temple (on the) holy [or to sacred and holy] Garizim [Argarizin] honoured Menippus (son) of Artemidorus, from Heraclion, himself and his descendants, for constructing and dedicating from his own funds in a prayer [ = vow] of God...... and crowned with a golden wreath and ..... .28 [The] Israelites who make first-fruit offerings to holy Argarizein honour Menippos, son of Artemidoros of Heraclea, himself as well as his descendants, to have established and dedicated its expenses, for an offering/prayer [to God], [- - - - - - - ] and [- - - - -] and crowned it with a golden crown and [- - -].29 Israelites who send the temple tax to sacred, holy Argarizein honour Menippos, son of Artemidoros, from Herakleion, himself and his family, (him), who built and kept of his own means a synagogue to Go[d], who … whole and the … with a golden cr[o]wn and ??30 The Israelites [on Delos?] who make first-fruit offerings to the temple on holy Mt. Gerizim honour Menippus, son of Artemidorus, of Herakleion, both himself and his descendants, for constructing and dedicating from his own funds for the prayer hall of God the … and the … and crown him with a golden crown and….31 The Israelites on Delos who contribute towards the holy temple on (or: contribute towards sacred and holy) Gerizim honored Menippos son of Artemidoros from Herakleia, himself and his descendents, who furnished and dedicated from his own resources on account of a prayer (proseuchē) of God (or: in fulfillment of a vow to God; or: for the prayer-house of God) . . . (about two lines missing) and they crowned him with a gold crown and . . . (the rest of inscription missing).32 Israeliten, die Opfer(gaben) darbringen zum/auf den heiligen, geweihten Garizim (oder: zum Tempel auf den heiligen Garizim), ehren Menippos, Sohn des Artemidoros aus Heraklion, ihn und seine Nachkommen, der ausgestattet und gewidmet hat aus den eigenen Mitteln für eine Proseuche/zur Anbetung Gottes …… und haben bekränzt mit einem goldenen Kranz und … 33

 27

Bruneau 1982: 474. IJO I: 229, Ach66. 29 Matassa 2007: 91–92. The second part of the translation makes little sense. 30 Kartveit 2009: 217. 31 Runesson, Binder, and Olsson 2008: 129–130, no. 100. 32 Philip Harland, available at: http://philipharland.com/greco-roman-associations/222-honors-byisraelites-for-benefactors/ (accessed 31 July 2018). 33 Böhm 2018: 181. 28

90

Monika Trümper

Fig. 3. Samaritan stele no 2; © EFA: Philippe Bruneau; cf. Bruneau 1982: 470, fig. 4.

The interpretation of the wording ἐπὶ προσευχῇ is much debated, identified as reference to either a building34 or as an offering/prayer.35 The wording seems unique to Delos, because in other instances the definite article is added after ἐπὶ,  34 Kartveit 2009: 217; Runesson, Binder, and Olsson 2008: 129–130, no. 100; cf. also Nielsen 2005: 184; Catto 2007: 64. 35 Bruneau 1982: 474; IJO I Ach66; Matassa 2007: 91–92; cf. also Collar 2013: 250. – Böhm 2018: 181 and Philip Harland, available at: http://philipharland.com/greco-roman-associations/ 222-honors-byisraelites-for-benefactors/ (accessed 31 July 2018), list both options.

The Synagogue in Delos Revisited

91

thus ἐπὶ τῆς προσευχῆς.36 Since κατασκευάσαντα καὶ ἀναθέντα clearly refer to something that Menippos had constructed and dedicated for the Israelites, there must have been a monument or building specific to the Israelites. It seems most plausible that this was an assembly or prayer house, a proseuchē.37 Thus, I would translate as follows: (The) Israelites who make offerings to the temple on the holy Mount Garizim have honoured Menippos, son of Artemidoros, from Herakleion, himself and his descendants, for constructing and dedicating from his own funds (something) for the prayer hall of God … and have crowned (him) with a golden wreath and …

The second marble stela (H. 0.48 m, W. 0.33–0.405 m, D. 0.075–0.11 m) is fully preserved at the bottom, but mutilated at the top (Fig. 4). It records that Sarapion from Knossos was honoured by the Israelites with a gold crown for benefactions, which are not specified any closer. Ὁι ἐν Δήλῳ Ἰσραελεῖται οἱ ἀπαρχόμενοι εἰς ἰερòν Ἀργαριζεὶν στεφανοῦσιν χρυσῷ στεφάνῳ Σαραπίωνα Ἰάσονος Κνώσιον εὐεργεςίας ἕνεκεν τῆς εἰς ἑαυτούς.38

On palaeographical grounds, Bruneau dated the stelae to 250–175 BCE (stela for Menippos) and 150–50 BCE (stela for Serapion), respectively, which causes quite a remarkable chronological gap. Magnar Kartveit recently argued, however, that the palaeography speaks for a similar date of both stelae in the first half of the second century BCE, and that the stela for Serapion was even made earlier than that for Menippos, which was reused and inscribed less carefully.39 If the parts are compared that are preserved in both stelae, the following picture emerges:

 36 See, e.g., four inscriptions from the Black Sea, all dated to the first century CE: CIRB 70, 71, 985, 1128. Cf. Bruneau 1982: 475 who argues that the wording ἐπὶ προσευχῇ would not be a local phenomenon. However, he does not provide any other examples. 37 As also argued by Bruneau 1982: 486, even if he translated ἐπὶ προσευχῇ as “en ex-voto” (474). For a recent exhaustive discussion of the term προσευχή, see Catto 2007: 14–48. 38 Bruneau 1982: 467–471, no. 1; IJO I Ach67. The translation is not debated, except for the initial lines (see inscription no. 2); Runesson, Binder, and Olsson 2008: 130–131, no. 101: “The Israelites on Delos, who make first-fruit offerings to the temple on holy Mt. Gerizim, crown with a gold crown Serapion, son of Jason, from Knossos, for his benefactions toward them.” 39 Bruneau 1982: 481–485; Kartveit 2009: 216–225: he also discusses other potential criteria for dating, such as the execution of the wreaths, the question of letter cutters, and the length of the texts or specific formulas, which would not yield conclusive answers at this point, however. Cf. also IJO I: 233 suggesting that “the two inscriptions may be closer to each other in time than Bruneau believes,” however without providing any arguments.

92

Monika Trümper

Criteria  Width of inscribed part Depth of inscribed part Size of letters Lines of inscription  Format of letters  Honour  Reason for honour

Stela for Menippos  52 cm 6–9.5 cm 12 (only omikron 8) cm 9 (or more?), 4 of them barely preserved Mostly carved, but some obviously only painted (alternating lines in black and red) Crowning with golden crown Construction and dedication ἐπὶ προσευχῇ

Stela for Sarapion  34 cm 7.5 cm 15 (sometimes 10) cm 6 (fully preserved) All carved (alternating lines in black and red) Crowning with golden crown  Benefactions towards Israelites

Fig. 4. Samaritan stele no. 1; © EFA: Philippe Bruneau; Bruneau 1982: 468, fig. 2.

The Synagogue in Delos Revisited

93

The stela for Menippos was larger, the inscription significantly longer and its letters smaller; while its wreath was certainly more elaborately carved than that for Sarapion,40 the letters of the inscription were barely incised and partially even only painted. The discrepancy in the quality of carving can be explained with fact that the stela was reused. The text of the inscriptions suggests that Menippos was more generous and important because he had funded construction works, whereas Sarapion was honoured more mundanely for generic benefactions. Construction works may also have preceded the unknown benefactions bestowed by Sarapion. The differences between the two stelae do not reveal a coherent pattern and are hard to explain. The Israelites may first have honoured Menippos for construction works, with a large, but reused stela and an inscription that was carved rather sloppily in an older style. The reuse and hasty execution could betray modest financial means of the Israelite community at this time, which needed financial support of a wealthy benefactor. Sarapion may have received, probably not long afterwards, a simpler smaller stela for somewhat lesser benefactions, but with a more modern carving style. In general, a date of both stelae after 167/6 BCE seems possible and would fit better with the general development of the quarter, as currently reconstructed.41 More importantly, the type of the monuments and the honours recorded have not yet received due attention. While Bruneau assumes that the two stelae were originally inserted into a wall, this is highly unlikely, particularly if they were really made at completely different ages. Instead, display in front of some wall and on top of a base seems much more likely.42 Different groups and associations in Delos commonly honoured their generous benefactors with a painted portrait or a statue, set up in their meeting place. For example, the Agora of the Italians (GD 52) and the Establishment of the Poseidoniasts of Berytos (GD 57) were filled with such honorary statues of generous euergetai.43 By comparison, a gold crown and simple stela seem more modest, suggesting either limited financial means, restricted facilities or different honorary practices. On the other hand, the crowning with a golden crown ranged among the honours typically bestowed upon benefactors in the Hellenistic world,44 and was often combined with a public proclamation. In the fully preserved dedication for Sarapion, the present tense of the verb crowning (στεφανοῦσιν) is used; according to John Ma, this means that the “proclamation is not just a historical statement about a past grant of honours, but also an honorific act in its own right, a performative enacted in front of  40

Contra Kartveit 2009: 219. Trümper 1998: 217–224; Trümper 2004: 578–581; Bruneau and Ducat 2005: 247–256, nos. GD 74– 80; Zarmakoupi 2014–2015. 42 See in detail Trümper 2004: 586–587. 43 ID 1520; Kreeb 1988: 21–29; Trümper 2002: 314–330; 2008: 138–225; 2011: 53–58, 65–68; 2014. 44 Gauthier 1985: passim, esp. 113–116; McLean 2002: 238–241; IJO I: 231; Ma 2013: esp. 34–38. For the practice of depicting wreaths on honorific stelae in Delos, Butz 2007. – I am much indebted to Daniel Kah for a discussion of the crowning practice. 41

94

Monika Trümper

an audience.”45 In contrast, Bruneau reconstructed for Menippos’ stela the verb crowning in the aorist tense (ἐστεφάνωσαν), in analogy to the preserved ἐτίμησαν.46 Ma interprets the aorist tense as a deliberate choice to make clear that the proclamation of honours “took place once, as an act of communication and publicity.”47 The honorary practice of crowning was not limited to public institutions, but also adopted by private groups and associations. For example, in Delos the associations of the Herakleists from Tyre and the Poseidoniasts from Berytos crowned their benefactors even regularly, during annual sacrifices.48 Since the Israelites also embraced the honouring format of crowns for their benefactors, it should be questioned whether they also adopted the practice of one-time or regular proclamations and crowning rituals. Such a practice would have required an appropriate setting, specific occasions or meetings and possibly a specific venue. The identity of the benefactors—Hellenized Jews with Greek names49 or nonJewish Greeks—may have played a role in the choice of honours; but even if the Israelites had opted to honour their non-Jewish Greek benefactors with standard

 45

Ma 2013: 35. Bruneau 1982: 471, 474. Accordingly, he correctly translated “have honoured … and crowned him” (“ont honoré … et l’ont couronné”), whereas most of the other translations cited above do not respect the tenses, except for IJO I Ach66 and Philip Harland, available at: http://philipharland.com/grecoroman-associations/222-honors-by-israelites-for-benefactors/ (accessed 31 July 2018). 47 Ma 2013: 34. 48 ID 1519, l. 35–51: “It was resolved by the association of the Tyrian Herakleists of merchants and shippers to praise Patron son of Dorotheos and to crown (στεφανῶσαι) him with a gold crown each year during the performance of the sacrifices to Poseidon on account of the virtue and goodness which he continues to have towards the association of the Tyrian merchants and shippers. (...) Let the appointed leaders of the society, the treasurers, and the secretary take care of proclaiming the following proclamation during the sacrifices as they are taking place and in the synods: ‘The synod of the Tyrian merchants and shippers crown (στεφανοῖ) Patron son of Dorotheos, the benefactor.’” Translation by Philip Harland, available at: http://philipharland.com/greco-roman-associations/223-honorary-decree-of-tyrian-immigrants-for-a-member-and-priest-2/ (accessed 31 July 2018). ID 1520, l. 34–53: “Also, let one day each year be celebrated for him on the day following the procession of the festivities in honor of Apollo, and let him invite two people of his choice. Now a gold crown will be placed upon him, upon him who ‘crowned’ the synod. Let the following be proclaimed during the festivities in honor of Poseidon: ‘The association crowns (στ[ε]-[φ]ανοῖ) Marcus Minatius son of Sextus with a gold crown on account of the virtue and goodwill which he continues to have towards the association. For good fortune!’ Furthermore, proclaim the following on the same day: ‘The association crowns (στεφανοῖ) Marcus Minatius son of Sextus and celebrates a day in his honor, both now and forever, on account of the virtue and goodwill which he continues to have towards the association. For good fortune!’ And in the monthly synods proclaim: ‘The association crowns (στεφανοῖ) Marcus Minatius son of Sextus with a gold crown, being a benefactor of the synod. For good fortune!’” Translation by Philip Harland, available at: http://philipharland.com/greco-roman-associations/224-honors-by-berytian-immigrantsfor-a-roman-banker/ (accessed 31 July 2018). 49 For this phenomenon see below, n. 55. 46

The Synagogue in Delos Revisited

95

“Greek pagan” formats, the honours still had to be proclaimed and advertised in some place.50 In sum, the two inscriptions suggest that a community of Israelites was resident in Delos in the second century BCE and had a common venue where they could have displayed honorary monuments for their benefactors and possibly also have carried out honorary rituals.51 The second inscription that mentions προσευχή was not found in GD 80 either: The inscription ID 2329 was carved on a block of white marble (H. 0.345 m, W. 0.17 m, D. 0.185 m) that shows a dowel hole (Dm. 0.06, D. 0.02 m) with remains of lead on its upper surface (Fig. 5). The block was found in house A of insula II of the Stadium Quarter (Quartier du Stade) and includes a dedication by Agathokles and Lysimachos ἐπὶ προσευχῆι, dated to the early first century BCE on palaeographic grounds.

Fig. 5. ID 2329; M. Trümper, based on Bruneau 1982: 499, fig. 13. Ἀγαθοκλῆς καὶ Λυσίμαχος vv ἐπὶ προσευχῆι.52

 50 Cf. IJO I: 231 where the findspot of the two stelae is identified as a building of the Samaritans: “It is not clear where exactly Menippus’ crowning took place – in the Samaritan establishment or during a religious festival or other public feast on Delos.” 51 As also argued by Kartveit 2009: 222. 52 ID 2329; IJO I Ach65.

96

Monika Trümper

The same problems arise as mentioned above: While Matassa translates ἐπὶ προσευχῆι as “for an offering/prayer”, Runesson, Binder, and Olsson translate “for the prayer hall”.53 Irrespective of the precise translation in this case, Matassa does not comment upon the intimate link of the term προσευχή with Jewish cult that is commonly emphasized in literature.54 Matassa argues that the names are not typically Jewish but would find parallels in other Greek inscriptions from Delos. This had already been noted by Bruneau, who maintained, however, that this would not speak against a Jewish identity of the donors, because Hellenized Jews would often have had Greek names.55 Since the cutting on the top of the marble stela still contains remnants of lead, this must, according to Matassa, have held a statue or votive offering, which would not be part of any known Jewish custom. The marble base is untypical within the Delian context, however, because it is completely unadorned, not decorated with mouldings at the base and the top, and is significantly higher than typical bases of statuettes, which were found in situ in wall niches. In contrast, it is currently not possible to safely identify the potential location of statuettes set up on high bases outside of those niches. Thus, it cannot be safely determined that this base necessarily supported a statuette and not some other object like a small basin or table.56 While Plassart had argued that this base had originally been set up in GD 80, Bruneau in 1982 suggested that it belonged to its findspot, the house. He identified this house as a Jewish residence because it includes an unusual reservoir with staircase, which may have served as a miqweh.57 This house has two distinct phases, however (Fig. 6). Originally, it included only the courtyard, one large room and three smaller rooms as well as the reservoir with staircase; the latter was directly accessible from the northern street. This reservoir was fed by groundwater like a well and possibly also with rainwater collected from the roofs. In a second phase, the building was enlarged and provided  53 Both readings have followers, but no complete list is provided here; cf. above nn. 34–35: Matassa 2007: 87–88; thus also, e.g., Bruneau 1970: 488; Bruneau 1982: 475; IJO I Ach65. – Runesson, Binder, and Olsson 2008: 128–129, no. 99; thus also, e.g., Plassart 1914: 530–531; White 1987: 142–143; Binder 1999: 305; Catto 2007: 64. 54 For example, Bruneau 1970: 488; 1982: 475; IJO I: 227; Catto 2007: 14–48. 55 Bruneau 1970: 488. Matassa 2007: 87–88 does, however, not refer to Bruneau 1970: 488 and his argument. The authors of IJO I: 224, 228 point out that the name Agathokles occurs in Jewish inscriptions from Egypt and Cyrenaica, and the name Lysimachos in a Jewish inscription from Cyrenaica. 56 For bases of statuettes or other votives from Delos, see: Peribolos Street, House E, vestibule, niche with unadorned marble base in situ; H. 0.24 m, W. 0.27 m, D. 0.27 m; on top small round dowel hole for a statuette or votive; dedication of Sporios Stertenios to Artemis Soteira (ID 2378); Kreeb 1988: 351, pl. 8.1. Theatre Quarter, Insula VI, House D, room c: niche with unadorned uninscribed marble base in situ; H: 0.065 m, W 0.27 m, D. 0.26 m; Kreeb 1984: 330, Figs. 12–13. For supports of basins and tables in Delos, see Deonna 1938: 34–36, 48–53, 75–77: these are commonly also adorned at the bottom and top, however, and have average heights of about 0.50–0.70 m. 57 Plassart 1916: 242; Bruneau 1982: 499–502.

The Synagogue in Delos Revisited

97

with a vestibule and new main entrance in the south, a latrine, a kitchen-bathroom, a peristyle, and an upper story. Direct access to the reservoir and courtyard from north was blocked and the new southern entrance was adorned with an altar. Such altars were most often combined with niches and liturgical paintings for the cult of the lares compitales, a cult usually performed by slaves and liberti of Italians or Romans. Even if the building had originally been constructed by Jewish patrons, as a house or even small assembly building, the transformation into a properly endowed house could have been initiated by Italians or Romans, or alternatively by a particularly liberal Jew whose family practiced pagan cults.58

Fig. 6. House A, Insula II, Stadium Quarter: reconstruction of first phase; based on Trümper 1998: 222, fig. 22.

Thus, assessment of ID 2329 is even much more complicated than recognized anywhere in literature, and the following two scenarios must be considered: If ID 2329 was found in situ, what did the term προσευχή refer to? Would the dedication simply have been left there when owners and their religious preferences changed in the building’s second major phase of use? If ID 2329 was not found in situ, from where, and particularly when and why was it transported to the house? While there is currently no safe answer to any of these questions, the first  58 For the findspot of the inscription in the northwest corner of the courtyard, Plassart 1916: 242; for the building history of the house, Trümper 1998: 222–223, Figs. 22, 24. Cf. IJO I: 226–227, which questions use of this house as a Jewish residence, without reference to Trümper 1998. The reconstruction of the first phase (here Fig. 6) is necessarily schematic; the building most likely included pavements and thresholds, which are not preserved or cannot be safely be attributed to the first phase.

98

Monika Trümper

scenario is the lectio facilior. Similar scenarios must have occurred in many houses of the Late Hellenistic trade port when inhabitants changed, probably quite frequently: elements of the fixed or even movable decoration were kept, when new people with possibly different cultural traditions moved in.59 As mentioned above, the wording ἐπὶ προσευχῇ/ἐπὶ προσευχῆι appears to have been a notable local practice. This suggests local knowledge and some communication or relationship between the dedicators of both inscriptions. While not impossible, it seems at least remarkable that, according to current dating of the inscriptions, traditions would have been transmitted over about 80 to 160 years, and possibly even between different “Israelite” groups (Jews vs. Samaritans).60 Reassessment of both inscriptions might reduce the chronological gap between them considerably, to maybe 40–50 years. Furthermore, both inscriptions with the wording ἐπὶ προσευχῇ/ἐπὶ προσευχῆι were not found in GD 80; their find spots should be treated equally, thus as spaces where the monuments may have originally been set up. While some scholars accepted that the Samaritan stelae may have been found at their point of use, this was commonly questioned for ID 2329,61 although this building, with its accessible reservoir, includes an architectural anomaly that might be considered the strongest architectural clue for Jewish ownership currently available in Delos. Following Bruneau, it is argued here that the inscription ID 2329 and the unusual water reservoir may have belonged to a small Jewish (or Samaritan) assembly space, built probably in the second half of the second century BCE, that was later transformed into a house, most likely before 88 BCE.62 Plassart found five inscribed monuments in rooms A and B of GD 80, along the walls and on benches, and he identified them as furniture pertaining to GD 80 and not as material intended to be burnt in the above-mentioned limekiln. Matassa criticizes Plassart for not explaining his reasoning, and identifies the pieces as fuel for the kiln, which had been brought from some unknown location to GD 80 and thus could not be cited for reconstructing the function of this building.63 This notwithstanding, Matassa then still spends considerable space and effort on a discussion of these inscriptions, trying to prove further from their content that they cannot be securely connected with Jews.  59

Trümper 1998: 124–137; 2003; 2006. For example, portrait statues set up in houses cannot always be attributed to the last phase of use of these houses, nor is it certain that these houses remained in the same family for generations, or at least after the erection of a portrait statue. Cf. for example the history of the famous House of Kleopatra, which was certainly remodelled and used for another 50 years after she had set up portrait statues of her husband and herself; Trümper 1998: 273–274. Cf. also Bruneau 1970: 450–451 for inscriptions found in houses. 60 See above for the dating of ID 2329 vs. the Samaritan stela no. 2: Bruneau 1970: 484; 1982: 481–485. 61 Cf., e.g., IJO I: 227 vs. 229. 62 Bruneau 1982: 502–504. That both the Samaritans stelae and ID 2329 were by chance removed from GD 80, is certainly the lectio difficilior. 63 Matassa 2007: 104.

The Synagogue in Delos Revisited

99

Irrespective of Matassa’s criticism, Plassart had clearly and convincingly argued that the presumably conveniently available benches in rooms A and B had largely escaped the lime kiln; this very fact would suggest that the walls next to these benches had already collapsed, burying the benches and monuments on top of them, when the lime kiln was installed.64 Therefore, these monuments were most likely found in their original context: if not in situ where they had actually been set up in the last phase of use of GD 80, then at least pertaining to the furniture of the building’s last phase of use.65 Their inscriptions and formats remain crucial for identifying the function of GD 80. Three are dedications to Theos Hypsistos, one is a dedication to Hypsistos. The monument with the inscription ID 2328 is a small column of white marble (H. 0.865 m, Dm. 0.175–0.21 m) that includes on its surface a dowel hole (Dm. 0.03 m, D. 0.03 m) in the centre of a round cavity (Dm. 0.07 m, D. 0.01 m). The column was found at the foot of the west wall in room B and commemorates a dedication of Lysimachos on behalf of himself to Theos Hypsistos, as thankoffering, carved in the early first century BCE according to the lettering (Fig. 7). Λυσίμαχος ὑπὲρ ἑαυτοῦ Θεῷ Ὑψίστῳ χαριστήριον. 66

While Matassa does not comment upon the function of the monument, several scholars argue that a sizeable hole with melted lead on top of the column must have served for fixing a statuette. The unadorned column with a height of 0.86 m is again highly untypical as a support of a statuette, however.67 As argued for the rectangular support with ID 2329, the height of which the column even exceeds by 0.50 m, it is far from certain that this column necessarily supported a statuette, and not some other object, like a basin.  64 Plassart 1914: 526. “Ces marbres n’ont pas été apportés d’un autre endroit pour alimenter le four à chaux: en effet, les petits bases 3, 4, et 5 on été découvertes, avec trois autres du même type, qui ne poretnt (sic) pas d’inscription gravée, sur un des bancs à l’Ouest de la salle A; or, si les bancs ont, en grande partie, échappé aux chaufourniers, c’est que la chute des parties hautes des murs les avait recouverts de décombres. La colonnette 2 a été, de même, trouvée au pied du mur Ouest de la salle B.” 65 The question of how and where exactly votive offerings were set up or stored in sanctuaries, assembly places or other contexts has not been comprehensively studied for Delos. While this important topic deserves and requires its own full study, this is beyond the scope of this paper. 66 ID 2328; IJO I Ach63. – Matassa 2007: 88: “Lysimachos for himself [to] God Most High [for a] votive/thank-offering.” IJO I: 223: “Lysimachus, on his own behalf, to the Highest God, a thank-offering.” Runesson, Binder, and Olsson 2008: 127–128, no. 97: “Lysimachus on behalf of himself to God Most High, a thank-offering.” 67 IJO I: 224; Runesson, Binder, and Olsson 2008: 127–128, no. 97. The column is clearly visible on the photo Plassart 1914: 524, but no detailed photo of the monument has been published so far. Lead in the dowel hole is not mentioned in any other publication. For the sizes of statuette bases, see above n. 56. No comparable monument is published in the Inscriptions de Délos, suggesting that this format was unique on Delos.

100

Monika Trümper

Fig. 7. Synagogue building GD 80, view from NE in 1913, column ID 2328 in upper left corner; Plassart 1913, pl. V.

The marble base (H. 0.25 m, W. 0.165 m, D. 0.12 m) with ID 2330 was found in room A on a bench in the west. It is decorated with mouldings at the bottom and top on all sides, except for the back, suggesting that it was set up in front of a wall. The inscription records a dedication of Laodike to Theos Hypsistos in fulfilment of a vow after having been saved by his therapies. Λαοδίκη Θεῶι Ὑψίστωι σωθεῖσα ταῖς ὑφ’ αὑτοῦ θαραπήαις, εὐχήν.68

It is dated to the early first century BCE, according to lettering. Matassa wrongly states that the donor was a man, whose name would not be specifically Jewish and who would have set up this monument around 108/107 BCE.69 She does not  68

ID 2330; IJO I Ach62. – IJO I: 222: “Laodice to the Highest God, having been saved by (medical) treatments by him, (made) a vow.” Runesson, Binder, and Olsson 2008: 127, no. 96: “Laodice, to God Most High, having been saved by his therapies (in fulfillment of) a vow,” offering by a woman. No photo of this monument has been published. 69 Matassa 2007: 89: “Laodike to God Most High for healing him of his infirmities, an offering.” IJO I: 223 point out that the participle ending σωθεῖσα alone suggests a female dedicator. The Lexicon of Greek Personal Names has 65 entries for Laodike between the fourth century BCE and the second century

The Synagogue in Delos Revisited

101

discuss the function or possible location of this monument, which is too shallow to have supported anything, and must have served as an independent votive offering. Two small marble bases with mouldings, horns and bowl-shaped cuttings had the shape of incense-burners. They were found on a bench in the west of room A and carry dedications, dated to the first/second century CE based on lettering: by Zosas of Paros to Theos Hypsistos (Fig. 8), and by Markia to Hypsistos.

Fig. 8. ID 2331; © EFA: Philippe Bruneau; cf. Bruneau 1970, pl. IX, fig. 5. Ζωσᾶς Παρίος Θεῷ Ὑψίστῳ εὐχήν.70



CE, all of which are women (among them four in Delos): ID 2230; ID 2628 b II 15 (108/7 BCE); ID 2628 b III, 16 (108/7 BCE); Couilloud 1974, 91 no. 77. – For the argument of Jews using Greek names, see above n. 55. 70 ID 2331, IJO I Ach60: base of white marble, with mouldings at top and bottom; 4 feet, horns and round carved bowl on upper surface; H. 0.18 m, W. and D. 0.085–0.10 m: Matassa 2007: 89: “Zozas (sic) of Paros to the God Most High, an offering.” IJO I: 220: “Zosas of Paros to the Highest God (made) a vow.” Runesson, Binder, and Olsson 2008: 125–126, no. 94: “Zosas of Paros to God Most High, (in fulfillment of) a vow.” IJO I: 220 and Runesson, Binder, and Olsson 2008: 125–126, no. 94 mention a mounting hole on the upper surface which would have held a votive; this is, however, possibly a misunderstanding of the phrase “creusée en godet”, first used by Plassart 1914: 527, and then resumed in ID 2331 and Bruneau 1970: 484. The discussion by Deonna 1938: 373–374, 383 shows, however, that this was a small cavity that held the incense or served as a fireplace. Mattassa 2007: 89 states that this inscription “is dated to the first century BC,” and that the name Zosas does not elsewhere appear in ID. But the inscription has been dated “d’époque bien plus basse” by

102

Monika Trümper Ὑψίστῳ εὐχὴν Μαρκία. 71

Numerous examples of such monuments have been found in Delos, most of which show distinct traces of burning at the top,72 thus testifying to actual use.

Fig. 9.

Uninscribed monument from GD 80; M. Trümper based on Deonna 1938, pl. CX, figs. 969–970 (proportions corrected).

Such signs of use are not mentioned for the two examples from GD 80, however (Fig. 9).73  Plassart 1914: 527, specified by Bruneau 1970: 484: “me semblent devoir être datées du Ier ou du IIe siècle ap. J.-C.” Cf. IJO I: 221. The name Ζωσᾶς is recorded on three grave monuments from Rheneia, dated to the second and first century BCE; Couilloud 1974: 60, no. 3, 90–91, no. 76, 147, no. 261. 71 ID 2332, IJO I Ach61: base of white marble; H. 0.17 m, W. and D. 0.08–0.10 m: Matassa 2007: 90: “[The] Most High [from] Markia”, again wrongly dated to the first century BCE, whereas the French scholars had dated it to the first/second century CE, see previous note. IJO I: 221: “To the Highest, Marcia (made) a vow.” Runesson, Binder, and Olsson 2008: 126, no. 95: “Marcia to the Most High (in fulfillment of) a vow.” 72 These incense-burners were made of marble, stone, terracotta, tufa, or metal; they are round and rectangular; and can be as small as 0.10 x 0.08 x 0.06 m; Deonna 1938: 371–389, esp. 373 for traces of burning. 73 Deonna 1938 frequently specifically mentions traces of burning in his catalogue, but not for the examples from GD 80: 388: s.n. (uninscribed) and A 3049 (ID 2332); 389: A 3050 (ID 2331). Matassa

The Synagogue in Delos Revisited

103

While the fifth inscribed monument found in GD 80 may also have been dedicated to Theos Hypsistos,74 this cannot be safely determined because the inscription is only partially preserved (Fig. 10).

Fig. 10. ID 2333; © EFA: Philippe Bruneau; cf. Bruneau 1970, pl. IX, fig. 6.

— — — — Ι# — — — rosette   rosette γενόμενος ἐλεύθερος.75



2007: 90 states: “Furthermore, there were other inscription bases found in GD80 which neither Plassart nor subsequent scholars have chosen to mention, and whose texts are illegible.” She supports this in note 25 on p. 133 with reference to “Deonna 1938: Pl. CXII, photographs 969 and 970.” These photos are actually published on Deonna 1938, pl. CX and show two sides of one and the same monument (albeit, irritatingly, at different scales), which was most likely never inscribed like most of these small altar/ incense-burners found in Delos; the monument is mentioned in Trümper 2004: 585 n. 171 with reference to Deonna 1934: 424, no. 166, figs. 31, 32 and Deonna 1938: 388, remarking that, with 0.37 m height, it is significantly larger than the inscribed examples. Of the “encensoirs, petits autels, brûle-parfums” published in Deonna 1938: 371–389, only very few carry inscriptions, in addition to those from GD 80: ID 1910, 2111, 2294, 234, 2348, 2355, 2369, 2370, 2384, 2390, 2403, 2410, 2593; IG XI 1274, 1359. 74 As proposed by IJO I: 225, “although personal dedications of this kind to the deity have not been found so far.” 75 ID 2333; IJO I Ach64: marble base with profiles at the top and bottom; H. 0.33 m, W. 0.18–0.20 m, D 0.24–0.26 m; for the date Bruneau 1970, 484. – Matassa 2007, 90: “…. became free.” IJO I: 224: “….

104

Monika Trümper

The small marble base is adorned with profiles at the top and bottom and two rosettes on the front. Because the upper surface did not yield any conclusive traces, its precise function (e.g., altar/incense-burner, base of something) cannot be identified, but it may simply have served as a votive on its own. The inscription records the manumission of an unknown person and has been dated to the first/ second century CE. While the four monuments dedicated to Theos Hypsistos/Hypsistos have typological comparisons in the local context they are at the same time somewhat unusual, possibly precisely because they did not serve for standard functions. The discussion about the identity of Theos Hypsistos is in many aspects similar to that about the meaning of proseuchē. Matassa goes back to De Mazur’s arguments, barely taking the later discussion into account, and argues that the dedications were not necessarily addressed to the Jewish god. Instead, they may also refer to the Greek deity Zeus Hypsistos, whose cult (a healing cult, and a more likely association given the physical form of the inscription bases) also used these epithets to describe their chief deity. The sanctuary of the cult of Zeus Hypsistos was located on Mt. Cynthus, less than 500 m from GD 80.76

Scholars debate whether hypsistos is a generic adjective applicable to many forms of divinity77 or whether Theos Hypsistos was a specific, if abstracted nameless supreme god whose aniconic cult was “basically monotheistic or henotheistic in nature, closely linked to the monotheism of the Jews.”78 There is, however, agreement that the denomination at least in some cases referred specifically to the Jewish god, as proven by the third century BCE Septuagint and epigraphic sources from the Hellenistic period to the fourth century CE.79 Particularly important are two inscriptions: first, from Alexandria, dated to the second century BCE, which records that an unknown donor dedicated a sacred precinct, a προσευχή and its appurtenances to Theos Hypsistos; and second, from Athribis, dated to 181–45 BCE, which records that, on behalf of king Ptolemy and queen Cleopatra, Ptolemy son of Epikydes, chief of police, and the Jews in Athribis (dedicated) the προσευχή to Theos Hypsistos.80 Recently, Anna Collar even argued that “all Hellenistic  having become free.” Runesson, Binder, and Olsson 2008: 128, no. 98 “…. having been set free”. The space above the rosettes would have sufficed for two lines with letters of similar size as the preserved ones. 76 Matassa 2007: 93. 77 Belayche 2010; cf. also Stein 2001. 78 Citation in Collar 2013: 225, referring to Mitchell 1999; 2010. For a recent summary of the debate, Collar 2013: 224–286, who supports Mitchell’s argument. Cf. also IJO I: 217–218, largely supporting Mitchell’s argument. 79 Mitchell 1999: 133–146, nos. 84–88, 207, 282, 283, 285; 2010: 186 n. 74. 80 Mitchell 1999: 146, nos. 283, 285; Cf. Catto 2007: 72–74. Matassa 2007: 109 criticized Trümper 2004: 569 for using later evidence to support earlier data in her discussion of the evidence for Theos Hypsistos. However, the Septuagint as well as the inscriptions from Alexandria and Athribis clearly testify to the use of Theos Hypsistos for the Jewish god in the Hellenistic period. Moreover, ID 2331 and 2332 have been dated to the first/second century CE, complying with the date of many testimonies collected by Mitchell, 1999: 132–147, nos. 1–293; and Mitchell 2010: 198–208, nos. A1–83. But Matassa 2007: 89–90 has dated

The Synagogue in Delos Revisited

105

attestations of Theos Hypsistos (…) have explicit Jewish connotations.”81 Among the earliest testimonies are two grave stelae from Rheneia, the burial island of Delos, dated to the late second century BCE, which invoke Theos Hypsistos with phrases that are clearly related to the Septuagint.82 Matassa does not cite these grave monuments at all although they clearly link Theos Hypsistos to Jews in Delos. The debated nature of Hypsistos and Theos Hypsistos notwithstanding, scholars currently agree that the inscriptions from GD 80 referred to the Jewish god.83 Using network theory, Collar even went a step further: Delos played a pivotal role in the introduction of the cult term “Hypsistos” to the Greek world, and possibly the place where the cult of Theos Hypsistos was formed, in dialogue between Jews, Samaritans and pagans. Delos occupies a central position in the epigraphic and network analyses, and as an international cult centre, slavetrading hub and free port in the Hellenistic period, people of considerable status from across the Mediterranean were frequent visitors. The opportunity for the exchange of religious ideas is clear in such a context, and the high-status individuals who visited the island were in a good position to bring new ideas back to their homelands, and have the social power to influence others.84

While Collar argues that GD 80 may have been a Jewish synagogue or a Greek sanctuary, because the donors of ID 2328 and 2330 did not have clearly Jewish names, she hypothesizes that the Samaritan community on Delos frequented, if not built and operated, the sanctuary of Zeus Hypsistos on the Kynthos; this may have been an international sanctuary where the Samaritans welcomed Greeks and introduced them to the cult term Hypsistos.85 Both Matassa and Collar’s arguments are problematic, for various reasons. The sanctuary of Zeus Hypsistos on Mt. Kynthos (GD 106) was founded by a private person of debated identity, a Poses Posidoniou or a Gaios Fabios Gaiou Romaios, who dedicated a peribolos and the fittings within to Zeus Hypsistos and to the gods to whom altars have been raised. According to the lettering, the dedication is dated to the first century BCE.86 It is nowhere discussed for how long  both of these erroneously to the first century BCE, see above n. 70–71. 81 Collar 2013: 244, 246 (citation). 82 ID 2532; Plassart 1914: 532–533; Bruneau 1970: 486; Couilloud 1974: 214–215, no. 485; Mitchell 1999: 135, no. 110; IJO I Ach70, 71; Collar 2013: 245–246. 83 Belayche 2010: 155 n. 89; Mitchell 2010: 186 n. 75; Collar 2013: 245–250. – Somewhat reserved IJO I: 218–219, although the inscriptions are all included in the corpus, IJO I Ach60–63. 84 Collar 2013: 284. 85 Collar 2013: 251. 86 ID 2306: Γάιος Φάβιος Γαίου Ῥωμ[αῖ]ος ἐφ’ ὁδῶι τὸμ περίβολον [καὶ τὰ ἐν] αὐτῶι χρηστήρια Διὶ Ὑψίστῳ καὶ θεοῖς οἷς τοὺς βωμοὺς ἱδρύσατο, ἀνέθηκεν.

106

Monika Trümper

the sanctuary was used, but the Kynthos seems to have been largely abandoned in the Imperial period.87 The small sanctuary (barely 100 m2) includes an open air precinct with a central shrine and four separate altars, which were most likely dedicated to Zeus Hypsistos and three other deities, mentioned in the inscription without specific names. The finds recorded in the publication include a round marble altar decorated with garlands and four lead figurines that served to cast some spell similar to tabellae defixiones, and that do not agree with a healing cult. No small uninscribed or inscribed monuments similar to those discovered in GD 80 were found here, however. It seems little likely that, by chance, all such votives from GD 106 would have been transported to GD 80 just to be burnt in a lime kiln,88 and that they were all dedicated to (Theos) Hypsistos, whereas the sanctuary GD 106 itself was specifically dedicated to Zeus Hypsistos. The healing powers of Theos Hypsistos and Zeus Hypsistos have been repeatedly discussed.89 While Laodike thanked Theos Hypsistos expressively for having been saved by his therapies or medical treatment (ID 2330), the other donors use common terms for generic thanks offerings: εὐχή and χαριστήριον. These terms were widely used in Delos for a broad variety of deities that have no specific healing powers.90 The idea that the inscribed bases from GD 80 and the nearby house supported votives was first brought up by Mazur who compared them to those from the sanctuary of Zeus Hypsistos on the Pnyx in Athens, which show healed parts of the body.91 Bruneau has already responded at length that there is no evidence for similar depictions in the sanctuary of Zeus Hypsistos on Delos or in GD 80.92 The notion of votives—showing body parts or other images—is somewhat tenaciously maintained in scholarship because of the questionable interpretation of traces on the upper surfaces of the inscribed monuments, esp. of ID 2328, ID 2329, and 2331. Even if it must remain open for now what the unusual  Plassart 1928: 289–293: the marble block (H. 0.22 m, W. 0.46 m, D. 0.17 m) was originally inserted into a wall; the donor is cautiously identified as Ποσ[ῆς] Ποσι[δ]ων[ί]ου. Bruneau 1970: 240–241, 649– 655; IJO I: 218 incorrectly states that this sanctuary did not yield any epigraphic evidence; Belayche 2010: 156 n. 92. For the debated donor, see Bruneau 1970: 241. 243 who favours the reading in ID 2306. 87 Bruneau and Ducat 2005: 44–45. 88 Lime kilns were commonly built at sites where many marble pieces were readily available. The small size of the inscribed monuments ID 2328, 2330–2332 would have facilitated transport, but the realistically expectable yield would hardly have justified the effort of carrying them down the quite steep slopes of the Kynthos (the sanctuary is even at the south-eastern edge of Kynthos). Marble pieces that were much more likely stored in GD 80 to feed the lime kiln, such as the drums and capital of a Corinthian column, were monumental; Trümper 2004: 552. 89 IJO I: 220–224; Matassa 2007: 93. 90 A search in the Packard Humanities Database for Delos yields 65 matches in 62 texts for εὐχή, and 82 matches in 81 texts for χαριστήριον, in connection with the following deities: Agathe Tyche, Anubis, Aphrodite, Aphrodite Hagne, Apollo, Artemis, Athena Kynthia, Dionysos, Harpokrates, Herakles, Hermes, the Muses, Isis, Sarapis, Samothracian Gods, Theois Synnaois, Zeus Helios, Zeus Ourios, and others. 91 Mazur 1935: 21. 92 Bruneau 1970: 487.

The Synagogue in Delos Revisited

107

monuments ID 2328 and 2329 carried, their inscriptions do not specifically mention healing.93 Next to Asklepios, several deities in Delos, such as Apollon Oulios, Isis, and Sarapis were occasionally venerated for their healing powers, but also had other qualities and areas of influence—and similar multiple competences may have been assigned to Theos Hypsistos.94 Collar’s argument, which largely ignores the evidence from GD 80 but identifies the sanctuary of Zeus Hypsistos as a Samaritan foundation, is equally unconvincing. The two stelae of the Israelites and the dedicatory inscriptions of the sanctuary were dated to different periods, the second and first century BCE, respectively. The Israelites mention their contribution to (the temple on) Mt. Garizim, but no Theos or Zeus Hypsistos in their inscriptions, let alone other deities. It can hardly be concluded from this evidence that the Samaritans, significantly later, operated a sanctuary with five altars to Zeus Hypsistos and other gods; the more so, because this sanctuary was dedicated by an individual who cannot be identified as a Samaritan, but was probably a Roman.95 In sum, the inscriptions and the archaeological evidence suggest that Zeus Hypsistos and Theos Hypsistos/Hypsistos were two distinct deities on Delos, linked to two entirely different buildings and to different religious practices: notably a small open-air sanctuary with altars for sacrifices to different gods (GD 106) versus GD 80 with its large roofed hall where altars for sacrifices are conspicuously absent. Thus, the Delian evidence would support the interpretation that, at least in the Hellenistic period, the adjective hypsistos was used for different deities.96 To summarize the discussion of the epigraphic evidence: The discussion of recent research has confirmed that all eight inscriptions found in and close to GD 80 have Jewish connotations. They were all dated solely on palaeographic grounds, to an astonishingly long period from about the second (or even third) century BCE to about the second century CE. A critical reassessment of this dating would, first of all, require extensive comparisons in the local context, both regarding the script and the monuments themselves. This has not yet been undertaken by any expert of Delian epigraphy and material culture, and is beyond the scope of this paper. The term προσευχή used in two inscriptions had a Jewish connotation, if it did not even refer specifically to buildings of the Jewish or Samaritan community, as  93

Bruneau 1970: 487 and discussion above. Bruneau 1970: 165, 371–373, 464, 487. 95 That the two stelae were transported down from Kynthos to the shore of the Stadium Quarter seems as little likely as for ID 2328, 2330–2332; see n. 88. Collar’s argumentation, 2013: 248–251, starts from the fact that, following Bar Kokhba, the emperor “Hadrian built a temple to Zeus Hypsistos on Mount Garizim” (quote: 248). There is no evidence from the Late Hellenistic period, however, that Zeus Hypsistos was associated with the Samaritan God. 96 Plassart 1928: 289–293 and Bruneau 1970: 240–241, 479, 655 argue convincingly, based on the location and plan of the sanctuary, that Zeus Hypsistos did not designate the Greek Zeus, but an oriental Baal. 94

108

Monika Trümper

preferred here. In Hellenistic Delos, Theos Hypsistos was clearly distinct from Zeus Hypsistos and was certainly equated with the Jewish god in the case of two Jewish grave stones. The lectio facilior is that the four dedications to Theos Hypsistos or Hypsistos, found in GD 80 and dated between the first century BCE and the second century CE, were also dedicated to the Jewish god and not to Zeus Hypsistos, whose sanctuary on the Kynthos was most likely already abandoned during the first century BCE. While the format and function of most of the eight inscribed monuments find parallels in the local context certain idiosyncrasies have been pointed out, which may go back to specific norms and practices of their donors. The Samaritan community was in charge of two monuments, whereas five others were donated by persons with Greek names or names that cannot safely be identified as Jewish (Agathokles, Laodike, Lysimachos, Markia, Zosas). Following Bruneau, these donors were most likely Hellenized Jews using Greek names, rather than Greeks who adopted Jewish terminology (proseuchē) or venerated the Jewish God, Theos Hypsistos.

Archaeological Evidence Matassa reassesses the archaeological evidence based on her visit to the site and provides a sketch of GD 80 based on Mazur’s plan from 1935 (Fig. 11), strangely ignoring the excellent state plan published by Bruneau.97 She discusses the physical evidence of GD 80, the cistern, the lime kiln, and finally the various interpretations of GD 80, which, however, were always based on both the written and the archaeological evidence. In assessing the various contributions by André Plassart, Belle D. Mazur, Eleazar Lipa Sukenik, Philippe Bruneau, Michael White, Thomas Kraabel, Hudson McLean, Peter Richardson, Donald Binder, Lee Levine, and myself, Matassa does not sufficiently engage with facts and arguments that are crucial for the debate. In her description of the physical evidence, Matassa acknowledges that some walls did not belong to the original plan, such as the partition wall of rooms A and B.98 A second phase of the building is mentioned that could be dated to after 74–63 BCE when the nearby gymnasium was plundered during the “pirate raids

 97

Matassa 2007: 95, Fig. 2; the overall crudely made sketch lacks a scale and includes many mistakes, which a comparison with Bruneau 1970, pl. B easily reveals. Matassa’s discussion of the archaeological evidence is overall not very coherent, and difficult to follow; it focuses on insignificant details, but omits facts central to understanding her argumentation. 98 Matassa 2007: 95; repeated on p. 100 where she erroneously states that this wall included inscribed marble bases and that all three doorways of this wall were found walled up in 1912/1913; see Trümper 2004: 537 nn. 54–55.

The Synagogue in Delos Revisited

109

of the Mithridatic wars” and its inscribed marble bases became available as building material for GD 80.99 The number and nature of phases as well as their consequences for the identification of the building, however, are never clearly and fully described. The reconstruction of the building history was one of the main issues of recent scholarship, but Matassa does not engage with any of it.100 She acknowledges only briefly and summarily that scholars “interpreted the physical layout of the first phase and second phases of GD80 in several ways, none of which really has much bearing on its identification as a synagogue, other than the fact that in the final phase of the structure it had benches arranged around the walls of the two main areas and that the final phase is oriented towards the east.”101

Fig. 11. GD 80, schematic plan; M. Trümper, based on Matassa 2007: 95, fig. 2.

The history of the building included certainly more than two phases, and this has implications for the identification and function of GD 80, as already argued by Michael White in 1987.102 While one could challenge the various reconstructions of the building’s history, some facts are undeniable and ought to be taken  99 Matassa 2007: 97: it is unclear why she indicates the period of 74–63 BCE. It is usually assumed that the building GD 76, commonly referred to as the gymnasium of Delos, was already abandoned after 88 BCE, and that pirates under Athenodoros raided Delos in 69 BCE; cf. Bruneau and Ducat 2005: 42–43; IJO I: 210. 100 See the summary in IJO I: 211–219; the building history was the main starting point and purpose of Trümper 2004. 101 Matassa 2007: 97–98. 102 White 1987.

110

Monika Trümper

into account. To cite just one particularly illustrative and important example: the north walls of room A and room C are made entirely differently, and the north wall of C abuts stucco on the west wall of C (= east wall of A) (Fig. 12). The north wall of C was obviously built later than both the north and the east wall of room A; since this wall crucially defines room C this evidence has consequences for the reconstruction of room C. The original building may not have included room C, which would have a major impact on the interpretation of GD 80.103 Thus, Matassa’s neglect of the development of GD 80 is a step back in the debate, which renders her argumentation largely obsolete.

Fig. 12. Synagogue building GD 80, N wall of room C abuts stucco on W wall; Trümper 2004: 532, fig. 14.

Matassa dedicates quite a long subchapter to the reservoir of GD 80 that consists of a natural gap in the rock with a total length of 6.08 m; it is located beneath  103 For the analysis of different building materials and wall techniques as well as the relationship of walls to one another, see Trümper 2004: 521–556; the joint with stucco between the north and west walls of C is shown on 532, Fig. 14.

The Synagogue in Delos Revisited

111

rooms B and D1, and was partly built up and roofed with a Poros vault.104 The reservoir was fully excavated by Bruneau who provided a description and detailed drawings. He argued that the reservoir was open to the sky in room D1 and could have been accessed by a wooden ladder or some stairs and thus have served for ritual ablutions.105 This idea has been resumed by Binder.106 Matassa challenges this idea, repeatedly criticizing Binder for having misunderstood Bruneau.107 She correctly argues that the reservoir was not accessible and thus could not have served as a miqweh. This is, however, not at all new but has already been shown before and with more substantial arguments.108 Going beyond the problematic archaeological evidence, it has been pointed out that by Jewish standards, the Delian water reservoir would not have qualified as a miqweh; it could, at best, have served as natural body of water because it was filled by groundwater like a well. Given the proximity to the sea, which could have served the same function, the identification of the water reservoir as a bathing pool becomes even more questionable.109

In sum, challenging the identification of the reservoir as a miqweh does not exclude the use of GD 80 as a synagogue, because the immediate proximity to the sea remains remarkable and may have been chosen for easy access to a natural body of water. In discussing the original building, Matassa seems to favour Mazur’s proposal who reconstructed a large peristyle courtyard (18 m x 18 m) and a second open

 104 Matassa 2007: 100–103; she consistently refers to this as a cistern, which is incorrect, as already pointed out by Bruneau 1970: 482. 105 Bruenau 1970: 481–482. 490–491; Bruneau 1982: 502. 106 Binder 1999: 306–317; cf. Trümper 2004: 575–578. 107 Matassa 2007: 102–103. Binder actually cited Bruneau mostly correctly, like also IJO I: 214. Contra Matassa 2007: 103 and rehabilitating Binder, human access to reservoirs, cisterns or wells in Delos was highly unusual; cf. Trümper 2004: 576 n. 134. 108 Trümper 2004: 575–578; also IJO I: 214 which states without any arguments that use of the reservoir for ritual cleansing seems unlikely. Both are not at all cited by Matassa 2007: 100–103. 109 Trümper 2004: 577–578. – Similarly, in her largely anecdotal discussion of the lime kiln Matassa 2007: 103–105 does not refer to Trümper 2004: 556. As shown above, the marble pieces found in GD 80 require a differentiated discussion and cannot be identified sweepingly as feed for the kiln. In a similar vein, the issue of niches shall briefly be addressed here, although this is one of the insignificant details alluded to above, without any impact on the interpretation of GD 80. Matassa 2007: 110 criticizes me for having identified the niche in room A as “crudely made”, Trümper 2004: 585. She argues that this is incorrect “as anyone who has looked at other structures on the island will see.” She would have found numerous such niches that may have been used for lampes or as shrines of some sort, and provides on p. 110 Fig. 13 three photos of generically labeled examples (Theatre District Niche 1–3). Based on many years of fieldwork, in Trümper 1998: 68–76, table 2, referred to in Trümper 2004: 538 n. 57, I have provided a table of all niches in Delian houses and dedicated an entire chapter to their analysis. Niches were constructed in different dimensions, positions, and numbers (single or in carefully planned groups), and clearly had different functions. I firmly maintain my original evaluation that the niche in room A of GD 80 is crudely made and was most likely conceived as a lamp niche.

112

Monika Trümper

courtyard at the site of rooms A/B, based on comparisons with the Establishment of the Poseidoniasts of Berytos (GD 57).110 She herself observed on site that the northern and southern walls of the existing structure (note by author: it is unclear which structure is meant) extend to almost the same point of collapse into the sea, some 1.50m beyond the stylobate, and rooftiles were found along the inside of these perimeter walls indicating that they were at least partially covered.111

This observation cannot be reconciled with the archaeological record, which has been thoroughly studied by others, and it does not lead to a full four-sided peristyle courtyard.112 Matassa acknowledges later, however, that a reconstruction with a portico, as proposed by Plassart, would equally be possible.113 For comparing GD 80 in the local context, she provides a sketch that shows three porticoes in room C (or a tristoa) (Fig. 13).114

Fig. 13. Comparison of GD 80 with Delian houses and the Establishment of the Poseidoniasts; M. Trümper, based on Matassa 2007: 106, fig. 12.

 110

Matassa 2007: 97: “… stylobate leading out into what was originally a peristyle courtyard. The peristyle would have measured approximately 18 m x 18 m, but has now been destroyed by the sea almost up to the line of the stylobate”; Matassa 2007: 100, where she states that the “throne in the benched area of room A sits on what was originally one of two courtyards of the house.” 111 Matassa 2007: 97. 112 For a detailed discussion of the evidence, see Trümper 2004: 542–556, with a proposed reconstruction based on the evidence: 552, Fig. 42; not referred to by Matassa 2007: 97. 113 Matassa 2007: 105. 114 Matassa 2007: 106, Fig. 12.

The Synagogue in Delos Revisited

113

Fig. 14. Scale-to-scale comparison of GD 57, 80, 89, 111; M. Trümper based on Bruneau 1970, pl. B; Delorme 1953, pls. 45, 47; Trümper 1998: 247, fig. 35; Trümper 2006: 23, fig. 2.

This comparison is key to her assessing the function of GD 80. “Having looked at the extant houses on the island,” she argues that the ground floor plans of the

114

Monika Trümper

House of Hermes (GD 89), the House of the Dolphins (GD 111) and the “House of the Poseidoniasts” (sic) (GD 57) would be very similar. This is not analysed and interpreted any further, but simply followed by the statement: “In any event, without any evidence to corroborate its identification, none of these interpretations of the original layout of GD80 have anything to do with it ever having been a synagogue.”115 She also briefly analysed the furniture of GD 80 in the local context because this played a major role in the identification of GD 80 as a synagogue. Rooms with benches or benches in general would have been common all over the island, and not at all specific to GD 80. For example, such benches are visible in the gymnasium (GD 76), two of the Sarapeia, the Heraion, the Agora of the Italians, the orchestra of the theatre and the Sanctuary of Apollo (a semi-circular exedra). She recognizes that the throne may well come from the theatre, but argues that it was simply reused to reduce the costs of furnishing.116 Methodologically correct comparisons and assessments of the local context yield an entirely different picture. A scale-to-scale comparison of the buildings cited by Matassa shows that the buildings differ significantly in size, orientation, and plan (Fig. 14). If comparable elements, such as the largest room, are highlighted, discrepancies become even much more obvious: The large rooms of GD 80 and GD 57, which were both certainly roofed contra Mazur and Matassa,117 are much larger (242 and 211 m2, respectively) and not combined with annex rooms as in the houses GD 89 and GD 111 (66 and 68 m2, respectively). If original plans are compared, GD 80 clearly stands out in its design (Fig. 15). While the Establishment of the Poseidoniasts and the houses include shrines and altars, latrines, and other rooms, all of these are conspicuously absent in GD 80.118 The same applies when GD 80 is compared, scale-to-scale, to safely identified pagan sanctuaries, such as the Sarapieion A and B or the Samothrakeion, which include large multifunctional rooms (with or without benches), but also always features such as shrines, altars, or water crypts (Fig. 16). Nobody has ever claimed that benches were unique to GD 80 and would alone suffice to identify this building as a synagogue. While they are a somewhat generic feature in Delian architecture and Greek architecture in general, rooms of houses were not commonly decorated with benches along all of their walls, nor was this standard of the large assembly rooms in the meeting places of groups and associations.119 None of the buildings cited by Matassa includes a combination  115

Matassa 2007: 105–106. Matassa 2007: 96–97. 117 See Trümper 2004: 524 n. 24. 118 Cf. already Trümper 2004: 560 nn. 103–105 for these comparisons and the dimensions of large rooms in Delian houses and buildings of associations. 119 For Delian houses, Trümper 1998: 34–35, 38, 63, 213–214, 234: benches were only found in vestibules, and one bath room. The main room E of the Establishment of the Poseidoniasts or the main room 116

The Synagogue in Delos Revisited

115

Fig. 15. Scale-to-scale comparison of GD 57, 80, 89, 111; M. Trümper based on Bruneau 1970, pl. B; Delorme 1953, pls. 45, 47; Trümper 1998: 247, fig. 35; Trümper 2006: 23, fig. 2.

 f of the House of the Diadumenos (GD 61; identified as the meeting place of an association in Trümper 1998: 202; 2006: 17; 2011: 52 n. 12) were not decorated with benches.

116

Monika Trümper

Fig. 16. Scale-to-scale comparison of GD 80 and sanctuaries GD 91, 93, 96, 106, all before 88 BCE; M. Trümper based on Roussel 1916: 35, pl. II; Plassart 1928: 288, fig. 234; Trümper 2004: 559, fig. 34; Bruneau & Ducat 2005: 269, fig. 83; 271, fig. 85.

of benches with a centrally placed throne, however, and the use of thrones was exceptional in other Delian buildings, actually confined to the orchestra of the

The Synagogue in Delos Revisited

117

theatre.120 Moving the throne all the way from the theatre just in order to reduce the costs of furnishing is unreasonable because other seating furniture was available much closer by, in the gymnasium (GD 76) or in the Agora of the Italians (GD 52).121 Thus, the throne clearly mattered in the furnishing of GD 80 in its last phase of use, and suggests an intended hierarchy among the users of room A. It was moved with considerable efforts over a long distance, which was most likely only possible after the abandonment of the theatre in 69 BCE.122 Matassa concludes her article stating that, “while there is nothing that would exclude GD 80 from being a synagogue, there is not one piece of evidence that would suggest that it actually was a synagogue.”123 Other than the Samaritan inscriptions, nothing specifically pertaining to Jews or Samaritans would have been found on the island, and the only names safely associated with a Jewish or Samaritan context on Delos would be “Jason of Knossos and Artemidoros of Heraclea” (sic).124 One would like to know what such a piece of evidence might be in the second and first centuries BCE. While there is no inscription that safely identifies GD 80 as a προσευχή or συναγωγή,125 the architecture is much more idiosyncratic and revealing in the local context than Matassa recognized. Well-founded typological comparisons show that GD 80 shares some general features with local buildings (wall techniques, pavements and stucco, type of the furniture, Greek language, inscribed monuments etc.), which is not astonishing in the cosmopolitan setting of the trade port. At the same time, however, the design of GD 80 in all of its different phases is unique and suggests that GD 80 was not:  a private house;126  a meeting place of a pagan cultic association;127  a private pagan sanctuary;  or a public pagan sanctuary.128  120 IJO I: 215 propose that the throne could have come from the gymnasium GD 76, but there is no evidence for this: no throne has ever been found in a gymnasium whereas they are typical of theatres; cf. Fraisse and Moretti 2007: 73–74. 121 A large number of benches was found in GD 76; cf. Audiat 1970: 38. 101–118; Moretti 1997: 149; as well as in the Agora of the Italians GD 52; cf. Trümper 2008: 138–142. 122 Fraisse and Moretti 2007: 244–247; p. 246–247 they provide a list of pieces from the theatre reused in various buildings and areas in Delos. Most of the pieces were used in buildings of the northern part of the Theatre Quarter and buildings located around the Agorai of the Competaliasts (GD 2) and Theophrastos (GD 47), which were inhabited after 69 BCE and relatively close to the theatre; except for an architrave, which ended up in Mykonos, the throne certainly is the piece that travelled farthest on Delos. 123 Matassa 2007: 111. 124 Matassa 2007: 112; the correct names are Sarapion son of Jason and Menippos son of Artemidoros. 125 Such as a dedicatory inscription on some piece of architecture or a separate marble slab; or an honorary inscription for a generous donor who has financed the construction or some remodelling of the proseuchē, in a grammatically uncontested wording. 126 For the almost 100 houses in Delos, see Trümper 1998. 127 For the meeting places of foreign associations and ethnic communities in Delos, see Trümper 2002; 2006; 2011; and Trümper forthcoming. 128 The best overview of sanctuaries is still provided by Bruneau 1970.

118

Monika Trümper

GD 80 may not have been a synagogue either, but then it needs to be explained what it was used for. I argue that the remarkable mixture of typically local-Hellenistic features and conspicuously idiosyncratic traits may go back to patrons who were partially well integrated into the local community, and partially kept their own distinct traditions and customs. These may very well have been Jews or Samaritans, who built GD 80 as their meeting place. In my opinion, Matassa did not succeed in unravelling the myth of the synagogue on Delos with any convincing arguments, neither regarding the epigraphic nor the archaeological evidence. Her contribution confirms what has repeatedly become apparent in the debate about GD 80: a comprehensive (re)evaluation of this building requires manifold competences and should ideally be carried out by an interdisciplinary team, including scholars familiar with the specific architecture and archaeology of Delos and specialists in Jewish culture and history.129

Conclusion Since 2004, no publication has appeared that reassesses the identification of GD 80 based on substantial new fieldwork or a new examination of finds (including the inscribed monuments). Therefore, the debate circles around questions that obviously cannot be satisfyingly solved from published evidence, such as the nature of Theos Hypsistos or the significance of the wording ἐπὶ προσευχῇ/ἐπὶ προσευχῆι. In 2004, I made several suggestions for further research, which, unfortunately, have not lost their relevance in the years passed since and hence are restated here with only few qualifications.130 1. The top priority remains further excavation, for example to the north of room A and the to the west of the rooms A/B, in order to clarify the absolute chronology and history, and possibly function of GD 80. 2. Another top priority is a comprehensive study of the extant architecture, architectural elements, and associated artefacts. This has not yet been done and requires study of the inventories and storerooms. Two examples may illustrate the importance and potential of such research. Plassart briefly mentions the fragment of a sundial found in GD 80, which could not safely be identified in later publications of sundials. Further research may clarify this question and discuss the importance of this rather prestigious find for GD  129 Cf. Trümper 2004: 517 where I clearly outlined the limits of my expertise and the limited purpose of my study. 130 Cf. Trümper 2004: 594. Cf. also IJO I: 214.

The Synagogue in Delos Revisited

119

80.131 Recently, a vessel appliqué in the shape of a Nubian head, dated to the Imperial period, has been attributed to GD 80. This would add to the debated repertoire of finds with figured decoration found in GD 80, but the reference to its provenance is somewhat dubious because it is recorded as a find made in 1914, when no excavations took place in GD 80 or the Stadium Quarter.132 3. A trial trench at the findspot of the two Samaritan stelae or even full excavation of this building should be carried out to address the question of a possible second synagogue. 4. An in-depth study of features submerged along the coast is required. While submerged features had already been recorded in the early 20th century, and a geological project lately confirmed that the sea level has risen for about 2.50 m since antiquity, systematic underwater research has only recently been begun by Mantha Zarmakoupi. She identified a warehouse building with amphorae in situ, which suggests that the east coast housed important harbour and commercial facilities, similar to the west coast with the main harbour.133 Nothing has ever been noted to the east of GD 80, however, and nothing is visible on Google Earth, but future underwater investigation may yield important insights. 5. The period of use of GD 80 has received astonishingly little attention but is in fact remarkable. Because of finds Bruneau dated the abandonment of GD 80 to the end of the second or early third century CE.134 While the Imperial period has long been neglected in research as a period of presumably utter decline, it has recently attracted more attention. It has been pointed out that urban life in Delos was revitalized in the Augustan period.135 In 2016, Alain Bouet and Enora Le Quéré published two of the four Roman-type baths that were found in the city centre on the western coast, dating them to the fourth century CE.136 Further research might eventually allow for a comprehensive evaluation of the  131

Plassart 1914: 531: “Un cadran solaire conique, de marbre blanc, a été trouvé près du mur oriental; il est brisé en bas, a droite, en avant, mais trois lignes horaires sont conservées, ainsi qu’une partie des lignes solstitielles et équinoxiale, et permettent de constater qu’il a bien été établi pour la latitude de Délos.” Deonna 1938: 191 lists a fragmentary sundial (0.25 m maximum dimension) from the Synagogue under B 4669, without any further description; he then refers with “S.n.” to the fragment described by Plassart, which he could not, however, identify in the storerooms. The Berlin Sundial Collaboration database includes B 4669: Berlin Sundial Collaboration, Ancient Sundials, Dialface ID 503, Delos, Inv. Nr. B4669, 2014, Edition Topoi, DOI: 10.17171/1-1-4089, however, without reference to its findspot. This fragment shows three hour-lines and several solstitial and equinoctial lines, and thus fits Plassart’s description. 132 Barrett 2011: 391–392, 586–587 with reference to Chatzidakis 2004: 387: Delos, Museum B 5738; from 1914 (?) excavations. For the excavations Plassart 1914, 1916: 10–11. For the finds from GD 80, Bruneau 1970: 484–485. 133 Zarmakoupi 2014–2015: 125, Fig. 132. 134 Bruneau 1970: 484–491: the conclusive finds are the inscriptions ID 2331–2333, but particularly 41 lamps that are dated to the Imperial period. 135 Fraisse and Moretti 2007: 247–248 n. 26. 136 Bouet and Quéré 2015–2016.

120

Monika Trümper

entire Imperial period, particularly also with view to continuous use of the Stadium Quarter. So far, GD 80 seems to have been completely isolated in the Imperial period, which seems strange and hardly convincing. While there is no guarantee that the measures proposed here would yield conclusive results for determining the function of GD 80, they offer the only possibility to provide new evidence and thereby advance the endless debate about GD 80. Until then, I would side with the optimists rather than with the pessimists and would identify GD 80 as a synagogue.

Figures Fig. 1. Fig. 2. Fig. 3. Fig. 4. Fig. 5. Fig. 6. Fig. 7. Fig. 8. Fig. 9. Fig. 10. Fig. 11. Fig. 12. Fig. 13. Fig. 14.

Fig. 15.

Fig. 16.

Delos, Synagogue building GD 80: state plan; © EFA: Philippe Bruneau; cf. Bruneau 1970, pl. B. Synagogue building GD 80: reconstruction of phases 1, 3, 4, 5; Trümper 2004: 559, fig. 34; 563, fig. 37; 566, fig. 38; 568, fig. 39. Samaritan stele no 2; © EFA: Philippe Bruneau; cf. Bruneau 1982: 470 fig. 4. Samaritan stele no. 1; © EFA: Philippe Bruneau; Bruneau 1982: 468, fig. 2. ID 2329; M. Trümper, based on Bruneau 1982: 499, fig. 13. House A, Insula II, Quarter of the stade: reconstruction of first phase; based on Trümper 1998: 222, fig. 22. Synagogue building GD 80, view from NE in 1913, column ID 2328 in upper left corner; Plassart 1913, pl. V. ID 2331; © EFA: Philippe Bruneau; cf. Bruneau 1970, pl. IX, fig. 5. Uninscribed monument from GD 80; M. Trümper after Deonna 1938, pl. CX, figs. 969–970 (proportions corrected). ID 2333; © EFA: Philippe Bruneau; cf. Bruneau 1970, pl. IX, fig. 6. GD 80, schematic plan; M. Trümper, based on Matassa 2007: 95, fig. 2. Synagogue building GD 80, N wall of room C abuts stucco on W wall; Trümper 2004: 532, fig. 14. Comparison of GD 80 with Delian houses and the Establishment of the Poseidoniasts; M. Trümper, based on Matassa 2007: 106, fig. 12. Scale-to-scale comparison of GD 57, 80, 89, 111; M. Trümper based on Bruneau 1970, pl. B; Delorme 1953, pls. 45, 47; Trümper 1998: 247, fig. 35; Trümper 2006: 23, fig. 2. Scale-to-scale comparison of GD 57, 80, 89, 111, all before 88 BCE; M. Trümper based on Delorme 1953, pls. 45, 47; Trümper 1998: 247, fig. 35; Trümper 2004: 559, fig. 34; Trümper 2011: 83, fig. 30. Scale-to-scale comparison of GD 80 and sanctuaries GD 91, 93, 96, 106, all before 88 BCE; M. Trümper based on Roussel 1916: 35, pl. II; Plassart 1928: 288, fig. 234; Trümper 2004: 559, fig. 34; Bruneau and Ducat 2005: 269, fig. 83; 271, fig. 85.

The Synagogue in Delos Revisited

121

Bibliography Audiat, J. 1970. Le Gymnase. Exploration archéologique de Délos 28. Paris: Boccard. Barrett, C. E. 2011. Egyptianizing Figurines from Delos: A Study in Hellenistic Religion. Leiden: Brill. Belayche, N. 2011. “Hypsistos: A Way of Exalting the Gods in Graeco-Roman Polytheism”. In: North, J. A. and Price, S. R. F. (eds.), The Religious History of the Roman Empire: Pagans, Jews and Christians. Oxford: Oxford University Press, 139–174.  Binder, D. D. 1999. Into the Temple Courts. The Place of the Synagogues in the Second Temple Period. SBLDS 169. Atlanta, GA: SBL. Böhm, M. 2018. “Samaritanische Diaspora im Imperium Romanum bis ca. 200 n.Chr.” In: Alkier, S. and Leppin, H. (eds.), Juden – Heiden – Christen? Religiöse Inklusionen und Exklusionen im Römischen Kleinasien bis Decius. WUNT 400. Tübingen: Mohr Siebeck, 171–196. Bouet, A. and Le Quéré, E. 2015–2016. “Les thermes impériaux de Délos: L’infrastructure publique d’une ville ἄδηλoς?”. BCH 139–140: 417–462.  Bruneau, P. 1970. Recherches sur les cultes de Délos à l’époque hellénistique et à l’époque impériale. BEFAR 217. Paris: Boccard. Bruneau, P. 1982. “Les Israélites de Délos et la juiverie délienne”. BCH 106: 465–504. Bruneau, P. and Ducat, J. 2005. Guide de Délos. 4th ed. Athens: École française d’Athènes. Butz, P.A. 2007. “Inscribed Wreaths: The Interaction between Text and Monument in Two Euergetistic Stelae from Delos”. In: Mayer i Olivé, M., Baratta, G., and Guzmán Almagro, A. (eds.), XII Congressus Internationalis Epigraphiae Graecae et Latine: Provinciae Imperii Romani inscriptionibus descriptae. Barcelona, 3–8 Septembris 2002, Acta I. Barcelona: Institut d’Estudis Catalans, Univ. de Barcelona, 211–216. Catto, S. 2007. Reconstructing the First-Century Synagogue. A Critical Analysis of Current Research. London: T&T Clark. Chatzidakis, P. 2004. “Ειδωλιόμορφα σκεύη από τη Δήλο”. In: Drougou, S. (ed.), ΣT’ Eπιστημoνική Συνάντηση για την Eλληνιστική Kεραμική. Athens: ΤΑΠΑ, 367–392. CIRB = Struve, V. V. (ed.) 1965. Corpus Inscriptionum Regni Bosporiani. Moscow: Izdat. “Nauka”. Collar, A. 2013. Religious Networks in the Roman Empire: The Spread of New Ideas. Cambridge: Cambridge University Press.  Delorme, J. 1953. “La maison dite de l’Hermès, à Délos: Étude architecturale”. BCH 77: 444– 496. Deonna, W. 1934. “Mobilier délien”. BCH 58: 381–447. Deonna, W. 1938. Le mobilier délien. Exploration archéologique de Délos 18. Paris: Boccard. Fraisse, P. and Moretti, J.-C. 2007. Le théâtre. Exploration archéologique de Délos 42. Athens: École française d’Athènes. Gauthier, P. 1985. Les cités grecques et leurs bienfaiteurs. BCH Sup. 12. Paris: Boccard. ID = Inscriptions de Délos. IJO I = Noy, D., Panayotov, A., and Bloedhorn, H. 2004. Inscriptiones Judaicae Orientis, Vol. I: Eastern Europe. TSAJ 101.Tübingen: Mohr Siebeck. Kartveit, M. 2009. The Origin of the Samaritans. VTSup 128. Leiden: Brill. Kreeb, M. 1984. “Studien zur figürlichen Ausstattung delischer Privathäuser”. BCH 108: 317–343. Kreeb, M. 1988. Untersuchungen zur figürlichen Ausstattung delischer Privathäuser. Chicago: Ares.

122

Monika Trümper

Levine, L. I. 2005. The Ancient Synagogue: The First Thousand Years. 2nd rev. ed. New Haven, CT: Yale University Press. Ma, J. 2013. Statues and Cities: Honorific Portraits and Civic Identity in the Hellenistic World. Oxford: Oxford University Press. Matassa, L. 2006. “The Myth of the Synagogue on Delos”. In: Day, J. et al. (eds.), SOMA 2004: Symposium on Mediterranean Archaeology. Proceedings of the Eighth Annual Meeting of Postgraduate Researchers, School of Classics, Trinity College Dublin, 20–22 February 2004. BAR International Series 1514. Oxford: Archaeopress, 105–114. Matassa, L. 2007. “Unravelling the Myth of the Synagogue on Delos”. BAIAS 25: 81–115.  Matassa, L. D. 2010. “The Invention of the First Century Synagogue: A Critical Examination of the Archaeological Evidence”. PhD thesis, University of Dublin. Available at: http://www.tara.tcd.ie/xmlui/handle/2262/78073 (accessed 31 July 2018). Matassa, L. D. 2018. Invention of the First-Century Synagogue. Ed. by J. M. Silverman and J. M. Watson. ANEM 22. Atlanta, GA: SBL Press. Mazur, B. D. 1935. Studies on Jewry in Greece. Athens: Printing Office Hestia. McLean, B. H. 2002. An Introduction to Greek Epigraphy of the Hellenistic and Roman Periods from Alexander the Great down to the Reign of Constantine (323 B.C. – A.D. 337). Ann Arbor, MI: Michigan University Press.  Mitchell, S. 1999. “The Cult of Theos Hypsistos between Pagans, Jews, and Christians”. In: Athanassiadi, P. and Frede, M. (eds.), Pagan Monotheism in Late Antiquity. Oxford: Clarendon Press, 81–148. Mitchell, S. 2010. “Further Thoughts on the Cult of Theos Hypsistos”. In: idem and Van Nuffelen, P. (eds.), One God: Pagan Monotheism in the Roman Empire. Cambridge: Cambridge University Press, 167–208. Moretti, J.-C. 1997. “Les inventaires du gymnase de Délos”. BCH 121: 125–152. Nielsen, I. 2005. “Synagogue (synagogé) and Prayerhouse (proseuché): The Relationship between Jewish Religious Architecture in Palestine and the Diaspora”. Hephaistos 23: 63–111. Plassart, A. 1913. “La Synagogue juive de Délos”. In: Mélanges Holleaux. Paris: Picard, 199– 215. Plassart, A. 1914. “La Synagogue juive de Délos”. RB 23: 523–534. Plassart, A. 1916. “Fouilles de Délos: Quartier d’habitations privées à l’est du stade, 1912– 1913”. BCH 40: 145–256. Plassart, A. 1928. Les sanctuaires et les cultes du Mont Cynthe. Exploration archéologique de Délos 28. Paris: Boccard. Runesson, A., Binder, D. D., and Olsson, B. 2008. The Ancient Synagogue from its Origins to 200 C.E.: A Source Book. AJEC 72. Leiden: Brill. Roussel, P. 1916. Les cultes égyptiens à Délos du IIIe au Ier siècle av. J.-C. Nancy: BergerLevrault.  Shanks, H. 1979. Judaism in Stone: The Archaeology of Ancient Synagogues. New York: Harper & Row. Spigel, C. S. 2012. Ancient Synagogue Seating Capacities: Methodology, Analysis and Limits. TSAJ 149. Tübingen: Mohr Siebeck. Stein, M. 2001. “Die Verehrung des Theos Hypsistos: Ein allumfassender pagan-jüdischer Synkretismus?”. Epigraphica Anatolica 33: 119–125. Trümper, M. 1998. Wohnen in Delos: Eine baugeschichtliche Untersuchung zum Wandel der Wohnkultur in hellenistischer Zeit. Internationale Archäologie 46. Rahden/Westf.: Marie Leidorf. Trümper, M. 2002. “Das Sanktuarium des ‘Établissement des Poseidoniastes de Bérytos’ in Delos: Zur Baugeschichte eines griechischen Vereinsheiligtums”. BCH 126: 267–332.

The Synagogue in Delos Revisited

123

Trümper, M. 2006. “Negotiating Religious and Ethnic Identity: The Case of Clubhouses in Late Hellenistic Delos”. In: Nielsen, I. (ed.), Zwischen Kult und Gesellschaft: Kosmopolitische Zentren des antiken Mittelmeerraums als Aktionsraum von Kultvereinen und Religionsgemeinschaften = Hephaistos 24: 113–150. Trümper, M. 2008. Die “Agora des Italiens” in Delos: Baugeschichte, Architektur, Ausstattung und Funktion einer späthellenistischen Porticus-Anlage. Internationale Archäologie 104. Rahden/Westf.: Marie Leidorf. Trümper, M. 2011. “Where the Non-Delians Met in Delos: The Meeting-Places of Foreign Associations and Ethnic Communities in Late Hellenistic Delos”. In: van Nijf, O. and Alston, R. (eds.), Political Culture in the Greek City after the Classical Age. GroningenRoyal Holloway Studies on the Greek City after the Classical Age. Leuven: Peeters, 49– 100. Trümper, M. 2014. “The Honorific Practice of the ‘Agora of the Italians’ in Delos”. In: Griesbach, J. (ed.), Polis und Porträt: Standbilder als Medien der öffentlichen Repräsentation im hellenistischen Osten. Studien zur antiken Stadt 13. Wiesbaden: Reichert, 69–85. Trümper, M. forthcoming. “Cult in Clubhouses of Delian Associations”. In: Cazemier, A. and Skaltsa, S. (eds.), Associations in Context: Rethinking Associations and Religion in the Post-Classical Polis. White, L. M. 1987. “The Delos Synagogue Revisited: Recent Fieldwork in the GraecoRoman Diaspora”. HTR 80: 133–160. White, L. M. 1996. The Social Origin of Christian Architecture, Vol. I: Building God’s House in the Roman World: Architectural Adaptation among Pagans, Jews and Christians. Valley Forge, PA: Trinity Press International. White, L. M. 1997. The Social Origin of Christian Architecture, Vol. II: Texts and Monuments for the Christian Domus Ecclesiae in its Environment. Valley Forge, PA: Trinity Press International. Zarmakoupi, M. 2014–2015. “Archaeology in Greece 2014–2015: Hellenistic & Roman Delos. The City & its Emporion”. Archaeological Reports 61: 115–132.

II. Interpreting Material Remains and Literary Sources

The Synagogue at Magdala: Between Localized Practice and Reference to the Temple

Lutz Doering

1. The Magdala Synagogue as a Late Second Temple Synagogue In 2009, when part of the site of Migdal or Magdala was about to be converted into a pilgrimage centre by the Roman Catholic order of the Legionaries of Christ, a building was discovered that was subsequently identified as an ancient synagogue. At the time of writing this chapter, we are still awaiting the final archaeological report on the Magdala synagogue,1 hence much of our discussion is still tentative. However, since there has already been lively debate about the Magdala synagogue, with partly far reaching claims about the function of synagogues during the late Second Temple period, interaction with and indeed participation in this debate cannot be further delayed. Based on the brief note that the excavators on behalf of the Israel Antiquities Authority, Dina Avshalom-Gorni and Arfan Najar, published in 2013,2 we can assume that the building existed during three phases of construction. From the earliest phase, in the middle of the first century BCE, date the remains of a building that according to the excavators “was probably not used as a synagogue”. However, it served “as such in the two later phases”. The excavators have not yet released significant data about the second phase. However, the third phase is the one they describe in detail. The pottery of this stratum is dated from ca. 50 CE to the end of the first century, including vessels resembling those of the workshop of Kefar ananiah, limestone “measuring cups” as well as (so-called “Herodian”) clay lamps with a knife-pared nozzle. Regarding indicative coins, a coin from the year 43 CE was found in the foundation of the mosaic in the main room of the synagogue. Another coin, from the second year of the First Revolt, was found on the street, and a coin dating to 80 CE was discovered on the collapsed ceiling of  1

See http://www.magdala.org/2017/05/magdala-stone-featured-in-rome/ (accessed 27 February 2020), where the following was announced in 2017: “Currently, the archaeological team is looking forward to publishing a full report of their research about the Art of the Synagogue at the end of 2017.” To my knowledge, the report has not yet been published. 2 Avshalom-Gorni and Najar 2013. In the following, quotations attributed to “the excavators” refer to this Internet publication. In addition, a preliminary report about the “Mexican” excavations in 2010– 2012 has been published: Zapata-Meza, Garza Diaz Barriga, and Sanz-Rincón 2018.

128

Lutz Doering

the synagogue. In the eastern corridor of the main hall, a mosaic was discovered with a rosette in its centre, flanked by two panels with meander pattern. Since both panels are cut off at the connection with the northern and southern corridors, the excavators argue that the “the construction of the mosaic was interrupted shortly after the work had begun”. The synagogue and the surrounding quarter were abandoned and partly dismantled in order to build fortifications during the First Revolt. Following the excavators’ view, the third phase of the building identified as a synagogue thus operated during the final decades of the Second Temple period. Stefano De Luca has questioned this scenario and suggested that the synagogue might have been renovated and finally destroyed in the Bar Kokhba Revolt, although he presents no decisive evidence (Bauckham and De Luca 2015: 107–108). The building at stage three consists of a main hall, a narrow vestibule to its west, and a small room to the south. The situation of the entrance is unclear; currently, it is reconstructed in the west, but more likely it was placed at the southern side of the building (see Ryan 2017: 74), next to a decorated small room, about which more will be said. The main hall, measuring ca. 120 sq.m.,3 featured raised corridors on all sides. At their outward sides, the corridors were lined by a stone bench, which according to the excavators included “architectural elements in secondary use in its construction”. The inner sides are framed with basalt stones that might have served as a bench as well. A stylobate runs along all four of these inner sides; two fragmentary columns were discovered in situ on the stylobate, a third one was found in the middle of the room. According to Ehud Netzer, the main room would originally have featured six columns.4 The columns and part of the walls show remnants of panels of coloured fresco. Apart from the mosaic in the eastern corridor, the floor of the inner space of the main room was found covered with a “level of tamped stones”, which the excavators interpret as “the foundation of a mosaic pavement”. On it they found, in the south-eastern inner part of the main hall, the now famous decorated Magdala stone. The vestibule also had a corridor on all sides, lined by basalt stones. It seems that this room, too, has a bench on all sides. Furthermore, it features a limestone block with grooves, to be discussed later as well. A further stone of this kind was discovered in secondary use in the main hall. The small room in the south featured a mosaic and walls decorated with a colourful fresco. Over the past fifteen years or so, the archetypal criteria for identifying early synagogues have been discussed afresh. It is now clear that no consistent orientation can be determined for these synagogues, either to Jerusalem or to any other direction. It is also apparent that there is no special outward decoration applied to these buildings. While these two observations are sometimes referred to as “criteria” for identifying an early synagogue (for example by Netzer 2003: 282;  3 4

Avshalom-Gorni and Najar 2013. Binder 2014: 23: “approximately 11 x 11 m.” As reported by Avshalom-Gorni and Najar 2013.

The Synagogue at Magdala: Localized Practice and Reference to the Temple

129

2004: 22), they are of course of little help in establishing the identification and cannot be deemed to be operable criteria. The chief criterion is in fact the feature of a central hall surrounded by benches and columns, which create an unobstructed space in the middle of the hall. Perhaps every communal building in a Jewish settlement, unless it shows clear features pointing to a different use, may thus be identified as a “synagogue”.5 Netzer (2003: 282) had formulated two further features that were, however, disputed by David Amit (2007: 27–29): the presence of a niche like in Gamla or Jericho and the connection to a water supply, including a miqweh. Amit has shown that these cannot be taken as criteria given that most early synagogues feature no niche, and that the use postulated for it by Netzer, the storage of scrolls, could also have been fulfilled by a wooden cupboard or a side-room, as probably in the Masada synagogue. Moreover, the regular and specific connection of a synagogue with a miqweh in the vicinity is difficult to demonstrate. In this respect, Magdala presents inconclusive evidence. Preliminary reports suggest that at a distance of ca. 70 m from the synagogue there were three miqwa’ot—the first ones of a type, previously unknown in the Land of Israel, that was supplied by ground water (Reich and Zapata Meza 2014). However, these distant pools were certainly not built for the synagogue, and it is unclear in which relation they stand with it. There are indications of further pools in the area, but it is unclear whether this will modify the picture significantly. At most, it can be said that the Magdala building is situated in the broader vicinity of several miqwa’ot. On the basis of the main hall surrounded by benches and columns, which clearly indicates a public rather than a private building, in a settlement that by its material remains suggests a Jewish population, the Magdala building can be identified as an early synagogue. And based on the stratigraphic analysis presented above we are, until further notice, justified in assuming that the Magdala synagogue is indeed a first-century CE synagogue and probably operated in the final decades of the Second Temple period. Significantly, it is the first Second Temple synagogue in the Galilee excavated so far. It joins the list of first-century CE synagogues, which has grown in recent years due to new finds and excavations, and now includes Gamla6 in the Golan as well as Masada,7 Qiryat Sefer (Khirbet Badd

 5 The difficulty of clear-cut criteria for defining a building as an early synagogue is the particula veri of the methodological critique by Matassa (2018), who however overlooks the combined evidence of comparable buildings, which already by the time of completion of the study (2010) had exceeded the five cases (or four, if one takes out Jericho as unlikely) she discusses. See below. For the refutation of the earlier claim by Kee 1990 that there were no distinct synagogue buildings before 70 CE, see van der Horst 2002: 55–62 (with further literature). 6 According to Yavor 2010: 60–61, the synagogue is not earlier than the turn of the era, whereas S. Gutmann had suggested a date in the middle of the first century BCE (Levine 2014: 135–136). 7 Secondary use as synagogue by the rebels, after use as a stable in the first phase under Herod. Thus Netzer and Stiebel 2008: 1937.

130

Lutz Doering

‘Isa),8 Modi‘in (Khirbet Umm el-‘Umdan),9 Khirbet e-uwani,10 and possibly Herodium11 in Judaea. The synagogue referred to in the Theodotus inscription (CIIP I 9 = CIJ II 1404), according to the palaeography and the archaeological context of the inscription, dates from the late first century BCE or the early first century CE.12 A first-century CE date is discussed for the basalt floor (and a building to which it originally belonged) underneath the limestone synagogue at Capernaum.13 Identification of other buildings as early synagogues is debated14 or still tentative, as in the case of the “possible synagogue” at Khirbet Qana, which however may have been built after 70 CE, when refugees from the south arrived (McCollough 2015: 141–142). It is conceivable that further finds might be added in the future, although the actual spread of Second Temple synagogues continues to remain unclear.15 In what ways do the additional architectural features of the Magdala building align with other synagogues from the Hellenistic and early Roman period? The small, decorated square room to the south can be structurally compared with the room in the Masada synagogue (for which see Fine 1997: 30; Levine 2005: 62). The rich decoration of this room at Magdala underlines its importance. It may be considered to have been a place in which important objects, probably scrolls, might have been stored (Avshalom-Gorni and Najar 2013). In addition, attention should be drawn to the long vestibule with benches to the west. There is a similar room adjacent to the main hall, though opposite the main entrance, in the Gamla synagogue, which may have been used for study in small groups, although other communal functions would not be excluded (Yavor 2010: 55–56). Similarly, the “possible synagogue” at Khirbet Qana, though potentially a few decades later,  8 See Magen, Tzionit, and Sirkis 2004, for the date 205: “built during the reign of Herod—the late first century BCE to the early first century CE—and was in use until the Bar-Kokhba Revolt.” 9 See On and Weksler-Bdolah 2005, who distinguish a synagogue from the Hasmonean period, built on a hall from the Seleucid period, from a subsequent Herodian-period synagogue that was damaged in the lead-up to the First Jewish Revolt in ca. 66 CE but was rebuilt afterwards, as a Neronian coin dating to 68 CE in the refurbished floor suggests. 10 See Har-Even 2012. 11 Unless the use as synagogue dates only from the time of the Bar Kokhba Revolt, as suggested by Corbo, though most scholars opt for the First Revolt; see Runesson, Binder, and Olsson 2008: 35. 12 See Kloppenborg Verbin 2000: 277: “attests a synagogue building in Jerusalem, probably constructed in the early first century CE”, taking issue with the late dating suggested by Kee 1999. Jonathan Price comments on CIIP I 9: “palaeography, together with the archaeological context, have securely dated the inscription, and therefore the building to which it was attached, to the late 1 c. BCE or early 1 c. CE” (in Cotton et al. 2010: 54). 13 See Runesson, Binder, and Olsson 2008: 29, 32. 14 E.g., orvat Etri (Zissu and Ganor 2009), Khirbet Diab (Har-Even 2016) or Tel Rekhesh (Aviam et al. 2019), on which see the chapter by Zeev Weiss in the present volume. The suggestion of a Hasmonaean synagogue at Jericho (Netzer 1999, 2004: 11–17) has been widely questioned (Levine 2005: 72–74, with further literature), although Runesson, Binder, and Olsson 2008: 40–42 tend to accept it, and Hachlili 2013: 42 lists Jericho among Second-Temple synagogues. 15 Cf. Richardson 2004: 68, speaking of “a relatively limited presence of synagogues in first-century C.E. towns.”

The Synagogue at Magdala: Localized Practice and Reference to the Temple

131

may have featured such a study room as well (Richardson 2004: 66). Moreover, the Early Roman “public building” preceding the “Galilean”-type synagogue at Khirbet Wadi amam also had a similar side room.16 Intriguingly, the Theodotus inscription makes mention of δώματα, most likely “halls” or “other rooms”,17 which were connected with this synagogue in Jerusalem. In sum, the Magdala building is structurally an assembly hall with adjacent rooms for which there are parallels in other early synagogues. Due to its location at the northern limit of the city and its modest size it may have been only one of several local synagogues in Magdala (Bauckham and De Luca 2015: 109). It should be admitted, though, that there is so far not a single site from the Early or Middle Roman period in which more than one synagogue has been discovered, but then most excavated sites are not as extensive as Magdala.

2. The “Magdala Stone Table” I shall now turn to the “Magdala stone table”. This box-shaped stone, which measures ca. 67 cm x 57 cm and 40 cm in height,18 was found not far from the south-eastern corner of the interior space in the main hall, with its four legs standing on a bed for a mosaic, the shorter sides facing north and south, and the southern-facing legs being slightly higher. We cannot be absolutely certain that this was the stone’s original site; however, the placing on a mosaic bed and the calibrated position of the stone (on which presently) suggests a deliberate placement during the phase of renovation before its abandonment. The most striking feature of this stone table is the depiction of the seven-branched menorah together with other vessels. According to Rina Talgam, who has announced a comprehensive study of the stone’s decoration, this item is a three-dimensional depiction of the Herodian Temple.19 We shall have to await her publication for her interpretation of the details. Similar views, though with varying differences in detail, have been voiced by Mordechai Aviam, Donald Binder, and Richard Bauckham.20 Steven Fine (2017) published a strong critique of these approaches, finding only the menorah as indicative of the Temple (31) and taking the other  16

See the contribution by Uzi Leibner in this volume. See Price in Cotton et al. 2010: 55: “It is not clear what the δώματα were, but obviously ancillary rooms which were part of the guest-house.” Cf. LSJ s.v.: “house, … also: chief room, hall”; but probably not “housetop”, as in Deut 22:8 LXX and in the New Testament, e.g., Mark 13:15. 18 See Zeev Weiss’ contribution in this volume, and there Fig. 5. Amit 2013: 174 gives 67 x 55 cm and 33 cm in height, Hachlili 2013: 40 has “50 x 60 x 50”; Hachlili 2017: 247, “h-35 cm, w-55 cm, l-65 cm”; Aviam 2013b: 208: “0.6 m. long, circa 0.5 m. wide and 0.4 m. high”. – I saw the original stone both in the Vatican exhibition mentioned below and, earlier, in the repository of the National Treasures. I thank Dr. Dina Avshalom-Gorni for allowing me to see the stone there. 19 As per Kershner 2015. – I thank Professor Talgam for two conversions we had about the stone in the spring of 2015. 20 Aviam 2013a; 2013b; 2014; Binder 2014; Bauckham 2015; partly revised in Bauckham 2018. 17

132

Lutz Doering

designs as decorative elements. Similarly, Rachel Hachlili (2013: 40–41; 2017) deemed the art programme of the stone table predominantly decorative but in her later publication suggested that the menorah “probably signified not the Temple but the priestly offices and their duties. … hence, the Migdal stone might also have had some connection to a priest’s family or office” (2017: 252). In the following evaluation, I shall engage with these approaches and explain my own, which treads somewhat of a middle ground. I shall give a tentative interpretation based on the most evident features of the stone and also raise the question of its function. The depiction of the seven-branched menorah on a triangular stand in the middle of the front panel is undisputed among those who have commented on the stone.21 A few weeks after the Münster conference, the original stone was shown in an exhibition on the menorah throughout the ages in the Vatican (Leone 2017: 122). The occurrence of the menorah on the stone table is significant, given its rather sparse representation in extant Second Temple period contexts.22 However, the interpretation and significance of this occurrence of the menorah is debated. While Aviam, Binder, Bauckham, and to some extent Fine view it as a reference to the Jerusalem Temple (with differences in emphasis), Hachlili interprets it, as we have seen, as a reference to priests and their office. This is echoed by other scholars regarding some of the earlier discoveries of the menorah in Second Temple contexts, especially in Jerusalem and its environs (e.g., Rahmani 1994: 51: an “emblem” for the Temple priesthood). For Magdala, the presence of priests would in part depend on the date and historical value of the list of twenty-four priestly courses mentioning “Midgal Nunayya” as a place of the priestly rotation of Yehezkel; while some (e.g., Bauckham 2018; Notley 2014) affirm it as reflecting a testimony from the Hasmonean period, others (e.g., Leibner 2009: 416) question this claim. However, even if the presence of some priests at Magdala were assumed this would not, without qualification, explain the presence of the menorah on the stone. After all, in contrast to images in a priestly mansion in the Upper City of Jerusalem (today, the Jewish Quarter of the Old City) or on an ossuary potentially belonging to a priest, the Magdala menorah  21

In addition to the references in nn. 2, 19–20 see Hachlili 2013: 40; Fine 2017: 36–37; Schiffman 2017: 163–164; and see the contributions by Zeev Weiss and Judith Newman in the present volume. 22 The following are to be noted (cf. Hachlili 2001: 41–46): the coin of Mattathias Antigonus (Meshorer 1982: 92–94), five graffiti in Jason’s Tomb (Rahmani 1967: 73–74), a Jewish Quarter graffito on a plastered wall (Habas 2003), a Herodian sundial from the Temple Mount excavations (Yarden 1991: Plate 33), two charcoal menorot in the grottos of el-‘Aleiliyât in Naal Mikhmas (Wadi es-Suweinit) in the historical territory of Benjamin (Hachlili 2001: 46; Patrich 1985), three graffiti of a menorah at the entrance to an oil press at orvat Beit Loya in the Judaean Shephelah (Hachlili 2018: 17), the stoneetching of a five-branched menorah discovered close to the Temple Mount in 2011 (Reich 2014), a small number of menorot (some five-, some seven-branched) incised on ossuaries (Hachlili 2001: 45–46), of which some are however dated to the period between 70 and 135 CE by Rahmani (1994: 51–52), and possibly a graffito from a tomb at Kafr Mukhmas, again in the territory of Benjamin, which may however as well be from the second century CE (Raviv 2018).

The Synagogue at Magdala: Localized Practice and Reference to the Temple

133

presumably stood in a public space and thus can hardly be seen as marking out a priestly object. More generally, Rahmani’s view of the menorah as priestly “emblem” might require some nuance. Thus, the menorah and showbread table on the Antigonus coins, whether minted at the end or—as has been suggested lately23—at the beginning of his reign, while pointing to Antigonus’ priestly lineage in contrast to his rival Herod, is probably not merely a priestly “emblem” but emphasizes his rule as high priest and king (see the inscription), and this also means: his rule at and over the Temple. Moreover, on the Antigonus coins, the Jewish Quarter graffito, and the Magdala stone the menorah does not appear in isolation, which may suggest that it should be seen in the context of the other objects depicted. What are these objects? On the obverse of the Antigonus coin, it is obviously the showbread table, on the Jewish Quarter graffito, it is equally the showbread table and (as tentatively restored) the incense altar (see Hachlili 2018: 6–8, 19). There is thus precedence for the menorah alongside other Temple objects. While I agree with Steven Fine’s critique of the “monistic” interpretations of the Magdala stone by Aviam, Binder, Bauckham, and Talgam, I am nevertheless not convinced by his suggestion that the objects accompanying the menorah are a mere assemblage of motifs appearing disjointedly elsewhere in the art of late Second Temple Judaism (Fine 2017: 35–36). The above-mentioned examples provide some justification for the claim that the items on the front panel of the Magdala stone are an ensemble of Temple vessels. Another example might be the fivebranched menorah on an ossuary flanked by two amphorae (see below). Apart from these artistic presentations, we can also point to the fact that the sevenbranched candelabrum is often presented in association with other temple vessels, such as the showbread table or the incense altar, in literary sources ranging from the Torah (Exod 25:23–40; 37:10–28) to texts from the Second Temple period (e.g., Josephus, B.J. 1.152; C. Ap. 1.198–199 [Ps.-Hecataeus]; 2.106).24 Hence, I find Bauckham’s suggestion (2015: 116–117) attractive that the amphorae on both sides of the menorah represent libation vessels, perhaps modelled on the golden flagons (keshawot, keshot ha-nesekh) mentioned in Exod 25:29; 37:16 (cf. Num 4:7) within passages describing both the showbread table and the menorah, and given by Josephus under their Greek name σπονδεῖα (“libation vessels”; B.J. 1.152; A.J. 14.72). While amphorae may also serve decorative purposes in Second-Temple art,25 the depiction on the Magdala stone table seems to point to a more specific context. It may be relevant that amphorae and flagons on Jewish coins from the First and the Bar Kokhba Revolt have been similarly understood as Temple vessels (Meshorer 2001: 121, 146–147), as have the amphorae  23

See Hachlili 2017: 251 for reference. Methodologically speaking, Fine is correct that one should begin “with the object” before exploring “literary sources for parallels” (Fine 2017: 35), but I would argue that having started with a comparison of material objects one might then cautiously also consider literary sources for contextualization. 25 See the overview in Gersht and Gendelman 2016. 24

134

Lutz Doering

on later synagogue lintels (Gersht and Gendelman 2016: 178). Gersht and Gendelman (ibid.) have recently suggested that the amphorae on the Magdala stone table represent, more specifically, those σπονδεῖα (“libation vessels”) that, according to Josephus (A.J. 12.5), were made under the high priest Eleazar from the gold and precious stones sent by Ptolemy II Philadelphus. In my view, this is difficult to prove but would similarly point to a representation of Temple vessels on the stone table. That the Magdala stone shows two amphorae, one on the right of the menorah and the other one on its left, may be owed to symmetry. I am thus critical of Aviam’s “holistic interpretation” of the decoration of the stone, and in this respect his identification of the amphorae as the two bnei ha-yihar associated with the “lampstand (menorah) all of gold” of Zech 4:1–14 (Aviam 2013b: 212). There are several problems with this suggestion, regarding both the text and its application to the stone.26 Although zeitim in Zech 4:2–3 is sometimes explained as referring to jugs with olive oil (e.g., Haran 1968: 22) rather than, as is more usual, olive trees (e.g., Petersen 1984: 223–224), in the continuation of the pericope itself these zeitim symbolize two persons, usually taken as Joshua and Zerubbabel. In addition, the golden lampstand in Zech 4:2 is described as having a bowl on its top with seven “lamps” and probably seven “spouts”, “above” or possibly “by” (‘aleyha) which the two zeitim are placed, one on the right of the bowl and the other one on its left. This is clearly a different form of lampstand from the Temple menorah (e.g., Petersen 1984: 222–223) depicted on the stone. It is therefore doubtful that the jugs accompanying the seven-branched Temple menorah would recall the zeitim or bnei ha-yihar of Zech 4. What is the object underneath the depiction of the menorah? While a few authors have identified it as a square podium for the menorah,27 this would be a rather atypical setting: it is uncertain whether the menorah ever stood on a podium before its presentation in the Ara Pacis in Rome. The extant Second Temple period graffiti of the menorah that appear to show a square base, such as two of the graffiti from Jason’s tomb, the menorah on a sundial, and the fivebranched menorah recently discovered near the City of David, are potentially understood to depict a pyramidical menorah base as viewed from above (Reich 2014); alternatively, it has been suggested that they feature the menorah behind the showbread table (Notley 2014: 156). Notably, the Magdala stone shows a triangular base in addition to the square item, and Notley has pointed out that, for the five-branched City of David menorah, “a three-pronged base is visible within the etched square” underneath the candelabrum (ibid.). Hence, the suggestion that the square object underneath the Magdala menorah is either the showbread table (apart from Notley, see also Schiffman 2017: 164) or the incense altar (Aviam 2013b: 211–212; Aviam and Bauckham 2018: 139), shown from above  26 27

Aviam’s interpretation is deemed “somewhat bizarre” by Gersht and Gendelman 2016: 178. Hachlili 2013: 40; 2017: 252; Binder 2014: 27; De Luca and Lena 2015: 313–314; Fine 2017: 37.

The Synagogue at Magdala: Localized Practice and Reference to the Temple

135

and not in natural perspective, deserves some consideration. The slightly oblong shape might fit the showbread table better—according to Exod 37:10, it was two cubits long and one wide—and the border might correspond to the “rim” (misgeret) mentioned in this passage.28 On the other hand, it could be argued that the incense altar, of a square layout (Exod 37:25: one cubit long and one wide), was squeezed somewhat to fit the stone, and it has been suggested that the small squares in the edges might represent the horns of the incense altar.29 The whole front panel is flanked by two fluted columns carrying a vault, thereby creating the impression of a view onto a façade30 and into a building. We are probably on solid ground to interpret this as a stylized view into the Temple hall (heikhal).31 This does not require that the entire stone table be interpreted as a three-dimensional model of the Jerusalem Temple (see further below). If not on the front, as suggested by Notley and Schiffman, should we expect to find the showbread table elsewhere on the stone, since it is closely connected with the menorah and partly the incense altar in literary texts,32 Hasmonaean coins,33 and probably the Jewish Quarter graffito? Apart from Hachlili (2013; 2017) and Fine (2017), most previously published contributions to the discussion have identified the showbread table on the upper panel of the stone, although the suggestions vary greatly.34 The most sophisticated theory is that of Bauckham (2015: 121–128), who takes the twelve multiform objects as showbread loaves of different shapes that symbolize the four sub-groups among the sons of Joseph (that is, the Leah, Rachel, Bilha and Zilpah tribes). Although there are depictions elsewhere showing the showbread in slightly diverse forms,35 this theory is beyond  28

There are however depictions of the table in various shapes in ancient Jewish art. Cf. Exod 30:2–3. See Aviam 2013b: 212; Bauckham: 2015: 115–116; Aviam and Bauckham 2018: 139–140, who suggest that the lozenge-shaped insets might represent the burning incense. 30 So also Hachlili 2013: 40. 31 The columns represent either the Temple façade or the entry to the heikhal proper. On Bar Kokhba coins, the former is depicted with two columns on each side, while the one example of a silver didrachm shows only one column on each side and may represent the entrance to the heikhal; see Aviam and Bauckham 2018: 139. 32 1 Kgs 7:48-50; 2 Chr 4:19-22; 1 Macc 1:21–24; 4:49–51; Josephus, A.J. 3.139–150; 12. 250,318; C. Ap. 2.106; Philo, Mos. 2.101–105; associated with the menorah only: Exod 25:23–40 (but for the ensemble read Exod 26:35 together with 30:1–6); B.J. 1.152; A.J. 14.72. 33 The obverse of the Antigonus coin features it: Meshorer 1982: 94–97. 34 Aviam (2013b: 215–216; 2014: 138; 2018: 146) refers to the twelve variously shaped items as representing the loaves, a view on which Bauckham (2015: 121–128) builds his more elaborate one; see presently. Schiffman 2017: 164 appears to distinguish between the showbread table on the front and the depiction of the loaves on the top of the stone. Binder 2014: 39–40 limits the depiction of the showbread table to the two rectangles in the middle of the top side with diamond-shaped forms, the latter representing the showbread. He interprets the two bag-shaped items in the front as the incense altar, visually split apart, as if seen from two sides. One might as well interpret these bag-shaped items as the showbread table depicted this way, and the rectangles on top as the two piles of showbread that we find, for example, on the Antigonus coin. 35 E.g., in the Samaritan synagogue of el-Khirbeh, probably 4th century CE. 29

136

Lutz Doering

any comparative evidence, and Bauckham has retracted it in the meantime (see below). Talgam appears to interpret the large rosette in the centre as the curtain (parokhet) behind the showbread table, dividing between the heikhal and the devir, the Holy of Holies (Kershner 2015); it would be fastened to something that looks like palm-trees. However, I would like to point out that the decoration of the Magdala stone has schematic features: not every element seems to correspond directly to a distinct object in the Temple compound. On this reading of the Magdala stone, we would find selected references in a structure that is suited to the dimensions and shape of the stone. Hence, I would question the assumption that this is a model of the Temple. Rather, it seems to me a stone block with references to the Temple. Thus, the rosette is a popular motif in Jewish art of the period. It is found in connection with the decoration of the Herodian Temple, but also on other objects such as lintels, ossuaries, and stone tables. Assuming that the identification of the two items next to the large rosette as palm trees is correct, it may be noteworthy that a six-petalled rosette flanked by schematic palm trees is also depicted on two lintels from Gamla: one comes from the “basilica” area (perhaps from the “basilica” itself), the other one, broken and now lost, with mere remnants of palm branches, was found in the synagogue (Peleg-Barkat 2010: 167–169). A few ossuaries from about the first century CE show two or three six-petalled rosettes set off by one or more stylized palm trees (e.g., Rahmani 1994: nos. 226, 442F, 522, 848). Moreover, an eight-petalled rosette is found on the mosaic floor of the Magdala synagogue, not far from where the stone was found. A rosette flanked by palm trees might therefore indicate a connection with a design that was prominent in the Herodian Temple but also in other “Jewish” art of the period; it might evoke an association with the Temple in particular or, more generally, be a symbolic statement of Jewish self-identification (Peleg-Barkat 2016: 41–43).36 If we therefore allow for some schematic features of the stone, we consequently do not have to decide definitively whether the vaults on the long sides permit a view into the Temple courts, as Bauckham (2015: 120) and reportedly Talgam37 claim, or rather into the Holy of Holies, as Aviam (2013b: 212) and Binder (2014: 30–31) argue. I tend to agree with Aviam that the relief looks most like “an arcade behind an arcade”, rather than with Binder, who thinks of streams of fire issued from the heavenly throne (cf. Dan 7:10). But arcades might simply be the decoration of the outward structure given to the stone. Without base in my view is Bauckham’s earlier suggestion (2015: 120–122) that the round items towards the front are two of the rings by which the showbread table is carried (Exod 25:26–  36

Binder 2014: 33(–39) attributes to the rosette flanked by palm trees “cultic import” on account of Solomon’s decoration of the First Temple with such motifs (1 Kgs 6:29, 32, 35). Such an extrapolation to the First Temple remains doubtful given the prominence of the Second Temple period finds. 37 See Kershner 2015: “architecture of the Temple”.

The Synagogue at Magdala: Localized Practice and Reference to the Temple

137

27), thereby rendering the entire block a “representation of the table of the showbread” (121), a view Bauckham no longer entertains. Fine (2017: 37) agrees with Binder’s balanced conclusion (2014: 31–32) that these round objects are indeed what they look like, namely Herodian oil lamps, rather than censers, as Aviam (2013b: 121–213) has it. While such oil lamps would have been prevalent in Jerusalem, they were also used in the Galilee, usually having been brought there from Jerusalem where they were manufactured—as we have seen above, such “knife-pared” lamps were found in the latest stratum of the Magdala synagogue as well. The appearance of such lamps on the stone might lend them a symbolic significance “related to Jerusalem and the Temple” (Peleg-Barkat 2016: 40), but in doing so it would also connect local objects with references to the Temple. In a more recent contribution, Bauckham has changed his mind about the interpretation of the upper side. Taking all objects on this side, apart from the rosette, to occur in pairs, he suggests that there are seven pairs representing the seven kind of produce in the Land of Israel of which first-fruits were offered (Aviam and Bauckham 2018: 150–155). Bauckham takes the two objects looking like palm-trees (but interpreted as rakes by Aviam 2013b: 214–215) to represent dates, and the two bag-shaped items—viewed from the other side than previously suggested—to depict kantharoi symbolizing wine. The two sets of rectangular shapes are taken as barley and wheat loaves, and the three pairs of (ivy- or heartshaped) leaves are seen as representing pomegranates, figs, and olives. However, this suggestion is problematic on several counts. While the identification of the palm-trees is possible (but contested), it is uncertain whether the bag-shaped objects can really be interpreted as kantharoi. Moreover, it remains unclear whether the first-fruits of barley were indeed offered in the form of loaves (Lev 23:15 speaks merely of a “sheaf of wave-offering”), thus raising doubts about the representation of barley by rectangular “loaves” forms. Finally, it seems difficult to see in the similarly-shaped leaves representations of fruits as vastly different as figs, pomegranates,38 and olives. Bauckham’s conflicting theories show how difficult a consistent interpretation of the top side is. While being aware of decorative purposes of rosettes, Bauckham tends to view the rosette on the top side, overall formed by six petals emerging from a centre and six further petals linking the tips of the former—hence, altogether twelve petals39—as a symbolic hint at Israel. I would now like to turn to the narrow back side of the stone table. Set between three columns and placed in the inter-columnar space directly underneath the vaults between the columns, we find here two wheels or perhaps rosettes, with twelve pie-quarter shapes joined in pairs to form six semicircles in the lower  38

Why is not the fruit itself depicted, shown, e.g., on Herodian coins (Meshorer 1982: 2.27–28)? For representations of pomegranates and figs on ossuaries see Rahmani 1994: 43–44. 39 Bauckham lists some comparative examples for this type of rosette (Aviam and Bauckham 2018: 154), e.g., Rahmani 1994: nos. 224, 358.

138

Lutz Doering

intercolumnar space lining the columns and the floor. According to Aviam, Binder, Bauckham, and reportedly Talgam, this represents the wheels of the merkavah and two sets of six flames underneath.40 Set within a vaulted structure this is taken to be a representation of the Holy of Holies. In contrast, Hachlili (2013: 41) and Fine (2017: 36) opine that the wheels are in fact rosettes, in line with the ornamental programme of Second Temple period Jewish art. Nevertheless, I tend to side with the merkavah interpretation here. First, the two “rosettes” are not placed in the centre of the panel but appear in the upper part right underneath the vaults formed by the columns, which is a rather remarkable position. Second, the placement of the pie-quarter shapes underneath the “rosettes” requires an explanation but is not addressed by Fine.41 Third, these “rosettes” are very different from the large “petal rosette” on the top side of the stone table: the objects on the back appear to feature continuous rims, which might justify an interpretation as wheels.42 Moreover, Aviam (2013b: 213) has justly observed that the knobs at the joining point of the “spokes” are different, the left one being convex, the right one, concave. Thus, I see here some justification for considering these round objects “wheels” rather than ornamental “rosettes”. Given the uncertainty of the matter I shall consider a weaker and a stronger interpretation of the imagery: the weaker being a display of some Temple vessels, notably the menorah and some vessels connected with it, the stronger adding a depiction of the merkavah over flames, probably a reference to the Holy of Holies. All of this is set in vaults, which might represent some structures of the Temple but also lend structure to the decoration of the stone. There is some resemblance with ossuary decoration but, in detail, it is not as close a fit as Hachlili (2017) suggested.  40 Aviam 2013b: 213; Binder 2014: 28–30; Bauckham 2015: 118–120; Aviam and Bauckham 2018: 141– 143; Kershner 2015. See also Judith Newman’s contribution in the present volume. 41 They are not really similar to the three semicircles on an ossuary (Rahmani 1994: no. 222L, plate 32), as suggested by Hachlili 2017: 255. Aviam and Bauckham (2018: 141) point to the resemblance of the pie-shape elements to what they interpret as “flames” on the incense altar underneath the menorah. For a depiction of flames on bas-reliefs in Early Roman Jewish contexts see, e.g., those above the branches of the menorah and a crossbar on decorated “Darom” lamps from between the First and the Bar Kokhba Revolt; see Sussman 1982: 20, 31; Hachlili 2001: 441. 42 Rosettes on ossuaries (see overview in Rahmani 1994: 39–41) often have circles in a shape different from the inner petals, or multiple concentric circles (apart from few specimens where the petal-tips are linked by further petals, similar to the rosette on top of the Magdala stone; see above, n. 39). They look mostly different from the “spoke-and-rim” design of the two objects on the stone. The “wheel/rosette” from the Ophel discoveries by Raymond Weill (Shanks 2004: 119, Pl. XXVB) referenced by Hachlili (2013: 41) is not really similar. For the clear depiction of two wheels on bas-reliefs see the famous chariot motif in the synagogue frieze at Capernaum. While more widely interpreted as the Ark of the Covenant returning from the Philistines (see the object on wheels in the painting from Dura Europos, Fig. 6 in the contribution of Katrin Kogman-Appel in the present volume), Aviam suggests that this motif represents “the combination of the Temple and the divine chariot, similarly to, but also differently from, the depiction of the Magdala stone” (Aviam and Bauckham 2018: 149).

The Synagogue at Magdala: Localized Practice and Reference to the Temple

139

3. Does the Stone Bring Us into the Temple Courts? Contextualizing the Evidence What are we to make of this so far? The iconographic programme, as interpreted above, provides a link with the Jerusalem Temple. After all, then, this seems to vindicate Binder’s earlier argument that synagogues led their visitors “into the Temple courts” (Binder 1999, title) or, more precisely, that they “served as subsidiary sacred precincts that extended spatially the sacrality of the Temple shrine and allowed Jews everywhere participation within the central cult” (Binder 1999: 32; for critique see Levine 2004: 81–83). Not without gratification, Binder writes in 2014 that, based on the Magdala stone, theories dating the notion of the sanctity of the synagogue to the Late Roman and Byzantine periods “need to be revised: that even while the Jerusalem temple stood, the synagogues also were viewed as sacred places—indeed, as extensions of the temple itself” (Binder 2014: 44). Bauckham (2015: 130, 132) has made similar comments. Finally, Talgam is reported to hold a slightly weaker version of such a view, namely, that the stone “might have been intended to give the space an aura of holiness ‘like a lesser temple’” (Kershner 2015). Here, Ezek 11:16 is applied to the Second Temple synagogue as the rabbis would do later (e.g., b. Meg. 29a).43 I find these conclusions problematic. First of all, the Magdala stone, thus far the only specimen of its kind (see further below), should not be taken as invalidating the overall tenor of the literary texts from the Second Temple period or shortly thereafter. On the whole, these texts do not portray synagogues as extensions of the Jerusalem Temple. Instead, as is well known, they describe the activities in Jewish places of assembly, both in the Diaspora and in the Land of Israel, as chiefly consisting of the communal reading of the Torah (and, rarely, the Prophets) as well as the instruction in, and study of, scripture on the Sabbath, in various Jewish circles.44 In addition, we know that civic debates and even political intrigues might take place in such a building, as in the great proseuchē in nearby Tiberias (Josephus, Vita 277–291).45 Although the evidence is not fully clear,  43 Note that Amit 2013: 175, aware of the Magdala stone, clearly assigns the shift to the perception of the synagogue as miqdash me‘a to the time after the temple destruction and connects it with rabbinic enactments regarding statutory prayer. 44 See Josephus, C. Ap. 2.175; A.J. 16.43 (Nicolaus of Damascus on behalf of the Jews in Ionia); Philo, Spec. 2.62; Mos. 2.216; Hypoth. 7.12–13; Somn. 2.127 (Egypt); Legat. 156–157 (Rome); Prob. 81–82 (Essaioi [Essenes] in Judaea); Contempl. 30–31 (Therapeutae in Egypt); Theodotus inscription (CIIP I 9 = CIJ II 1404; Jerusalem); Luke 4:16–30; Acts 13:15. Cf. the permission of public reading of the Torah according to 4Q264a 1 4–5 par. 4Q421 13+2+8 2–3 as reconstructed by Noam and Qimron 2009: 57, 80– 88; further, perhaps the statement “to search/expound and to read in the book on the S[abb]ath” according to 4Q251 1–1 5 (if this is a positive statement), see Doering 1999: 246–248; finally, the disqualification of priests with a defective voice from reading (out) the Torah according to 4QDa (= 4Q266) 5 ii 1–3 par, 4QDb (= 4Q267) 5 iii 3–5 par. 4QDh (= 4Q273) 2 1. For the entire issue see Levine 2005: 147–158; Schiffman 1999. 45 There is no doubt that a categorical distinction between synagōgē and proseuchē cannot be made (contra Kee 1990); see Fine 1997: 26, who argues that by the first century CE the terms were synonyms.

140

Lutz Doering

there are reasons to believe that, apart from individual prayer that could optionally take place in synagogues (as well as at street corners: Matt 6:5), some form of communal prayer, at least in some places on some occasions, took place there as well. Pieter van der Horst (2002: 62–82) has discussed some of the evidence, including the use of the name proseuchē for Jewish synagogues, and Lee Levine (2005: 162–169) has suggested that there may have been general differences between the Diaspora, where communal prayer might have played a larger role in synagogues, and Judaea, where it seems to have played “little or no role”, while its introduction in distant places as the Galilee “in some limited fashion is a moot issue” (169). Levine has also pointed out that some groups in the Land of Israel such as the Qumran yaad developed communal forms of fixed prayer in the late Second Temple period (Levine 2005: 167). In addition, scholars like Esther Chazon and Daniel Falk have suggested that some prayer texts found at Qumran, such as Divre Hame’orot (4Q504–506) or Festival Prayers (4Q509), are nonsectarian;46 thus, communal prayer in Second Temple Judaea cannot be associated with sectarian circles alone, and it may have taken place in some of the synagogues. On the other hand, such prayer ought to be distinguished from the statutory, thrice-daily prayer that became mandatory in rabbinic Judaism and in which shaarit and minah are aligned with the times of the pre-70 morning and afternoon tamid in the Temple. In sum, the literary evidence suggests that later Second Temple synagogues were assembly halls for Jews with a number of purposes, chiefly the reading of, and instruction in, the Torah (and to some degree the Prophets), but including civic discussions and some form of private and communal prayer in some places, by some groups, and at certain times. Some Greek papyri and inscriptions from Ptolemaic Egypt call proseuchai or their compounds “holy”,47 and while it is likely that some similarity with nonJewish temples was perceived, the precise source of this ascribed holiness remains unclear. Similar terms are used in some passages by Philo and Josephus.48 For Philo, we should account for the attempt to present Jewish places of prayer as  46

Chazon 1992; Falk 1998: 61–63, 156–157. CPJ I 134 (late 2nd century BCE); JIGRE 9 (2nd century BCE [?]), using ἱερός; cf. JIGRE 24 (140– 116 BCE), speaking of a pylōn and thus evoking temple architecture. Some inscriptions dedicate proseuchai to the Ptolemaic ruler(s): JIGRE 13 (37 BCE [?]), 22 (246–221 BCE), 24, 25 (140–116 BCE), 27 (2nd or 1st century BCE), 28 (2nd or 1st century BCE, referring to the exedra), 117 (246–221 BCE). JIGRE 125 (47–31 BCE, replacement of a 145–116 BCE original) calls the respective proseuchē a place of asylum, similar to Egyptian sanctuaries. See Fine 1997: 26–27, 177–178. Roman inscriptions from Egypt call the synagogue a “holy place”, using ἅγιος: JIGRE 17 (Roman), cf. the later Roman inscriptions 16 and possibly 127 and, from Macedonia, the Stobi inscription IJO I Mac1 (second half 2nd–first half 3rd century CE). 48 For Philo, see Flacc. 48, where he calls the prayer houses “sacred precincts” (ἱεροὺς περιβόλους) in which the Jews show their gratitude to their benefactors; and Prob. 81, where he says of the Essenes that they “arrive at sacred places that are called synagogues” (εἰς ἱεροὺς ἀφικνούμενοι τόπους, οἳ καλοῦνται συναγωγαί). For Josephus, see his deployment of terminology referring to the “sacred” in B.J. 4.407–408 (ἱερά); 7.44–45 (ἱερόν), 143–144 (ἱερά); cf. B.J. 2.289: the pagan bird sacrifice in front of the Caesarea synagogue “defiled the place” (μεμιασμένου δὲ τοῦ χωρίου). 47

The Synagogue at Magdala: Localized Practice and Reference to the Temple

141

compatible with Graeco-Roman categories.49 Hence, the “holiness” of synagogues in the Diaspora may be part of an attempt to invest them with an aura that would be respected in the Greek and Roman world. Moreover, Andrew Krause has argued that the rhetorical dimension should also be accounted for regarding the presentation of synagogues as “holy” in Josephus’ Bellum Judaicum insofar as Josephus “is magnifying the importance of these structures for the sake of his intended Roman audience”.50 For the Land of Israel, however, the evidence for the “holiness” of synagogues is much sparser. Steve Fine (1997: 33) has suggested that [t]he two possible forms of synagogue holiness in latter Second Temple period Palestinian synagogues were the sanctity of the Sacred Scripture and the application of Temple forms of communal worship. The extant evidence stems from marginal groups within Jewish society, the defenders of Masada, the Qumran sectarians, and Philo’s Essenes.

However, the relevant evidence will be even more limited if we omit Philo’s comments on the holiness of the Essene synagogues: this is more readily explicable as this Diaspora author’s view than as deriving from Judaean Essenes proper. As to “the application of Temple forms of communal worship”, it is clear that the yaad attested to in Dead Sea Scrolls transposed aspects of Temple worship to the community.51 Whether this can be applied, in terms of locale, to the sect’s “synagogue” is unclear. Intriguingly, Josephus’ report on the Essenes speaks of their approach of the dining room “as if it were some [kind of] sanctuary”.52 In the Dead Sea Scrolls, the most relevant passage is CD-A 11:21–12:1: No one who enters the 11:22 House of Prostration (‫ )בית השתחות‬shall enter as washed impure (‫)טמא כבוס‬. When the trumpets for assembly are blown, 23 let him advance or retreat, but they should not stop the whole service (‫[ )את העׄב ׄודׄה כולה‬in the h]ouse— 12:1 it is (a place of) holiness (‫)]?[ בב[ ׄיתקודש הוא‬.53

 49 Thus, contra Fine 1997: 29, 31, the statement in Prob. 81 about the synagogues of the Essenes as “sacred places” says more about Philo’s presentation of these places than about the sanctity that Essenes ascribed to their synagogues. 50 Krause 2017: 175–188, quotation: 188. In contrast, the mention of “sacrifices” (θυσίας) in the acta pro Iudaeis in A.J. 14.227, 260, according to Krause (86–87), may refer to some kind of activity deemed sacrificial, for which Jewish sacrifice at other places outside Jerusalem, e.g., Leontopolis, may be compared. 51 See 4Q174 (= 4QFlor) 1+2+21 i 6–7 (“temple of men/man/Adam”); 1QS V 4–7; IX 3–6; XI 7–9, each applying the image of a holy building to the community; further 1QS VIII 4–11, potentially referring to an elite group (“a holy house for Israel and a foundation of a holy of holies for Aaron”). See brief discussion with further references in Doering 2016: 258–262. 52 B.J. 2.129: καθάπερ εἰς ἅγιόν τι τέμενος. 53 My reading cautiously follows Solomon (1997: 19 with n. 68), where he accepts the reconstruction ‫ ב[ ׄית‬suggested by Steudel (1993: 51–52) but rejects her further, widely accepted restoration ‫כי ב[ ׄיתקודש‬ ‫“( הוא‬for it is a house of holiness”) because he claims there is space for only one further character before ‫ב[ ׄית‬. Due to varying letter spaces in the manuscript from the Cairo Genizah, this matter cannot be settled definitively, though one would have to assume some erasure to have taken place in upholding Steudel’s

142

Lutz Doering

This passage of the Damascus Document Law Code most likely does not come from the yaad but rather from a preceding group. Like the law on the Sabbath sacrifice preceding it (CD-A 11:17–21) and the ban on sex in the Temple City following it (12:1–2), it probably does not (yet) suggest a complete withdrawal from Temple service. On the contrary, CD-A 11:21–12:1 is replete with references to the Temple and seems to relate, not to the community as temple, but to a building called “House of Prostration”, with 4QDf (= 4Q271) 5 i 15 supplying the definitive article “the House of Prostration” (‫)בית ההשתחוות‬. Most likely, this refers to the Jerusalem Temple, and not to a “sectarian synagogue”.54 If so, explicit evidence for synagogue-style assembly rooms of the yaad or their forerunners as “holy” is lacking, although some meetings of the yaad would clearly have adopted functions of the Temple. What remains of Fine’s suggestions is the possibility that the sacredness of scriptures stored, handled, and read in early synagogues may have lent them an aura of “holiness” (van der Horst 2002: 77–80). The scriptures are explicitly referred to as “sacred” in several of the texts dealing with ancient synagogues,55 and even when the texts simply refer to “laws” or the like, it is clear that these were regarded as holy by the Jews.56 Fine (1997: 30) points to the “obtrusive position” of the room in the Masada synagogue in which fragments of Deuteronomy and Ezekiel were found, which according to Fine may not only have been a genizah but also a room housing the biblical books of the synagogue. At the very least this would emphasize the centrality of scriptures for the Masada synagogue.57 It has also been suggested that one of the functions of miqwa’ot in the vicinity of  suggestion (see an image of the manuscript at http://cudl.lib.cam.ac.uk/view/ MS-TS-00010-K-00006/11, accessed 27 February 2020). 54 See Solomon 1997; Anderson 2000: 991–992; cf. Baumgarten in Charlesworth et al. 1995: 51 n. 178: “an area of the Temple”. Even Steudel, who suggested that the term “houses of prostration” refers to “duplicates of the Temple” (note the use of the plural here!), observed the unusual terminology for sectarian assembly rooms and remained vague regarding their connection with evidence from the Serekh texts (1993: 57–58). Hempel 1998: 153–155 assigns this rule to the somewhat elusive category of “miscellaneous halakhah”, which is however not clearly sectarian. Schofield 2016 deems the reference of the term “House of Prostration” to an otherwise unknown synagogue “unlikely” (134); however, her cautious suggestion that it might refer to an altar at Qumran (134–135), as considered by Magness 2016, meets the objection that the regulations of the CD Law Code do not necessarily refer to the site of Khirbet Qumran. 55 See Philo, Hypoth. 7.13; Somn. 2.127; Josephus, A.J. 16.164 (in Augustus’ letter). Cf. Philo’s statement on the “divine inspiration” (κατοκωχῆς ἐνθέου) of the national laws in Prob. 80, immediately preceding the mention of the synagogues of the Essenes. 56 “Law”: Josephus, C. Ap. 2.175; Theodotus inscription (CIIP I 9 = CIJ II 1404); “customs and laws”: A.J. 16.43; “national laws”: Philo, Prob. 80; “national philosophy”: Mos. 2.216; Legat. 156; cf. Spec. 2.62. See Wischmeyer 2004. 57 There is some debate about the place where scriptures were stored. It is sometimes claimed that the Masada synagogue might have been the exception in providing a room for storage of scrolls, and that otherwise these would have been kept by wealthy patrons (e.g., Wise 2015: 489 n. 64). However, the Magdala synagogue might provide further evidence for such a room (the decorated small side room), and the reference to “rooms” (δώματα) in the Theodotus inscription (see above) should also be noticed in this regard. Magen, Tzionit, and Sirkis 2004: 205 report also the evidence of two side rooms at the Qiryat Sefer synagogue, one of which “may have served for the storage of holy books.”

The Synagogue at Magdala: Localized Practice and Reference to the Temple

143

synagogues was for men intending to handle the Torah scrolls, who had to purify from short-time impurities such as seminal emission.58 Importantly, therefore, if any sacredness was ascribed to synagogues in the late Second Temple period— and the evidence remains sparse—it was most likely conferred by the holiness of scriptures used and—perhaps—stored in them. It is thus not the space of the synagogue as such59 that is holy, at least in Palestine, nor is there evidence that the synagogue was deemed holy by derivation from the sacred space of the Jerusalem Temple. In terms of the placement of the Magdala stone table (if found in situ) it should also be observed that, while the side with the menorah is visible to a person facing Jerusalem, the side with the potential chariot is in the opposite direction. Thus, the stone with its all-round decoration is in keeping with, and potentially even reinforces, the orientation of Second Temple synagogues towards the middle of the hall. This is suggestive of the idea that the stone played a role in the localized practice of the congregation gathered whilst exhibiting a reference to the Temple. What localized practice? This leads to the quest for the function of the stone.

4. The Function of the Stone Table According to Fine (2016: 22; 2017: 38), the function of the Magdala ashlar is “still a mystery”. However, I think we ought to be more exerting here. One relevant suggestion is that it served as a bema. This might be meant by Avshalom Gorni and Najar 2013 when they speak of a “prayer table”, although prayer, as we have seen, was not the main activity in Palestinian synagogues in the late Second Temple period. Hachlili thought it may be possible “that the reader knelt when reading the scroll” (2013: 41). However, this is unlikely: if Luke 4:16 is not completely misguided about this detail it shows that the reader “stood up to read” (ἀνέστη ἀναγνῶναι; here: Jesus, reading from the Prophets). Alternatively, Aviam (2013a; 2013b) has suggested that the stone was a base for a lectern that would have rested on four wooden or stone legs fixed to the corners of the stone, which show round-shaped rims. Meyers and Chancey (2012: 211–212, 216) make a similar suggestion. Aviam’s view is deemed “plausible” by Bauckham (2015: 114), although he and De Luca also question whether the stone would have any function beyond making the connection with the Temple visible.60 However, Binder (2014: 42) has justly questioned whether the fastening of  58 Thus Adler 2008, although Adler now thinks that this was no particular reason for the placement of miqwa’ot near synagogues, but that stepped pools were so common that they should also occur close to synagogues (paper in the Hellenistic Judaism section at the 2019 SBL Annual Meeting in San Diego). 59 As suggested by Catto 2007: 110–112. 60 Bauckham and De Luca 2015: 111. According to Zangenberg (2016: 75* n. 9), De Luca, in a personal communication September or October 2013, called the stone an “offering table” (cf. also the following

144

Lutz Doering

the table legs postulated by Aviam would have been stable enough. In addition, there is in my view the distinct possibility that the rims at the corners are in fact remnants of horns that have been chipped off at some point. These might be a reference to the incense altar (Exod 30:2) or the burnt offering altar (Exod 27:2) but may serve also decorative or other functional purposes (see below). If the rims in the corners really come from horns with which the stone has been originally fitted, then the theory that any sort of legs were fixed to it in the corners becomes implausible. Observing the “broken corners” as well, Hachlili initially (2013: 41) wondered whether the stone itself might have served as an incense altar. Binder (2014: 42– 43) suggests as further functional possibilities: a base for a lampstand, a seat for the archisynagōgos, and a base for a charity box. These latter theories, together with Hachlili’s suggestion that the stone served as incense altar, are problematic, since such uses may cover, or threaten to destroy, the surface. In addition, we have no evidence for the liturgical use of incense outside the Temple in the late Second Temple period (Aviam and Bauckham 2018: 156). While Bauckham originally thought that the function of the stone was simply to represent the Temple (Bauckham and De Luca 2015: 111), he has more recently suggested that it served as a table on, or next to, which the first-fruits were gathered to be taken to the Temple (Aviam and Bauckham 2018: 155–157). However, while it is not implausible that first-fruits were gathered regionally, we find no (other) evidence that this would have taken place within synagogues in the Land of Israel: the rabbinic imagination in m. Bik. 3:2 even states that the inhabitants of a region (ma‘amad) would assemble in the main city of the region with their first-fruits, spend the night outside in the street, and then make their way to Jerusalem early in the morning. Bauckham further suggests that the original horns of the stone might have indicated that the stone represented the sacrificial altar, before, or next to, which the first-fruits would have been placed in Jerusalem (Deut 24:4; m. Bik. 3:6). However, this would be a far-reaching conclusion, not fully consistent with the iconographic programme of the stone. Most importantly, however, it would depend on Bauckham’s interpretation of the upper side of the ashlar, which is highly uncertain and even unlikely. What, then, might be a plausible function of the stone? While I take issue with Aviam’s suggestion that the stone might have been a base for a Torah reading table, I agree with him on the crucial functional connection between the stone and the Torah. In my view, the most likely suggestion is that the stone served as a table for the temporary placement of items,61 in particular the scripture scrolls  note), but this view is not found in Bauckham and De Luca 2015: 111, where the explanation that “offerings were placed” on the stone “before being taken to the temple” is mentioned as a suggestion made by “[o]thers” and contrasted by the view stated above. 61 Cf. the suggestion by De Luca and Lena 2015: 317: “an offering table used for different purposes”.

The Synagogue at Magdala: Localized Practice and Reference to the Temple

145

that had been brought in, most likely from the small decorated side room, a suggestion made in different ways by the late David Amit, the late Rachel Hachlili in her second publication on the stone, and now in detail by Zeev Weiss.62 One might suggest that the “horns” in the corners prevented such scrolls from falling off. Can we find comparative evidence for such a use? Possibilities appear to exist: Rahmani (1990: 208–213) and Spigel (2012: 40) have suggested a similar use of a stone seat, the so-called “cathedra of Moses”, appearing in later synagogues (e.g., Chorazin), a suggestion explicitly adopted for the Magdala stone by Hachlili (2017: 263), who even takes the broken corners as remnants of handrails similar to the Chorazin “cathedra of Moses”. Among Second Temple synagogues we find a raised, “special” seat in the northern stone bench of the Herodian phase of the synagogue at Khirbet Umm el-‘Umdan, which according to the excavators may have served for prohedric seating or for placing the Torah scrolls during services (On and Weksler-Bdolah 2005: 111). Another installation that might be relevant for comparison is the so-called “central stylobate” of the Gamla synagogue, which may have been related to the place of the Torah reading, although its precise function is debated (Yavor 2010: 47, 51). Brief reference must be made here to the stone table discovered in the synagogue of orvat Kur in 2012, in secondary use from the late sixth/early seventh century at the place it was found, but of uncertain date and primary function. Aviam (2016) suggests that this was (in his view: also) a base for a Torah reading table, here in the form of a monopodium.63 In contrast, Jürgen Zangenberg (2016; 2019) cautions against the assumption of too close a relation between the orvat Kur stone and the Magdala stone. Zangenberg argues in detail that the iconography of the orvat Kur stone most closely relates to Graeco-Roman banquets and tentatively interprets it as an “offering table”, perhaps related to dining in one of the rooms of the synagogue. In view of the different iconographic programme and also its uncertain date, I am sceptical that the orvat Kur stone table can help us much in determining the function of the Magdala stone table—except perhaps for the general observation that “stone tables” were a potential though not very frequent element of synagogue furnishing, from the Early Roman to the Byzantine period. In this connection, we need to reflect briefly on two other stone blocks that were discovered in the synagogue building in Magdala. As mentioned above, these are limestone blocks with grooves on both of the narrow sides of their upper surface but otherwise undecorated. The first one appears to be in secondary use  62

Amit 2013: 175; Hachlili 2017: 262–263; and see Zeev Weiss’ contribution to the present volume. Aviam’s argument is not free from circular reasoning: “I am absolutely convinced that there was a Torah reading table in every synagogue, of which we have no evidence. In most cases these would be simple wooden tables, but some were probably beautifully decorated. The suggestion that there were decorated bases for wooden reading tables is strengthened by the two newly discovered decorated stone bases …” (Aviam 2016: 81*). 63

146

Lutz Doering

behind the column closest to the decorated stone table, the other one in the narrow vestibule of the synagogue, in the middle of a floor of basalt flagstones. The excavators suggest that the latter stone was used “in secondary use … as a base for a chair or table” (Avshalom-Gorni and Najar 2013). An exalted chair in a narrow vestibule surrounded by benches would perhaps make less sense than a table; one could also consider the possibility that the block itself served as such as a table. It may be suggested that the table in the vestibule was used to deposit items, such as scrolls, lamps or vessels during meetings. Depending on whether this stone was put here in secondary use, as the excavators suggest, or originally, the grooves on the sides may have been used or designed to prevent items from falling off, or perhaps for the affixation of some wooden structure on which items would have been placed.64 This could be accommodated by the suggestion that the bench-encircled side room was used for small study groups, as has similarly been suggested for the ancillary room of the Gamla synagogue (Yavor 2010: 56– 57). As to the limestone specimen now placed in the main hall of Magdala synagogue, one might speculate whether it was the predecessor of the current one in the vestibule or even the predecessor of the decorated stone table. Coming back to the latter, the Magdala stone, in my view, links particularly the localized communal practice of reading and studying the Torah with the Jerusalem Temple. This does not confer the sanctity of the Temple to synagogues nor does it allow Jews at a distance to participate in the central (sacrificial) cult. However, there is a connection between the distinct activities of assemblies in synagogues and the work of the Jerusalem Temple, and this link was likely provided by the Torah that was present in both. To be sure, Torah reading was not the dominant element of the priestly-led activities in the Temple, although the Torah certainly played a role in the Second Temple, in various respects (Hayward 2013: 335–344). First of all, it is likely that master copies of Torah scrolls were deposited and kept in the Temple, from which other Torah scrolls would have been corrected; this would create a notional link between the Torah scrolls deposited in the Temple and those used at various other places, inter alia, in synagogues (Tov 2008: 177–182). In addition to the Torah, the Temple library contained other authoritative writings that could be consulted there as well (2 Macc 2:13–15). According to Josephus, a Torah scroll from the Temple spoils was paraded in Vespasian and Titus’s triumph in Rome (A.J. 7.148–150, 162). Rabbinic texts reflect memories of one or more Torah scrolls that had been kept in the Temple compound.65 In terms of public reading of the Torah or portions from it, the key passage Deut 31:10–13 has Moses command the Israelites to assemble (Hebrew: haqhel)  64 Alternatively, Ryan 2017: 74 has suggested that the grooves might have held the rollers of a scroll opened for reading (similarly Anders Runesson, in private communication), although this seems more speculative. 65 E.g., m. Kel. 15:6: a Torah scroll “of the Temple courtyard” (‫ ;)ַהֲע ָז ָרה‬y. Ta‘an. 4:2, 68a: three Torah scrolls in the Temple courtyard.

The Synagogue at Magdala: Localized Practice and Reference to the Temple

147

the people at the Temple during the festival of Sukkot at the end of the Sabbatical year, in order to read “this law before all Israel in their hearing”. Ezra is said to have convened such a reading, according to First Esdras and Josephus clearly at the Temple, when he read the Torah from a wooden platform, and the Levites instructed the people in the Torah.66 Later sources state that during the haqhel ceremony in the Temple the high priest (Josephus, A.J. 4.209–211) or the king (m. Soah 7:8) read (as per the Mishnah: portions from) the Torah.67 While on account of the mention by Josephus the haqhel reading is usually taken as historically plausible during the late Second Temple period, greater uncertainty surrounds further references in the Mishnah to the high priest’s reading of portions from the Torah in the Temple on Yom Kippur (m. Yoma 7:1–3; m. Soah 7:7),68 as well as to the officiating priests’ reciting of short Torah portions in the Chamber of the Hewn Stones after the slaughter and preparation of the tamid sacrifice (m. Tamid 5:1).69 Therefore, while some details of actual practice in the Temple must remain uncertain, there can be no doubt that the Temple was seen as the preeminent locus of Torah, with which the localized practice of Torah reading in synagogues could notionally connect whilst being distinct from the specific sacrificial activities dominant in the Jerusalem Temple. A comparison with the institution of the ma‘amadot (m. Ta‘an. 4:2–3), which has been seen by some at the beginning of communal Torah reading at the citygates (e.g., Levine 2005: 38–40), is instructive here. According to the Mishnah, the twenty-four mishmarot of priests and Levites (cf. 1 Chr 24) had corresponding sections of ordinary Israelites, and those of them who were unable to “stand by” the sacrifices (this seems to be the meaning of ma‘amad) administered by their respective mishmar of priests in the Temple would locally fast from Monday to Thursday of the week of service (which would occur about twice a year) and assemble for sequential reading of the creation account (Gen 1:1–2:3) from Sunday to Friday of that week. Here, we have a more direct link with Temple service, though the local activity of the ordinary Israelites is still distinct from the practices in Jerusalem by its focus on fasting and Torah reading (and does not, e.g.,  66

Neh 8:1–8, 18 (on the square before the Water Gate, of uncertain location, but perhaps not far from the Temple); 1 Esdr 9:3–55 (on the square before the eastern gate of the Temple); Josephus, A.J. 11.154– 155 (in the open court of the Temple near the east-facing gate). – Cf. the reading of the law book in the Temple under Josiah according to 2 Kgs 23:2; 2 Chr 34:30; Josephus, A.J. 10.58–63. 67 According to m. Soah 7:8, the following portions from Deuteronomy were read: Deut 1:1–6:3; 6:4– 9; 11:13–21; 14:22–29; 26:12–15; 17:14–20; 27:15–28:69. For the relation between Josephus’ account and m. Soah 7:8, see the comments by L. H. Feldman in Flavius Josephus Online, ed. S. Mason: https://referenceworks.brillonline.com/entries/flavius-josephus-online/*AJ_4_00209 (accessed 23 March 2020). 68 The passages are related to the ritual complex of Yom Kippur: Lev 16; 23:26–32; Num 29:7–11; cf. Runesson 2001: 209–212, who deems this reading a post-70 CE rabbinic invention, in contrast to the haqhel reading. 69 The passages are the Decalogue (Exod 20:1–17; Deut 5:6–21), the shema‘ (Deut 6:4–9), Deut 11:13– 21, and Num 15:37–41. For the location of the recitation see m. Tamid 4:3.

148

Lutz Doering

entail reflection on the sacrifices).70 In addition, this reading is apparently unrelated to Sabbath gatherings, and it applies only to the Israelite ma‘amad of the respective priestly mishmar, thus cannot be generalized for regular sabbatical Torah reading in Second Temple synagogues. It should be noted that it cannot be firmly established that the reading practice of the ma‘amadot goes back to the Second Temple period (cf. Runesson 2001: 311), from which we have no direct testimonies about it. In sum, the regular synagogal assemblies on Sabbaths (and festivals) do not extend the Temple cult to the synagogue, but rather connect the local practice of Torah reading and interpretation with the Temple, where— alongside the cult—master copies of the Torah were kept and read as well as interpreted by leading members of the Jewish people. I wish to conclude this section with a few comments on the potential depiction of the merkavah. If this interpretation is valid, that is, if we allow for the “strong” interpretation, then this panel is based on the merkavah visions in Ezek 1:4–28; 3:12–13; 10:1–22.71 However, this would be a very special way of depicting the Holy of Holies. In fact, while these visions point to God’s presence, the very implement of the chariot does not suggest the fixity of God’s abode in the Temple but rather its mobility. This prompts me to the raise the question whether the reference in the imagery is solely the Jerusalem Temple, or whether there might be some reference to God’s abode in the celestial temple here as well. In ancient Jewish literature, the heavenly merkavah is mentioned, for example, in some apocalyptic texts, beginning with the Book of Watchers (1 En. 14:8–25), through the Book of Daniel (Dan 7:9), the Parables of Enoch (1 En. 71:5–12) to the Apocalypse of Moses (Apoc. Mos. 22:3; 33:2–3).72 Significantly, it features in prayers, especially in the Songs of Sabbath Sacrifice. The twelfth song calls the praying community to join in the praise of God in his celestial tabernacle, surrounded by Cherubim and Ophannim, thereby featuring a depiction of the Holy of Holies in terms of the heavenly merkavah (4Q405 20 ii–22 6–14). While the provenance of the Songs of Sabbath Sacrifice is debated (it might be sectarian), there are also non-sectarian texts that take up Ezek 1 and 3, such as 11Q5 XXVI 9–15 (= 11QHymn to the Creator) and perhaps 4Q503 (= 4QDaily Prayers).73 If there was some hint to the celestial abode of God in the depiction of the merkavah on the  70

For the choice of the creation account, see the suggestions by Levine 2005: 39 n. 70, pointing to a connection between the tabernacle and creation (the tabernacle was finished on New Year’s day [Exod 40:17], traditionally associated with the Sabbath after creation) or, alternatively, to traditions about the beginning of creation at the Temple Mount, most clearly in late rabbinic traditions about the even shetiyyah and the creation of Adam from the dust of the Temple Mount; one might add that this is preceded by the correlation between the Garden of Eden and Mt. Zion in Jub. 4:26; 8:19, and the link of Mt. Zion with the new creation in Jub. 1:19. 71 Ezekiel 1:4–28; 3:12–13; 10:1–22. 72 The Apocalypse of Moses possibly dates from the first century CE: Dochhorn 2005: 172, who is now willing to opt for an even earlier date (personal communication, October 2015). 73 Chazon 1999, 2003; though Falk 1998: 21–29 tends to connect 4Q503 cautiously with the yaad.

The Synagogue at Magdala: Localized Practice and Reference to the Temple

149

Magdala stone table, then we might have a triangular relation here: local synagogue—Jerusalem Temple—heavenly sanctuary. The localized activity in the synagogue would then also have included a reference to God’s heavenly presence. But this would seem to depend on the appropriateness of the identification of the merkavah on the back of the stone.

5. Conclusion The Magdala evidence largely fits other Second Temple-period synagogues and does not challenge our views on the predominant activities in a synagogue, though it creates a link with the Jerusalem Temple; the link likely involves the Torah scrolls; and it might involve some connotation of the celestial temple. We have to account for localized Jewish worship distinct from the Jerusalem Temple but in some connection with it, which found its place within, or perhaps at the fringe, of a lively, multiculturally minded city in early Roman Galilee.

Bibliography Adler, Y. 2008. “‫ הממצא הארכאולוגי וזיקתו להלכה קדומה‬:‫”בתי כנסת עתיקים ומקוואות טהרה‬. Cathedra 128: 51–72. Amit, D. 2007. “‫”בתי כנסת כפריים מימי בית שני‬, Michmanim 20: 21–32. Amit, D. 2013. “ ‫השוני התפקודי בין בתי הכנסת של ימי הבית השני לאלה שלאחר החורבן‬ ‫”והשתקפותו באדריכלות‬, Jerusalem and Eretz-Israel 8–9 [FS A. Kloner, ed. B. Zissu]: 171– 184. Anderson, G. A. 2000. “Worship, Qumran Sect”. EDSS [2.]991–996. Aviam, M. 2013a. “Mesa de piedra para la lectura de la Tora, adornada con simbolismos judaicos en la sinagoga de Magdala”. Revista el Pensador 5: 46–50. Aviam, M. 2013b. “The Decorated Stone from the Synagogue at Migdal: A Holistic Interpretation and a Glimpse into the Life of Galilean Jews at the Time of Jesus”. NovT 55: 205–220. Aviam, M. 2013c. “The Book of Enoch and the Galilean Archaeology and Landscape”. In: Bock, D. and Charlesworth, J. H. (eds.), Parables of Enoch: A Paradigm Shift. London: Bloomsbury T&T Clark, 159–169. Aviam, M. 2014. “Reverence for Jerusalem and the Temple in Galilean Society”. In: Charlesworth, J. H. (ed.), Jesus and Temple. Minneapolis, MN: Fortress Press, 123–144. Aviam, M. 2016. “Another Reading Table Base from a Galilean Synagogue: Some Comments on the Stone Table from orvat Kur”. In: Patrich, J., Peleg-Barkat, O., and BenYosef, E. (eds.), Arise, Walk through the Land: Studies in the Archaeology and History of the Land of Israel in Memory of Y. Hirschfeld on the Tenth Anniversary of his Demise. Jerusalem: Israel Exploration Society, 79*–82*. Aviam, M. 2018. “The Synagogue”. In: Bauckham, R. (ed.), Magdala of Galilee: A Jewish City in the Hellenistic and Roman Period. Waco, TX: Baylor University Press, 127–133. Aviam, M. and Bauckham, R. 2018. “The Synagogue Stone”. In: Bauckham, R. (ed.), Magdala of Galilee: A Jewish City in the Hellenistic and Roman Period. Waco, TX: Baylor University Press, 135–159.

150

Lutz Doering

Aviam, M. et al. 2019. “A 1st–2nd Century CE Assembly Room (Synagogue?) in a Jewish Estate at Tel Rekhesh, Lower Galilee”. Tel Aviv 46: 128–142. Avshalom-Gorni, D. and Najar, A. 2013. “Migdal”. Hadashot Arkheologiyot 125. http://www.hadashot-esi.org.il/Report_Detail_Eng.aspx?id=2304&mag_id=120 (accessed 27 February 2020). Bauckham, R. 2015. “Further Thoughts on the Migdal Synagogue Stone”. NovT 57: 113– 135. Bauckham, R. 2018. “Magdala in the List of the Twenty-Four Priestly Settlements”. In: idem (ed.), Magdala of Galilee: A Jewish City in the Hellenistic and Roman Period. Waco, TX: Baylor University Press, 287–305. Bauckham, R. and De Luca, S. 2015. “Magdala as We Now Know It”. Early Christianity 6: 91–118. Binder, D. D. 1999. Into the Temple Courts: The Place of the Synagogues in the Second Temple Period. SBLDS 169. Atlanta, GA: Society of Biblical Literature. Binder, D. D. 2014. “The Mystery of the Magdala Stone”. In: Warner, D. A. and Binder, D. D. (eds.), A City Set on a Hill: Essays in Honor of J. F. Strange. Mountain Home, AR: BorderStonePress, 17–48. Catto, S. K. 2007. Reconstructing the First-Century Synagogue. LNTS 363. London: T&T Clark. Charlesworth, J. H. et al. 1995. Damascus Document, War Scroll, and Related Documents. PTSDSSP 2. Tübingen: Mohr Siebeck and Louisville, KY: Westminster John Knox Press. Chazon, E. G. 1992. “Is Divre ha-me’orot a Sectarian Prayer?” In: Dimant, D. and Rappaport, U. (eds.), The Dead Sea Scrolls: Forty Years of Research. STDJ 10. Leiden: Brill and Jerusalem: Magnes, 3–17. Chazon, E. G. 1999. “The Qedushah Liturgy and its History in Light of the Dead Sea Scrolls”. In: Tabory, J. (ed.), From Qumran to Cairo: Studies in the History of Prayer. Jerusalem: Orhot Press, 7–17. Chazon, E. G. 2003. “The Use of the Bible as a Key to Meaning in Psalms from Qumran”. In: Paul, S. M. et al. (eds.), Emanuel: Studies in Hebrew Bible, Septuagint and Dead Sea Scrolls in Honor of E. Tov. Leiden: Brill, 85–96. Cotton, H. et al. (eds.) 2010. Corpus Inscriptionum Iudaeae/Palaestinae, Vol. 1: Jerusalem, Part 1: 1–704. Berlin: De Gruyter. De Luca, S. and Lena, A. (2015). “Magdala/Taricheae”. In: Fiensy, D. A. and Strange, J. R. (eds.), Galilee in the Late Second Temple and Mishnaic Periods, Vol. 2: The Archaeological Record from Cities, Towns, and Villages. Minneapolis, MN: Fortress Press, 280–342. Dochhorn, J. 2005. Die Apokalypse des Mose: Text, Übersetzung, Kommentar. TSAJ 106. Tübingen: Mohr Siebeck. Doering, L. 1999. Schabbat: Sabbathalacha und -praxis im antiken Judentum und Urchristentum. TSAJ 78. Tübingen: Mohr Siebeck. Doering, L. 2016. “‘You Are a Chosen Stock …’: The Use of Israel Epithets for the Addressees in First Peter”. In: Furstenberg, Y. (ed.), Jewish and Christian Communal Identities in the Roman World. AJEC 94. Leiden: Brill, 243–276. Falk, D. K. 1998. Daily, Sabbath, and Festival Prayers in the Dead Sea Scrolls. STDJ 27. Leiden: Brill. Fine, S. 1997. This Holy Place: On the Sanctity of the Synagogue during the Greco-Roman Period. Christianity and Judaism in Antiquity Series 11. Notre Dame, IN: The University of Notre Dame Press. Fine, S. 2016. The Menorah: From the Bible to Modern Israel. Cambridge, MA: Harvard University Press. Fine, S. 2017. “From Synagogue Furnishing to Media Event: The Magdala Ashlar”. Ars Judaica 13: 27–38.

The Synagogue at Magdala: Localized Practice and Reference to the Temple

151

Gersht, R. and Gendelman, P. 2016. “The Amphora and the Krater in Ancient Jewish Art in the Land of Israel”. In: Killebrew, A. E. and Faßbeck, G. (eds.), Viewing Ancient Jewish Art and Archaeology: Vehinnei Rachel. Essays in Honor of R. Hachlili. JSJSup 172. Leiden: Brill, 151–185. Habas, L. 2003. “An Incised Depiction of the Temple Menorah”. In: Geva, H. (ed.), Jewish Quarter Excavations in the Old City of Jerusalem, Vol. II. Jerusalem: Israel Exploration Society, 329–342, and Photo 12.1–12.2. Hachlili, R. 2001. The Menorah, the Ancient Seven-Armed Candelabrum: Origin, Form, Significance. JSJSup 68. Leiden: Brill. Hachlili, R. 2013. Ancient Synagogues—Archaeology and Art: New Discoveries and Current Research. HdO I 105. Leiden: Brill. Hachlili, R. 2017. “The Migdal Stone and its Ornamentation”. RB 124: 245–272. Hachlili, R. 2018. The Menorah: Evolving into the Most Important Jewish Symbol. Leiden: Brill. Har-Even, B. 2012. “ ‫ביזנטית והתקופה‬-‫ התקופה הרומית‬,‫ יישוב מימי הבית השני‬:‫חורבת א־טוואני‬ ‫”המוסלמית הקדומה‬. In: ‫ דברי הכנס השביע‬:‫הְספר והמדבר בארץ ישראל‬, 15–29. Available from: http://www.atarsusya.co.il/pages/articles.asp (accessed 27 February 2020). Har-Even, B. 2016. “A Second Temple Synagogue at Khirbet Diab in Western Benjamin”. Qadmoniot 151, 49–53 (in Hebrew). Hayward, C. T. R. 2013. “Scripture in the Jerusalem Temple”. In: Carleton Paget, J., Schaper, J. (eds.), The New Cambridge History of the Bible, Vol. 1: From the Beginnings to 600. Cambridge: Cambridge University Press, 321–344. Hempel, C. 1998. The Laws of the Damascus Document: Sources, Tradition, and Redaction. STDJ 29. Leiden: Brill. Kee, H. C. 1990. “The Transformation of the Synagogue after 70 C.E.”. NTS 36: 1–24. Kee, H. C. 1999. “Defining the First-Century CE Synagogue: Problems and Progress”. In: idem and Cohick, L. (eds.), Evolution of the Synagogue: Problems and Progress. Harrisburg, PA: Trinity Press International, 7–26. Kershner, I. (2015). “A Carved Stone Block Found in Israel Upends Assumptions About Ancient Judaism”. The New York Times, 9 December 2015: page A4. Online version available from: https://www.nytimes.com/2015/12/09/world/middleeast/magdalastone-israel-judaism.html (accessed 27 February 2020). Kloppenborg Verbin, J. S. 2000. “Dating Theodotos”. JJS 51: 243–280. Krause, A. R. 2017. Synagogues in the Works of Josephus: Rhetoric, Spatiality, and FirstCentury Jewish Institutions. AJEC 97. Leiden: Brill. Leibner, U. 2009. Settlement and History in Hellenistic, Roman, and Byzantine Galilee: An Archaeological Survey of the Eastern Galilee. TSAJ 127. Tübingen: Mohr Siebeck. Leone, F. (ed.) 2017. La menorà: Culto, storia e mito. Milan: Skira. Levine, L. I. 2004. “The First-Century Synagogue: Critical Reassessments and Assessments of the Critical”. In: Edwards, D. R. (ed.), Religion and Society in Roman Palestine: Old Questions, New Approaches. New York: Routledge, 70–102. Levine, L. I. 2005. The Ancient Synagogue: The First Thousand Years. 2nd ed. New Haven, CT: Yale University Press. Magen Y., Tzionit, Y., and Sirkis, O. 2004. “Khirbet Badd ‘Isa—Qiryat Sefer”. In: Magen, Y. et al., The Land of Benjamin. Judea and Samaria Publications 3. Jerusalem: Israel Antiquities Authority, 179–241. Magness, J. 2016. “Were Sacrifices Offered at Qumran? The Animal Deposits Bone Reconsidered”. JAJ 7: 5–34. Matassa, L. D. 2018. Invention of the First-Century Synagogue. Ed. by J. M. Silverman and J. M. Watson. ANEM 22. Atlanta, GA: SBL Press.

152

Lutz Doering

McCollough, C. T. 2015. “Khirbet Qana”, in: Fiensy, D. A. and Strange, J. R. (eds.), Galilee in the Late Second Temple and Mishnaic Periods, Vol. 2: The Archaeological Record from Cities, Towns, and Villages. Minneapolis, MN: Fortress Press, 127–145. Meshorer, Y. 1982. Ancient Jewish Coinage. 2 vols. New York: Amphora. Meshorer, Y. 2001. A Treasury of Jewish Coins: From the Persian Period to Bar Kokhba. Jerusalem: Yad Ben-Zvi Press. Meyers, E. M. and Chancey, M. A. 2012. Archaeology of the Land of the Bible, Vol. 3: Alexander to Constantine. AYBRL. New Haven: Yale University Press. Netzer, E. 1999. “A Synagogue From the Hasmonean Period Recently Exposed in the Western Plain of Jericho”. IEJ 49: 219–221. Netzer, E. 2003. “The Synagogues from the Second Temple Period: According to Archaeological Finds and in Light of the Literary Sources”. In: Bottini, C. G., Di Segni, L., and Chrupcała, D. (eds.), One Land – Many Cultures: Archaeological Studies in Honour of S. Loffreda OFM. SBFCMa 41. Jerusalem: Franciscan Printing Press, 277–285. Netzer, E. 2004. “ ‫בתי הכנסת מימי הבית השני והשפעתם על התפתחות בתי הכנסת של תקופת‬ ‫”המשנה והתלמוד – סקירה אדריכלית‬. In: Eshel, Y. et al. (eds.), ‫עשו לי מקדש – בתי הכנסת‬ ‫מימי קדם ועד ימינו‬. Ariel: College of Judea and Samaria, 9–24. Netzer, E. and Stiebel, G. 2008. “Masada”. NEAEHL 5: 1935–37. Noam, V. and Qimron, E. 2009. “A Qumran Composition of Sabbath Laws and Its Contribution to the Study of Early Halakah”. DSD 16: 55–96. Notley, R. S. 2014. “Genesis Rabba 98, 17 – ‘And Why Is It Called Gennosar?’ Recent Discoveries at Magdala and Jewish Life on the Plain of Gennosar in the Early Roman Period”, In: Fine, S. and Koller, A. (eds.), Talmuda de-Eretz Israel: Archaeology and the Rabbis in Late Antique Palestine. SJ 73. Berlin: De Gruyter, 141–157. On, A. and Weksler-Bdolah, S. 2005. “Khirbet Um el-Umdan – a Jewish Village with a Second-Temple Synagogue in Modiin”. Qadmoniot 130: 107–116 (in Hebrew). Patrich, J. 1985 “Inscriptions araméennes juives dans les grottes d’el-‘Aleiliyât: Wadi Suweinit (Na#al Michmas)”. RB 92: 265–273. Peleg-Barkat, O. 2010. “Architectural Decoration”. In: Syon, D. and Yavor, Z. (eds.), Gamla II: The Shmarya Gutmann Excavations 1976–1989. The Architecture. IAA Reports 44. Jerusalem: Israel Antiquities Authority, 159–174. Peleg-Barkat, O. 2016. “Interpreting the Uninterpreted: Art as a Means of Expressing Identity in Early Roman Judaea”. In: Leibner, U. and Hezser, C. (eds.), Jewish Art in Its Late Antique Context. TSAJ 163. Tübingen: Mohr Siebeck, 27–48. Petersen, D. L. 1984. Haggai and Zechariah 1–8. OTL. Philadelphia, PA: Westminster Press. Rahmani, L. Y. 1967. “Jason’s Tomb”. IEJ 17: 61–100. Rahmani, L. Y. 1990. “Stone Synagogue Chairs: Their Identification, Use and Significance”. IEJ 40: 192–214. Rahmani, L. Y. 1994. A Catalogue of Jewish Ossuaries in the Collection of the State of Israel. Jerusalem: Israel Antiquities Authority and Israel Academy of Sciences and Humanities. Raviv, D. 2018. “A Seven-Branched Menorah Graffito from Kafr Mukhmas”. Strata 36: 87– 99. Reich, R. 2014. “A Depiction of a Menorah Found Near the Temple Mount and the Shape of Its Base.” ZDPV 130: 96–101. Reich, R. and Zapata Meza, M. 2014. “A Preliminary Report on the Miqwa’ot of Migdal”. IEJ 64: 63–71. Richardson, P. 2004. Building Jewish in the Roman East. Waco, TX: Baylor University Press. Runesson, A. 2001. The Origins of the Synagogue: A Socio-Historical Study. ConBNT 37. Stockholm: Almqvist & Wiksell. Runesson, A., Binder, D. D., and Olsson, B. 2008. The Ancient Synagogue from its Origins to 200 C.E.: A Source Book. AJEC 72. Leiden: Brill.

The Synagogue at Magdala: Localized Practice and Reference to the Temple

153

Schiffman, L. H. 1999. “The Early History of Public Reading of the Torah”. In: Fine, S. (ed.), Jews, Christians, and Polytheists in the Ancient Synagogue: Cultural Interaction during the Greco-Roman Period. London: Routledge, 44–56. Schiffman, L. H. 2017. “The Magdala Stone”. Ami Magazine May 28, 2017: 162–164. Available from: http://lawrenceschiffman.com/wp-content/uploads/2017/06/ magdalastone.pdf (accessed 27 February 2020). Schofield. A. 2016. “An Altar in the Desert? A Response to J. Magness, ‘Were Sacrifices Offered at Qumran?’”. JAJ 7: 123–135. Shanks, H. (ed.) 2004. The City of David: Revisiting Early Excavations. English Translations of Reports by R. Weill and L.-H. Vincent. Notes and Comments by R. Reich. Washington, DC: Biblical Archaeology Society. Solomon, A. 1997. “The Prohibition against &evul Yom and Defilement of the Daily Whole Offering in the Jerusalem Temple in CD 11:21–12:1: A New Understanding”. DSD 4: 1– 20. Spigel, C. S. 2012. Ancient Synagogue Seating Capacities: Methodology, Analysis and Limits. TSAJ 149. Tübingen: Mohr Siebeck. Steudel, A. 1993. “The Houses of Prostration CD XI, 21–XII, 1 – Duplicates of the Temple”. RevQ 16/61: 59–68. Sussman, V. 1982: Ornamented Jewish Oil-lamps: From the Destruction of the Second Temple through the Bar-Kokhba Revolt. Warminster: Aris & Phillips. Tabory, J. 2003. “The Precursors of the ‘Amidah”. In: Gerhards, A., Doeker, A. and Ebenbauer, P. (eds.), Identität durch Gebet: Zur gemeinschaftsbildenden Funktion institutionalisierten Betens in Judentum und Christentum. Paderborn: Schöningh, 113–125. Tov, E. 2008. “The Text of the Hebrew/Aramaic and Greek Bible Used in the Ancient Synagogues”. In: idem, Hebrew Bible, Greek Bible, and Qumran: Collected Essays. TSAJ 121. Tübingen: Mohr Siebeck, 171–188. Van der Horst, P. W. 2002 [1999]. “Was the Synagogue a Place of Sabbath Worship Before 70 CE?”. In: idem, Japheth in the Tents of Shem: Studies on Jewish Hellenism in Antiquity. CBET 32. Leuven: Peeters, 55–82. Wischmeyer, O. 2004. “Das heilige Buch im Judentum des Zweiten Tempels”. In: eadem, Von Ben Sira zu Paulus: Gesammelte Aufsätze zu Texten, Theologie und Hermeneutik des Frühjudentums und des Neuen Testaments. Ed. Becker, E.-M. WUNT 173. Tübingen: Mohr Siebeck, 15–38. Yarden, L. 1991. The Spoils of Jerusalem in the Arch of Titus: A Re-Investigation. Stockholm: Svenska Institutet i Rom. Yavor, Z. 2010. “The Architecture and Stratigraphy of the Eastern and Western Quarters”. In: Syon, D. and idem (eds.), Gamla II: The Shmarya Gutmann Excavations 1976–1989. The Architecture. IAA Reports 44. Jerusalem: Israel Antiquities Authority, 13–112. Zangenberg, J. K. 2016. “A Basalt Stone Tabe from the Byzantine Synagogue at "orvat Kur, Galilee: Publication and Preliminary Interpretation”. In: Patrich, J., Peleg-Barkat, O., and Ben-Yosef, E. (eds.), Arise, Walk through the Land: Studies in the Archaeology and History of the Land of Israel in Memory of Yizhar Hirschfeld on the Tenth Anniversary of his Demise. Jerusalem: Israel Exploration Society, 61*–78*. Zangenberg, J. K. 2019. “New Observations on the ‘Basalt Stone Table’ from Horvat Kur, Galilee”. Strata 37: 95–111. Zapata-Meza, M., Garza Diaz Barriga, A., and Sanz-Rincón, R. 2018. “The Magdala Archaeological Project (2010–2012): A Preliminary Report of the Excavations at Migdal”. ‘Atiqot 90: 83–126. Zissu, B. and Ganor, A. 2009. “Horvat Etri – A Jewish Village from the Second Temple Period and the Bar Kokhba Revolt in the Judean Foothills”. JJS 60: 90–136.

Contextualizing the Magdala Synagogue Stone in Its Place: An Exercise in Liturgical Imagination

Judith H. Newman How might a public space become a sacred place in antiquity? The discovery of a richly carved ashlar stone in the center of a recently excavated synagogue in the Galilean town of ancient Magdala provides a new possibility for answering that question. Dating from the first century CE before the destruction of the Temple, the synagogue stone is unique, to date, in that it contains cultic imagery from the Temple-tabernacle, including a menorah, in the land, but at some distance from Jerusalem. The stone has important implications for understanding the role of synagogues and their diversity in Israel. In this essay I build on the work of scholars who have already interpreted the imagery of the stone. In the first part of the paper, I will review the findings of major essays that have been published on the iconography of the Magdala stone. While a connection to the Temple in Jerusalem is clearly apparent, the way in which the Temple is represented is not as straightforward as an eyewitness, photographic rendering. I propose that the iconography should not be seen as a detailed replica of the Temple in Jerusalem but that its imagery served as a diachronic assemblage to conjure not only the Temple in Jerusalem, but also the mobile wilderness tabernacle. In the second part, I will bolster this assertion by arguing that the space of the synagogue becomes a sacred place through memoryconstruction mediated through both the eye and the ear. The two were intertwined. Cultural memory of a mobile divine presence was constructed through the synagogue’s regular reading of its sacred texts. In that way, a generic space was made a local place through the specificity of Jewish traditions.

1. The Iconography of the Stone A. The Synagogue of Magdala What kind of place was Magdala? While the final excavation report on the archaeological site of Magdala has not yet been published, the character of the region had already become evident since excavations began in the 1970s. Jürgen

156

Judith H. Newman

Zangenberg has observed that “it more and more becomes apparent how ‘urban’ and ‘Mediterranean’ this site has been since its foundation in the 2nd century BCE. It is entirely justified to see Magdala in the same category as the large Hellenistic cities in Greece or Asia Minor” (Zangenberg 2010: 475). Magdala was a wealthy and urbanized center. Given its large harbor, the largest and best-preserved discovered to date on the Galilean Lake, the economy was likely driven both by fishing and trade with the Decapolis on the eastern side of the lake. Excavated in 2009 by Dina Avshalom-Gorni and Arfan Najar (2010, 2013), Magdala’s synagogue dates to the beginning of the first century CE. It covers 120 meters square. Its main hall is 11 meters by 11 meters with stone benches along the walls dating from the late second Temple period and later. The so-called Magdala stone with its rich iconography was discovered near the center of the main hall. It is 0.6 m long, and roughly 0.5 m wide and 0.4 m high and made out of a single piece of limestone. It has four short stout legs of which the two front ones are shorter than the back ones, making the top surface slighted raked. Three scholarly essays have appeared that treat the iconography of the Magdala stone in some detail. The earliest was published four years ago by Mordechai Aviam (2013a), who first identified the stone in connection with the Temple in Jerusalem. A second essay by Donald Binder (2014) is part of a Festschrift for James Strange in which he makes reference to Aviam’s work and offers his own interpretation of the iconography. A third essay by Richard Bauckham (2015) appeared in the same journal in which Aviam published his first essay and builds explicitly on it, but Bauckham makes no mention of Binder’s article and seems not to have read it.

B. Initial Interpretations I begin by reviewing the respective views on the iconography. a. The north (front) face of the stone (Fig. 1) contains the most noted and unmistakable feature of the iconography: a seven-branched menorah on a tripod stand. [N.B., following Binder, I identify the four faces according to cardinal direction as the stone was found in situ.] On the ashlar, the menorah stands on a square base and is flanked by two amphorae. The menorah and amphorae stand under an arch supported by two fluted columns. While the menorah and its association with the Temple were already known from its depiction on a coin minted in the first century BCE by Mattathias Antigonus, this is the earliest depiction of the menorah appearing in an architectural format in a public space. It is the earliest menorah from the Second Temple period to be discovered outside of Jerusalem. As well, it is the earliest menorah to be found in a synagogue that predates the destruction of the Second Temple.

Contextualizing the Magdala Stone in Its Place

157

Fig. 1. The Magdala stone, north (front) face (line drawing by Anne Kaye).

While all three agree on the menorah, as have all others who have seen the stone, there is more disagreement on the other objects. Aviam interprets the square below the menorah as the gold incense altar, which would in fact have been in front of the menorah. He is followed in that interpretation by Bauckham with some qualifications, who would have expected to see some representation of the showbread table on this face because it was one of the three central cultic objects alongside the Gold Altar and the menorah (Bauckham 2015: 116). These three cultic objects are found most famously in Rome on the Arch of Titus in its depiction of the sack of Jerusalem. Binder does not see the square below the menorah as anything other than a base. Binder believes the incense altar is depicted elsewhere on the stone. Binder understands the amphorae as representing the oil that was burned in the lamps of the Temple (Exod 27:20, Lev 24:2; Zech 4:1–14) as does Bauckham. Aviam identifies them as two “sons of oil” figures in connection with Zechariah’s vision. The depiction of amphorae is rare in the Second Temple period, but they do appear on ossuaries usually next to rosettes. As for the face as a whole, all scholars concur that this represents the holy place, the hekhal, of the Jerusalem Temple. To inform his interpretation, Bauckham draws on two references in the Talmud to priests lifting the cult objects at the entrance to the holy place, so that the people could see them. Accordingly for Bauckham, the north face is thus a presentation of viewing the inside of the Holy

158

Judith H. Newman

Place, a view that might have been accessible to those on pilgrimage for the great festivals.

Fig. 2. The Magdala stone, south (back) face (line drawing by Anne Kaye).

b. The opposing south (back) face of the stone depicts three fluted columns similar to those that appear on the north face (Fig. 2). Two circular objects appear on either side of the central column. While the excavator interpreted the two circles as rosettes, Aviam rightly noticed that the center of each circle was different. On the left seems to be a depiction of a protruding knob and on the right it is recessed. He thus understands them not as ornamental rosettes but as two chariot wheels with spokes, the first as seen from the outside, the second from the inside, the place in the socket where the shaft would be inserted (Aviam 2013a: 213). He draws on a comparison between the chariot wheels on the Magdala stone with a decorated stone found in the Capernaum synagogue. That Capernaum stone clearly depicts a building carried on four wheels. Although early archaeologists understood this as a depiction of the ark of the covenant on its return from the Philistines (1 Sam 6:10–17), the absence of oxen leading it and its similarities to the Magdala stone caused Aviam (2013a) to reinterpret the depiction as a chariot. Aviam understands the set of curved, triangular shaped elements below the wheels on the Magdala stone as schematic flames of fire. He thus suggests this is a representation of the wheels of the divine chariot throne, described in slightly different ways in Daniel, Ezekiel, Zechariah, and 1 Enoch. Dan 7:9 may be the most relevant: “his throne was a fiery flame, and its wheels were burning fire.” Binder and Bauckham agree with this characterization. Binder affirms a more

Contextualizing the Magdala Stone in Its Place

159

programmatic statement: “The interpretation of these images as ‘wheels of fire’ fits well within the larger context of the Magdala stone” (Binder 2014: 29). Given the location of this panel in relation to its opposite, it makes spatial sense that this area ‘behind” the menorah panel should depict the holy of holies. Aviam, building on an earlier essay on the provenance of 1 Enoch (2013b), understands this face to represent “a mystical, allegorical view directly into the Holy of Holies and the Divine Spirit in the Jerusalem Temple” (Aviam 2013a: 213). This is an intriguing observation, and it will be considered at greater length below. c. The east and west sides of the stone are mirror images of each other. I have only included a depiction of the west side (Fig 3). Each panel contains a series of four arches. At the end of each one is a round vessel with an indentation in the center and a protruding tube. Within three of the arches there are additional curved lines. While the excavators understood these to be sheaves of corn, Aviam argued that they were in fact architectural arches depicting “an arcade within an arcade” or a “building within a building,” which presents the Holy of Holies inside the Temple (Aviam 2013a: 212). The hanging object in the fourth arch he understands not as an oil lamp as do the excavators, but as an incense vessel, used in the daily offerings and according to Josephus placed on top of the showbread table.

Fig. 3. The Magdala stone, west side (line drawing by Anne Kaye).

Binder agrees with Aviam in regard to the arches, though draws on the Songs of the Sabbath Sacrifice to suggest that the lines in the arches represent “streams

160

Judith H. Newman

of fire” cascading down (Binder 2014: 31). Bauckham, in contrast to both Aviam and Binder, does not think these interior arcades depict the Holy of Holies, but rather the Temple courts. On that basis, Bauckham understands that “the four sides of the stone would then depict the three spaces that made up the Temple: the courts, the Holy Place, and the Holy of Holies (Bauckham 2015: 120). As for the object in the fourth arch, Bauckham proffers that Aviam’s view that this is an incense dish does not “fit well” with his own suggestion that the arcades “represent the Temple courts” (Bauckham 2015: 120–121). Therein lies part of the hazard of allowing a holistic interpretation to guide one’s understanding of its individual elements. d. The top face of the stone (Fig. 4) is the most crowded and complex of the entire iconographic program and the interpretation of its elements is the most debated. Its most prominent feature is a central rosette with six petals. On two sides of the rosette are two long shapes with fan-shaped tops. There are six heart-shaped elements, three on both the north and south sides of the rosette. One “heart” on each side is larger than the other two. There are also two rectangular objects with rectangles inside inscribed with two crossed lines within. The stone seems to have suffered some damage so not all of the carved imagery is intact.

Fig. 4. The Magdala stone, top face (line drawing by Anne Kaye).

Contextualizing the Magdala Stone in Its Place

161

Rosettes are a particularly common motif on ossuaries of the HellenisticRoman period, although there is no consensus on their meaning. Levi Rahmani (1994) and Rachel Hachlili (2017) are known for their shared view that there was no symbolic meaning attached to these rosettes; they were simply decorative. Aviam’s understanding of the circles on the south face as chariot wheels is made more convincing because they differ from the rosette on the top of the stone. This large rosette has “petals” whose points intersect with the circle on the outside perimeter forming six elongated lozenge shapes in a circle, whereas the chariot wheels are continuous. Aviam and Bauckham root the meaning of the rosette according to Josephus and Philo as having cosmic symbolism, related to the twelve signs of the zodiac. Bauckham, however, points out that a weakness in that cosmological interpretation is that it is not connected to the motifs on the top surface of the stone as a representation of the showbread table. There is also disagreement about the long shapes on either side of the rosette. Aviam understands these as rakes, i.e., implements used to clear ashes and burnt bones from the Temple’s main altar and mentioned in m. Tamid 2:1 (Aviam 2013a: 214; cf. Binder 2014: 37). Both Binder and Bauckham understand these as date palm trees. Binder (2014: 33) sees the rosette and palm tree combination as variants known from other synagogues. He views them as “laden with cultic import” stemming in part from the description of Solomon’s decoration of the Jerusalem Temple in 1 Kings. As for the six heart-shaped elements on the top, Aviam draws on descriptions of the showbread table in the Talmud and Josephus to suggest that these are loaves of bread that were kept on the table, likely two piles of six loaves each. Initially, Bauckham (2015: 123) found this likewise convincing and, on that basis, shaped his own interpretation to understand that the top surface of the stone, and indeed, the entire stone was meant to represent the showbread table. In this interpretation, he connected the loaves to the importance of the twelve tribes. From this perspective, the rosette with its six petals and six surrounding lozenges represents the twelve tribes of Israel that complement the loaves. More recently, Bauckham has proposed a different interpretation of the upper side. He now distinguishes seven objects, apart from the rosette, as occurring in pairs and identifies them as the seven species from which first-fruits were offered (Aviam and Bauckham 2018: 150–155).

Evaluating the Interpretations Each scholar has offered detailed descriptions of their reasoning behind their identification of all of the iconography, but it is not necessary to go into all their differences. I have gone into some detail about the iconographical identifications of these scholars for two reasons. I agree with them there is cultic significance to this imagery. From my perspective as well, there are still open questions about

162

Judith H. Newman

the identification and significance of many of the iconographic elements especially the shapes on the top panel. I think there is in fact the risk of over-interpretation without some controls, whether from another similar artefact or an explicit description of such a synagogue stone in a literary source. But at this point, there is only one stone like the Magdala stone. While each scholar provides an exhaustive treatment, an important methodological distinction is that while Aviam and Bauckham draw on biblical sources, early Jewish literature, Josephus, Philo and the Mishnah and Talmud to inform their identification of the iconographic elements; Binder uses only contemporaneous and other pre-70 sources. Binder’s methodological rigor is commendable and serves as an important reminder not to think teleologically as if later descriptions of the Temple in rabbinic literature necessarily and accurately describe the Temple or other aspects of life in the Second Temple era. Historiographical scholarship should proceed without having “Christians or rabbis” always in mind (Runesson 2017), that is, being shaped unduly by the normative discourses of a later era. In spite of differences, they agree on the idea that this is a representation of the Temple in Jerusalem, whether in whole or in part.

C. Other Perspectives on the Stone The interpretations of the stone as a “mini-Temple” have not gone unchallenged. Before elaborating my own view of the stone and its role in the public space of the synagogue, I want to turn to three additional scholars who have subsequently evaluated the stone. Rina Talgam (see Kershner 2015), Rachel Hachlili (2017), and Steven Fine (2017) have offered interpretations of the stone as well. Though she has not yet published her interpretation, the art historian Talgam’s perspective is known from a New York Times article by Isabel Kershner on the synagogue and its stone. Talgam, like Aviam, concludes that the stone is a three-dimensional depiction of the Temple of Herod, including its most sacred inner sanctum, the Holy of Holies. Given Talgam’s expertise with images and archaeology of the Byzantine period, a more in-depth treatment of her own may strengthen the case she makes in this journalistic interview. In contrast, the archaeologist Rachel Hachlili has leveled the strongest criticism of the interpretation of the stone’s imagery as Temple-related. She denies that the images carved on the stone have any symbolic meaning whatsoever. In her words: The Migdal stone repertoire of ornamental motifs reflects a rigid aniconic and nonsymbolic choice of floral, geometric and architectural patterns which are part of an ensemble of decorative designs connected to the fashionable art of the Second Temple period. Hence, no symbols are depicted on the Migdal stone or on the ossuaries, nor do any of the rendered motifs represent the Temple (Hachlili 2017: 263).

Contextualizing the Magdala Stone in Its Place

163

While Hachlili’s understanding of the terms symbol, and symbolic is not made explicit in her work, I take her to mean that there is no abstract or representational meaning beyond the shape itself as decoration. Hachlili emphasizes parallels between the rosettes found on ossuaries during the Second Temple period and the wheel shapes on the Magdala stone. While she recognizes two different types of rosettes on the stone front and south (back) faces, the schematic drawing included in her article does not distinguish between the centers of the two wheels found in the south (back) face of the stone as does Aviam in his identification of chariot wheels. She acknowledges the appearance of a menorah, but rather than connecting it to the Temple in Jerusalem, she suggests “the simply incised menorot were meant to represent a priest’s family house, a priest’s tomb, or a priest’s sundial; hence the Migdal stone might also have had some connection to a priest’s family or office” (Hachlili 2017: 252). Her contention that the menorah might represent the priestly office and their duties stands in tension with her claim that no secondary “symbolic” meanings are attached to the imagery. Given the location of priestly duties is the Temple, her insistence on disconnecting the menorah from the sanctuary is puzzling. A second essay by Steven Fine (2017) articulates a methodologically informed critique of scholarship on the Migdal stone. He holds a certain skepticism regarding what he characterizes as a “media event” surrounding the stone and synagogue which he views as designed to appeal to Christian pilgrims’ interest in tracing Jesus’ footsteps in the Holy Land. He also offers an evaluation of earlier attempts to interpret the stone’s imagery. His criticism of the approach is largely two-fold. The first relates to the nature of interpretation of the iconography. He objects to what he terms a “monistic” interpretation that seeks to link literary texts with a singular interpretation of the stone’s images. In his words: “To associate this rosette with Ezekiel’s chariot (Ezek. 1) is to jump from a relatively common iconography to a very specific application of a biblical text” (Fine 2017: 35). The second related objection is also rooted in methodological critique. Fine thinks it is essential to contextualize and compare the material object with other findings from material culture. Fine misses the point that the authors are not always arguing for specific texts that illustrate the iconography, though in some cases, particularly with Bauckham, this is the case. Rather, it is compelling to use the larger body of interpretive tradition that is reflected in particular texts. That is to say Ezekiel 1 offers just one instantiated example of the chariot throne from a much larger body of tradition that depicts the divine throne. Fine may himself be too narrowly focused on the biblical text in his own evaluation rather than taking the larger Jewish literary world of the Hellenistic-Roman era into account. Fine rightly cautions against a one-to-one relationship between text and image, or what might be called eisegesis of images stemming from exegesis of texts. Yet missing in the discussion is consideration of dynamic textual traditions that lie off the page, as well as the social context of “lived religion” in which the stone was

164

Judith H. Newman

used (Newman 2018). The former, interpretative tradition as an oral phenomenon, can be assumed, but we have no direct scholarly access to it. The social context of the synagogue is more easily constructed. Up to this point, I have discussed the imagery on the stone and its interpretation. Before evaluating it, some mention of the presumed function of the stone within the synagogue will help to situate further discussion. The archaeologists and scholars who have interpreted the stone’s iconography have identified the function of the stone in various ways. Avshalom-Gorni and Najar (2010, 2013) state succinctly and without elaboration that the stone represents the Second Temple and served as a prayer table. Aviam (2013) thinks the stone itself symbolizes the Temple in Jerusalem. He understands the stone as a base for a reading table on which Torah scrolls would have been placed. Though there is no material nor comparative evidence for it, he presumes that a wooden structure would have been on top of the stone. Binder (2014) considers four possibilities: as a table for reading scripture, as a base for a lamp, as a base for an offering vessel, or as a seat for the leader of the synagogue. Other suggestions are an incense altar or offering table (De Luca and Lena 2015), a table for the gathering of first-fruits to be taken to the Temple (Aviam and Bauckham 2018: 155–157), or no specific function other than to represent the Temple (Bauckham and De Luca 2015). The immediate practical use to which the stone was put is thus unclear. But just as the practical use of a stained glass window in a cathedral is to let in light, the object and its imagery can also and simultaneously serve other narrative and ideational purposes. I thus want to return to the imagery itself. From the foregoing review, I find two aspects of the identification of the imagery convincing. The first is the menorah as a symbol of the holy place or Temple, an understanding that is recognized by all but one scholar. Some of the interior of the Temple is also represented. The second is the identification of the chariot wheels as a depiction of the base of the divine throne with flames of fire below them. The summoning of the divine chariot, known from Ezekiel and other texts, suggests the immanent presence of the divine. The degree to which this should be assessed according to the heuristic descriptors “mystical” and “allegorical” as does Aviam, is a matter that requires more attention elsewhere. At the same time, I think the singular emphasis on the Temple in Jerusalem is misplaced for its emphasis both on the synchronic aspect of the idea of a single location-Jerusalem—at a single time—the present. This temple is also evocative of the past and the future in its role in the history of Israel from the pre-exilic period through its post-exilic reconstruction and beyond. Indeed, hope for the restoration of the First Temple resulted in imaginative envisioning found in such texts as Ezekiel 40–48. In the remainder of the paper, I would like to argue that the stone served to evoke its context away from Jerusalem. The stone is much more than the relocated Temple in Jerusalem; rather it serves as the “templetabernacle” in Magdala in which the divine presence might dwell with the people.

Contextualizing the Magdala Stone in Its Place

165

I want to attend to the way in which public space of the synagogue in Magdala might become a meaningful cultural place of sanctity over time. In making a distinction between space and place, I borrow from the work of the humanistic geographer Yi-Fu Tuan (1979). In his formulation, space is a neutral concept, a reality devoid of special meaning. Place, by contrast, is a space permeated by human experience, whether this comes through the physical senses, the cultural mediation of memory, or some other means. If space is universal and anonymous; place is local and particular. We must thus consider some of the particulars associated with the synagogue space that transform it into a place of gathering for Jews in Magdala. Our understanding of the stone’s significance and role should be focused as much on the local context and what the object means in its local setting in Magdala as part of the lived religion or lived culture of a local community, rather than simply identifying the elements that are inscribed on it without regard for the local social context. With the stone in mind, we will contextualize it briefly not only as the individual creation of a stoneworker, but as a collection of images that helps to create a sacred place for a textual community through reading practices.

2. Contextualizing the Magdala Stone within a Textual Community Space becomes place in part through the role of memory. Missing in these initial scholarly treatments of the iconography is any discussion of the role of memory and memorializing that occurs in relation to this stone. Assessing the stone’s visual poetics requires attention to memory processes both in the construction of the imagery on this stone on the part of those who made it, but also on the part of the congregation who views this iconography as the community gathered in that place to take part in the activities of a textual community.

A. The Selectivity of Memory Memory is a selective act of construction. In this regard, two aspects of memory formation are illustrative. First, there are the neurophysiological aspects of individual memory and its memory processes. We know that memory is not mechanical. Visual memory is not as simple as a pair of eyes that records and recalls the details of an image with photographic accuracy. Nor do human ears record for later comprehensive recall all information they hear like tape recorders—the sound of the door shutting, the baby crying, the gray noise of the air conditioner. Our eyes do not remember everything we see like video recorders—all the leaves of the trees we see as we walk along the street to work, the name on every street sign registered as we pass by, and so on. Our brains would simply explode from

166

Judith H. Newman

being too full of information. In recalling previous events, there is a selective process at work. As Mary Carruthers (2008) has discussed in her masterful work on memory in medieval Christianity, all accounts after Aristotle have separated memory into two processes: that of storage, and that of recollection. The assemblage of images is important not because the artist had actually seen the Temple in Jerusalem with his eyes and tried to recreate it in its many particulars to provide a replica. The artisan had shaped these memories into a potent ritual object for this space in Magdala through memory processes that were both biological and social. In short, the implication is that the images on the stone are not supposed to be like the Temple in the so-called “Holyland model” of Jerusalem, a detailed model following Josephus’s description of first century Jerusalem and originally displayed at the site of the former Holyland Hotel in Jerusalem, now on the grounds of the Israel Museum. The “Holyland model” presents a space; it is a synchronic space with its universal comprehensive view. The model does not allow for the lived religion and experiences of a community who moved through that space. Moreover, we recognize now that the account of Josephus itself, whether in his discourse about the Temple or about synagogues, must be evaluated according to his rhetorical aims rather than taken at face value (Krause 2017). The images on the stone are much more limited in scope. In their arrangement on the stone, the assemblage of images thus might better be likened to the constellation of images found on a tourist postcard: “Greetings from Vatican City!”—“Greetings from Jerusalem!” Such images are not meant to be exhaustive of the site, but to conjure some essential aspects of the history and culture of the location. A “Saluti dal Vaticano” postcard would surely include an image of St. Peter’s basilica, whether from the front or a bird’s eye view, but also the keyshaped portico by Bernini, an image of the current pope, perhaps a member of the Swiss guard, and Michelangelo’s Sistine Chapel ceiling. They are not meant as a complete rendering of all there is to see. Rather they are images that convey a whole, but one that itself was constituted through time. These images contribute to the identity of residents of that site, and the nation as a whole that renders it a meaningful place of cultural memory. The analogy is inexact of course, given a postcard’s ephemeral character on flat cardstock, as opposed to the threedimensional permanence of the stone. The imagery on the stone, while far from refined carving, clearly represents an investment of time and expense on the part of those who sponsored it. It reflects no doubt the wealthier setting of Magdala as an urban center whose residents might afford such a disbursement of resources. In addition to individual brain processes we have also learned from Maurice Halbwachs (1925, 1950) and other social memory theorists, that memory is both a powerful faculty of the imagination, the “mind’s eye,” and a social phenomenon. Any recollection of Jerusalem or Rome, both by individuals and communi-

Contextualizing the Magdala Stone in Its Place

167

ties, involves integrating personal experiences of the location into the larger cultural memory, whether it is the sight and feel of the Western Wall or participation in a papal Mass. Both aspects are important.

B. Making a Synagogue Place through Memories of Divine Presence What else then was involved in shaping the memory of the Temple in Magdala in the public space of the synagogue? Not only physical objects, but textual traditions informed the communal memory. The synagogue served many functions in antiquity, from hostel for travellers to political meeting site, but it was preeminently a location where scriptures were read on the Sabbath (Josephus C. Ap. 2.175; Philo, Legat. 156; Luke 4:16–17; Acts 13:14–15; 15:21). Cessation from work and synagogue gathering accompanied other practices designed to separate the seventh day from the rest of the week (Doering 1999). As the common constituents of Jewish scripture in the late second Temple period, the Torah of Moses and the prophets were no doubt also read, studied, discussed, and interpreted in Magdala. We can identify those gathered within the synagogue in terms of what Brian Stock (1983) has called a “textual community”. The synagogue gathering was one in which the cultural value of texts and their study had great currency. While private, individual copies of manuscripts would not have been in the possession of the gathered group, the value of the text and its study was greatly esteemed. Nor would all members of the community need to be literate as readers and writers to be invested in this textual community. In the predominately oral cultures of Mediterranean antiquity, the ear and the tongue played an integral role in sustaining a textual community. We might consider then how the textual repertoire informs the cultural memory of the Magdala congregation. In his book Time Maps, Eviatar Zerubavel (2003) discusses some of the many ways the presence of the past continues to shape the present, a phenomenon he calls bridging, that allows for “mnemonic connectedness”. These connections might be visible in certain memorial sites and architectural monuments or in conceiving of events on analogy with other earlier momentous events. Another means of connecting the past with the present in ancient Israel was to recall not a particular location or a particular moment in time, but the intangible promise of divine presence in a sanctuary. The memory of divine presence was inscribed then, not only in the Magdala stone chariot throne, but also within the diet of texts consumed in the synagogue precincts. The Torah is the obvious text with which to begin because it formed Jewish identity on a weekly basis. At the center of the Torah, however, stands not the Temple, but the divine instructions for building the Wilderness tabernacle and its subsequent construction within the Sinai account. Of course, we have no knowledge of a lectionary cycle in the first century, and indeed the exact shape of

168

Judith H. Newman

the “Torah” in this era is an open question. Nevertheless, the tabernacle narrative would no doubt have been encountered with some regularity. The sources for this representation of the tabernacle come from a distinctive priestly, and also, an exilic theological strand in the Bible, both in the Torah and in the prophets. The Exile was pre-eminently a period of relocation and mobility in which the promise of divine presence took a new turn. We see in the priestly source of the Pentateuch a re-envisioning of both divinity and sanctity such that the mobile presence of the divine takes on a heightened prominence. The tabernacle moves with Israel through the Wilderness, along with the glorious manifestation of the divine in the pillar of cloud and fire. The Priestly account of the instructions for the tabernacle and its construction are also acts of time travel, of projecting into the past an account of a later sanctuary. The divine command to Moses about the tabernacle includes the divine promise of eternal, abiding presence in the Israelites’ midst (Exod 25:1–2). That holy divine presence moves with them. This divine dwelling comes, however, with an obligation on the part of the people. According to the Priestly source in Exodus 31:12, the conclusion of the covenant is the obligation for Israel to observe the Sabbath (Blenkinsopp 1976). Memories of the Temple cannot be separated from the desert tabernacle and its appurtenances in the cultural memory. Jon Levenson (1989) has illustrated the multiple entwined linkages of desert tabernacle and Temple especially in the priestly literature of Jewish scripture and tradition. Moreover, he convincingly portrays the deep-seated connection between cosmology and Temple, such that portrayals of the construction of both tabernacle and Temple are homologous to the created order of the world in Gen 1:1–2:4. Similar to common Ancient Near Eastern prototypes, the Temple-tabernacle was conceived as a microcosm. It is conceived as a culminating capstone to world-building, a seal on the omphalos mundi. In his words: “World building and temple building are not parallel projects but are mutually implicated. Neither is complete alone” (Levenson 1989: 99). By extension, then, the synagogue at Magdala with its chariot-inscribed stone might also be considered such a place of world-building and completion, especially on the designated day that served as a sign of Jewish observance, the Sabbath. The synagogue was a place in which the instruction of Moses was discussed on a weekly basis. Not only the Torah but the prophets contribute to cementing a mnemonic connection to the immanent divine dwelling. In Ezekiel, the Temple structure does not move; it is only the kabod adonai on the chariot throne that moves with the people into Exile in Babylon. At the end of the book, the prophet is transported through vision, and views the return of the mobile divine presence to a restored temple in Jerusalem: 26

I will make a covenant of peace with them; it shall be an everlasting covenant with them; and I will bless them and multiply them, and will set my sanctuary among them forevermore. 27My dwelling place shall be with them; and I will be their God,

Contextualizing the Magdala Stone in Its Place

169

and they shall be my people. 28Then the nations shall know that I the LORD sanctify Israel, when my sanctuary is among them forevermore. (Ezek 37:26–28 [NRSV])

C. The Sage as Mediator of Divine Presence The Torah and the Prophets as known from the medieval Masoretic Text offer one way to conceive of mnemonic connectedness through the promise of divine presence. But given what we know of the open-ended character of scriptures in their entwinement with Jewish liturgical practices (Newman 2018), not to mention the comparative wealth of the city of Magdala, the library of the synagogue might well have contained more than the Torah of Moses and the Prophets. One can imagine that a range of literature was also read or studied, discussed and interpreted there. I want to treat briefly a textual assemblage drawn from the sapiential anthology called Ben Sira that recalls the Tabernacle-Temple. The work is thought to have originated in Jerusalem, because the end of the work describes the activities of the high priest Simon. Yet we know the text travelled. Not only does the prologue describe Ben Sira’s grandson taking the work to Egypt in order to translate it for Jewish communities there, but closer to Jerusalem, one partial manuscript dating to the first century was discovered in a wall of the synagogue at Masada. A mnemonic bridge to the promise and fulfilment of divine presence occurs in Ben Sira 24, a hymn in which the female figure of Wisdom praises herself. There is a continual and paradoxical juxtaposition knit into this passage, which contrasts remoteness and availability, the universally available with the local and particular. The spirit of wisdom is both in heaven with the Most High in the assembly with his hosts, and at the same time is praising herself in the midst of her people. She proclaims: “I came forth from the mouth of the Most High and covered the earth like a mist.” (Sir 24:3) She looks for somewhere to dwell and moves across the world, then travels in the pillar of cloud and fire until ultimately finally received a divine command: 8

Then the creator of all commanded me, and he who created me put down my tent (σκηνήν μου) and said, “Encamp (κατασκήνωσον) in Jacob, and in Israel let your inheritance be.” 9Before the ages, from beginning, he created me, and until the ages I shall not cease. 10In a holy tent (ἐν σκηνῇ ἁγίᾳ) I ministered before him, and thus in Zion I was established. (Sir 24:8–10 LXX)

The language of encampment and the tent clearly derive from the tradition of the wilderness wanderings and the tabernacle during the Mosaic period. She is mobile and fluid. Though Wisdom comes to rest in Zion, a single city, and there clearly is a reference to the Temple in Jerusalem, she nonetheless is described as remaining in a tent. Moreover she encamps with the people.

170

Judith H. Newman

The well-known identification of Wisdom and her instruction (“all this”) with the Torah of Moses appears in the middle of the passage: 23

All this (Ταῦτα πάντα) is the book of the covenant of the Most High God, the law that Moses commanded us as an inheritance for the congregations of Jacob. It overflows, like the Pishon, with wisdom, and like the Tigris at the time of the first fruits. (Sir 24:23 LXX)

We thus see the textualization of the Wisdom tradition, or the sapientialization of Israelite history in a priestly vein. From this point on, water metaphors prevail. The end of the passage identifies the sage as the ultimate recipient of wisdom. He is a conduit who not only receives the inheritance of wisdom issuing from Torah as it enlarges from rivulet to river to sea, but also communicates it to the assembly. Thus wisdom is available both in the particular places of wilderness tabernacle and Temple in Jerusalem, but available to the people as a whole as an inheritance in the all-encompassing Torah mediated by the sage in local contexts. The enmeshment of Temple and tabernacle not to mention it relocation remains a paradoxical possibility. This passage is an apt analogy to the assemblage of images on the Magdala stone because it conflates references to the tabernacle and Temple through mnemonic bridging. Wisdom, as the spirit of the divine present at creation, represents the promise of divine presence, however malleable the manifestation may be. Thus the passage also thematizes and reflects the paradoxical tension between the permanent and the mobile, the central and the peripheral or local, the stability of a text with the fluidity of its interpretive extension. Through the medium of the Torah as well as in Ben Sira’s own teaching, the Tabernacle-Temple is accessible to all. It is accessible, that is, provided of course there is a learned sage who can dispense such wisdom as “the inheritance for the congregations of Jacob.” If the Magdala stone did indeed serve as a base for a reading stand, or a stand for scrolls of the Torah, I can well imagine in my own mind’s eye, the scroll of Ben Sira sitting on top of it, or being studied in the adjacent hall. After all, the book did travel—to Masada, not to mention the remote locale of Egypt. The well-known passage in Ben Sira is not the only textual site where Temple and tabernacle are conflated. The identification of the sanctuary and tabernacle appears in three prayers of the Hellenistic-early Roman era: Judith 9:8; Wisd 9:8 and Tobit 13:10 (LXX 13:11) (cf. 2 Macc 2:4–8). Another highly relevant text is the Songs of the Sabbath Sacrifice, known in manuscripts both from the Qumran caves and Masada as well. The extent of its dissemination and use throughout the land is not known and a consideration of its contents is not possible here given space constraints. Yet, that liturgical text, too, is replete not only with the Temple-tabernacle, but also with images of the divine chariot throne and the immanent presence of divine glory among the people (Newman 2008).

Contextualizing the Magdala Stone in Its Place

171

Conclusion The carved stone discovered in a first century CE synagogue in the Galilean town of Magdala is a unique find—to date—for its representation of Temple-tabernacle iconography in the land of Israel while the Second Temple was still standing. Foundational scholarly work has offered both cogent identifications of the iconography as well as cautions about over-interpretation. While scholars differ on many details, their agreements point to clear connections to the “Temple in Jerusalem”. Yet, the way in which the poetics of the iconography relates to the lived religion of the local populace has not yet been discussed in any depth. How does this assemblage of images contribute not simply to memorializing the Temple in Jerusalem, but to shaping a congregation in order to make a sacred place in Magdala? The making of “place” as I have described it with the help of Yi-Fu Tuan is only partly available to us in the present. The entire iconographic program of the Magdala synagogue, as well as all the activities that took place in it, are not definitively known to us. Yet we can imagine a textual community that gathered regularly in the synagogue on Sabbaths for instruction in scriptures. I have made a preliminary attempt to delineate the way in which the visual mosaic of Templetabernacle memories inscribed on the stone might have been reinforced with a verbal mosaic within a textual community to shape a congregation. A central promise from the priestly perspective was that the glorious divine presence would dwell with the people whether in the tabernacle or Temple, or even beyond, among a people dispersed. That thread of the memory of the promise of divine presence serves as a mnemonic connection to remember both past in the present and anticipate a future. In conceiving the repertoire of the Magdala synagogue’s textual community, I have considered not only the Torah and Prophets, but have assessed a verbal assemblage from Ben Sira which associates the mobile spirit of Wisdom with Temple-tabernacle and teaching sage. Other texts, like the Qumran Songs of the Sabbath Sacrifice or the Letter of Aristeas, are surely relevant to our understanding of divine presence among textual communities. The Magdala stone is itself a product of a constructive liturgical imagination that invites connection with the mobile presence of the divine.

Bibliography Aviam, M. 2013a. “The Decorated Stone from the Synagogue at Migdal: A Holistic Interpretation and a Glimpse into the Life of Galilean Jews at the Time of Jesus”. NovT 55: 205–220. Aviam, M. 2013b. “The Book of Enoch and the Galilean Archaeology and Landscape”. In: Charlesworth, J. H. and Bock, D. L. (eds.), Parables of Enoch: A Paradigm Shift. Jewish

172

Judith H. Newman

and Christian Texts in Contexts and Related Studies Series 11. London: Bloomsbury, 159–169. Aviam, M. and Bauckham, R. 2018. “The Synagogue Stone”. In: Bauckham, R. (ed.), Magdala of Galilee: A Jewish City in the Hellenistic and Roman Period. Waco, TX: Baylor University Press, 135–159. Avshalom-Gorni, D. and Najar, A. 2010. “The Stone Decorated with Reliefs from the Second Temple Period Synagogue at Migdal: First Impressions”. Studies in Archaeology and Ancient Art; Third Conference: Past and Present – Old Theories and New Finds. Jerusalem (in Hebrew). Avshalom-Gorni, D. and Najar, A. 2013. “Migdal: Preliminary Report”. Hadashot Arkheologiyot 125. Available from: http://www.hadashot-esi.org.il/report_detail_ eng.aspx?id=2304 (accessed 28 February 2020). Bauckham, R. 2015. “Further Thoughts on the Migdal Synagogue Stone”. NovT 57: 113– 135. Bauckham, R. and De Luca, S. 2015. “Magdala As We Now Know It”. Early Christianity 6: 91–118. Binder, D. D. 2014. “The Mystery of the Magdala Stone”. In: Warner, D. A. and Binder, D. D. (eds.), A City Set on a Hill: Essays in Honor of J. F. Strange. Mountain Home, AR: BorderStonePress, 17–48. Blenkinsopp, J. 1976. “The Structure of P”. CBQ 38: 275–292. Doering, L. 1999. Schabbat: Sabbathalacha und -praxis im antiken Judentum und Urchristentum. TSAJ 78. Tübingen: Mohr Siebeck. Fine, S. 2017. “From Synagogue Furnishing to Media Event: The Magdala Ashlar”. Ars Judaica 13: 27–38. Hachlili, R. 2017. “The Migdal Stone and its Ornamentation”. RB 124: 245–272. Halbwachs, M. 1925. Les cadres sociaux de la mémoire. Paris: Librairie Félix Alcan. Halbwachs, M. 1950. La mémoire collective. Paris: Les Presses Universitaires de France. Hayward, C.T. R. 1996. The Jewish Temple: A Non-Biblical Sourcebook. London: Routledge. Kershner, I. 2015. “A Carved Stone Block Upends Assumptions About Ancient Judaism”. The New York Times. December 8, 2015. Available from: https:// www.nytimes.com/2015/12/09/world/middleeast/magdala-stone-israel-judaism.html (accessed 28 February 2020). Krause, A. R. 2017. Synagogues in the Worlds of Flavius Josephus: Rhetoric, Spatiality, and First-Century Jewish Institutions. AJEC 97. Leiden: Brill. Levenson, J. D. 1989. Creation and the Persistence of Evil: The Jewish Drama of Divine Omnipotence. San Francisco: Harper & Row. Newman, J. H. 2008. “Priestly Prophets at Qumran: Summoning Sinai through the Songs of the Sabbath Sacrifice”. In: Brooke, G. J., Najman, H. and Stuckenbruck, L. T. (eds.), The Significance of Sinai: Traditions about Sinai and Divine Revelation in Judaism and Christianity. TBN 13. Leiden: Brill, 29–72. Newman, J. H. 2012. “Liturgical Imagination in the Composition of Ben Sira”. In: Penner, J., Penner, K. and Wassen, C. (eds.), Prayer and Poetry in the Dead Dea Scrolls and Related Literature: Essays in Honor of Eileen Schuller on the Occasion of her 65th Birthday. STDJ 98. Leiden: Brill, 323–338. Newman, J. H. 2018. Before the Bible: The Liturgical Body and the Formation of Scriptures in Early Judaism. New York: Oxford University Press. Rahmani, L. Y. 1994. A Catalog of Jewish Ossuaries in the Collections of the State of Israel. Jerusalem: The Israel Antiquities Authority. Runesson, A. 2017. “Synagogues without Rabbis or Christians? Ancient Institutions Beyond Normative Discourses”. Journal of Beliefs & Values 38: 159–172.

Contextualizing the Magdala Stone in Its Place

173

Stock, B. 1983. The Implications of Literacy: Written Language and Models of Interpretation in the Eleventh and Twelfth Centuries. Princeton, NJ: Princeton University Press. Talgam, R. 2013. “The Representation of the Temple and Jerusalem in Jewish and Christian Houses of Prayer in the Holy Land in Late Antiquity”. In: Dohrmann, N. B. and Reed, A. Y. (eds.), Jews, Christians, and the Roman Empire. Philadelphia, PA: University of Pennsylvania Press, 222–248. Tuan, Y.-F. 1979. Space and Place: The Perspective of Experience. Minneapolis, MS: University of Minnesota Press. Tweed, T. A. 2006. Crossing and Dwelling: A Theory of Religion. Cambridge, MA: Harvard University Press. Tweed, T. A. 2011. “Space”. Material Religion 7: 116–123. Zangenberg, J. K. 2010. “Archaeological News from the Galilee”. Early Christianity 1: 471– 484. Zerubavel, E. 2003. Time Maps: Collective Memory and the Social Shape of the Past. Chicago: University of Chicago Press.

The Rhetoric of Synagogue Space: Theoretical Issues in the Study of Jewish Institutions in Literary Sources

Andrew R. Krause The use of literary sources for the reconstruction of synagogue origins and development in the period of Classical Antiquity is fraught with issues of historicity and ideology. Especially with narratives that purportedly recount historical events, we must show due care in studying our sources with an understanding of both the Tendenz and context of the author. However, unlike the positivistic historians of past eras, we cannot merely discredit and disregard sources that evince such ideological coloring, but must instead seek to understand and to analyse the compositional elements of a text if we are to understand the historical truth claims being made. We cannot merely disregard such elements as literary artifice, but rather we must interpret such qualitative and evaluative aspects of the text if we are to understand fully the community spaces being described. As we seek to study such elements in a systematic, responsible fashion, one particularly effective tool is Critical Spatial Theory. Critical Spatial Theory, or social geography, acknowledges that the way we define, construct, understand, and remember space is different from how we do so with any other objects. Spaces both constrain action and mark boundaries. Different spaces can affect different moral and ethical requirements, and even possess different ontological properties (e.g., sacred, liminal, impure, etc.). In this paper, I will argue that Critical Spatial Theory and its various implications are not only a profitable tool for historians of the ideologically-loaded space of the synagogue, they are necessary if we are to understand what is being communicated. Thus, I will begin with a short survey of productive spatial theories. Following this, I will treat two passages from the works of the first-century Roman-Jewish historian Flavius Josephus, including one early historical narrative from Bellum Judaicum and a passage from his apologetical treatise Contra Apionem. I will then conclude with a brief treatment of the first-century Theodotus inscription (CIJ II 1404 = CIIP I 9). In each case, I will argue that spatial theory allows a more detailed and thorough examination and description of the data provided in each text.

176

Andrew R. Krause

1. Relevant Spatial Theory1 Scholars of the humanities and social sciences have increasingly acknowledged that we must move beyond purely conceptual and philological study if we are to understand the production of space within a given society and the literary works generated within such a society. Spaces are uniquely able both to constrain and to define our actions through the boundaries that they provide. Spaces are both implicitly understood and explicitly built with these effects in mind. This means that how we think about space must be both material and conceptual. No scholars have had a more enduring influence on the critical spatial theory than Henri Lefebvre and Edward Soja. When studying space, scholars had previously sought to differentiate the ideal from the real. However, Lefebvre and Soja challenged this binary opposition. Henri Lefebvre, for his part, rejected this duality by adding a third field of spatial experience. Lefebvre argues that there is ‘an-Other’ spatial element, which he refers to as ‘representational space’. This space is not merely a mixture of perceived space (real) and conceived space (ideal), but a deconstruction and reconstruction of what he calls the double illusion of the ideal and real, as they are encountered in the disorderly lives of individuals holding specific ideals, which constantly interact with the spaces ‘as they exist’ in the lived experience of the community (Lefebvre 1991: 229–291). Lefebvre’s theories were adopted and built upon by Edward Soja, who in turn offered his thirdspace theory. While following Lefebvre’s tripartite theory of space, Soja seeks to make the fields more systematic. For Soja, the inherently social nature of space manifests itself in the ordering and use of said space based on the actions and expectations of those in the space (Lefebvre 1991: 68–168). Soja’s three axes are the actual or perceived space (“firstspace”), the ideal or conceived space (“secondspace”), and the lived experience of space as the first two spaces are combined in the cognition of those in the space (“thirdspace”) (Soja 1996: 73–82). Playing on the English translation of Lefebvre’s French original, Soja notes that the thirdspace is not a representational space, but a space of representation. This theory is based on the notion that we have specific assumptions about how a space should be ideally used and understood, which will inevitably affect how we act in and experience a given space. To state this simply, we do not merely encounter the space itself, but we experience it as a disorderly combination of the space “as it is” and how we represent or expect it to be. This distinction between the space itself and the experience of the space by individuals and communities is also important because it forces us to acknowledge that, as with any form of religious experience, the ritual text is not an objective inscription of the religious experience. Rather, we possess a textual articulation of the writer’s experience within their liturgical and literary traditions, which is complicated by a complex matrix of ideals and expectations. It is precisely in this experience that  1

Sections 1–3 are abridged and adapted from Krause 2017.

Rhetoric of Synagogue Space

177

Soja and Lefebvre’s theories help us to understand how spaces are so central to socio-political dissent, as marginalized groups re-present their lived space in order to negotiate power within this space. Beyond the more systematic aspect of his theories, Soja omits the Marxist nature of Lefebvre’s spaces of dissent. Instead, he relies on Michel Foucault’s theory of the heterotopia. According to Foucault, the Enlightenment notion of a utopia fails specifically because it refers to no actual space; utopias are merely an inverted analogy to societies as they exist, rather than a concrete and workable use of space.2 For Foucault, the true space of dissent is the heterotopia. Heterotopias are “counter-sites, a kind of effectively enacted utopia in which the real sites that can be found within the culture… are simultaneously represented, contested, and inverted.”3 They are those spaces used to contest power through reuse and new representation, as they overlay a specific location with multiple historical meanings and uses at once. In other words, heterotopias effectively overlay the people, actions, ideas, and things that have never occurred together, though which are present in the place’s meaning for those using it as a space of dissent (Schofield 2012: 474). This highly idealized notion of spaces of protest forms a central component of Soja’s thirdspace theory and is congruent with his own post-colonial ideals, as the marginalized and oppressed are able to appeal to the history of their places in their fight against oppression (Soja 1996: 154–163). For our current purposes, it is important to note that heterotopias are intrinsically tied to the founding of the society in question, as such foundational acts have lasting meaning and often draw dissatisfied members of the society to harken back to the initial ideals of the group (as the dissenters understand them). Thus, for Foucault and Soja, the memory associated with these places is a representation of the community’s identity, which Mathieu de Bakker clarifies as “a spatial realm where memory offers itself in a … continuous dialogue with the present” (de Bakker 2013: 178). Thus, while Lefebvre cautioned against the use of these theories in the interpretation of literary and historiographic texts, Soja and Foucault’s attention to memory and representation should sufficiently obviate his initial concerns.

2. The Caesarea Synagogue as a Constructed and Bound Space In the first example to be discussed, we find an extended narrative regarding a series of events that, according to Josephus, led directly to the First Jewish Revolt. In this text, local Greeks (a vague ethnic trope used by Josephus for any people  2 Foucault 1986. According to Foucault, utopias fail to be actualized because they create a “placeless place”. 3 Foucault 1986: 24. Jo Heirman gives the intriguing example of the common imagery of a sea voyage associated with Greek symposia in the Greek lyrical poets as a heterotopia that dissociated the group from their mundane surroundings and acted as a unifying representation. See Heirman 2013.

178

Andrew R. Krause

who neighbor the Judaeans) invalidate the usage of a synagogue in Caesarea Maritima. While this may seem like just another story of cultural antagonism and retributive violence, Josephus uses this text and its various spaces in order to illustrate the encroachment of outsiders as a key factor in the initial conflagrations that led to open revolt. In this narrative, both the community space of the synagogue and the geographic space of Caesarea Maritima are key to understanding the narratival progression. Caesarea was a medial, border city that held meaning and importance for both Jews and the so-called ‘Greeks’ This text uses the real, historical space (firstspace), though understood through the spatial ideals and memories of Josephus (secondspace), as he communicates the implications of these events in this space for his (and likely at least some of the Jewish communities’) experience. This was a space that was already physically marginalized and a site of oppression of the Jews. It should be noted that this text stands out against other texts from Bellum regarding synagogues, as it refers to this space exclusively as a synagogue (συναγωγη ) rather than as a temple. In other texts such as B.J. 4.407–408, 6.128, and (likely) 7.143–144, Josephus refers to these spaces as holy places (ἱερά). This was likely done in order to maximize the transgression leveled against the Jewish rebels who he sought to condemn, as the synagogues that they plundered would be portrayed as sacred space in these narratives in the reading of the wider Roman audience of this history (Krause 2017: 175–181). As we know from the writings of the first-century BCE, Roman military engineer Vitruvius, Romans did not merely view the planning of cities, especially the placement and construction of holy places, as a matter of pure expediency. In a rather uncharacteristic way, this empire, which has become known for its pragmatism, took the symbolism regarding planning of cities, provinces, and public buildings very seriously. This symbolism must be taken into account as we address the pivotal scene in the Caesarean synagogue (B.J. 2.285–292). This text has been taken by some synagogue scholars as proof of an inherent sacrality of the building due to the fact that the space was susceptible to impurity. However, a better understanding of Roman city-craft should lead us to nuance Josephus’ actual claims regarding the nature of this space. This text relies on such assumptions of symbolic construction and bounded nature of the Jewish public assembly space in order to communicate the severity of the religious insult paid to the Jews. The hostilities in Caesarea Maritima were the result of a protracted conflict between the Syrians and the Jews in Caesarea beginning in B.J. 2.266. The Jews of the city claimed to control the polis based on the founding of the city by Herod.4 In response, the Syrians pointed out the previously Syrian nature of Strato’s Tower (i.e., the previous settlement) and that Herod’s erection of nonJewish statues and temples would call the Jewish nature of the city into question.  4

On the ideological importance of the founder of a city, see Fai 2000: 268–272. The obvious example of Romulus and Remus in Livy book 1 is instructive.

Rhetoric of Synagogue Space

179

This led to internecine strife, which had to be put down by Felix. Subsequent procurators failed to quell the tension. Following this, Josephus reports the scene that would officially open the Revolt (B.J. 2.284–292): a Jewish synagogue, which had been unsatisfactorily built in a marginal space due to the refusal of neighbours to sell the Jewish community the necessary land, was targeted by Syrian troublemakers. . . the Judeans in Caesarea, having a meeting [place] (συναγωγη) beside a site whose owner was a certain Caesarean Greek, tried hard and often to acquire the spot, offering a price many times its worth; but while disdaining their appeals, with added insult, he himself built across the site, constructing workshops. He was thus leaving them a passageway that was both narrow and constrained in every direction.5

Thus, the placement of this synagogue, whether it is to be understood as a civic or sacred edifice, is highly marginalizing. However, this sense of marginalization was exacerbated when, on the Sabbath, the Jewish community approached the alleyway that led to the synagogue and found a Syrian youth sacrificing a bird on an overturned pot at the synagogue’s entrance. Despite pleas for peace on the part of the Jewish leaders, Jewish youths engaged in battle with the Syrians who had orchestrated the sacrifice. After a previous failed attempt to bribe Florus for aid— Florus took the money but did nothing to help—the Jews were left to remove their Law scroll and flee, for which they were arrested on the charge of removing sacred paraphernalia from the city. This strife is staccatoed with issues of city and building craft, especially as it relates to the sacred. That Josephus was deeply concerned with the religious geography and architecture of Israel in Bellum is the subject of a monograph by Jason von Ehrenkrook entitled Sculpting Idolatry in Flavian Rome: (An)Iconic Rhetoric in the Writings of Flavius Josephus. With regards to Bellum, von Ehrenkrook argues that Josephus consistently inverts the common, Graeco-Roman treatment of statues as delimiting sacred space; he treats them instead as a profaning element, becoming a boundary marker both religiously and politically (von Ehrenkrook 2012: 101– 102). While we should indeed take care not to retroject a modern geographic understanding of such space, there is good reason to acknowledge that Josephus accepted an idea of general national boundaries. As in the works of Vitruvius (Arch. 1.7.1–2), the placement of statues and temples would demonstrate the nature of a given city and its relation to the larger ‘bodies’ of the province and the whole Empire.6 In terms of Caesarea, von Ehrenkrook points out the prominence of the statues mentioned by the Syrians in B.J. 2.266 and the fact that this city stands on the very boundaries of the Jewish heartland.7 Indeed, we know that the  5

B.J. 2.285–286. Mason 2008: 232–234. Fai 2000: 274; McEwen 2003: 279; and Kloppenborg 2000: 237. 7 See von Ehrenkrook 2012: 107–108, 127. Von Ehrenkrook aptly points to the difference between B.J. 2.266 and its parallel text in A.J. 20.173, in which the issue is the lack of Jews at the founding of Strato’s 6

180

Andrew R. Krause

history of Caesarea Maritima was fraught with the issues of being a connection point between the Greek and Jewish communities in the north, as it acted as a conduit for Roman imperial presence, military force, and goods eastward (Kloppenborg 2000: 230). Von Ehrenkrook is thus correct to use Caesarea as a key boundary and a source of potential profanation from the outside world. Transgression against the Jewish community in Caesarea would have been felt by all Jews due to the medial placement of this city. Such civic and geographic marginalization through spatial transgression is referred to by Tim Cresswell as “cultural negation”, a process by which a nation or group is slowly pushed back geospatially with both symbolic and political consequences (Cresswell 1996: 149– 50). Such loss was emphasized in Josephus’ telling of the story, as he presents this sacrilege as a “last straw” for the Jews. However, von Ehrenkrook ignores the possible implications of these issues for the synagogue pericope under discussion, despite some instructive correlations. Given the ideological importance of the placement of public and sacred buildings for both Josephus and Vitruvius, the relegation of the Jewish synagogue to a remote, industrial space should be viewed as an act of derision. That the Syrian populace in B.J. 2.284–285 forced the Jewish population to follow an alleyway to reach their assembly space would be marginalizing for Vitruvius, as their synagogue was removed from main streets and the harbour, as well as obscuring the sight of the space. As Vitruvius states in Arch. 1.7.1 (LCL), The choice of sites [for public buildings] for the convenience and common use of citizens is to be explained. . . if the ramparts are by the sea, a site where the forum is to be put is to be chosen next to the harbour; but if inland, in the middle of the town. But for sacred buildings of the gods under whose protection the city most seems to be, both for Jupiter and Juno and Minerva, the sites are to be distributed on the highest ground from which the most of the ramparts is to be seen.8

Thus, whether the synagogue is to be understood as a civic or sacred edifice, its construction and placement do not meet with any of Vitruvius’ standards for the building of sacred or civic assembly spaces.9 To add to this, irregular space (i.e., space lacking perfectly round or quadrangular shape) was also a sign of lack of divine approval.10 All of this would have led to a sense that the Jewish community was being symbolically, as well as politically, marginalized.

 Tower and whether or not the Jews had claim to ἰσοπολιτεία. Regarding the different texts, see Levine 1974: 381–397; and Kasher 1977: 16–27. 8 Arch. 7.1.2 goes on to describe the importance of space for the individual deities, with prominence of place correlating to prominence of the deity. See also Arch. 3.2.1. 9 While a synagogue excavated in the north fortification wall near the sea was found, this synagogue dates to ca. 100 years later, so should not be taken as a possible site; Mason 2000: 232–233 n. 1829. 10 Arch. 3.1.3. See McEwen 2003: 166–167, 181.

Rhetoric of Synagogue Space

181

Many scholars have viewed the mention of sacrilege in this scene as further proof of synagogue sanctity,11 though such an assertion is rather vague. This is partly because there is little consensus regarding the exact nature of the sacrilege. Frowald Hüttenmeister attempts to connect the sacrifice of two birds with the sacrifice of birds for removal of leprosy in Lev 14:4 (Hüttenmeister 2000: 83–84). This answer, however, expects too much knowledge of Jewish texts and practices on the part of both the expected Roman reader and Syrian youth, and it runs counter to Josephus’ practice of explaining all relevant Jewish practices in Bellum. Brighton is probably closer to the mark in claiming that Josephus presents the resultant στάσις as being the polluting agent, as Josephus does elsewhere.12 Indeed, Josephus consistently presents those who seek peace (in this case the Jewish elite) as intrinsically positive, while condemning those who cause war.13 However, we may also add that the act of sacrificing something was intrinsically an act of claiming in the Roman world (e.g., Sophocles, Ant. 256, 775). Once a place had been thus claimed, religious convention would necessitate strict ceremonies to approach the space in question, otherwise this act would pollute the one approaching (Hersey 1988: 43). Thus, the synagogue need not have been previously sacral to have been viewed as polluted by the youth for a Roman readership. In this reading, the alleyway is that which would have been rendered inaccessible to the Jewish community. This sacrilege would thus bring to completion the explicit programme of the Syrians to bar Jews from their synagogue (Bilde 1999: 19). Thus, at the very start of the Revolt, we find Josephus narrating a version of the Caesarean crisis which places a premium on the location of the Jewish synagogue on the margins of what was itself a city on the boundary of the Land. The Jews were being pushed out of a city in a way that encroached both on their local community space and their land as a whole. That this took place in synagogue environs, which elsewhere have been treated as sacred space, is more narratively symbolic than historical. It is a statement of their spatial marginalization and the work of the Syrians to keep them from creating a meaningful assembly space in such a way as they conceive of such space. The Jews did not seek out war (see B.J. 7.361), but were forced by the marginalizing actions of those around them to fight for their rights. The religious and political boundaries of the Jewish nation are thus of key importance here and the greatest insult symbolically takes place in, or at least in front of, the synagogue.

 11

E.g., Sanders 1997: 10; and Binder 1999: 156. Cf. B.J. 2.424; Brighton 2009: 73. 13 McLaren 1998: 102–103. We should also note that in the closing speech of Eleazar ben Yair at Masada in B.J. 7.361, the Caesarean Jews are specifically said to have not even contemplated starting this Revolt. 12

182

Andrew R. Krause

3. Proper and Improper Synagogues in Contra Apionem 2 The second example comes from Josephus’ latest extant work and his only treatise: Contra Apionem. This apologetical work is intended to counter the various aspersions cast upon Judaism by Greek, Roman, and Egyptian philosophers and historians (C. Ap. 1.6–2.144) and to present a more agreeable picture of Judaism within these more dominant cultures (2.145–286). This work is thus rife with ethnic, colonial, and other political arguments and implications that will inevitably affect the construction and portrayal of a Jewish institution such as the synagogue. This would be important for numerous reasons, especially given the placement of the synagogue within both Jewish culture and Roman cities at the time. As we saw in the previous example, synagogues were both targets for anti-Jewish sentiment and action, and centres of Jewish identity. Both this Jewish ideology and the urban geography of the Jewish institutions made their spatial construction important. For Josephus, it was important to show how various aspects of Jewish life made this nation ideal subjects of Roman rule. One of the primary ways in which this goal was accomplished was by showing that Jewish Law made Jews uniquely suited as citizens of the Empire, rather than as the deviants that certain philosophers had claimed. Thus, Jews were the very picture of legal knowledge and education. The negative portrayal of the synagogue is the first issue raised by Apion against the Jews, as Apion presents Jewish προσευχαί as originating under Moses in Heliopolis, Moses, as I have heard from old people in Egypt, was a native of Heliopolis, who, being pledged to the customs of his country, erected prayer houses (προσευχαί), open to the air, in the various precincts of the city, all facing eastwards; such being the orientation also of Heliopolis. In place of obelisks, he set up pillars, beneath which was a model of a boat; and the shadow cast on this basin by the statue described a circle corresponding to the course of the sun in the heavens. (C. Ap. 2.10–11; LCL)

Based on the grammatical errors in the text, Barclay states that this passage was likely a truncated summary from another, earlier writer such as Manetho (who shared Apion’s negative aetiology of the Jews) (Barclay 2001: 174 n. 31). It should be noted here that not all elements of Apion’s description were without warrant. The Leontopolis Temple of Onias (B.J. 7.421–432; A.J. 13.62–73)14 had still been standing during Apion’s day and was said to have been preceded by many different Jewish “temples” (A.J. 13.66). While the προσευχή in C. Ap. 2.11 did not entirely match the Leontopolis Temple in form, there were similarities, especially in terms of the Egyptian sun iconography. This may be seen both in the sun dial in Apion’s version and the hanging, golden lamp that took the place of the  14

See discussion in Krause 2016.

Rhetoric of Synagogue Space

183

lampstand in B.J. 7.429. Likewise, Egyptian temples in Heliopolis were said by Herodotus (Hist. 2.111) to have obelisks, which Moses is said to have replaced with pillars. If this claim were true, it would make this precinct deviant to the Egyptians and present Moses as a modifier of tradition.15 This is in line with the accusations against Moses as a deviant leader of former Egyptians afflicted with tumors of the groin (σαββώ, based on Apion’s etymology of Sabbath), which Josephus would refute in C. Ap. 2.12–40. Likewise, certain texts present Jews as being celestially-oriented in prayer,16 though whether this was a synagogue practice has been vigorously questioned.17 Praying with hands toward the sky is even referred to in C. Ap. 1.208–209, After telling this story and mocking Stratonice for her superstition, Agatharchides uses as an example a story about us, and writes as follows: Those called Judeans inhabit the best fortified city of all, which, it happens, the natives call Hierosolyma, and it is their custom to do no work every seventh day—neither to carry weapons on the occasions mentioned, nor to put their hands to any agriculture, nor to attend to any public service—but to pray in the temples until evening, with hands outstretched. (Barclay 2001: 118–119)

This is another citation which some have taken as a historically trustworthy synagogue text.18 However, a proper, contextual reading of this text illustrates that the Jews are being caricatured as passive, because they spurn proper work for improper worship practices. This negative portrayal runs contrary to the ideal synagogue conceived of in Josephus’ positive exposition of the Law, and should be rejected as a historical truth claim. Given Apion’s negative account and the solar aspects it describes, it is more likely that this vague celestial focus is being offered by Josephus’ interlocutors as an insult. Given this negative presentation, we should not be surprised that when Josephus turns to the positive presentation of the synagogue as an ideal educational institution, all such superstition and negative portrayal of the Sabbath are opposed. This is done in a section outlining the work of Moses and the structure of his constitution (C. Ap. 2.151–189; see Barclay 2001: xxi). In C. Ap. 2.175, Josephus states, [Moses] appointed the Law to be the most excellent and necessary form of instruction (παίδευμα), 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 (συλλεγω) 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. (LCL)

 15

Barclay 2001: 175 n. 38. E.g., Philo, Flacc. 121; Sir 35:20–21. 17 Zeitlin 1973: 1.92–133; Levine 2005: 163–169; and Reif 1993: 44–52. 18 E.g., Binder 1999: 412. Binder states that the plural hiera is enough to prove that we are dealing with a legitimate synagogue text. 16

184

Andrew R. Krause

Many will note that this passage does not specifically state that these readings take place in the synagogue. However, the practice of reading the Law seems to have been a common and well-known practice in synagogues during the Sabbath at this time.19 As Stefan Reif aptly states, the study of ancient worship must move beyond conventional, contemporary ideas of liturgy and include the study of texts as a primary aspect of liturgy, alongside prayers, amulets, and the like.20 Specifically, we must note the liturgical nature of reading sacred texts in a communal setting, rather than retrojecting views of education as separate from ritual observance. According to Philo, the expressed purpose of education in the Jewish Law is to make all Jews “believe that the Father and Creator of the universe is one God” (Legat. 115).21 As Runesson argues, reading of the Law stands out beyond all other liturgical aspects of synagogue practice and was the uniquely Jewish practice of the synagogue.22 Thus, both here and in Nicolas’ speech of defense (A.J. 16.42–43), Josephus counters accusations that the Jews are secretive and anti-social by lauding the communal knowledge of the Law and its positive effects on the Jews as citizens when performed as part of the synagogue liturgy (Cohen 1987: 421–422). The Jews are here said to have complete knowledge of their Law, and the entire community is entrusted with the responsibility of educating the young (C. Ap. 1.60; 2.188). This leads to the concern to pray for the welfare of the whole community, which is contrasted specifically with the kind of self-serving licentiousness and debauchery found in mystery associations (C. Ap. 2.195–196). Thus, Josephus is presenting this legal teaching as a practical educational performance within the community and for the community’s benefit. Philo echoes Josephus’ interest in the synagogue as a place of education in Legat. 156, stating, He knew therefore that they have houses of prayer and meet together in them, particularly on the sacred sabbaths when they receive as a body a training in their ancestral philosophy. (LCL)

However, Josephus extends the institution to Moses.23 It should be noted that neither treats these assemblies as specifically sacred places, but rather both emphasize the supra-local nature of the reading of the Law, in order to reach all Jews.24  19 E.g., Philo’s Legat. 156; Hypoth. 7.11–14; Somn. 2.127; Prob. 81; Mos. 2.215–216; as well as Luke 4:16–22; Acts 13:14–15, 15:21. See Levine 2005: 148–149; Levine also notes that this is how Philo presents the Essenes as worshiping in their synagogues (Prob. 81–82) and that Suetonius presents one Diogenes as lecturing in Rhodes on the Sabbath (Tib. 32.2). It is notable that Legat. 156 presents Caesar Augustus as acknowledging the importance of the study and instruction of the Law in Jewish culture. 20 Reif 2004: 442. 21 See Borgen 2001: 91. 22 Runesson 2001: 191–192, 451; see also Levine 2005: 148. 23 See Goodman 2007: 220. 24 For discussion of synagogue sacrality in history, see Krause 2016.

Rhetoric of Synagogue Space

185

This conception fits the description of so many other sources regarding the reading of the Law in synagogues, so Josephus’ ideal (secondspace) seems firmly set in the synagogue ‘as it was’ (firstspace), which should lead us to see this as Josephus’ lived experience of the synagogue (thirdspace). The history with which Josephus overlays the synagogue is much more complex here than in his other works. Josephus utilizes both Roman and Jewish ideals, including legal and pedagogical archetypes, in the construction of the synagogue as heterotopia in this work. Also, Josephus needed to counter the overlay of negative, etic historical meanings which were attached to this institution by its enemies. For Josephus, the importance of the synagogue lay in its communal nature and scripture reading, as he creates a supra-local heterotopia, which he connects to Moses through the reading of the Mosaic Law.

4. Inscribed Spatial Construction in the Theodotus Inscription The final example is that of the so-called Theodotus inscription (CIJ II 1404 = CIIP I 9). This short text was found in the 1913–14 season by the French archaeologist Raymond Weill in a cistern on the eastern slope of Ophel, as part of the City of David excavations. This inscription commemorates the building of a synagogue by Theodotus son of Vettenus somewhere in the vicinity of Jerusalem.25 Theodotus claims that both his father and grandfather were archisynagōgoi and priests.26 As with all inscriptions, this text is both literary and material. It’s use in the building of the synagogue makes this text uniquely spatial in its very nature. As such, it has different properties and implications that affect its reading and use. However, as with the Josephan texts discussed previously, it makes specific historical truth claims that constitute important data for synagogue research. This inscription is important as it enumerates several key features and purposes for the synagogue of which it was originally a part. This short text states, Theodotus, son of Vettenus, priest and ruler of the synagogue (ἀρχισυνάγωγος), son of a ruler of the synagogue (ἀρχισυνάγωγος), grandson of a ruler of the synagogue (ἀρχισυνάγωγος), built the synagogue (συναγωγή) for the reading of the law and the teaching of the commandments, and also the guest chamber and the upper rooms and the ritual pools of water for accommodating those needing them from abroad, which his fathers, the elders (πρεσβύτεροι) and Simonides founded.

 25 It would be unlikely for this stone to have been moved any further than needed, so the initial provenance should be set in the vicinity of Ophel. See Levine 2005: 57. 26 For an interesting analogy to this listing of an individual with his father and grandfather, see the recently excavated menorah mosaic in the 5th century CE synagogue at orvat Kur, which lists the three names above the menorah. Unfortunately, the connections between this family is not explicitly stated, though some level of benefaction can reasonably be inferred. See Zangenberg 2017: 114–115.

186

Andrew R. Krause

Immediately after stating that he had the synagogue built, Theodotus specifies that the synagogue was chiefly built for the reading of the law, teaching of commandments, and providing lodgings and miqwa’ot for those travelling from abroad. Beyond his claims to stature as a priest and in the long line of archisynagōgoi (not unlike Josephus’ own claims to familial standing in Vita 1– 12), this text explicitly states this synagogue’s raison d’être (Runesson 2001: 315– 317). However, the fact that this needed to be stated so explicitly should force us to keep from understanding these purposes as entirely normative. That is to say, these purposes represent the ideals of the space (secondspace) of this individual or family, rather than necessarily representing a generalized purpose. The other point worthy of mention here is that this inscription was found as spolia, and therefore not in situ. How we read this text is deeply affected by a lack of spatial context. As Levine warns, we do not even know if the lodging and bathing were done in the same building or if another building was connected (Levine 2005: 59). According to Peter Richardson (2004: 128), the lack of discernible synagogue remains should force us to be wary of definitive interpretations. However, I would go even further than Richardson. This inscription is defined by its intended context just as much as it defines the purpose of Theodotus’ synagogue. Critical Spatial Theory forces us to recognize the importance of the actual space and its connection to the ideals communicated here, and we must thus treat this text as incomplete outside of its provenance. At most, we can say that Theodotus intended the synagogue that he had built to house the same sort of teaching found in other synagogues mentioned in our second example, and that it should welcome outsiders. However, the space itself (or lack thereof) severely hampers what can be deduced from this text.

Conclusion In all three examples, we have found that literary texts must be interpreted with care when using them in the historical reconstruction of synagogue development in the first centuries. In these texts—even those recording purportedly historical events—the spaces “as they were” and the ideals of those remembering or building them are intricately entwined in the literary representations of these synagogues. Thus, we can neither credulously take these texts as normative descriptions of the spaces and their intended usage, nor ignore them as mere literary artifice as we seek to integrate all known data from this axial time in synagogue development. Josephus used various ideals and experiences of historical synagogues in his various tendentious works, just as Theodotus was seeking selfaggrandizement for his benefaction. However, both of these individuals betray a high level of cultural importance in the use and construction of synagogues. As we move forward, we must seek to understand the memories of these spaces as

Rhetoric of Synagogue Space

187

inscribed in these texts as important and meaningful enough to record, though be careful not to take these texts as meaningful outside of their rhetorical and spatial contexts.

Bibliography Barclay, J. M. G. 2001. Against Apion: Translation and Commentary. Leiden: Brill. Bilde, P. 1999. “Was hat Josephus über die Synagogue zu sagen”. In: Kalms, J. U. (ed.), Internationales Josephus-Kolloquium Brüssel 1998. Münster: LIT Verlag, 15–35. Binder, D. D. 1999. Into the Temple Courts: The Place of the Synagogue in the Second Temple Period. Atlanta: Society of Biblical Literature. Borgen, P. 2001. “Application of and Commitment to the Laws of Moses: Observations on Philo’s Treatise”. On the Embassy to Gaius. SPhilo 13: 86–101. Brighton, M. A. 2009. The Sicarii in Josephus’ Judean War: Rhetorical Analysis and Historical Observations. Atlanta: SBL Press. Cohen, S. J. D. 1987. “Respect for Judaism by Gentiles according to Josephus”. HTR 80.4: 409–30. Cresswell, T. 1996. In Place/Out of Place: Geography, Ideology, and Transgression. Minneapolis: University of Minnesota Press. de Bakker, M. 2013. “The Epitaphios, Civic Ideology, and the Cityscape of Classical Athens”. In: Klooster, J. and Heirman, J. (eds.), The Ideologies of Lived Spaces in Literary Texts, Ancient and Modern. Gent: Gingko Academia Press, 175–99. Fai, S. 2000. “Architecture and Conflict in Caesarea Maritima”. In: Donaldson, T. L. (ed.), Religious Rivalries and The Struggle for Success in Caesarea Maritima. Waterloo: Wilfrid Laurier University Press, 268–72. Foucault, M. 1986. “Of Other Spaces”. Diacritics 16.1: 22–27. Goodman, M. 2007. Judaism in the Roman World: Collected Essays. AJEC 67. Leiden, Brill. Heirman, J. 2013. “Symbolic ‘Lived Spaces’ in Ancient Greek Lyric and the Heterotopia of the Symposium”. In: Klooster, J. and Heirman, J. (eds.), The Ideologies of Lived Spaces in Literary Texts, Ancient and Modern. Gent: Gingko Academia Press, 83–94. Hersey, G. 1988. The Lost Meaning of Classical Architecture. Cambridge, MIT Press. Hüttenmeister, F. G. 2000. “Συναωή, προσευχη, und τόπος bei Josephus und der rabbinische Hintergrund”. In: Kalms, J. U. (ed.), Internationales Josephus-Kolloquim Aarhus 1999. Münster: LIT Verlag, 77-96. Kasher, A. 1977. “The Isopoliteia Question in Caesarea Maritima”. JQR 68.1: 16–27. Kloppenborg, J. S. 2000. “Ethnic and Political Factors in the Conflict at Caesarea Maritima”. In: Donaldson, T. L. (ed.), Religious Rivalries and The Struggle for Success in Caesarea Maritima. Waterloo: Wilfrid Laurier University Press, 227–248. Krause, A. R. 2016. “Diaspora Synagogues, Leontopolis, and the Other Jewish Temples of Egypt in the Histories of Josephus”. JAH 4.1: 88–112. Krause, A. R. 2017. Synagogues in the Works of Flavius Josephus: Rhetoric, Spatiality, and First-Century Jewish Institutions. Leiden: Brill. Lefebvre, H. 1991. The Production of Space. trans. D. Nicholson-Smith. London: Blackwell. Levine, L. I. 1974. “The Jewish-Greek Conflict in First Century Caesarea”. JJS 25.3: 381–97. Levine, L. I. 2005. The Ancient Synagogue: The First Thousand Years. 2nd ed. New Haven: Yale University Press. Mason, S. 2008. Judean War 2: Translation and Commentary. Leiden: Brill. McEwan, I. K. 2003. Vitruvius: Writing the Body of Architecture. Boston: MIT Press.

188

Andrew R. Krause

McLaren, J. S. 1998. Turbulent Times? Josephus and Scholarship on Judaea in the First Century CE. Sheffield: Sheffield Academic Press. Reif, S. 1993. Judaism and Hebrew Prayer: New Perspectives on Jewish Liturgical History. Cambridge: Cambridge University Press. Reif, S. 2004. “Prayer in Early Judaism”. In: Egger-Wenzel, R. and Corley, J. (eds.), Prayer from Tobit to Qumran. Berlin: De Gruyter, 439–64. Richardson, P. 2004. Building Jewish in the Roman East. JSJSup 92. Leiden: Brill. Runesson, A. 2001. The Origins of the Synagogue: A Socio-Historical Study. ConBNT 37. Stockholm: Almqvist & Wiksell. Sanders, E. P. 1997. “Common Judaism and the Synagogue in the First Century”. In: Fine, S. (ed.), Jews, Christians, and Polytheists in the Ancient Synagogue: Cultural Interaction during the Greco-Roman Period. London: Routledge, 1–17. Schofield, A. 2012. “Re-Placing Priestly Space: The Wilderness as Heterotopia in the Dead Sea Scrolls”. In: Mason, E. F. et al. (eds.), A Teacher for All Generations: Essays in Honor of J. C. VanderKam, Vol. 1. JSJSup 153,1. Leiden: Brill, 469–90. Soja, E. W. 1996. Thirdspace: Journeys to Los Angeles and Other Real-and-Imagined Places. London: Routledge. von Ehrenkrook, J. 2012. Sculpting Idolatry in Flavian Rome: (An)Iconic Rhetoric in the Writings of Flavian Rome. Atlanta: SBL Press. Zangenberg, J. 2017. “The Menorah on the Mosaic Floor from the Late Roman/Early Byzantine Synagogue at "orvat Kur”. IEJ 67.1: 110–26. Zeitlin, S. 1973. Studies in the Early History of Judaism. New York: Ktav.

The Contributions of Historical and Archaeological Study of Early Synagogues to Historical Jesus Research

Jordan J. Ryan The contemporary quest for the historical Jesus is marked by the concern to recover a historical portrait of a truly Jewish Jesus, one that is centered and rooted in the world and thought of Second Temple Judaism. As current research has shown, the synagogue was one of the premiere institutions of Jewish society in the Land of Israel during the Second Temple period. As attested by the contributions to the present volume, synagogue scholarship has continued to grow in leaps and bound in recent years, contributing fresh insights to our comprehension of the social, religious, and political world of early Judaism. Despite this, synagogue scholarship and archaeology has had little impact on and reception in research on the historical Jesus. The potential for connection between research on synagogues and the study of the historical Jesus is clear, since synagogues are featured prominently in the canonical Gospel narratives. Just as the synagogue was a vital dimension of early Palestinian Jewish life and society in antiquity, so too was it a vital dimension of Jesus’ life and ministry as they are narrated by the evangelists. We can sketch a brief portrait of Jesus’ involvement with synagogues as detailed by the canonical Gospels. According to the data presented by the evangelists, Jesus customarily attended synagogue gatherings on the Sabbath (Luke 4:16), and local synagogues were the primary locus of his teaching and proclamation of the Kingdom of God (Mark 1:38; Matt 4:23; Luke 4:14–15, 43–44; John 18:20), at least while he was outside of Jerusalem. Moreover, according to the canonical Gospels, Jesus performed exorcisms and healings in synagogue settings (Mark 1:21–28, cf. Luke 4:31–37; Mark 3:1–6, cf. Matt 12:9–14, Luke 6:6–11; Luke 13:10–17), participated in synagogue proceedings by reading from the scroll of Isaiah in his hometown (Luke 4:16–30, cf. Mark 6:1–6, Matt 13:54–58), discussed and debated the interpretation and practice of Torah in synagogue gatherings (Mark 3:4, cf. Luke 6:9; Matt 12:10–12; Luke 13:14–17; John 6:30–33, 48–50, 58), told parables in a synagogue (Luke 13:18–21), interacted with synagogue officials (Mark 5:22, cf. Matt 9:18, Luke 8:41, possibly also John 4:46; Luke 4:20), as well as with a synagogue benefactor (Luke 7:1–10, esp. v. 5), and discussed inappropriate behavior in synagogues in his teaching (Matt 6:2, 5). Jesus was met with

190

Jordan J. Ryan

varied responses to his teaching and actions in synagogues, ranging from rejoicing (Luke 13:17) and recognition of his authority (Mark 1:27) on one end of the spectrum to unbelief (Mark 6:6), rage (Luke 4:28), and rejection resulting in a loss of followers (John 6:60, 66). The Gospels also claim that Jesus envisioned his followers receiving corporal punishment at the hands of local councils in synagogues (Mark 13:9 and parallels), and being excluded from participation in synagogue assemblies (John 16:2, cf. 9:22, 12:42, see Bernier 2013). Whether or not we accept any of this data as literally reflecting historical reality, we must at least admit that the evidence provided by the Gospels concerning Jesus’ interaction with and participation in synagogue assemblies is rich and well attested. It also spans several streams of tradition, as the data above is drawn from both Mark (assuming Markan priority) and John, as well as unique Lukan and Matthean material. Although the criterion of multiple attestation has rightly been subject to critique (e.g., Goodacre 2012) along with the rest of the criteria of authenticity (e.g., Hooker 1972; Keith and Le Donne 2012; Allison 2011), the fact that synagogue material is found across tradition streams is noteworthy. When we consider the data from a historical perspective, each datum constitutes a node of evidence, which we can connect with strands of logical inference and historical imagination (Collingwood 1993; cf. Ryan 2015), thus creating a robust picture of the past. Because the evidence is so plentiful and rich, and because it is corroborated (as opposed to “multiply attested”) and supported by multiple data points, the most logical conclusion to draw is that the synagogue material in the Gospels is indicative of a strong memory in the early tradition of Jesus being active in synagogues, teaching and performing actions that were understood as miraculous by his first century audience.1 Ultimately, this means that the breadth and depth of the evidence demands the responsible historian to take the evangelists’ testimony to Jesus’ activity in synagogues seriously as a reflection of the past. In the words of Michael Oakeshott, “what really happened” must be replaced by “what the evidence obliges us to believe” as the goal of history, since “all that history has is ‘the evidence’; outside this lies nothing at all” (Oakeshott 1994: 107–108).  1 Belief in the miraculous was fairly widespread in antiquity. Other examples from early Judaism outside of the Jesus movement include Tob 8:1–3; 4Q560; 11Q5 XXVII 2–11; 11Q11 I 4, 6; IV 4; Jub. 10:1– 14; Josephus, A.J. 8.45–48; m. Ber. 5:5; b. Ber. 34b; m. Sanh. 10:1. Furthermore, on physicians, see Sir 38:1–8. In the Hebrew Bible, see the healing miracles in the Elisha cycle, in 2 Kgs 4:17–37 and 5:1–14. A well-known example from Greco-Roman tradition is that of Apollonius of Tyana, in Philostratus, Vit. Apoll. 3.39; 4.20. Because of the widespread belief in the miraculous, it is reasonable to hold that the healings and exorcisms performed by Jesus in synagogue settings were understood to be miraculous by ancient peoples, whether or not modern historians would see them in the same way. From the perspective of the modern science of history, these sorts of “mighty deeds” can be understood in terms of anthropological, sociological, or psychosomatic phenomena. See, for example, Funk and the Jesus Seminar 1998: 531; Bond 2012: 106–108; Eve 2002: 350–360; Lewis 1971; Witmer 2012: 22–60; Guijarro 1999; and Miquel 2010.

Contributions of Early Synagogues to Historical Jesus Research

191

Given that this is the case, it is all the more puzzling that the academic study of the synagogue has been absent from the history of scholarship on the historical Jesus for so long.2 That is not to say that the synagogue itself has been absent from research on Jesus and the Gospels. However, as Anders Runesson has put the matter, Some scholars assume a relationship between the fields of New Testament and synagogue studies but never argue or develop the case taking into account the specifics of recent synagogue research. Others simply ignore the question, while yet others may even understand synagogue studies as peripheral to the field of New Testament studies, such that publications in the former field are judged to be of marginal relevance for the latter. Common to several of the studies which do address the place of the synagogue in relation to Jesus and the Gospels is, unfortunately, a lack of attention to recent critical developments in the study of ancient synagogues. (Runesson 2014: 265–266)

Runesson’s observations in the passage quoted above are important to note and to take into consideration, both for scholars who research synagogues and for scholars who work primarily on the Gospels. Unfortunately, inaccuracies concerning early synagogues persist in scholarship on Jesus and the Gospels. This is especially curious given the recognition of the importance of understanding Jesus within Judaism in contemporary scholarship, particularly since the influential work of Geza Vermes (1973; cf. Vermes 1993) and E. P. Sanders (1985). Perhaps a few of the more common issues can be addressed here. A survey of recent commentaries reveals that such things as the separation between men and women and the presence of the raised platform (bema) are sometimes presented as though they were found in early synagogues.3 However, as previous scholars have shown through examination of both literary and archaeological evidence, the separation of men and women in synagogues is not attested until the medieval period (Safrai 1964; Brooten 2020: 103–138; Levine 2005: 502–505). Furthermore, the raised platform with Torah shrine is a development that appears in later synagogues, but not in synagogues of the Second Temple period (see Hachlili 2013: 163–198). Likewise, questions concerning the existence of synagogue buildings in the Second Temple period are frequently discussed, despite the fact that the issue is now considered to be definitively and clearly settled in favor of the existence of synagogue buildings (see, e.g., Kloppenborg 2006; Catto 2007; Magness 2012: 286–292; Ryan 2017a: 102–108). It is safe to say that the question about the existence of synagogue buildings in the Second Temple period belongs to the history of scholarship rather than to current research.  2

A complete review is not possible here. Refer to Ryan 2017a: 95–121. For the sake of collegiality, it is perhaps not productive to name particular works that are guilty of these issues. Instead, it is preferable to note the phenomenon in general, and to concentrate our efforts on how we might move forward. 3

192

Jordan J. Ryan

Given the evident distance between synagogue scholarship and scholarship on Jesus and the Gospels, there is a responsibility for synagogue scholars to engage with New Testament scholarship and vice versa, in order to assure that we are presenting the past as accurately as we are able. It serves no purpose to dwell further on what is problematic, and in my opinion, it is better to be constructive at this juncture. Thus, the remainder of this essay will explore some of the opportunities for the historical and archaeological study of early synagogues to enrich and inform scholarship on the Gospels, and especially on the historical Jesus.

Historical Context Historical Jesus research stands to benefit greatly from the context offered by synagogue scholarship to the evangelists’ narratives of Jesus’ ministry in general and to individual episodes in the Gospels set in synagogues. Given the prominence of the synagogue in the Gospels and its role as the setting for Jesus’ teaching and proclamation, as well as for key episodes such as the Bread of Life discourse (John 6:25–59), the incident at Nazareth (Luke 4:16–30 and parallels), and the controversy about the Sabbath (Mark 3:1–6; Luke 13:14–17; cf. also Luke 6:9; Matt 12:10–12), it stands to reason that understanding the institutional setting of the synagogue may have some bearing on how to best interpret these passages from a historical perspective. In fact, as the summary statements of Jesus’ ministry in the Gospels (Mark 1:38; Matt 4:23; Luke 4:14–15, 43–44; John 18:20) suggest, and as I have endeavored to demonstrate elsewhere (in Ryan 2017a), the evangelists indicate that the synagogue was intrinsic rather than incidental to Jesus’ mission. If we hope to understand the mission, and thus the intentions of the historical Jesus (cf. Meyer 2002), we must consider the setting in which he attempted to carry it out. The distinction between public and association synagogues (cf. Runesson 2001) helps to clarify the functions of the synagogues with which Jesus interacted and the roles that they played in Jewish society.4 The synagogues in the Gospel narratives are apparently of the public variety (Runesson 2001: 355–357; Twelftree 2011: 3133; Ryan 2017a: 32–33). There is no clear indication that the synagogues that Jesus participates in belong to a particular group. They are depicted as gathering places belonging to the local populations and open for outsiders, such as Jesus, to enter, participate in services, and interact with the people gathered within.5 The evangelists present Jesus freely entering synagogues across  4 As argued by Runesson 2001. This distinction has been accepted and employed in various forms by, e.g., Twelftree 2011: 3109–3110; Bernier 2013: 57–60; Olsson 2003: 133; Binder 2003: 122–123; Stewart 2009: 139; Meyers and Chancey 2012: 203–204; Ascough 2015: 39–40; Korner 2015: 60–61; Krause 2017; and Ryan 2017a. Cf. also Hachlili 2013: 16; and Levine 2005: 44. 5 Mark 1:21, 23, 39; 3:1; 6:2; Matt 4:23; 9:35; 12:9; 13:54; Luke 4:15, 16, 31–33, 44; 6:6; 7:5; 13:10; John 6:59; 18:20.

Contributions of Early Synagogues to Historical Jesus Research

193

Galilee (Matt 4:23; 9:35; Mark 1:39; Luke 4:15, cf. John 18:20). This presumes that the synagogues that Jesus interacted with were public places, open to the general populace. In fact, Luke 7:5 describes the synagogue in Capernaum as though it were owned by the town (cf. m. Ned. 5:5) rather than by an association. Recognizing the institutional nature and function of the public synagogues in the Gospel narratives sheds a great deal of light on them. Public synagogues can be described as “town halls with Torah” (to borrow the phrase from Bernier 2013: 59), meaning that they were local-official institutions whose signature function was the reading, interpretation, and discussion of Torah. The communal nature of the public synagogue is clear (cf. Levine 2005). Evidence from the Land indicates that public synagogues functioned, for example, as legal-judicial courts,6 political institutions operating at the municipal level,7 and as the place where Jewish scripture was publicly read and interpreted.8 The public synagogue was thus a primary venue for legal and political deliberation and debate. As Carl Mosser (2013) has demonstrated, the main activity at synagogue gatherings following the reading of Torah (and possibly also the Nevi’im) was open discussion in which “anyone could offer insights or dispute the interpretive claims of others” (550).9 Recent scholarship indicates that by the Hellenistic period, the Torah was being treated similar to functional law in the Land (Watts 2013; Kazen 2013: 296– 298; Runesson 2016: 15–20; Lefebvre 2006: 182, 239–240, 258–260; cf. Alexander 1988). As James Watts (2013) has shown, we see the Torah being applied to the governance of such diverse aspects as marriage contracts (Tob 1:8; 7:12–13), battle plans (1 Macc 3:48), Sabbath observance (1 Macc 2:34–41), and criminal justice (OG Sus 62).10 Furthermore, when we look at the evidence stemming from the Roman period, it is clear that transgression of the statutes of the Torah could have very real consequences, including corporal and even capital punishment (Josephus, C. Ap. 2.176–177; A.J. 4.210; m. Mak. 3:1–15; John 10:31–33, cf. Lev 24:10–23). The Torah thus governed everyday life, and not just “religious” or

 6 Sir 23:24; OG Sus 28; Mark 13:9 [cf. Luke 21:12, Matt 10:17–18]; Matt 23:34; Luke 12:11–12; Acts 22:19; m. Mak. 3:12; cf. Levine 2005: 395–396. 7 E.g., Josephus, Vita 276–303; John 18:20; Judith 6:16; Sir 38:33; 41:18; 42:11; 44:15. Cf., e.g., Levine 2005; Runesson 2001; Binder 1999: 204–226; and Bernier 2013: 65–68. 8 Acts 15:21; Josephus, C. Ap. 2.175; A.J. 16.43; B.J. 2.292; CIJ II 1404; Philo, Prob. 80–83; Legat. 156. 9 E.g., Neh 8:1–8; Luke 4:22–30; Mark 6:2. Discussion of teaching based on passages from the Torah also occurs in John 6:25–59. Later tradition also mentions interpretation (targum) following a reading, as in t. Meg. 3:20. Philo mentions the interpretation of the reading in Hypoth. 7.13, but also states that the multitude remained silent. This curious statement is at odds with all of our other evidence. We must remember that Philo is writing in the Diaspora, for an audience that is unlikely to include Palestinian Jews, and that his statements reflect an association rather than a public setting. Active discussion certainly seems more pressing and likely in a public setting, where local-official practice is at stake. It is also possible that Philo here presents an unrealistic idealized notion of a synagogue assembly. 10 As argued by Watts 2013: 357.

194

Jordan J. Ryan

“cultic” life. It is in light of this that we should note that Josephus makes no distinction between “religious” law and secular (“political”) law in C. Ap. 2.170–171, placing the Torah on the same level as Roman law (C. Ap. 2.176–178). The synagogue was the place where the people of the Land most regularly encountered Torah, as it was the place where it was read, interpreted, and discussed. As Runesson observes, A certain village or town could thus be dominated by an influential group striving to control the local-official level ideologically. Such a struggle for domination is likely to have been the case in many of the villages and towns of Galilee (and elsewhere in Palestine) and it is in this context that we are to understand the mission of Jesus and other groups such as Judas the Galilean and his followers, the Herodians, the Sadducees, and the Pharisees. (Runesson 2001: 220–221)

It is in light of this institutional background that we should approach the passages involving synagogues in the Gospels. By participating and teaching in synagogue assemblies, Jesus is not engaging with a partisan or sectarian group, but rather, with mainstream, common Judaism. He is, in fact, participating in public Jewish life. Furthermore, the institutional nature of the public synagogue highlights the political dimension of Jesus’ synagogue ministry. By “political”, I do not mean “anti-Roman” or “nationalist”. Rather, Jesus’ activity in synagogues should be understood primarily on the local, or municipal level rather than on the national or imperial levels, as the decisions made within them had an impact on local practice and direction, but a decision made in the synagogue of Capernaum would typically be of little consequence to the Roman empire or to the Jerusalem elite. Commentators sometimes make much of the fact that the synoptic evangelists write that Jesus taught and proclaimed his message in “their synagogues” (εἰς τὰς συναγωγὰς αὐτῶν; Mark 1:39; cf. Luke 4:15; Matt 4:23; 9:35; 12:9; 13:54), especially in discussion of the Gospel of Matthew, where this phrase appears several times. The use of αὐτῶν (“their”) is often taken as a sort of outgroup designation, indicating distance between the synagogue and the evangelist, or even Jesus himself (see discussion in Davies and Allison 1988: 1:413–414). However, given the institutional nature of public synagogues, this is probably overstated, as by “their synagogues”, the evangelists probably mean to refer to the synagogues belonging to the Galilean towns and villages that Jesus visited. Why does Jesus enter into debate and discussion in synagogues in the passages mentioned above? If our only points of reference for this setting are modern synagogue or church services, in which such open discussion and controversy is not typical, this element of the synagogue narratives might strike us as odd. However, the matter comes into clearer focus when seen in light of the context provided by the institutional nature of the public synagogue. As local public assemblies in which the populace of the town actively participated, synagogues were a place where ideas were presented and discussed, resulting in acceptance or rejection.

Contributions of Early Synagogues to Historical Jesus Research

195

Hence, Ben Sira presents the local assembly as a place in which one can attain honor (e.g., Sir 38:33; 44:15) or be put to shame (41:18; 42:11).11 Being able to speak wisely in a synagogue setting was highly valued. Thus, according to Ben Sira, “The utterance of a sensible person is sought in the assembly, and they ponder his words in their minds” (21:17). Furthermore, if someone “holds to the law” (15:1), wisdom “will open his mouth in the midst of the assembly” (15:5). We should observe that in these passages, the sort of honor discussed (expressed by Ben Sira as wisdom) depends on public recognition. The public assembly is thus an arena in which the wisdom or folly of one’s speech would be judged by the opinion of the public, resulting in the accrual of honor or shame (cf. Crook 2009; Pitt-Rivers 1966). This understanding of the synagogue needs to be considered in order to contextualize and interpret the episodes in the Gospels involving synagogues discussed above. This is particularly relevant for the stories that depict Jesus teaching and discussing Jewish Scripture with interlocutors in synagogue settings (esp. Matt 12:9–14; Luke 4:16–30; 13:10–17; John 6:25–71). The portrait painted by modern scholarship of the public synagogue highlights what was at stake in the discussions of Jewish Scripture that took place in synagogues. In these instances, it is important to pay attention to the general public response to Jesus, which typically indicates either acceptance or rejection, not only of Jesus himself, but of his teachings and proclamations as well. All of this is illuminated by our previous discussion of the synagogue setting and what was at stake when entering the arena of synagogue discourse. It was, as Binder (1998: 403) writes, “not for the fainthearted”. Luke records mixed results, as Jesus is clearly met with rejection in the synagogue at Nazareth (4:28–30), but is apparently victorious in a dispute with an archisynagōgos taking place in an unidentified synagogue (Luke 13:17; see Ryan 2017b). In John’s telling of the Bread of Life discourse, which is set in the synagogue at Capernaum (John 6:59), there are some indications that the crowd is not taken with Jesus’ teaching (6:41–42, 52). In fact, in the epilogue of the discourse, we find that many of Jesus’ disciples found the teaching difficult (v. 60), and that they abandoned his cause as a result (v. 66). I would also suggest that the double-tradition saying pronouncing woes on Galilean towns (Luke 10:13–16; Matt 11:20–24) should be understood in light of synagogue discourse. Given that Jesus is said in the synoptic tradition to have been proclaiming his message of repentance in synagogues (Mark 1:15, 39), that he considered his message of repentance to have been corporately rejected by the three Galilean towns, and that the public synagogue represented the town, it is logical to conclude that the rejections mentioned in Luke 10:13–16 and Matt 11:20–24 likely refer to events that occurred in a public synagogue setting.  11

Ben Sira uses the terms synagōgē and ekklēsia interchangeably to refer to the public assembly, that is, the public synagogue. On this, see Ryan 2017a: 47–48; Ryan forthcoming b; cf. Korner 2015.

196

Jordan J. Ryan

Strangely, there is no description of these rejections in either Matthew or Luke. The mention of rejection by Capernaum is particularly odd, since Jesus is never met with rejection at Capernaum in Matthew or Luke. To the contrary, Luke 4:31–36 records an incident taking place in the synagogue at Capernaum that resulted in the assembly recognizing Jesus’ authority and power (v. 36). The only clear rejection situated in Capernaum is narrated in John 6:25–71. This may hold the solution. In my opinion, regardless of the literal “authenticity” or “inauthenticity” of the episode as a whole, there is some memory of Jesus’ teaching being rejected in the synagogue at Capernaum underlying the Bread of Life discourse. It is thus very likely that the woes pronounced on Capernaum in Luke 10:15 and Matt 11:23–24 remember the same incident. It is also essential to consider what the rhetorical effect of the healings and exorcisms that Jesus is reported to have performed would have had on his synagogue audiences, and how that might play into the honor and shame dynamic that we see present in the Ben Sira texts discussed above. For example, do Jesus’ miracles mentioned in Mark 1:21–28 and 6:1–6 (esp. v. 2) have any bearing on the recognition of Jesus’ authority and wisdom by the synagogue assemblies in those stories? It is also worth noting that, according to Mark 6:5–6, Jesus’ ability to perform is apparently tied to the belief of the assembly, as he is unable to perform deeds of power, only healing a few sick people. This is a striking example of the ancient attitude towards the tie between power and acceptance by the public synagogue assembly. The mentions in the Gospels of the presence of and interactions with Pharisees within synagogue settings (Mark 3:6; cf. Matt 12:14; Luke 6:7; John 12:42), as well as the sayings involving Pharisees and synagogues (Luke 12:43; cf. Matt 23:6; possibly Matt 23:1), are best understood in light of the distinction between association and public synagogues. Although some scholars have viewed synagogues as Pharisaic institutions (e.g., Kee 1990; Hengel and Deines 1995), current scholarship has clarified that, while Pharisees might have their own semi-public meeting places, which we might call “association synagogues”,12 local public synagogues did not belong to a particular subgroup, but to the community. The municipal ownership of public synagogues during the Roman period is presumed in Josephus’ narrative of the incident in the synagogue at Tiberias (Vita 276–303) and is explicit in m. Ned. 5:5 and m. Meg. 3:1–3.13 How should this shape our understanding of the interaction between Jesus and Pharisees involving synagogues? Incidents such as the one narrated in Mark 3:1–  12

A case has been made that the Matthean Jesus interacts with a Pharisaic association synagogue in Matt 12:9, since the antecedent of “their synagogue” (τὴν συναγωγὴν αὐτῶν) could be the Pharisees mentioned in 12:2. This has been argued by Runesson 2008. However, this feature is unique to Matthew’s version of the story, as Mark’s version (Mark 3:1 followed by Luke 6:6) lacks the third person plural possession pronoun (“their”; Gk. αὐτῶν). While it is thus possible to infer that the Matthean Jesus interacted with a Pharisaic association synagogue, it is difficult to make the same inference about the historical Jesus. 13 See discussion in Runesson, Binder, and Olsson 2008: nos. 82, 84.

Contributions of Early Synagogues to Historical Jesus Research

197

6 should be viewed in terms of competition between factions struggling to control the town ideologically on the local-official level (cf. Runesson 2001: 220) through successfully dominating the discourse in the synagogue in the eyes of the public. We should thus view the group composed of Jesus and his followers (Mark 1:16–20, 2:13–18, 23) as one such faction. The Pharisees should be viewed as another. It is in light of this background that we should probably also understand the dispute between Jesus and the archisynagōgos in Luke 13:10–17. That Jesus here opposes not just an individual but a group, or possibly a faction, is demonstrated by the fact that the incident is resolved in v. 17, stating that “all his opponents were put to shame”. Regardless of what we think about the literal historicity of the incident, I would also suggest that the narrative of John 6:25–59 should be viewed as describing a faction, led by Jesus, attempting to persuade the assembly and thus assert some control over the locale. However, as discussed above, the assembly is not persuaded, and Jesus ends up losing followers as a result. New light can also be shed on the aposynagōgos passage mentioning the Pharisees in John 12:42. Does this passage describe a fear of being “put out” of a Pharisaic association synagogue or of a public synagogue? Both possibilities have been proposed. Kloppenborg (2011) has argued that the passage is best understood within the context of disaffiliation in associations. However, Bernier (2013: 61– 74) has argued strongly against Kloppenborg’s stance, suggesting that all three aposynagōgos passages (John 9:22; 12:42; 16:2) refer to the public assembly of Jerusalem. Bernier’s hypothesis makes the most sense of the evidence. It is not necessary to reproduce the entirety of his argument here. However, it is important to note that, on both the narrative and historical levels, it is difficult to comprehend why non-Pharisees would be afraid of being “put out” of a Pharisaic association, whether they are the blind man’s parents (9:22), members of the Jerusalem ruling class (12:42), or members of Jesus’ own faction (16:2). Furthermore, John 16:2 associates the mechanism of being “put out” of the synagogue with violence, even lethal violence, which falls within the domain of public assemblies (cf., e.g., Sus 28, 41–43, 60–62;14 see Levine 2005: 143), but not semiprivate associations. How, then, should the influence of the Pharisees mentioned in John 12:42 be interpreted within the setting of a public synagogue? The synagogue need not belong to the Pharisees for them to wield influence within it, or even to dominate or exert control within it. The best way to understand John 12:42 in light of synagogue studies is to see it as a reference to the public assembly of Jerusalem, and  14 Note that the Old Greek (OG) version of this narrative is not set in the Diaspora. The Theodotion recension of Susanna sets the tale in Babylon (v. 1). However, the OG recension probably began at v. 5b. Thus, the setting of the OG recension is unclear. A setting in the Land for the Old Greek version cannot be excluded. Whatever the case may be, the text’s provenance is likely Palestinian (Moore 1977: 91–92). Thus, regardless of the fictional setting of the narrative, OG Susanna reflects the historical realia of Second-Temple period Palestine (Engel 1985: 12–15; Knibb 2001: 26–27).

198

Jordan J. Ryan

to imagine that the Pharisees as a faction have managed to situate themselves within it in such a way that they wield enough influence to convince the assembly to exclude certain people by means of violence. This coheres broadly with Josephus’ presentation of the Pharisees as commanding popular support who are “able greatly to persuade the body of the people” (A.J. 18.15). The aposynagōgos passages have had a notable impact on the history of scholarship on Jesus and the Gospels. According to J. Louis Martyn (2003 [1968]) and those who have followed him, the aposynagōgos passages reflect the history of the late first-century Johannine community, whose members have been excluded from participation in synagogues due to the institution of the birkat ha-minim (“Benediction against Heretics”). This hypothesis complicates the use of John as a historical source for the events of the life of Jesus, since it makes it very difficult to determine what elements of the narrative can or should be attributed to the time of Jesus, since the experiences of the community events have shaped the narrative of the Fourth Gospel. However, the Martynian hypothesis was developed prior to the significant advances made in synagogue scholarship in the years since the turn of the millennium. Indeed, Bernier’s study shows that the aposynagōgos passages fit most plausibly into the time and place of the life of Jesus. By examining the evidence in light of current synagogue research, Bernier has provided a much-needed evaluation of this longstanding hypothesis, and has advanced a more plausible hypothesis in its place. As we have seen thus far, engagement with synagogue studies offers much to the study of Jesus and the Gospels in terms of context, which opens up new and better avenues for interpretation and understanding. We have yet to discuss how archaeology might contribute further to this, and so, it is to that topic that we now turn.

Archaeology Archaeology has greatly increased our understanding of early synagogues, especially since the turn of the millennium. At least eight examples of early synagogue buildings dating either to the late Hellenistic or early Roman periods have been identified within the Land of Israel. These eight buildings are located at Capernaum (Corbo 1982; Loffreda 1993: 32–49; Runesson, Binder, and Olsson 2008: no. 8), Gamla (Syon 2010: 41–61; Runesson, Binder, and Olsson 2008: 10), Herodium (Runesson, Binder, and Olsson 2008: no. 11; on the date, see Foerster 1982), Jericho (Netzer 1999; Runesson, Binder, and Olsson 2008: no. 15), Magdala (Avshalom-Gorni and Najar 2013), Masada (Yadin 1982; Runesson, Binder, and Olsson 2008: no. 28), Modi‘in (Khirbet Umm el-Umdan) (Onn, Wexler-Bdolah, Rapuano, and Kanias 2002; Runesson, Binder, and Olsson 2008: no. 29), and Qiryat Sefer (Magen, Tzionit, and Sirkis 2004; and Runesson, Binder, and Olsson

Contributions of Early Synagogues to Historical Jesus Research

199

2008: no. 35). Of these eight, only the identifications of the structures at Capernaum and Jericho as synagogues are heavily debated. However, there is nevertheless good reason to think that both could be identified as synagogues.15 Beside these eight “canonical” synagogue buildings, there are several more discoveries that are either very new and awaiting publication or less certain, but that are fairly likely to be synagogues (cf. Ryan 2017a: 65–67). Two buildings that are very good candidates for identification as Roman-period synagogue buildings have been found, one at Tel Rekhesh in Galilee (Aviam et al. 2019) and the other at Majduliyya in the Golan. However, the synagogue at Majduliyya seems to have been used also beyond the time of the Gospels.16 Architectural fragments that likely belonged to an early synagogue have been found at Kefar Shikhin (Strange 2015: 103–105). Two other buildings recently identified as synagogues have also been uncovered at Khirbet e-uwani (Har-Even 2012) and Khirbet Diab (HarEven 2016), though the columns usually associated with early synagogue architecture have not been found at either site. A stone slab featuring a Greek inscription that belonged to a first century CE synagogue was also uncovered in the Weill excavations in the Ophel region of Jerusalem (Weill 1920: 98–100; Kloppenborg 2000; CIJ II 1404 = CIIP I 9; Runesson, Binder, and Olsson 2008: no. 26). Although the building itself is not extant, the architectural fragment of the inscription, which mentions that the synagogue had been “built”, is clear evidence of the existence of a synagogue building. A synagogue dated to the late first or early second century CE, and thus to the first decades of the middle Roman period, has been discovered at Khirbet Qana (McCollough 2015). Thus, while the Qana synagogue would not have existed at the time of Jesus, nor during the time of the Second Temple, its genesis is roughly contemporary with or slightly later than the writing of the Gospels. In total, we may have archaeological attestation for up to fifteen synagogue buildings that existed prior to the close of the first century CE. While this number is more than significant enough to draw conclusions about the existence, nature, and function of synagogue buildings, it is not an overwhelming number. It is not yet clear just how widespread the use of a building to house synagogue gatherings was in the first century CE. That having been said, it is worth noting that Qiryat  15 All that remains of the early Roman period synagogue at Capernaum is a black basalt pavement located beneath a monumental limestone synagogue building that dates to the Byzantine period. As such, its identification as a synagogue has been challenged, particularly by Magness 2001. However, as Runesson 2007 has argued, the excavators’ identification of the building as a synagogue is probably correct. I also argue the identification of the building as a synagogue in a forthcoming publication (Ryan forthcoming a). The identification of the synagogue at Jericho has also been challenged (Ma‘oz 1999; Stacey 2004). The challenge is based on the location of the synagogue in a semi-private location, the Hasmonean palace, apart from the main town. As Runesson 2004 has observed, this challenge only stands if we hold a definition of “synagogue” as solely a public institution, excluding the existence of semi-public association synagogues. 16 For Tel Rekhesh see now Aviam et al. 2019; and cf. the contribution by Zeev Weiss in this volume. For Majduliyya see the chapter by Mechael Osband and Benjamin Arubas in this volume.

200

Jordan J. Ryan

Sefer was a small, rural village, while Magdala was a very large Jewish town (we might even call it a “city”), showing that synagogue buildings could be found in the smallest of rural locales as well as in larger urban settings. A town did not necessarily need one of these benched buildings in order to have a “synagogue”. Public synagogue gatherings likely originated in open air public assemblies held at city gate complexes (cf., e.g., Levine 2005; Binder 1999; Runesson 2001; Bernier 2013; Krause 2017; Ryan 2017a), and appear to have begun to move into buildings in the Land during the late Hellenistic period, when the first synagogue buildings in Israel-Palestine appear in the archaeological record. However, it is reasonable to assume that the earlier practice of meetings in public space in the open air continued in the late Hellenistic and early Roman periods in some locales. This makes the most sense of the evidence, as we see numerous Jewish settlements in Galilee in the first century CE, but only a total of five buildings (at Capernaum, Magdala, Kefar Shikhin, Tel Rekhesh, and Khirbet Qana) identified as synagogues dated to this time, one of which (Tel Rekhesh) is likely to have been a semi-public synagogue, and another of which (Khirbet Qana) was not built until after the destruction of the Jerusalem Temple. The architecture of synagogue buildings reflects a common design, especially in their main rooms, which were clearly designed for public assembly. Early synagogue assembly rooms featured benches, usually stepped, located along three or four of the walls in a rectangular arrangement. The quadrilateral plan of the benches creates an open area in the center, which did not contain permanent (stone) benches. Columns are usually located in this central area. The practical function of the columns was likely to support a clerestory ceiling (Strange 2003), thus allowing light into the assembly room. These buildings were designed to accommodate larger public gatherings in a way that domestic structures were not. The particular quadrilateral arrangement of benches places the focus in the center of the room, which made it easier to engage in discussion with people seated along the other walls, especially those across the way. It is worth observing that this basic architectural plan, with benches and columns, continued to be used in late antiquity. Well-known examples include, for example, the late antique synagogues at Chorazin and Capernaum. The architectural plan shared by synagogue buildings helps to drive the historical imagination when reading and interpreting the Gospel passages that envision a synagogue setting. Although synagogue gatherings did not necessarily need to involve buildings, it is likely that some, if not many, of the towns visited by Jesus housed their public assemblies in buildings. This is the case with Capernaum, whose synagogue is mentioned several times in the Gospel narratives (independent references include Mark 1:21; 3:1; Luke 7:5; John 6:59; perhaps also Luke 10:15; cf. Matt 11:23). The architecture of synagogue buildings reflects their communal functions, and moreover, sheds some light on how discussion would have taken place. The layout of synagogue buildings is, as mentioned above,

Contributions of Early Synagogues to Historical Jesus Research

201

designed to facilitate public discussion and engagement with the people seated across the room, as opposed to those who might be seated nearby. The architectural configuration of the seating, with its focus placed in the center of the assembly room, makes it such that discussions held between people on opposite sides of the room would carry by necessity and be easily heard by all present. Thus, so far as the Gospel narratives of synagogue disputes are concerned, we might imagine Jesus and his followers seated on the benches along one side of the synagogue, and his interlocutors seated across the way. It is important for scholarship on Jesus and the Gospels to take this archaeological data seriously when attempting to situate Jesus within his Jewish environment. In particular, we should note how much new information about synagogues has become available through archaeological discoveries since the turn of the millennium. Because of the rapid pace at which the archaeological study of synagogues has advanced in recent years, it is essential for students of the New Testament to keep themselves up to date rather than relying solely on old reports. While some scholarship of the 1990s spoke of the lack of synagogue buildings in Galilee in the first century CE (e.g., Kee 1990; Horsley 1995; cf. Horsley 1996), this hypothesis must now be regarded as problematic, and should be abandoned in light of the available evidence. Since Magdala is located on the eastern shores of the Sea of Galilee, it was located within the region in which Jesus was active, making its synagogue particularly relevant for historical Jesus research. The identification of the public building at Magdala that was excavated in 2009 (not to be confused with the “springhouse” excavated in the 1970s) as a synagogue building is not presently controversial amongst archaeologists, nor should it be. However, given the history of questioning the identification of synagogue buildings in New Testament scholarship, even in cases that were not controversial amongst archaeologists (such as with the synagogue at Gamla), it is perhaps prudent to present the rationale for the identification of the building as a synagogue here.17 The architecture of the Magdala synagogue features stepped benches, and is clearly designed for public assembly. It is not typical domestic architecture, and in fact, differs substantially from the layout of houses of this period. The assembly room cannot be a courtyard, as it was roofed (cf. Avshalom-Gorni and Najar 2013; De Luca and Lena 2015: 313). It is located in a Jewish locale (Zapata-Meza, Garza Diaz Barriga, and Sanz-Rincón 2018). The Jewish character of the area is indicated by, for example the prominence of Jewish coins and pottery on site for the early Roman period, the existence of four miqwa’ot nearby, and by the presence of limestone vessels.  17 This argument is abbreviated from a forthcoming piece contributed to a forthcoming volume on the excavations at Magdala. Interested readers are encouraged to refer to that piece once it is available. Similar arguments have also been given by Donald Binder (2014).

202

Jordan J. Ryan

Perhaps the most publicized aspect of the discovery of the synagogue building at Magdala is a carved limestone block, popularly termed the “Magdala Stone”. This stone was found in the central area of the synagogue, and was likely part of the building’s furniture. The function of the Magdala Stone is unclear.18 Most scholars who have published on the Magdala Stone, including the excavators (Avshalom-Gorni and Najar 2013), consider the stone’s artwork to represent Jerusalem Temple imagery, though this has been disputed (Fine 2017; Hachlili 2017). Whether or not the Magdala Stone represents Temple imagery, it prominently features a menorah, which was certainly an important Jewish symbol in antiquity. Thus, at Magdala we have i) a public building designed for gatherings; ii) distinct from domestic architecture; iii) in a Jewish locale; iv) featuring an example of a Jewish symbol; v) architecturally identical in plan to a set of other public buildings from the early Roman period that are identified as synagogues; vi) and identical in architectural plan to late antique buildings that have been confidently identified as synagogues on the basis of their inscriptions and artwork. It is thus most reasonable to conclude that the building under discussion should be identified as a synagogue. The very existence of synagogue buildings calls us to question who would have been present at synagogue gatherings such as the ones described in the Gospels. As indicated by the literary evidence, some of which has been discussed above, the presence of a public element made up of the townspeople was a given at synagogue gatherings. This does not mean that everyone in the town would have had to have been present at every synagogue gathering, not even on the Sabbath. Chad S. Spigel’s recent study (2012) of synagogue seating capacities has shown that, except in cases of very small villages, it would not have been possible for synagogue buildings to accommodate the whole population of a town at once. This does not make synagogues any less of a public institution. Modern town hall meetings and the decisions made in them are understood to represent the public despite not literally being an assembly of every resident of the town. We should not expect the residents of first century CE Jewish towns to have attended every public assembly any more than we can expect every resident of modern towns to attend every town hall meeting.

Plausibility Historical Jesus research has begun to move away from traditional conceptions of “authenticity” and “inauthenticity” in recent years. Current research has  18 A number of studies have already been published discussing the stone, its artwork, and purpose (Aviam 2013a; Binder 2014; Bauckham 2015; Fine 2017; and Hachlili 2017). See in this volume the chapters by Lutz Doering and Judith Newman.

Contributions of Early Synagogues to Historical Jesus Research

203

instead tended to emphasize the place of plausibility in historical investigation and reconstruction (e.g., Theissen and Winter 1997; Winter 2012; Pitre 2015: 34– 52; Ryan 2017a). Although a scenario that is plausible is not necessarily historical in the sense of being “real”, a scenario that is implausible is unlikely to be historical. Evidence tempers the historian’s ability to determine whether a given hypothesis or reconstruction is plausible. It also helps to determine whether or not a statement provided by a source is plausible. Synagogue evidence can function in this way for historical Jesus research. It serves a dual purpose. The evidence provided by current synagogue scholarship and archaeology can help to determine whether the narratives involving synagogues in the Gospels are plausible.19 It can also help to determine whether a reconstruction or hypothesis pertaining to the historical Jesus is plausible or implausible insofar as synagogues are involved. We may demonstrate the importance of synagogue studies and evidence for determining plausibility with a case study here. Is the evangelists’ depiction of Jesus travelling from town to town and teaching in their public synagogues plausible? Would Jesus have been able to participate in the public synagogue assemblies of towns and villages of which he was not a resident? This question is sometimes raised in formal academic discussion in conference and presentation settings, and it is raised often enough that it is fitting to take the opportunity to address this issue here. Even if that were not the case, it is nevertheless worth exploring, since it helps us to better control and understand an important datum.20 There is no evidence contemporary with the Gospels that exactly corroborates the depiction of a non-resident participating in and even speaking aloud in a public synagogue. On the other hand, it must be also observed that there is no evidence that directly contradicts the Gospels in their depiction, showing nonresident Jews being barred from a public assembly simply for being non-residents. At most, the lack of corroborating data can only be used to formulate an argument from silence. However, that argument from silence only holds if we eliminate the Gospels as evidence that can potentially speak to the realia of early synagogues from the outset. In other words, the sources are only silent if we eliminate the most relevant sources, that is, the only sources that directly speak to the issue. Historical evidence for the first century CE synagogue is notoriously slim, and of that, only a fraction pertains to Sabbath synagogue gatherings. The relevant  19

Elsewhere, I have argued in favor of the general plausibility of the Gospel narratives involving synagogues (Ryan 2017a), at least so far as their relation to synagogues is concerned. This does not mean that all of these episodes are necessarily literally historical, that they are “authentic,” or that they “happened”. It simply means that they fit with what we know of ancient synagogues of the first century CE based on current synagogue studies. 20 Indeed, Ben F. Meyer (2002: 81–87) has highlighted the importance of controlling the data in historical Jesus research.

204

Jordan J. Ryan

pool of evidence becomes even slimmer once we eliminate anything coming from the New Testament. To expect confirmatory commentary on this particular practice is extremely problematic. In fact, there are no other narrative depictions dating to the Second Temple period of a specific Sabbath synagogue assembly in the Land apart from those in the New Testament. Typically, the other descriptions of Sabbath synagogue gatherings speak generally, and thus they describe what was typical, not occasions when an itinerant or guest teacher might be present. For whatever it is worth, the Gospels corroborate one another, insofar as they agree in their depictions of Jesus teaching in multiple synagogues, not only the synagogue of the town in which he is a resident (Mark 1:39; Matt 4:23; 9:35; Luke 4:15, 44; 13:10–17; John 18:20). It is worth observing that the only named locations of synagogues in the Gospels are Capernaum and Nazareth (Mark 1:21; 6:1–2; Luke 7:5; John 6:59). Based on the evidence provided by the Gospels, Jesus could be considered a resident of both of these locales. Nazareth is where Jesus was brought up (Luke 4:16), his “hometown” (πατρίς; Mark 6:1; cf. Luke 4:16). Moreover, Capernaum is where, according to Matt 4:13, Jesus “made his home”.21 Thus, Jesus’ presence in these particular assemblies is not at all curious, at least based on the evidence provided by the evangelists about Jesus’ places of residence. It is mildly curious that no other synagogues in named locations are found in the Gospel narratives.22 We should note, however, that the episode of the healing of the woman with the bent back and the subsequent dispute between Jesus and an archisynagōgos in Luke 13:10–17 cannot be set in Nazareth nor Capernaum, as it takes place during the travel narrative, and is apparently set somewhere in Judea. There is some important evidence that speaks directly to the existence of visiting teachers in synagogue settings on the Sabbath in the earliest Rabbinic literature. The evidence that pertains the most directly to the matter at hand is provided by m. ‘Erub. 3:5, which discusses rulings for where ‘eruvim can be drawn when a visiting sage has come to town on the Sabbath. Whether or not these particular rulings were ever followed is beside the point, as is the date of the compilation of the Mishnah at the beginning of the third century CE. What is important is that the fact that the ruling exists in the Mishnah at all is evidence that the practice of itinerant teachers visiting Jewish Sabbath assemblies was widespread enough by the late second century at the latest for the matter to have been discussed in the Mishnah at all. The presumption here is that the practice was common enough to require a ruling that is incidentally related to it. This tradition  Cf. Mark 2:1. However, we must note that the meaning of the Greek phrase “ἐν οἴκῳ ἐστίν,” which is a way of saying that someone is at home, is ambiguous, as it can also mean that Jesus was “in a house”. See the discussion by Boring 2006: 74. 22 However, this excludes the Jerusalem assembly, which appears to have met within the temple precincts (Neh 8:1–12; cf. 1 Esdr 9:38; Josephus, B.J. 1.122; 2.1–5; 2.294–295; 2.320–324; 4.336; A.J. 11.154– 158; 17.200–201; m. Yoma 7:1; m. Soah 7:7–8; m. Sanh. 11:2; see Ritmeyer 2014; Binder 1999: 219–220; Bernier 2013: 64–68; and Ryan 2017a: 33–34). 21

Contributions of Early Synagogues to Historical Jesus Research

205

had to begin at some point. The practice seems established by the time of the tradition preserved here in m. ‘Erub. 3:5. Thus, we have a practice of a visitor teaching in synagogues attested in first century texts, and a similar practice attested in a text compiled at the beginning of the third century, which probably reflects practices common by the late second century at the latest. Rather than placing the evidence from the Gospels and the Mishnah in separate silos, responsible historiography entails considering that the two stand in diachronic relation to one another. It is likely that the later text reflects the continuity of practices attested in the earlier ones. As I have argued elsewhere (Ryan 2017a: 89–93), the differences between first century and second century CE public synagogues are minimal, despite being sometimes overstated in scholarship. As Catherine Hezser (2011) has effectively demonstrated, there was a culture of mobility in early Judaism, as the Mishnah and Tosefta depict the sages as mobile teachers. In later literature, we even have evidence of sages travelling specifically to synagogues, as with R. Berekhiah’s visit to the synagogue at Beth Shean (y. Meg. 3:3–4, 74a). In early Roman Galilee, we see a culture of travel and mobility within Jewish territories reflected in the archaeological record, as well as a sense of interconnectivity and shared identity between Jewish locales. It is well known that a strong preference for wares made by Jews appears to be attested at Galilean sites during the late Hellenistic and early Roman periods (e.g., Berlin 2005: 445–448; Magness 2011: 60–61; Aviam 2013b). Those wares are typically made at a few specific workshops, such as Kefar ananya,23 and were then brought and sold in other places. Magdala makes for a good exemplar because it is typical of Jewish towns in this period. Petrographic analysis of the pottery at Magdala has shown that it was all produced at Kefar ananya and Shikhin, both Jewish towns in Galilee, save for just nine fragments of imported vessels (Zapata-Meza, Garza Diaz Barriga, and Sanz-Rincón 2018: 109). As for the numismatic evidence, the vast majority of the cleaned and identified coins from Magdala are Jewish coins minted in Jerusalem (Zapata-Meza, Garza Diaz Barriga, and Sanz-Rincón 2018: 117–119), located well over 150 kilometers south, despite the mint at Tiberias being about 5 kilometers away. If Jewish coins and pottery moved from Jewish town to Jewish town, so did Jewish people.24 Moreover, the shared culture witnessed by the archaeology of daily life is significant. The situation that emerges from the data is this: in the Roman period, we see relatively high levels of mobility between Jewish locales, and a sense of shared identity. Both pottery and coins travelled and circulated  23

Other examples of workshops existed at Shikhin and Yodefat. James F. Strange (2014) has recently produced a short study of the village pathways and travel between Galilean villages in the early Roman period. 24

206

Jordan J. Ryan

from place to place amongst Jews in Galilee. Because of this, it seems quite plausible that a Jew like Jesus from somewhere like Capernaum or Nazareth could travel to another Jewish town nearby, such as Chorazin, or Magdala, and find himself welcome based on his shared identity and the relative commonality of visitors from other Galilean Jewish towns. It is important to remember the significance of local Jewish networks in light of shared Jewish identity, especially in a region as small as Galilee. Synagogues were a part of this network (cf. Aviam 2013b). Their shared general architectural plan is indicative of something held in common, in this case, common purpose and function. As has been showed by viewshed analysis, the columns actually obstruct vision, leading James F. Strange (2003) to the striking conclusion that synagogue buildings were designed with hearing rather than seeing in mind. This makes a good deal of sense when we recall that the public reading of Torah was the synagogue’s signature function. As Runesson (2001: 261– 283) has shown, widespread Torah reading had the effect of forging and reinforcing a shared national identity, which he ties to the very origins of the synagogue. The public synagogue is itself an expression of that shared national identity through public Torah reading. The synagogue, we must remember, is not just a town hall, but a town hall with Torah, and Torah was central and foundational to shared Jewish identity in antiquity. That shared Jewish identity founded on Torah, especially within the Land, was the basis for public synagogue membership, not local residency. This, then, is a significant difference between the Jewish institution of the synagogue and local assemblies in other cultures in antiquity, such as the Greek dēmos. The idea that Jesus the Jew would be able to participate in a synagogue in his place of residence, but then be barred from a synagogue in a town just five or ten kilometers away stretches the imagination, and moreover, misses the very essence of the role and concept of the public synagogue in early Judaism. Closely related to the concept of shared national identity expressed by the public synagogue is the relationship between the synagogue and the Jerusalem Temple. While the synagogue was the local-official assembly, the Jerusalem Temple was the national assembly, especially during festivals. Some scholars (Binder 1999; Strange 2003) have argued that the architecture of synagogue buildings, especially the columns, evokes that of the Temple and its colonnades. Donald Binder (1999) has produced a forceful argument that the synagogue was considered to be an extension of the Jerusalem Temple, a hypothesis that has since been significantly bolstered by the discovery of the Magdala stone (Binder 2014), which may depict Temple imagery in a synagogue setting. Binder (1999: 219– 220) rightly observes that, with the exception of the sacrificial cult, which was not performed in synagogues, synagogue functions parallel Temple functions, including the public reading of Torah (cf. 1 Esdr 9:38; Josephus, A.J. 11.154–158), which took place in the Jerusalem Temple in the first century CE. If Binder is

Contributions of Early Synagogues to Historical Jesus Research

207

correct, then it is reasonable to think that local synagogue assemblies would naturally have been open to all Jews, much like the Temple courts. That synagogues could be entered freely by passers-by is demonstrated by t. Meg. 2:18, which states that one should not (emphasis my own) enter synagogues, among other things, to escape the heat, cold, or rain, nor should one stroll or relax in them. As Levine (2005) writes, 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 effecting 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

Thus, it would seem as though synagogue buildings were open to be entered. This speaks to their fully public nature during the Roman period. Some indirect evidence dating from the first century C.E. may help to further clarify the matter. Contrary to outsiders being barred from synagogues, there is evidence indicating that synagogues were places of hospitality for visitors (Levine 2005: 144). Most notably, the Theodotus inscription (CIJ II 1404 = CIIP I 9) states that the synagogue to which it was at one time attached functioned as a hostel (τὸν ξενῶνα). It is difficult to imagine synagogues offering hospitality to visiting Jews, but not allowing them to attend gatherings at which Scripture would be read. There is also at least one instance of an outsider speaking at a public synagogue gathering outside the Gospels, as Ionathes speaks in the Tiberian synagogue in Josephus’ account of the incident in Tiberias (Vita 276–303). Of course, this is in the context of public deliberation rather than teaching per se. Our examination of the evidence shows that the basic scenario presented in the Gospels of Jesus travelling to various public synagogues, especially in Galilee, is plausible. As we have seen, this determination comes out of engagement with current studies and understandings of synagogues and their context within Jewish Galilee in the Roman period. Of course, the determination of plausibility is a starting point rather than an end game. While it does not “authenticate” the Gospel synagogue narratives in itself, our discussion here helps us to control the data and understand how best to apply it within historical reconstruction.

Conclusion The study of the historical Jesus stands to benefit greatly from engaging with the historical and archaeological study of ancient synagogues. Synagogues were a key dimension of early Jewish society in Galilee during the early Roman period. This is reflected in their prominence within the Gospel narratives. Any historical study that aims to situate Jesus within his early Jewish context should take current

208

Jordan J. Ryan

understandings of ancient synagogues into consideration. As we have seen, synagogue data, both historical and archaeological, has the potential to shed important light on key Gospel evidence used in the historical investigation into the life of Jesus. Scholarship on the ancient synagogue has recently helped to evaluate hypotheses involving the synagogue that have enjoyed prominence in New Testament scholarship. In particular, the premises underlying the hypothesis of the nonexistence of synagogue buildings, as well those underlying the community approach to the aposynagōgos passages in John, have been called into question in light of present advances in our understanding of the synagogue, both from archaeological and historical quarters. This highlights the importance of dialogue between New Testament scholars and research on synagogues. There is a pressing need for scholars of Jesus and the Gospels to ensure that synagogues are accurately and appropriately presented in accordance with contemporary synagogue scholarship in their work. So doing will help us to recover a Jesus who is fully and truly situated within his Jewish context.

Bibliography Allison, D. C. 2011. “How to Marginalize the Traditional Criteria of Authenticity”. In: Holmén, T. and Porter, E. (eds.), Handbook for the Study of the Historical Jesus. Leiden and Boston: Brill, 3–30. Aviam, M. 2013a. “The Decorated Stone at Migdal: A Holistic Interpretation and a Glimpse into the Life of Galilean Jews at the Time of Jesus”. NovT 55: 205–220. Aviam, M. 2013b. “People, Land, Economy, and Belief in First-Century Galilee and Its Origins: A Comprehensive Archaeological Synthesis”. In: Fiensy, D. and Hawkins, R. (eds.), The Galilean Economy in the Time of Jesus. ECL 11. Atlanta: Society of Biblical Literature, 5–49. Aviam, M. et al. 2019. “A 1st–2nd Century CE Assembly Room (Synagogue?) in a Jewish Estate at Tel Rekhesh, Lower Galilee”. Tel Aviv 46: 128–142. Avshalom-Gorni, D. and Najar, A. 2013. Migdal: Preliminary Report. Hadashot. [Online] 125. Available from: http://www.hadashot-esi.org.il/report_detail_eng.aspx?id= 2304&mag_id=120 (accessed 27 February 2020). Bauckham, R. 2015. “Further Thoughts on the Migdal Synagogue Stone”. NovT 57: 113– 135. Berlin, A. 2005. “Jewish Life Before the Revolt: The Archaeological Evidence”. JSJ 36: 417– 470. Berlin, A. 2002. “Romanization and Anti-Romanization in Pre-Revolt Galilee.” In: Berlin, A. M. and Overman, J. A. (eds.), The First Jewish Revolt: Archaeology, History, and Ideology. London: Routledge, 57–73. Bernier, J. 2013. Aposynagōgos and the Historical Jesus in John: Rethinking the Historicity of the Johannine Expulsion Passages. BibInt 122. Leiden: Brill. Binder, D. D. 1999. Into the Temple Courts: The Place of the Synagogue in the Second Temple Period. SBLDS 169. Atlanta: Society of Biblical Literature.

Contributions of Early Synagogues to Historical Jesus Research

209

Binder, D. D. 2003. “The Origins of the Synagogue: An Evaluation”. In: Olsson, B. and Zetterholm, M. (eds.), The Ancient Synagogue from Its Origins until 200 C.E.: Papers Presented at an International Conference at Lund University. ConBNT 39. Stockholm: Almqvist & Wiksell, 118–131. Binder, D. D. 2014. “Mystery of the Magdala Stone”. In: Warner, D. A. and Binder, D. D. (eds.), A City Set on a Hill: Essays in Honor of James F. Strange. Mountain Home, AR: BorderStonePress, 17–48. Bond, H. 2012. The Historical Jesus: A Guide for the Perplexed. London: T&T Clark. Boring, E. 2006. Mark: A Commentary. NTL. Louisville: Westminster John Knox. Brooten, B. J. 2020. Women Leaders in the Ancient Synagogue: Inscriptional Evidence and Background Issues. BJS 36. 2nd ed. Atlanta, GA: Scholars Press. Available online from JSTOR: www.jstor.org/stable/j.ctvzpv5mr.7 (accessed 2 September 2020). Catto, S. 2007. Reconstructing the First-Century Synagogue: A Critical Analysis of Current Research. LNTS 363. London and New York: T&T Clark. Collingwood, R. 1993. The Idea of History. Oxford: Clarendon Press. Corbo, V. C. 1972. Cafarnao. Jerusalem: Franciscan Printing Press. Corbo, V. C. 1975. Cafarnao: Gli edifici della città. Jerusalem: Franciscan Printing Press. Corbo, V. C. 1982. “Resti della sinagoga del primo secolo a Cafarnao”. Studia Hierosolymitana 3: 314–57. Crook, Z. 2009. “Honor, Shame, and Social Status Revisited”. JBL 128.3: 591–611. Edwards, J. 2015. The Gospel According to Luke. PNTC. Grand Rapids, MI: Eerdmans. Engel, H. 1985. Die Susanna-Erzählung. OBO 61. Göttingen: Vandenhoeck & Ruprecht. Eve, E. 2002. The Jewish Context of Jesus’ Miracles. JSNTS: 231. Sheffield: Sheffield Academic. Fine, S. 2017. “From Synagogue Furnishing to Media Event: The Magdala Ashlar”. Ars Judaica 13: 27–38. Gideon, F. 1982. “The Synagogues at Masada and Herodium”. In: Levine, L. (ed.), Ancient Synagogues Revealed. Jerusalem, Israel Exploration Society, 24–29. Goodacre, M. 2012. “Criticizing the Criterion of Multiple Attestation: The Historical Jesus and the Question of Source”. In: Keith, C. and Le Donne A. (eds.), Jesus, Criteria, and the Demine of Authenticity. London and New York: T&T Clark, 152–169. Guijarro, S. 1999. “The Politics of Exorcism: Jesus’ Reaction to Negative Labels in the Beelzebul Controversy”. Biblical Theology Bulletin 29.3: 118–129. Hachlili, R. 2013. Ancient Synagogues: Archaeology and Art: New Discoveries and Current Research. HdO I 105. Leiden and Boston: Brill. Hachlili, R. 2017. “The Migdal Stone and Its Ornamentation”. RB 124.2: 245–272. Har-Even, B. 2012. “ ‫ביזנטית והתקופה‬-‫ התקופה הרומית‬,‫ יישוב מימי הבית השני‬:‫חורבת א־טוואני‬ ‫”המוסלמית הקדומה‬. In: ‫ דברי הכנס השביעי‬:‫הְספר והמדבר בארץ ישראל‬, 15–29. Available from: http://www.atarsusya.co.il/pages/articles.asp (accessed 27 February 2020). Har-Even, B. 2016. “A Second Temple Synagogue at Horvat Diab in Western Benjamin”. Qadmoniot 151: 49–53. Hengel, M. and Deines, R. 1995. “E. P. Sanders’ ‘Common Judaism”. JTS 46.1: 1–70. Hooker, M. 1972. “On Using the Wrong Tool”. Theology 75: 570–581. Horsley, R. 1995. Galilee: History, Politics, People. Valley Forge: Trinity Press International. Horsley, R. 1996. Archaeology, History, and Society in Galilee: The Social Context of Jesus and the Rabbis. Valley Forge: Trinity Press International. Loffreda, S. 1993. Recovering Capernaum. Jerusalem: Franciscan Printing Press. Kazen, T. 2013. Scripture, Interpretation, or Authority? Motives and Arguments in Jesus’ Halakic Conflicts. WUNT 320. Tübingen: Mohr Siebeck.

210

Jordan J. Ryan

Kee, H. 1990. “The Transformation of the Synagogue After 70 C.E.: Its Import for Early Christianity”. NTS 36: 1–24. Keith, C. and Le Donne, A. (eds.) 2012. Jesus, Criteria, and the Demise of Authenticity. London and New York: T&T Clark International. Kloppenborg, J. 2000. “Dating Theodotos” (CIJ II 1404). JJS 51.2: 243–280. Kloppenborg, J. 2011. “Disaffiliation in Associations and the ἀποσυναγωγός of John”. HTS Teologiese Studies 61: n.p. Knibb, M. 2001. “The Book of Daniel in its Context”. In: Collins, J. and Flint, P. (eds.), The Book of Daniel: Composition and Reception. 2 vols. Leiden and Boston, 1:16–35. Korner, R. 2015. “Ekklēsia as a Jewish Synagogue Term: Some Implications for Paul’s SocioReligious Location”. JJMJS 2: 53–78. Krause, A. R. 2017. Synagogues in the Works of Flavius Josephus: Rhetoric, Spatiality, and First-Century Jewish Institutions. Leiden: Brill. LeFebvre, M. 2006. Collections, Codes, and Torah: The Re-Characterization of Israel’s Written Law. OTS 451. London and New York: T&T Clark. Lewis, M. 1971. Ecstatic Religion: An Anthropological Study of Shamanism and Spirit Possession. Harmondsworth: Penguin. Magen, Y., Tzionit, Y., and Sirkis, O. 2004. “Khirbet Badd ‘Isa—Qiryat Sefer”. In: Magen, Y. et al., The Land of Benjamin. Judea and Samaria Publications 3. Jerusalem: Israel Antiquities Authority, 179–241. Magness, J. 2001. “The Question of the Synagogue: The Problem of Typology”. In: AveryPeck, A. and Neusner, J. (eds.), Judaism in Late Antiquity, part 3: Where We Stand: Issues and Debates in Ancient Judaism, vol. 4: The Special Problem of the Ancient Synagogue. HdO I 55. Leiden: Brill, 1–48. Magness, J. 2011. Stone and Dung, Oil and Spit: Jewish Daily Life in the Time of Jesus. Grand Rapids, MI: Eerdmans. Magness, J. 2012. The Archaeology of the Holy Land: From the Destruction of Solomon’s Temple to the Muslim Conquest. Cambridge: Cambridge University Press. Ma‘oz, Z. U. 1999. “The Syn[a]gogue that Never Existed in the Hasmonean Palace at Jericho: Remarks Concerning an Article by E. Netzer, Y. Kalman and R. Loris [sc. Laureys]”. Qadmoniot 118: 120–121 (in Hebrew). McCollough, C. T. 2015. “Khirbet Qana”. In: Fiensy, D. and Strange, J. R. (eds.), The Archaeological Record From Cities, Towns, and Villages. Vol. 2 of Galilee in the Late Second Temple and Mishnaic Periods. Minneapolis: Fortress, 127–145. Meyer, B. F. 2002. The Aims of Jesus. PTMS 48. Eugene: Pickwick. Meyers, E. and Chancey, M. 2012. Alexander to Constantine. Archaeology of the Land of the Bible 3. New Haven: Yale University Press. Miquel, E. 2010. “How to Discredit an Inconvenient Exorcist: Origin and Configuration of the Synoptic Controversies on Jesus’ Power as an Exorcist”. Biblical Theology Bulletin 40.4: 187–206. Modrzejewski, J. M. 2005. “What is Hellenistic Law? The Documents of the Judaean Desert in the Light of the Papyri from Egypt”. In: Katzoff, R. and Schaps, D. (eds.), Law in the Documents of the Judaean Desert. JSJSup 96. Leiden and Boston: Brill, 7–21. Moore, C. A. 1977. Daniel, Esther, and Jeremiah: The Additions. AB 44. New York: Doubleday. Mosser, C. 2013. “Torah Instruction, Discussion, and Prophecy in First-Century Synagogues.” In: Porter, S. E. and Pitts, A. W. (eds.), Christian Origins and Hellenistic Judaism: Social and Literary Contexts for the New Testament. Texts and Editions for New Testament Study 10. Leiden and Boston: Brill, 523–552. Netzer, E. 1999. “A Synagogue from the Hasmonean Period Recently Exposed in the Western Plain of Jericho”. Israel Exploration Journal 49: 203–221.

Contributions of Early Synagogues to Historical Jesus Research

211

Oakeshott, M. 1994. Experience and Its Modes. Repr. ed. Cambridge: Cambridge University Press. Olsson, B. 2003. “The Origins of the Synagogue: An Evaluation”. In: Olsson, B. and Zetterholm, M. (eds.), The Ancient Synagogue From Its Origins Until 200 C.E.: Papers Presented at an International Conference at Lund University, October 14-17, 2001. CBNTS 39. Stockholm: Almqvist & Wiskell, 132–138. Onn, A., Wexler-Bdolah, S., Rapuano, Y., and Kanias, Tz. 2002. “Khirbet Umm el-‘Umdan”. Hadashot Arkheologiyot 114: 64*–68*. Pitt-Rivers, J. 1966. “Honour and Social Status”. In: Peristiany, J. G. (ed.), Honour and Shame: The Values of Mediterranean Society. The Nature of Human Society Series. Chicago: University of Chicago Press, 21–77. Ritmeyer, L. 2014. “Imagining the Temple Known to Jesus and to Early Jews”. In: Charlesworth, J. H. (ed.), Jesus and Temple: Textual and Archaeological Explorations. Minneapolis: Fortress Press, 19–57. Runesson, A. 2001. The Origins of the Synagogue: A Socio-Historical Study. ConBNT 37. Stockholm: Almqvist & Wiksell. Runesson, A. 2004. The Nature and Origins of the First-Century Synagogue. Bible and Interpretation. [Online]. Available from: http://www.bibleinterp.com/articles/ 2004/07/run288001.shtml (accessed 24 August 2020). Runesson, A. 2007. “Architecture, Conflict, and Identity Formation: Jews and Christians in Capernaum from the 1st to the 6th Century”. In: Zangenberg, J., Attridge, H. W. and Martin, D. B. (eds.), Religion, Ethnicity, and Identity in Ancient Galilee: A Region in Transition. WUNT 210. Tübingen: Mohr Siebeck. Tübingen: Mohr Siebeck, 231–257. Runesson, A. 2008. “Rethinking Early Jewish-Christian Relations: Matthean Community History as Pharisaic Intragroup Conflict”. JBL 127: 95–132. Runesson, A. 2014. “The Historical Jesus, the Gospels, and First-Century Jewish Society: The Importance of the Synagogue for Understanding the New Testament”. In: Warner, D. A. and Binder, D. D. (eds.), A City Set on a Hill: Essays in Honor of James F. Strange. Mounatinhome: Borderstone, 265–297. Runesson, A. 2016. “Entering a Synagogue with Paul: First-Century Torah Observance”. In: Wendel, S. J. and Miller, D. M. (eds.), Torah Ethics and Early Christian Identity. Grand Rapids, MI: Eerdmans, 11–26. Runesson, A., Binder, D. D., and Olsson, B. 2008. The Ancient Synagogue From its Origins to 200 C.E.: A Source Book. Leiden and Boston: Brill. Ryan, J. J. 2015. “Jesus at the Crossroads of Inference and Imagination: The Relevance of R. G. Collingwood’s Philosophy of History for Current Methodological Discussions in Historical Jesus Research”. JSHJ 13: 66–89. Ryan, J. J. 2017a. The Role of the Synagogue in the Aims of Jesus. Minneapolis, MN: Fortress Press. Ryan, J. J. 2017b. “Jesus and Synagogue Disputes: Recovering the Institutional Context of Luke 13:10-17”. CBQ 79: 41–59. Ryan, J. J. forthcoming a. “‘He Said These Things While He Was Teaching in the Synagogue at Capernaum’: Capernaum in the Fourth Gospel in Light of Archaeology and Synagogue Studies”. In: Anderson, P. N. (ed.), Archaeology and the Fourth Gospel. Grand Rapids, MI: Eerdmans. Ryan, J. J. forthcoming b. “The Socio-Political Context of Public Synagogue Debates in the Second-Temple Period”. In: Bonnie, R., Hakola, R., and Tervahauta, U. (eds.), The Synagogue in Ancient Palestine: Current Issues and Emerging Trends. FRLANT. Göttingen: Vandenhoeck & Ruprecht. Sanders, E. P. 1985. Jesus and Judaism. Philadelphia: Fortress.

212

Jordan J. Ryan

Spigel, C. S. 2012. Ancient Synagogue Seating Capacities. TSAJ 149. Tübingen: Mohr Siebeck. Stacey, D. 2004. “Was There a Synagogue in Hasmonean Jericho?” Bible and Interpretation [Online]. Available from: http://www.bibleinterp.com/articles/Hasmonean_ Jericho.shtml (accessed 27 February 2020). Strange, J. F. 2003. “Archaeology and Ancient Synagogues up to about 200 C.E”. In: Olsson, B. and Zetterholm, M. (eds.), The Ancient Synagogue From Its Origins Until 200 C.E.: Papers Presented at an International Conference at Lund University, October 14-17, 2001. CBNTS 39. Stockholm: Almqvist & Wiskell, 37–62. Strange, J. F. 2014 “The Galilean Road System”. In: Fiensy. D. A. and Strange, J. R. (eds.), Galilee in the Late Second Temple and Mishnaic Periods. 2 vols. Minneapolis: Fortress, 1: 263–271. Theissen, G. and Winter, D. 1997. Die Kriterienfrage in der Jesusforschung: vom Differenzkriterium zum Plausibilita tskriterium. NTOA 34. Gottingen: Vandenhoeck & Ruprecht. Twelftree, G. H. 2011. “Jesus and the Synagogue”. In: Holmén, T. and Porter, S. E. (eds.), Handbook for the Study of the Historical Jesus. Leiden: Brill, 3105–3134. Vermes, G. 1973. Jesus the Jew: A Historian’s Reading of the Gospels. Philadelphia: Fortress. Vermes, G. 1993. The Religion of Jesus the Jew. Minneapolis: Fortress Press. Watts, J. W. 2013. “The Political and Legal Uses of Scripture”. In: Paget, J. C. and Schaper, J. (eds.), From the Beginnings to 600. Vol. 1 of The New Cambridge History of the Bible. Cambridge: Cambridge University Press, 345-364. Winter, D. 2012. “Saving the Quest for Authenticity from the Criterion of Dissimilarity: History and Plausibility”. In: Keith, C. and Le Donne, A. (eds.), Jesus, Criteria, and the Demise of Authenticity. London and New York: T&T Clark, 115–131. Witmer, A. 2012. Jesus, the Galilean Exorcist: His Exorcisms in Social and Political Context. LNTS 459. London and New York: T&T Clark. Yadin, Y. 1982. “The Synagogue at Masada”. In: Levine, L. I. (ed.), Ancient Synagogues Revealed. Jerusalem: Israel Exploration Society, 19–23. Zapata-Meza, M., Garza Diaz Barriga, A., and Sanz-Rincón, R. 2018. “The Magdala Archaeological Project (2010–2012): A Preliminary Report of the Excavations at Migdal”. Atiqot 90: 83–125.

III. Theorizing Practice in Ancient Synagogues

What Were They Doing in Second Temple Synagogues? Philo and the προσευχή

Jutta Leonhardt-Balzer

Introduction The ancient synagogue has been the object of much research. Yet, despite an increasing number of archaeological finds and the plausible expectation that such an important institution might have left a wealth of evidence, there is little information about what went on in Second Temple synagogues. It is easier to find evidence for the existence of the synagogues than for what people did there, either because what happened in the buildings varied locally or because it was taken for granted. The archaeological and epigraphic evidence does not provide much information about activities within. The synagogue liturgies are dated much later. Thus, scholarship must rely on the literary evidence from Second Temple times, which consists largely of Philo and Josephus (and the New Testament), and even here the evidence is not clear cut, but serves each author’s literary purpose.1 Josephus and his rhetorical intention in the references to the synagogue has been covered recently by Andrew R. Krause (2017). In the present contribution, I will focus mainly on Philo’s writings. Other evidence will be covered in other papers in this volume. None of the evidence provides a comprehensive overview of the function of the synagogue, but together they give us islands of evidence which may help us to draw a picture of some detail. Philo regards the synagogue as an ancient institution.2 The archaeological evidence for synagogues in Egypt goes back to the middle of the second century BCE, so by Philo’s time it had been around for about 300 years. For Philo and his contemporaries in Alexandria, already the translation of the Torah is clothed in legend, and so is this institution in which the Law is the centre. Consequently, Philo traces the origin of the synagogue back to the time of Moses.3 For Philo, the Torah has always been read publicly and interpreted by  1 See Levine 2005. A broad recent overview of the evidence can be found in Runesson, Binder, and Olsson 2008. 2 On Philo and the synagogue, see Leonhardt 2001a: 53–100. 3 See Mos. 2.215–216. Josephus, C. Ap. 2.175, also dates the synagogue back to Moses; cf. also Mann 1971: 452–453 and Sanders 1990: 78.

216

Jutta Leonhardt-Balzer

experts. In the present context I cannot discuss the issues of the origin of the synagogue.4 It is clear that—for Philo at least—the emergence of the synagogue is inseparably linked to the teaching of the Torah. In search of further details, I will here focus on Philo’s terms for the synagogue, his idea of the purpose of this institution, and Philo’s account of the proceedings in the synagogues of the Essenes and Therapeutae. I will then briefly look at some evidence from Josephus which draws a slightly different picture, before I finish with some conclusions about what Diaspora Jews were thinking they were doing in synagogues. Hence, the first of the following sections is devoted to an analysis of Philo’s use of the terms and what they say about his idea of the synagogue’s function.

The Terms for the Synagogue in Philo Philo does not often refer to the synagogue, most frequently in the historical treatises. He uses mainly two terms for the Jewish places of assembly. The term which became the standard term in later antiquity is “synagogue”, συναγωγή. In a nonJewish context, it is any “assembly”, of things (as in “harvest”), people (“assembly” or “army”), or thoughts (“conclusion”).5 Without political connotations it was the term for pagan associations and societies,6 but consistently denoting groups, not a place. The same use can be seen in the early Jewish sources. The Septuagint uses συναγωγή in Genesis and Numbers for all Israel (not so in Deuteronomy).7 Philo also uses the term for the congregation in a quotation of Num 27:16–17 in Post. 67 and Agr. 44. The related form συναγώγιον is only rarely used by Philo, thus the evidence may not be representative. It can have all the nuances of συναγωγή, but in the non-Jewish texts it usually refers to a gathering of people or its place (e.g., a “picnic”8 or the “place of a picnic”9). In Somn. 2.126–127, a member of the pagan government uses the term for the Jewish Sabbath assemblies. Likewise, in Legat. 311 Philo refers to a letter of Augustus, which grants the Jewish community meeting rights in their συναγώγια, but it is not clear whether these are buildings or assemblies.10  4

On the synagogue, see Krauss 1922; Gutmann 1975; Levine 1987; Juster 1914: 1.546–472; Schürer 1973–1987: 2.423–454; Schrage 1990; Safrai 1977; and Oesterley 1941: 211–218. On synagogue origins, see Gutmann 1975: 3–76; Kee 1990; 1994; 1995; Oster 1993; Riesner 1995, and Hegermann 1989; on the regional variety, see Kraabel 1992a, 1992b. 5 See Liddell and Scott 1996: 1691–1692. 6 Schrage 1990: 800–802. 7 Hatch and Redpath 1998: 1309–1310. 8 Menander 159, see Liddell and Scott 1996: 1692. 9 Polliander 6.7, see Liddell and Scott 1996: 1692. 10 Cf. Hengel 1971: 169.

Philo and the προσευχή

217

Prob. 81 contains the only unambiguous reference to συναγωγή as a building: the meeting house of the Essenes (“they proceed to sanctuaries, which are called synagogues” (εἰς ἱεροὺς ἀφικνούμενοι τόπους οἳ καλοῦνται συναγωγαί).11 Already Martin Hengel suggested that this term “synagogue” (συναγωγή), not proseuchē (προσευχή), was the common term for the Jewish meeting places in Palestine.12 This would mean that Philo’s account of the Essenes mirrors the Palestinian usage described first by Hengel.13 By contrast, in Philo’s time, the main term for the buildings for Jewish assemblies in the Diaspora was προσευχή. Derived from the meaning “prayer”,14 the term is used for the Temple in Isa 56:7 LXX, the “house of prayer”, οἴκος προσευχῆς, and likewise other “places of prayer” are also called προσευχαί, but only in the Jewish context.15 Philo uses the term nineteen times,16 more often than any of the other terms for the synagogue, and the cognate term προσευκτήριον once (Mos. 2.216). Compared with the roughly 1,500 preserved pages of Philo’s oeuvre, the number of occurrences is small, but this does not imply any lack of interest.17 It is largely due to the exegetical nature of Philo’s work.

The Purpose of the προσευχή in Philo In 1994 Heather McKay argued, largely based on Philo’s writings, that the Jewish Sabbath assemblies only served as schools without any religious purpose.18 Her theory has not found general acceptance,19 but as the language Philo uses gave rise to this claim it is necessary to outline the evidence here. McKay is correct in the sense that Philo sees the most important purpose of the proseuchē as the site of the Sabbath assemblies (Spec. 2.61–63): The abstention from manual labour on the Sabbath enables all the Jews to study philosophy (61), which is the study of the principles of virtue: For on the seventh day (ταῖς ἑβδόμαις) in every city (κατὰ πᾶσαν πόλιν) there are wide open thousands (μυρία) of schools of good sense, temperance, courage, justice, and all the other virtues (διδασκαλεῖα φρονήσεως καὶ σωφροσύνης καὶ ἀνδρείας καὶ δικαιοσύνης καὶ τῶν ἄλλων ἀρετῶν), in which they sit quietly in order (ἐν κόσμῳ

 11 McKay 1994: 75–96, distinguishes the term as referring to the specific meeting places of the Essenes from the proseuchai found elsewhere, but Philo’s account of the activities in both (see below on the Essenes) is identical. 12 Hengel 1971: 177. 13 Cf. Horbury 1998a: 301. 14 Liddell and Scott 1996: 1511. 15 Cf. Levinskaya 1996: 207–225. 16 For a list of passages see Borgen, Fuglseth, and Skarsten 2000: 299. 17 Cf. Martin 2000a. 18 McKay 1994. 19 Cf. van der Horst 1999; Leonhardt 2001a: 53–100.

218

Jutta Leonhardt-Balzer καθέζονται σὺν ἡσυχίᾳ), with pricked ears (τὰ ὦτα ἀνωρθιακότες) and all attention (μετὰ προσοχῆς πάσης) for the sake of their thirst for a drink of reason (ἕνεκα τοῦ διψῆν λόγων ποτίμων), while one of the most experienced stands up (ἀναστὰς δέ τις τῶν ἐμπειροτάτων) to instruct them in what is best and most useful (ὑφηγεῖται τὰ ἄριστα καὶ συνοίσοντα), by which the whole life can be handed over to improvement (οἷς ἅπας ὁ βίος ἐπιδώσει πρὸς τὸ βέλτιον). (Spec. 2.62)

This important text describes the content and the form of the Sabbath teaching in terms of Greek philosophy. Philo, just like 4 Macc 1:17–18; 2:23 and Wisd 8:7,20 identifies the Jewish teaching as instruction in the Greek virtues, ethics, and the practical use of philosophy for a good life.21 Not only among the Jews, but also in the Greek context the ethical behaviour of a philosopher or the members of a philosophical school was seen as proof of the value of their philosophical instruction. Consequently, when Philo describes the good practice among the Jews, this serves to prove the quality of the Jewish philosophy. Philo, like Mark 12:28–34, structures the Jewish principles of virtue under two headings: they are “two highly relevant principles, one towards God, through piety and holiness, and one towards human beings, through humane behaviour and justice” (δύο τὰ ἀνωτάτω κεφάλαια, τό τε πρὸς θεὸν δι εὐσεβείας καὶ ὁσιότητος καὶ τὸ πρὸς ἀνθρώπους διὰ φιλανθρωπίας καὶ δικαιοσύνης; Spec. 2.63). The idea of the proseuchē as a school of philosophy to improve one’s life is also expressed in Mos. 2.215 for the time of Moses using almost the same terminology, with a similar list of virtues (216). Although Philo calls these places of teaching “schools” (διδασκαλεῖα), here, he clearly does not refer to general education (unlike Cher. 98–106), but to the specific Sabbath assemblies. Based on Mos. 2.215, McKay claims that “Philo is painting a picture of educational gatherings in προσευκτήρια where religious, social and moral topics are discussed”,22 but she ignores the fact, that Philo and his Hellenistic contemporaries, when talking about the Sabbath teaching, translate Jewish traditions into Greek terminology.23 Philo frequently calls the Sabbath teaching the practice of philosophy (Opif. 128), but that does not mean that he regards it as secular: Greek philosophy also included teaching about the gods, for the separation between theology and philosophy did not exist at the time. In Philo’s apologetic context it was opportune to identify the content of the Sabbath instruction with Greek virtues.24 In Hypoth. 7.11 Philo uses unambiguous terms:25 the Sabbath meetings ensure that the Jews “have expert knowledge of the ancestral laws and customs” (καὶ τῶν  20

Cf. Winston 1990: 384. Cf. also 4 Macc. 1:17–18, 2:23. On Philo’s representation of the Jewish teachings as a varying number of Greek virtues with the addition of “piety and holiness” (cf. Mos. 2.215–216), see Cohen 1995: 86–99. 22 McKay 1994: 66. 23 On Philo’s references to “philosophy” as normal Jewish Sabbath teaching, see Mann 1971: 454. 24 Cf. Borgen 2001. 25 Sanders 1990, 197–202 matches Philo’s description of the Sabbath studies to that of other sources. 21

Philo and the προσευχή

219

πατρίων νόμων καὶ ἐθῶν ἐμπείρως ἔχειν). Pagan as well as Jewish texts use the terminology, “ancestral laws” for the recognized legislation of a nation, in this case it is the Torah.26 Thus, the passage does not refer to the general instruction in Greek philosophy but specifically to the teaching of the Torah and other Jewish traditions.27 The term “school”, used by Philo for the proseuchai, situates the Jewish assemblies in the intellectual and cultural environment of Alexandria,28 comparing their instruction with Graeco-Roman philosophical schools. Unlike the Hellenistic philosophical schools, the proseuchai, however, provide ethical teaching not just for a wealthy elite, but for every Jew. These schools, as Philo emphasizes, exist “in every city” (Spec. 2.62; Legat. 145), a claim also made by Josephus (C. Ap. 2.282). While naturally both authors exaggerate in their apologetic intention, epigraphic evidence proves that there were proseuchai even in such relatively small Egyptian settlements as Schedia and Crocodilopolis,29 and the building of a syngagogue in the middle of nowhere to commemorate a victory in battle in 3 Macc 7:20 also gives evidence of their wide distribution. Regarding the question of what went on in these schools, the above-mentioned text Spec. 2.62 emphasizes that the assembled audience listens quietly to expert instruction. The same is mentioned in Somn. 2.127, in a polemic account of the proceedings from the mouth of a hostile Roman official. He asks the Jews who insist on following their traditions despite his acts against them whether they would behave similarly in the case of a natural disaster and lists the activities in their proseuchai: And will you be seated (καθεδεῖσθε) in your assemblies (ἐν τοῖς συναγωγίοις ὑμῶν), meeting in your usual association (τὸν εἰωθότα θίασον ἀγείροντες), reading safely the holy books (ἀσφαλῶς τὰς ἱερὰς βίβλους ἀναγινώσκοντες), explaining if anything is obscure (κἂν εἴ τι μὴ τρανὲς εἴη διαπτύσσοντες), and passing your time and theorizing at great length over the ancestral philosophy (τῇ πατρίῳ φιλοσοφίᾳ διὰ μακρηγορίας ἐνευκαιροῦντές τε καὶ ἐνσχολάζοντες)? (Somn. 2.127)

It is clear from this account that there is a strict order, probably also relating to seating, a reading of the “holy books” and the exposition of difficult passages. Philo focusses on the reading of the bible and the interpretation of unclear pas  26 The term “ancestral laws” was used in the ancient world for accepted oral and written national cultic and political ordinances, see Schröder 1996. On the term in Philo: ibid. 216–224. 27 Sandmel 1978: 144–147, calls the synagogue “a school for adults” (147). Moore 1927: 1.285– 307, also sees its purpose as instruction. This does not imply, however, that the synagogue was a separate institution from the school, against Mann 1971: 499. 28 On the stabilizing social and political importance of the sage in Hellenistic times, see Fiore 1990: esp. 332. 29 Cf. Horbury and Noy 1992 (JIGRE): nos. 22–23, 117; see also the contribution of Kimberley Czajkowski in the present volume.

220

Jutta Leonhardt-Balzer

sages, including the discussion of related traditions. The list contains only activities which would be unreasonable in the case of a natural disaster. Other activities, such as prayer, might have been regarded as sensible reactions to a catastrophe, therefore the text does prove that prayer was not practiced in the service. It could merely have been omitted from this speech. The reference to the “reading of the holy books” (127) indicates that McKay’s distinction between “Sabbath observance” and “Sabbath worship”30 does not correspond to Philo’s own understanding of the proceedings. In Hypoth. 7.12–14 there are further details of the Sabbath teaching in the proseuchai: There is a seating order to facilitate undisturbed attention to the readings and instructions (12).31 The assembly remains “in silence, except when it is customary to reply something to what has been read” (σιωπῇ, πλὴν εἴ τι προσεπευφημῆσαι τοῖς ἀναγινωσκομένοις νομίζεται; 13). This indicates a structured exchange, but not a dialogical development of an exposition, rather it seems to imply a liturgy of fixed responses to the reading. This content of the reading might conceivably also include psalms or other prayers, associated with a communal response,32 such as the “Amen” mentioned in the rabbinic account of the service in the great synagogue in Alexandria.33 However, Philo calls the text that is read and interpreted the νόμος, which is the term for the Pentateuch. Therefore, it is more likely that Philo refers to Torah reading. The account then mentions that the Torah was not just read but also “interpreted point by point” by a “priest” or “elder” (τῶν ἱερέων δέ τις ὁ παρὼν ἢ τῶν γερόντων εἷς ἀναγινώσκει τοὺς ἱεροὺς νόμους αὐτοῖς καὶ καθ’ ἕκαστον ἐξηγεῖται), and that this exposition went on until the late afternoon (μέχρι σχεδὸν δείλης ὀψίας; 13). As chapter or verse division did not exist at the time the καθ’ ἕκαστον ἐξηγεῖται either refers to “each law”, as νόμος was the last noun before the phrase, or rather, as the Torah not only consists of laws but also narrative, genealogies etc., καθ’ ἕκαστον refers to each sentence or even each exegetical issue. There is no evidence for a reading  30

MacKay 1994: 63, 73, 75. The seating is also an important criterion in modern archaeology, where the existence of benches along the walls in excavated buildings is seen as evidence for the identification of the building as a synagogue; see the papers of Zeev Weiss as well as of Mechael Osband and Benjamin Arubas in this volume, as well as the Introduction by the editors. 32 Rabbinic evidence can only be used with difficulty to reconstruct pre-70 traditions, see Maier 1990: 548, 550. Moore 1927: 1.290, sees Philo as evidence for the continuity of the synagogue and its services, but although Philo attests to the existence of prayer, reading and interpretation in the proseuchē, he does not provide any information about the liturgy of his time. For reconstructions of the early synagogue service based on rabbinic material, see Elbogen 1993; Schürer 1973–1987: 2.447– 463; Sandmel 1978: 147–153; Safrai 1977: 78–83; Manns 1986: 49–51, 130–155; Salzmann 1994: 455– 456; Billerbeck 1964; Maier 1990: 556–560; and Moore 1927: 1.285–307. Reconstruction based on Second Temple, Rabbinic and Christian sources in Horbury 1998a. On regular communal prayer in Qumran (esp. 1QS VI 3, 8), see Sanders 1990: 197, 206, 351. On Qumran prayer and rabbinic evidence, see Maier 1990: 543–547; and Sarason 2003. 33 See y. Sukkah 5:1, 55a–b; cf. Mélèze Modrzejewski 1997: 91; Salzmann 1994: 456; Moore 1930: 3.91–92. 31

Philo and the προσευχή

221

cycle,34 but only for reading accompanied by a qualified exposition (14).35 Based on Philo’s allegorical texts, Peder Borgen has tried to reconstruct a “homiletic pattern”,36 to reflect the method of the Sabbath teaching: First the text is quoted (it can also be paraphrased or the quotation repeated throughout the exegesis). The subsequent interpretation focusses on details of the text and uses other biblical passages to develop an argument. “The closing statement of the homily refers back to the main statement at the beginning and at the same time sums up points from the entire homily.” Apart from the main text there can also be subordinate quotations.37 Borgen finds this pattern in Philo, Paul and Palestinian Midrashic texts,38 and while the extant evidence is exclusively of a literary nature, there is the possibility that it could also be applied to oral sermons.39 Philo appreciates the Sabbath instruction of the Torah as the reason why every Jew is enabled not only to live according to the Jewish customs, but also to defend them without help from experts and to pass the instruction on to women, children and slaves at home (Hypoth. 7.14).40 This comment, together with Philo’s wonder at the presence of women in the meetings of the Therapeutae (Contempl. 32–33) indicates that women probably were not present in the normal proseuchē meetings known to Philo. However, in Mos. 1.180 and Agr. 79–83, referring to Exod 15:20–21, women are described as joining the men in a double choir, comparable to pagan choirs. This could indicate that the Sabbath meetings contained one part for prayer—and possibly the reading of the Torah, as Neh 8 might seem to suggest—in which women were present, and another for the exegesis, when they left.41 The presence of women during the praise sections might be supported by Philo’s account of the Jewish praise after Flaccus’ arrest in Flacc. 122. This bipartite structure would correspond to the practice in ancient symposia, where it was also customary for the women to leave after the meal and before the entertainment or philosophical discussion,42 but based on Philo alone it is virtually impossible to prove.  34 There is no direct evidence for a pre-70 reading cycle (Manns 1986: 104; Salzmann 1994: 457; and Safrai 1977: 83–86), critical description in Perrot 1984:130–132. For an attempt of a reconstruction of Torah reading (based on Mos. 2.41), see Cohen 1995: 289–290; on the haftarah cycle, see Cohen 1997. 35 Perrot 1984: 125–128 suggests that either the reading of Scripture consisted of Torah and haftarah reading plus interpretation (Acts 13:15), Torah reading and sermon including elements from the prophets, or just Torah reading with sermon. The third option is evidenced in Philo. 36 See Borgen 1965: 28–58, using Leg. 3.162–168, Mut. 253–263 and John 6:31–58. 37 Borgen 1965: 42–43. 38 Borgen 1965: 46–58. 39 Cf. the contribution in this volume of Hermut Löhr on the petichah as an artful dialogue of texts with the Torah. 40 Josephus agrees with Philo on the importance of the Sabbath teaching for the ordinary Jews’ knowledge about the Torah (C. Ap. 2.175). 41 Horbury 1999: 363–367, finds in Second Temple sources evidence for different parts of the synagogue service: Torah reading, prayer, liturgy and exposition. 42 Cf. Leonhardt-Balzer 2017a: 253–257.

222

Jutta Leonhardt-Balzer

Philo’s first explicit piece of evidence for a clearly religious use of the proseuchai is his indication that they can be made unfit for Jewish use by placing in them statues of the emperor Gaius (Flacc. 41, 53). This proves that they are not merely secular community buildings or philosophical academies. Likewise, the name of the buildings—“places of prayer”—suggests a religious use. The term proseuchē is not an Egyptian peculiarity, only used for buildings in Egypt, but for those all over the Empire with the same sacred status, and Philo is aware of this (Flacc. 45, 47).43 The fact that the name is not merely a metaphor but refers to actual prayers inside is also proved by Philo’s account, for after this desecration the Jews pray outdoors (Flacc. 122), suggesting that normally they would have prayed inside the proseuchai. Philo describes the purpose of the proseuchai as “piety towards their benefactors” (εἰς τοὺς εὐεργέτας εὐσέβειαν), without them the Jews “do not have any holy enclosures” (οὐκ ἔχοντες ἱεροὺς περιβόλους) and places for “thanksgiving” (τὸ εὐχάριστον; Flacc. 48). Thus, Philo’s account states that thanksgiving and piety to the Jewish benefactors are the true purpose of the proseuchai.44 This euergetistic framework of the synagogues is also confirmed by the epigraphic evidence from Egypt and elsewhere in the Roman Empire.45 Thus, Philo presents the proseuchai as functioning within the Graeco-Roman civic context. The first benefactor to be commemorated is God. The Greek gods could also be described using euergetistic terms,46 but the Jewish people had particular cause to define themselves as the recipients of God’s gifts, which comprised the Exodus, the Torah, the Land of Israel and the Temple in Jerusalem. Because of these gifts it was a duty to commemorate the donor.47 Likewise, the rulers were already honoured in the earliest proseuchai in Egypt.48 Already the early inscriptions from the Fayum and the Delta and then all over Egypt call the Ptolemaic kings and queens “benefactors”.49 This demonstrates the Jewish loyalty to the current rulers, who in turn protect the buildings as “inviolate”, ἄσυλον, as sacred places, dedicated to a recognized deity, just like the temples of other cults.50 The inscriptions prove that the combination of respect for the rulers and Jewish monotheism as shown in Flacc. 48 was common  43 The sacred status seems to derive from the activities within, especially from reading the sacred texts, while the question of their relation to the sanctity of the Jerusalem Temple remains debated; see the contribution of Lutz Doering in this volume. In the Diaspora, the connection between the proseuchai and the Jerusalem Temple may be seen in the collection of the Temple tax. 44 Cf. McKay 1994: 67–69, 71–73. 45 See Rajak 1996; and Lifshitz 1967. On euergetism in general, see Veyne 1990; Winter 1994: 22, 25–40; and Gordon 1990. 46 See Neyrey 2005. 47 See Horbury 1986: 40–42. 48 See Cerfaux and Tondriau 1957: 264; and Nock 1972. 49 See Horbury and Noy 1992 (JIGRE): nos. 22, 24–25, 27–28, 117, 125–126. 50 The dedicatory inscriptions in the proseuchai are expressions of loyalty to the ruler, but the language of the ruler cult is also applied to the Jewish God, cf. Horbury 1998b: 74–76. Mélèze Modrzejewski 1997: 94–95; Horbury and Noy 1992 (JIGRE): no. 125.

Philo and the προσευχή

223

among the Jews in Egypt.51 The benefactor terminology plays elegantly with the expectation of ruler worship without compromising Jewish monotheism. When Philo adds that the Jews were willing to die for the sake of this place out of piety for their benefactors, he indicates that the Jews can only express their loyalty to the Roman rulers in their own worship. Thus, Philo can call the proseuchai centres of homage “to the Augustan house” (εἰς τὸν Σεβαστὸν οἶκον, Flacc. 49). On the other hand, the statues which Gaius has put up in the synagogue stand in breach of the ties of loyalty between the ruler and the loyal citizens.52 A similar picture is drawn in the Legatio ad Gaium. In the hope for royal favour, statues of Gaius were erected in those Alexandrian proseuchai which could not be torn down (Legat. 134, 137, 165, 191). Unlike Gaius (346), neither the Ptolemies (138) nor Caligula’s Roman predecessors (148, 156) had similar images erected. Augustus himself protected the Sabbath and the proseuchai in Rome,53 as well as the collections for the Temple there (156).54 The statues of Gaius inside the proseuchai effectively turn them into temples to Gaius (346). His actions, interfering with the means to live according to the Jewish customs in the Diaspora, are seen as an attack on the Jewish way of life (κατὰ τὰ πάτρια τῶν Ἰουδαίων, 371).55 In the account of the celebration after Flaccus’ arrest, Philo describes that the Jews pray and sing outdoors near the sea (Flacc. 120–123). This event is only singular56 in the sense that the Jews would normally have done the same in their proseuchai, not that the singing and praying was unusual. Josephus quotes a decree from Halicarnassus which refers to customary Jewish prayer near the sea (A.J. 14.256–258),57 and archaeological evidence from Galilee and the Diaspora situates synagogues near the water.58 Philo also describes the celebration of the translation of the LXX as occurring on the beach with “prayers and thanksgiving” (μετὰ δὲ τὰς εὐχὰς καὶ τὰς εὐχαριστίας, Mos. 2.42). This combined evidence supports the conclusion that the references to “houses of prayer” are not merely metaphorical.  51

Cf. Cerfaux and Tondriau 1957. See Juster 1914: 1.460. 53 Proseuchai in Rome are also attested in pagan sources, e.g., Ovid, Ars 1.76; see Horbury 1998a: 300. 54 See Juster 1914: 1.338–385. 55 Against McKay 1994: 87. 56 McKay 1994: 69–70 regards it as a one-off event. 57 The form προσευχαί is sometimes translated as “prayers”, see Horbury 1998a: 303, sometimes rendered as proseuchai, see Pucci ben Zeev 1998: 215–216. Both renditions attest to Jewish communal prayer at the shore. 58 Horbury 1998a: 303. Proseuchai were situated near beaches because of the need for facilities for purification rites. Their location near water must have been widely known, for in Acts 16:13 Paul and his companions, new to Philippi, go to the river expecting to find a προσευχή there. However, purification is rather a requirement to enter the building and congregation, not an activity within, and does not need to be discussed in greater detail here. 52

224

Jutta Leonhardt-Balzer

Philo’s account of the prayers and hymns at the celebration of Flaccus’ arrest, however, provide further information about the events in the proseuchai, for the combination of prayer and singing points towards psalmody, and there is strong evidence for Philo’s appreciation of the Psalms.59 Of the biblical books, the Psalms are the next most frequently quoted after the Pentateuch,60 albeit never as the main text, but only as a sub-quotation explaining the Torah in the allegorical commentary, usually in cases where there is no proof text for Philo’s argument in the Torah.61 Philo also quotes Pentateuchal psalms, e.g., Exod 15 and Deut 32 (Plant. 46–61; Leg. 3.105; Det. 114; Post. 121, 167, etc.). When quoting the psalter, he not only introduces the texts referring to the psalmists (Agr. 50) and their singing (Conf. 149), or prayer (Conf. 39), he also refers to the recital of the psalms when introducing a psalm quotation: “in the psalms it is sung/recited” (ἐν ὕμνοις ᾄδεται, Somn. 1.75; cf: ᾄδεται δὲ καὶ ἐν ὕμνοις ᾆσμα τοιοῦτον; Mut. 115). Philo places great value on human singing (Post. 106; Plant. 126, 135, Ebr. 94; Agr. 80– 81).62 He also mentions song as one form of praise (Somn. 2.38), especially in the form of hymns to God (Somn. 1.35), who can be called the ποιητής, the “creator” or “poet” of the universe (ὁ δὲ ποιητὴς τῶν ὅλων, ὁ τοῦ κόσμου πατήρ, Mos. 2.238), who deserves praise. Furthermore, Philo repeatedly emphasizes that the purpose of a hymn is thanksgiving (Spec. 2.199; Mos. 1.180; 2.256–257). In Virt. 72–75 Philo presents the great Song of Moses as a model of thanksgiving, in tune with human beings and angels. The song of Moses and Miriam at the sea in Exod 15 is described as a model of antiphonal liturgical singing of two choirs, one male and one female to give “thanks and honour” to God (Agr. 79–82 and Mos. [1.180]; 2.256–257).63 The language of thanksgiving in Mos. 2.256: “he honours the benefactor with hymns of thanksgiving” (εὐχαρίστοις ὕμνοις γεραίρει τὸν εὐεργέτην) is close to the way in which Philo describes the purpose of the proseuchai as thanksgiving and honour to the benefactors, even if the antiphonal chanting is only mentioned in Philo’s interpretation of a biblical narrative, not in an account of the practice of synagogal singing in his time. At the very least, the language shows that for Philo thanksgiving can also be expressed through hymnody. In the same way he describes the singing at the beach after the desecration of the proseuchai and the arrest of Flaccus: “stretching their hands towards the heavens they sang hymns of praise and led songs of triumph to the God who guards human affairs” (προτείνοντες τὰς χεῖρας εἰς οὐρανὸν ὕμνουν καὶ παιᾶνας ἐξῆρχον εἰς τὸν ἔφορον θεὸν τῶν ἀνθρωπίνων πραγμάτων; Flacc. 121–122). Most of Philo’s references to liturgical hymnody in his time refer to the Temple  59 See Leonhardt 2001a: 142–174. On Philo’s Psalms quotations as evidence of his attitude towards prayer, see Kaiser 2013. 60 In Cohn and Wendland 1986–1930: 7.43. 61 Thyen 1960: 59. 62 On Philo and music, see Feldman 1996; Levarie 1991: esp. 127 on Post. 104 and Cicero, Nat. d. 2.159. 63 See also Plant. 48; Ebr. 79, 111–118; Leg. 2.102–103; Conf. 35 and Somn. 2.268–271 on Exod 15.

Philo and the προσευχή

225

(Spec. 1.193, 224),64 but in the context of the Therapeutae, it is this hymn of Exod 15 which provides the model for the all nightly hymnody of men and women during their sober symposia at their great festival. Here we find an actual reference to the antiphonal singing of Exod 15. Beyond this biblical psalm Philo also mentions that they sing self-composed hymns (Contempl. 80–88). Here, for the Therapeutae, there is evidence that there was psalmody not only of the psalter, but also of other hymns, in the proseuchai. Regrettably for the attempt to prove regular psalm recital in synagogues, the fact that Philo needs to add that these hymns are self-composed seems to indicate that this was unusual among his contemporaries. Thus, Philo suggests that the Sabbath activities in the proseuchai consisted of Torah reading, exposition, prayer and psalm singing. Beyond the Sabbath meetings, the only evidence he provides for activities in the proseuchē is the collection of the Temple tax.65 Philo mentions that in the proseuchai of Rome the Jews collect the Temple dues and send them to Jerusalem (Legat. 156), a practice permitted by Augustus (157). In Legat. 256 Philo adds that the Temple dues are sent from all places of Jewish residence, inside and outside the Roman Empire. Philo also points out in the Legatio as well as in the Flaccus that any threat to the synagogues affects the synagogues elsewhere. This link between the synagogues and the Temple in Jerusalem is even recognized by the Romans, when Cicero comments on the half-shekel offering that it ties the Diaspora Jews to the city of Jerusalem (Flac. 69) and that any threat to the Jerusalem Temple also has consequences for the Jews in Rome.66 Thus, the evidence suggests that the proseuchai reinforce the link of the Diaspora Jews with this institution.67 In conclusion, we find that for Philo the proseuchai are centres of Jewish tradition in the Diaspora. Their teaching guarantees the continuation of the Jewish tradition. They do not rival but reinforce the Temple, nevertheless they are regarded as sacred sites, focus of worship of the Jewish God and loyalty to the ruler, and within them there is prayer and thanksgiving in addition to Torah reading and exposition.

Synagogues in Jewish Groups: Essene συναγωγαί and the προσευχαί of the Therapeutae In addition to the general references to the proseuchai, Philo also describes the meetings of two specific Jewish groups, the Essenes and the Therapeutae. The  64

Cf. prayers and hymns at Passover in Spec. 2.148, against Feldman 1996: 527. On the importance of the Temple tax, see also Leonhardt-Balzer 2017b. 66 See Schwartz 1996: 125. 67 On the Temple and the synagogue, see Leonhardt 2007. Against the theory of a competition between the two in Martin 2000b. 65

226

Jutta Leonhardt-Balzer

Essenes are most likely related to some group associated with the Qumran movement,68 and the Therapeutae are an ascetic Jewish group near Alexandria.69 Both groups represent for Philo an ideal type of Judaism, the Essenes that of the practical and the Therapeutae of the contemplative life of virtue (Contempl. 1). The Essene assemblies in their synagogues are described in the same way as those of other Jews: the seating order places the younger below the older members (Prob. 81). Unlike the other Jews, the Therapeutae allow women members, specially qualified and trained, into their congregations, albeit hidden behind a screen high enough for decency, but low enough to hear the exposition (Contempl. 32– 33).70 This means that Philo is aware of regional variation of the practices in the proseuchai, although, on the whole, his picture is quite uniform. One of those present reads the scriptures, then “one of the most experienced” (τις τῶν ἐμπειροτάτων)—precisely the same term as in Spec. 2.62—interprets the difficult passages (Prob. 82). The only difference is that, according to Philo, the Essenes use allegory for their exegesis (82). Compared with the Therapeutae, the Essenes emphasize ethical teaching more than theoretical contemplation (80). Like the other Jews in their proseuchē meetings, the Essenes teach the range of Greek virtues, in this context Philo uses three headings: “love of God, love of virtue and love of humankind” (ὅροις καὶ κανόσι τριττοῖς χρώμενοι, τῷ τε φιλοθέῳ καὶ φιλαρέτῳ καὶ φιλανθρώπῳ; 83). As proof of the value of their teaching Philo adds an account of the virtue of the Essenes’ lives and some details of their ethics (84–88). Thus, Philo presents the Essenes as more focussed in their practice on Torah observance and equipped with a special hermeneutic for reading the Torah, yet their practices in their synagogues do not seem to be fundamentally different from what other Jews did in their proseuchai. Therefore, based on Philo’s account, it cannot be argued that the Palestinian synagogue and the Diaspora proseuchē were two different institutions.71 Compared with the Essenes, the Therapeutae in their daily lives are less practically focussed. They emphasize contemplation: daily morning and evening prayer and, in between, solitary allegorical bible study (Contempl. 27–29). Their morning and evening prayers contain singing (29). The Sabbath assemblies break the social fast of the week in that they provide opportunity for communal gathering. Like all the other proseuchai meetings, the meetings of the Therapeutae have a seating order, and the Therapeutae even place their hands inside their clothes to prevent their improper use (30).72 Again, “the oldest and most experienced” provides their teaching (ὁ πρεσβύτατος καὶ τῶν δογμάτων ἐμπειρότατος;  68

For a thorough review of the debate and the evidence, see Taylor 2012. On the evidence for the Therapeutae, see Taylor 2013.  On the ancient ideas of propriety, see Kraemer 1992: 106–107, although she later (126) argues from silence against separation. 71 Against the argument of Kimberley Czajkowski in this volume. 72 For the connection of this posture with Sabbath law see Doering 1999: 379–380. 69

Philo and the προσευχή

227

31).73 The object of this teaching is the Torah and the other writings (28–29). The emphasis in the teaching is comprehensibility and the congregation, unlike that of the ‘normal’ Jews do not respond audibly, but only through body language (31). However, there is no reference to Torah reading in the Sabbath meetings.74 The Torah reading is practiced during the week, so that weekly life and Sabbath together form a continuous worship. It is not the reading of scripture alone, which is important, but the reading which is completed by the interpretation of a qualified expert. The Torah only exists with its interpretation; it is the interpretation which perfects the Torah. This concept also applies to the reading practices in the normal proseuchai, as can be seen in Philo’s above-mentioned view that the proseuchē already existed in Moses’ time, but in the separation of reading and exegesis among the Therapeutae Philo takes his view of scripture perfected by exegesis to a new level.75 Unlike for the other proseuchai, Philo even mentions the extent of the Therapeutae’s scriptural range: They read the scriptures (“holy writings”, 28) every day, and they are called “laws, oracles delivered through prophets, psalms and the other books through which knowledge and piety are supported and perfected” (νόμους καὶ λόγια θεσπισθέντα διὰ προφητῶν καὶ ὕμνους καὶ τὰ ἄλλα οἷς ἐπιστήμη καὶ εὐσέβεια συναύξονται καὶ τελειοῦνται; Contempl. 25). Without identifying specific books, this reflects the basic structure of Torah, prophets and writings with the separate reference to the Psalms. However, the Torah retains prominence.76 Like that of the Essenes, the Therapeutae’s exegetical method is allegory, and Philo adds that they use older commentaries (28). Thus, we see that the Jewish sects described by Philo only underline his view of the importance of the synagogue for the teaching of the Torah. They do not provide evidence for additional activities in the groups Philo describes, although the evidence for hymnody in the account of the symposia of the Therapeutae is considerably more explicit.

Contrasting Evidence from Josephus This is not the place for a detailed discussion of Josephus’ account of synagogues;77 however, he quotes one edict particularly, which is relevant for Philo,  73

On the similarities between the descriptions, see McKay 1994: 73–75. As noted by Doering 1999: 380. 75 Philo’s approach resembles the Roman attitude to literature: the reading and discussion of philosophy was regarded as more important than the library holdings of an individual, and the reading was done by a slave, interrupted at times to discuss a certain point, see the contribution of Clemens Leonhard in this volume. This comes very close to what Philo describes in the reading practices of the proseuchē: a separate reader and exegete who expounds on specific issues, with the difference that Philo emphasizes the strict order of the exchange. It is the exegete who decides when to stop and discuss a certain point. 76 See Perrot 1984: 126–128. 77 For this, see Krause 2017. 74

228

Jutta Leonhardt-Balzer

especially as it seems to paint a markedly different picture. It is an imperial decree on behalf of the Jews of Sardis, which allows them the right to meet and to perform their ancestral ‘prayers and sacrifices’ (A.J. 14.260–261): they are allowed to gather according to their legal custom, … [it is commanded] that they are given a place, in which they can assemble with their women and children, to offer the ancestral prayers and sacrifices to God. And the senate and the people have decreed that they are allowed to gather together on the appointed days to act according to their laws, and that such a place should be given to them by the praetors for them to build and inhabit as they see fit, and those who are in charge of the provisions of the city shall be responsible for the import of such food as they see fit for their consumption. ἵνα κατὰ τὰ νομιζόμενα ἔθη συνάγωνται …., δοθῇ τε καὶ τόπος αὐτοῖς, εἰς ὃν συλλεγόμενοι μετὰ γυναικῶν καὶ τέκνων ἐπιτελοῦσιν τὰς πατρίους εὐχὰς καὶ θυσίας τῷ θεῷ· §261 δεδόχθαι τῇ βουλῇ καὶ τῷ δήμῳ συγκεχωρῆσθαι αὐτοῖς συνερχομένοις ἐν ταῖς ἀποδεδειγμέναις μέραις πράσσειν τὰ κατὰ τοὺς αὐτῶν νόμους, ἀφορισθῆναι δ αὐτοῖς καὶ τόπον ὑπὸ τῶν στρατηγῶν εἰς οἰκοδομίαν καὶ οἴκησιν αὐτῶν, ὃν ἂν ὑπολάβωσιν πρὸς τοῦτ ἐπιτήδειον εἶναι, ὅπως τε τοῖς τῆς πόλεως ἀγορανόμοις ἐπιμελὲς ᾖ καὶ τὰ ἐκείνοις πρὸς τροφὴν ἐπιτήδεια ποιεῖν εἰσάγεσθαι.

The place, τόπος, in question is not named, but the decree clearly refers to the local proseuchē. Although it represents the Roman language, this decree gives rise to a number of questions: Were there sacrifices in the synagogues? Were there women and children? Were there meals? Why does it not mention Torah exegesis at all? Firstly: what does the phrase “prayers and sacrifices” mean. As shown above, the evidence from Philo for the existence of synagogal prayer is indirect, except for the name of the proseuchē. Most Philonic texts mentioning communal prayer refer to the Temple (e.g., Spec. 1.193). The prayer of the Therapeutae at their great festival asks for God’s blessing for the feast (Contempl. 66–67). In spite of the scarcity of evidence, it is reasonably certain, that the term “prayer” is not a metaphor but refers to actual prayer in the proseuchai. Does this mean that the Sardis decree provides evidence for the possibility that the Jewish community in Sardis practices sacrifices in their proseuchē, as Andrew Krause argues?78 This suggestion, however, does not have a basis in the text, because Josephus uses pagan language, reflecting the transferral of the status of a temple.79 From the Roman perspective, a sacred place required the performance of sacrifices, and  78 Krause 2017: 80–81, esp. 86–87, 195–196, where he lists as support for his reading that Philo describes a Jewish woman’s “prayers and sacrifices” in Spec. 3.171. However, in that context Philo contrasts general male and female behaviour and describes the rules of modesty for women in general. It is neither clear that the “prayers and sacrifices” refer to specific rites nor that the woman in question is Jewish. Certainly, the evidence from Philo and Josephus is not sufficient to claim that there are “multiple attestations of some form of sacrifice in synagogue institutions”; Krause 2017: 87. 79 See Pucci ben Zeev 1998: 223–225.

Philo and the προσευχή

229

the phrase “prayer and sacrifices” merely indicates the Roman acknowledgment of the practice of valid, officially recognized Jewish worship in this place.80 Furthermore, even in a Jewish context, the specific phrase “prayers and sacrifices” (εὐχαί καὶ θυσίαι) is used by Philo—who is one of the few sources for the Jewish use of this kind of language—for any type of public worship, not specifically for the performance of sacrifices. The passage in Spec. 3.171, which Krause uses as an argument for actual sacrifices, does not specify actual rites, but refers in general to the worship of women outside the house. It is neither clear that the women Philo mentions are Jewish nor that they actually perform any sacrifices at all, because even in the Temple in Jerusalem women did not perform the sacrifices themselves. The phrase merely refers to “worship rites” in general, and in that sense, Philo also uses it to describe the official worship of the High Priest (Mos. 1.149; 2.5; Somn. 1.214–215; Spec. 1.113; Spec. 3.131–132) without reference to clearly defined rites or specific sacrifices.81 There is another theory which supposes that the Passover lamb might have been awarded the status of a sacrifice, and that the decree refers to this rite. Yet there is no evidence to back up this theory either.82 Philo’s reference to the Passover celebration does not mention the proseuchai. While he writes that on the Passover every house “is invested with the appearance and semblance of a temple” (Spec. 2.148), because the lamb is slaughtered, the guests are ritually purified and there are “prayers and hymns”, it is important to note that Philo does not use the term θυσία in this context,83 and he emphasizes elsewhere that there is only one place where God permits sacrifices (Somn. 1.61–67).84 The same sentiment is expressed by Josephus in A.J. 13.65–68, where the Temple in Leontopolis is called “contrary” to what is proper. Archaeological evidence of incense holders has only been found in later synagogues, and whatever the Magdala stone may mean,85 it certainly does not provide evidence for sacrifices in the synagogue there. Therefore, there is no evidence to indicate that the term “prayers and sacrifices” in the Sardis decree refers to anything other than Jewish worship in general, which is afforded protection by the Roman state.  80 Krause 2017: 86 mentions in this context the use of the term θυσία by the Roman Dolabella in a letter to the Ephesian Jews (A.J. 14.260), however, the letter does not refer to “sacrifices”: In a list of Jewish activities which Dolabella permits and for the sake of which he exempts Jews from military duty, he lists among food and Sabbath observances the things relating to “the tributes/deductions for the sacrifices” (τῶν πρὸς τὰς θυσίας ἀφαιρεμάτων), i.e., the Temple dues, which were indeed also collected in the Diaspora. 81 Leonhardt 2001b. 82 Cf. Leonhardt-Balzer 2011. 83 Against Sanders 1992: 133–134. Criticism of Sanders in Mendelson 1994: 163–165. 84 Philo does not contradict this principle when he calls all creation a sanctuary in Spec. 1.66–67, because in this passage he immediately specifies that matching the spiritual worship of the mind and the angels there can only ever be one physical Temple, against Krause 2017: 87. 85 On the fascinating find in Magdala and its possible meaning(s), see the contributions of Lutz Doering and Judith Newman in this volume.

230

Jutta Leonhardt-Balzer

The reference to women and children in the assemblies is considerably more specific, yet it does not indicate that they needed to be present throughout the whole meeting. While it does not contradict Philo’s evidence, it does not add much to it either. In one sense, however, the decree provides a new perspective on Jewish practices in the proseuchai. The command to provide proper food for the Jewish assemblies implies the existence of synagogal meals.86 It might be conceivable that the officials in Sardis are merely charged with providing kosher food for the Jewish congregation’s daily use, and this would explain Philo’s silence about synagogal meals except in the context of the great festival of the Therapeutae (Contempl. 40–89). However, the decree’s reference to food occurs in the same sentence as the permission to assemble on their festive days, which suggests that the meals occurred in the context of the assemblies in the proseuchē,87 and the reference to the Therapeutae’s special feast, far from disproving the practice of synagogal meals, demonstrates that Jews could celebrate their festivals with special symposia, and the most natural place for this would be the proseuchē.88 What remains unclear, however, is whether these meals occurred in the regular Sabbath service, which according to Philo ended in the early evening, too early for dinner. So, either there was some regional variety in the frequency of meals in the proseuchai, or, more likely, they took place on the festivals, just as those of Philo’s Therapeutae. As for the question of why the edict does not mention Torah exegesis, this, once more, is due to its Roman provenance. For the Romans it did not matter which precise practices the worship in the synagogue entailed, merely that its activities were legal. What the decree does refer to repeatedly, however, are the Jewish laws and customs, which is a recognition of the importance of the Torah for the Jewish community. Altogether, the decree from Sardis shows that Philo’s picture of the proseuchai is focussed on the teaching side, while other activities are only hinted at, but the Sardis picture does not fundamentally contradict Philo’s evidence.

Conclusion: What Were They Actually Doing in the Synagogues? The recognition that Philo presents his own view on the proseuchai does not mean that it is impossible to draw any conclusions about the historical practices. The most prominent aspect of the synagogue activities for Philo was the teaching  86

Cf. Schürer 1973–1987: 3.1, 144–145. This is also noted by Krause 2017: 80–81. 88 On meals in Philo and Josephus, especially in the synagogues and among the Therapeutae, see Leonhardt-Balzer 2017a. 87

Philo and the προσευχή

231

of the Torah on the Sabbath.89 Philo presupposes that this is the activity that was practised everywhere. All other activities are only mentioned indirectly. This does not mean that they did not exist. However, for Philo, the teaching aspect is so crucial for the preservation of the Jewish way of life in the Diaspora, that the only alternative term for the proseuchai he uses is “school”.90 This is the background against which he reads them and therefore Philo consistently describes the order of the meetings, the discipline in the exchange, and the quality of the instruction. All this provides optimal conditions for the congregation to hear and understand the meaning of the text. In this picture, women and children do not seem to have been part of this part of the meetings, because, for Philo and his contemporary intellectuals, they only provide distractions. Based on Philo’s own exegetical practice and his conviction of the divine inspiration of the Greek translation (Mos. 2.25–45), the biblical text used was the Septuagint. The method of exegesis among the Essenes and Therapeutae was allegory, but if at all it was only practiced along with the literal exegesis among the other Jews, because allegory could not be understood by everyone, only by those highly trained in philosophy (Migr. 89–92). In whichever form, the interpretation of the Torah constitutes the core, the common denominator of all the proseuchai.91 Philo hints at a liturgy, but there is no evidence for a reading cycle—possibly because the practices varied regionally. Although it has been doubted that there was prayer in the proseuchai, the term itself and Philo’s reference to the thanksgiving to benefactors implies the existence of regular prayers. There is no evidence for the public recitation of the shema‘ from Second Temple sources.92 Neither is there clear evidence for a pre70 use of the Eighteen Benedictions.93 Although later prayers certainly may have their roots in Second Temple practices, this does not necessarily prove that these roots were in the context of the proseuchē. The Qumran texts give evidence of a liturgical use of psalms and of a fixed liturgy outside the Temple,94 but there is no  89 Against Safrai 1977: 88 and Salzmann 1994: 456, who doubt that there was a sermon in the synagogue service. 90 Thus, Sandmel 1978: 147 calls the ancient synagogue a “school for adults” and sees this aspect as relevant to this day. 91 Cohen 1995: 29–31 suggests a possible influence of the Diaspora on Palestine through guest speakers in the synagogue. 92 Cf. Sanders 1992: 197. Josephus merely refers to the daily individual, not communal recitation of the shema‘ (A.J. 4.212). Communal recitation is suggested by Manns 1986: 130–140; Salzmann 1994: 455; and Maier 1990: 556–558. 93 Against, e.g., Manns 1986: 49–51, 141–155; Salzmann 1994: 456; Maier 1990: 558–560. On the lack of rabbinic evidence for early synagogal communal prayer, see also Ruth Langer in her contribution to this volume. Even the Theodotus inscription, one of the most detailed inscriptions concerning the synagogue (CIJ II 1404 = CIIP I 9), does not mention liturgical details; cf. Riesner 1995. 94 See Falk 1998, who finds the roots of the practices in the Qumran texts in the Temple (89–92, 119–120) rather than in the synagogue. Independently of their origin, however, in the Qumran context the texts themselves attest to a use of the psalms outside the Temple. Wherever the origins of the rite, the Qumran texts do attest to regular communal prayer (especially 1QS VI 3, 8). For a comparison between the Qumran cult and liturgy and that according to rabbinic sources, see Maier 1990.

232

Jutta Leonhardt-Balzer

proof of the existence of these practices outside the yaad, and certainly not in all proseuchai, at all places. Likewise, Paul’s references to liturgical singing in 1 Cor 14:26 attest to the practice in the Christian communities, and even if they probably took it from Jewish communities before them, this does not necessarily mean that the practice was universal. All this collective evidence suggests, however, that prayer and psalmody seem to have been very diverse in Second Temple times,95 but they are widely attested, and Philo’s evidence, vague as it is, situates prayer and psalmody within the Sabbath activities of the proseuchē. There are several important activities associated with the synagogue that Philo does not mention, though. Thus, he remains completely silent about meals in the context of the ordinary proseuchai, they appear only in his account of the great festival of the Therapeutae. The presence of meals however, creates a strong resemblance to pagan associations,96 therefore it is noteworthy that Philo does not dwell on meals in the ordinary proseuchai, although he can use the term “association” for the synagogue meetings in Somn. 2.127 (τὸν εἰωθότα θίασον ἀγείροντες). Neither does Philo mention the proseuchai as hostels, in the way the Theodotus inscription from Jerusalem suggests (CIJ II 1404 = CIIP I 9).97 A possible reason for this is that in the Diaspora the need for accommodation was not as urgent as in Jerusalem with its regular stream of pilgrims. Philo’s silence about guests in the proseuchē certainly also indicates a lack of interest in this particular function. He does not mention any political or court or other non-religious meetings either, the existence of which Josephus, the Gospels and the epigraphic evidence imply for synagogues in Palestine.98 The practice of using the proseuchai as temporary treasuries for the Temple tax, which Philo does mention, implies a further use as archive, which again is supported by epigraphic evidence. This suggests, once again, that Philo does not mention everything that the proseuchē was used for, and that his evidence cannot be taken as comprehensive, not even for Alexandria. Nevertheless, his picture is valid and sheds some light on the practices of his time, even allowing for regional variation and different interpretations. An activity which neither Philo’s evidence nor any other supports, and which therefore probably did not take place in the proseuchai historically, is sacrifice. This means that while synagogues took on certain activities previously associated with the Temple, such as psalmody, prayer, Torah reading and teaching, and the collection of the Half-Shekel offering, all this does not lead to the conclusion that

 95

Cf. Maier 1990: 543–547. See the contribution of Benedikt Eckhardt to this volume, especially its initial paragraphs. 97 See also Lifshitz 1967: 70–71; McKay 1994: 222–223; and Riesner 1995:192–200; cf. Sanders 1990: 202. 98 Cf. Runesson, Binder, and Olsson 2008: 14–15. 96

Philo and the προσευχή

233

the Diaspora Jews only paid lip service to the Temple.99 To the contrary: all the activities that are taken over from the Temple are those that provide knowledge about the Jewish traditions, the commandments and the Temple in places far distant from Jerusalem. The purpose of the proseuchē therefore was to maintain the link to Jewish tradition in the Diaspora. The purpose of the Temple was different: to perform worship and sacrifices to God and unify all Jews everywhere. Thus, the two were not in competition with each other. Far from any rivalry between the proseuchai and the Temple in Jerusalem,100 the evidence from Philo suggests that the proseuchai served to reinforce the relevance of the Temple all over the Diaspora among people who may never have seen the land of Israel in their lives. The proseuchē thus was the centre of Jewish life abroad. If the Temple can be compared to the heart,101 the Second Temple proseuchai are the blood system, arteries coming from the heart and veins leading back to it—different from the heart but connected, not identical but not in competition either.

Bibliography Billerbeck, P. 1964. “Ein Synagogengottesdienst in Jesu Tagen”. ZNW 55: 143–161. Borgen, P. 1965. Bread from Heaven: An Exegetical Study of the Concept of Manna in the Gospel of John and the Writings of Philo. NTSup 10. Leiden: Brill. Borgen, P. 2001. “Greek Encyclical Education, Philosophy and the Synagogue: Observations from Philo of Alexandria’s Writings”. In: Matsson, O. et al. (eds.), Libens Merito: FS S. Strömholm. Uppsala: Kungl. Vetenskapssamhället, 61–71. Borgen, P., Fuglseth, K., and Skarsten, R. 2000. The Philo Index: A Complete Greek Word Index to the Writings of Philo of Alexandria. Leiden: Brill. Cerfaux, L. and Tondriau, J. 1957. Le culte des souverains dans la civilisation gréco-romaine. Bibiothèque de Théologie, Sér. III. 5. Tournai: Desclée & Co. Cohen, N. G. 1995. Philo Judaeus: His Universe of Discourse. Frankfurt am Main: Peter Lang. Cohen, N. G. 1997. “Earliest Evidence of the Haftarah Cycle for the Sabbaths between ‫י׳׳ז‬ ‫ בתמוז‬and ‫ סוכות‬in Philo”. JJS 48: 225–249. Cohn, L. and Wendland, P. 1896–1930. Philonis Alexandrini opera quae supersunt. 7 vols. Berlin: Reimer. Doering, L. 1999. Schabbat: Sabbathalacha und -praxis im antiken Judentum und Urchristentum. TSAJ 78. Tübingen: Mohr Siebeck. Elbogen, I. 1993. Jewish Liturgy: A Comprehensive History. Trans. R. P. Scheindlin. Philadelphia: Jewish Publication Society. Feldman, L. H. 1996: “Philo’s Views on Music”. In: idem (ed.), Studies in Hellenistic Judaism. Leiden: Brill, 504–528.

 99 Against Martin 2000. Runesson 2001, argues for different origins of the synagogue: in Israel the public reading of the Torah, in the Diaspora, however, from temple spaces. In view of the Philonic emphasis on reading in the Diaspora and the fact that he never identifies the proseuchai as temples, this theory does not match the ancient sources. 100 Leonhardt 2007: 29–53. See also Lutz Doering in this volume on the Magdala stone. He does not see it as evidence of a competition to the Temple either. 101 For this metaphor, see Leonhardt-Balzer, 2011: 146–147.

234

Jutta Leonhardt-Balzer

Falk, D. K. 1998. Daily, Sabbath, and Festival Prayers in the Dead Sea Scrolls. STDJ 27. Leiden: Brill. Fiore, B. 1990: “The Sage in Select Hellenistic and Roman Literary Genres (Philosophic Epistles, Political Discourses, History, Comedy and Romances)”. In: Gammie, J. G. and Perdue, L. G. (eds.), The Sage in Israel and the Ancient Near East. Winona Lake, ID: Eisenbrauns, 328–341. Gordon, R. 1990. “The Veil of Power: Emperors, Sacrificers and Benefactors”. In: Beard, M. and North, J. (eds.), Pagan Priests. London: Gerald Duckworth & Co., 201–231. Gutmann, J. 1975. The Synagogue: Studies in Origin, Archaeology and Architecture. New York: Ktav. Hatch, E., Redpath, H. A. et al. 1998. A Concordance to the Septuagint and the Other Greek Versions of the Old Testament: Including the Apocryphal Books. 2nd ed. Grand Rapids, MI: Baker Academic. Hegermann, H. 1989. “The Diaspora in the Hellenistic Age”. In: Davies, W. D., Finkelstein, L., and Sturdy, J. (eds.), The Cambridge History of Judaism, Vol. 2: The Hellenistic Age. Cambridge: Cambridge University Press, 115–166. Hengel, M. 1971. “Proseuche und Synagoge: Jüdische Gemeinde, Gotteshaus und Gottesdienst in der Diaspora und in Palästina”. In: Jeremias, G., Kuhn, H.-W., and Stegemann, H. (eds.), Tradition und Glaube: Das frühe Christentum und seine Umwelt. Festgabe für Karl Georg Kuhn. Göttingen: Vandenhoeck & Ruprecht, 157–184. Horbury, W. 1986. “Ezekiel Tragicus 106: δωρήματα”. VT 36: 37–51. Horbury, W. 1998a. “Early Christians on Synagogue Prayer and Imprecation”. In: Stanton, G. N. and Stroumsa, G. (eds.), Tolerance and Intolerance in Judaism and Christianity. Cambridge: Cambridge University Press, 296–317. Horbury, W. 1998b. Jewish Messianism and the Cult of Christ. London: SCM. Horbury, W. 1999. “Women in the Synagogue”. In: Davies, W. D., Horbury, W. et al. (eds.), The Cambridge History of Judaism, Vol. 3: The Early Roman Period. Cambridge: Cambridge University Press, 358–401. Juster, J. 1914. Les Juifs dans l’Empire Romain: Leur condition juridique, économique et sociale. 2 vols. Paris: Geuthner. Kaiser, O. 2013. “Das Gebet bei Philo von Alexandria”. In: Grund, A., Krüger. A., and Lippke, F. (eds.), Ich will dir danken unter den Völkern: Studien zur israelitischen und altorientalen Gebetsliteratur. FS für Bernd Janowski zum 70. Geburtstag. Gütersloh: Gütersloher Verlagshaus, 625–644. Kee, H. C. 1990. “The Transformation of the Synagogue After 70 CE: Its Import for Early Christianity”. NTS 36: 1–24. Kee, H. C. 1994. “The Changing Meaning of Synagogue: A Response to Richard Oster”. NTS 40: 281–283. Kee, H. C. 1995. “Defining the First Century Synagogue: Problems and Progress”. NTS 41: 481–500. Kraabel, A. T. 1992a. “Synagoga Caeca: Systematic Distortion in Gentile Interpretations of Evidence for Judaism in the Early Christian Period”. In: Overman, J. A. and MacLennan, R. S. (eds.), Diaspora Jews and Judaism: Essays in Honor of and in Dialogue with A. Thomas Kraabel. Atlanta: Scholars Press, 35–62. Kraabel, A. T. 1992b. “Unity and Diversity Among Diaspora Synagogues”. In: Overman, J. A. and MacLennan, R. S. (eds.), Diaspora Jews and Judaism: Essays in Honor of and in Dialogue with A. Thomas Kraabel. Atlanta: Scholars Press, 21–31. Kraemer, R. S. 1992. Her Share of the Blessings: Women’s Religions among Pagans, Jews and Christians in the Greco-Roman World. Oxford: Oxford University Press. Krause, A. R. 2017. Synagogues in the Works of Flavius Josephus: Rhetoric, Spatiality, and First Century Jewish Institutions. AJEC 97. Leiden: Brill.

Philo and the προσευχή

235

Leonhardt, J. 2001a. Jewish Worship in Philo of Alexandria. TSAJ 84. Tübingen: Mohr Siebeck. Leonhardt, J. 2001b. “Εὐχαὶ καὶ θυσίαι (A 14:260): Opfer in der jüdischen Synagoge von Sardes?”. In: Kalms, J. U. (ed.), Internationales Josephus-Kolloquium Amsterdam 2000. Münsteraner Judaistische Studien 10. Münster: LIT, 189–203. Leonhardt-Balzer, J. 2007. “Jewish Worship and Universal Identity in Philo of Alexandria”. In: Frey, J., Schwartz, D. R., and Gripentrog, S. (eds.), Jewish Identity in the Greco-Roman World / Jüdische Identität in der griechisch-römischen Welt. AJEC 71. Leiden: Brill, 29– 53. Leonhardt-Balzer, J. 2011. “Priests and Priesthood in Philo”. In: Schwartz, D. R. and Weiss, Z., with R. A. Clements (eds.), Was 70 CE a Watershed in Jewish History? On Jews and Judaism Before and After the Destruction of the Second Temple. AJEC 78. Leiden: Brill, 121–147. Leonhardt-Balzer, J. 2017a. “Die Funktion des Mahles für die Gemeinschaft bei Philo und Josephus”. In: Sänger, D. and Hellholm, D. (eds.), The Eucharist – Its Origins, and Contexts: Sacred Meal, Communal Meal, Table Fellowship in Late Antiquity, Early Judaism, and Early Christianity. 3 vols. Tübingen: Mohr Siebeck, 1.253–273. Leonhardt-Balzer, J. 2017b. “Diaspora Jewish Attitudes to Metropoleis: Philo and Paul on Balanced Personalities, Split Loyalties, Jerusalem, and Rome”. In: Walton, S., Trebilco, P. and Gill, D. W. J. (eds.), The Urban World and the First Christians. Grand Rapids, MI: Eerdmans, 86–98. Levine, L. I. 1987. The Synagogue in Late Antiquity. Philadelphia: The American Schools of Oriental Research. Levine, L. I. 2005. The Ancient Synagogue: The First Thousand Years. 2nd ed. New Haven, CT: Yale University Press. Levinskaya, I. 1996. The Book of Acts in its Diaspora Setting. BAFCS 5. Grand Rapids, MI: Eerdmans. Liddell, H. G. and Scott, R. 1996. A Greek-English Lexicon. Rev. and augmented by H. S. Jones et al. with new Supplement. Oxford: Clarendon Press. Lifshitz, B. 1967. Donateurs et fondateurs dans les synagogues juives: Répertoire des dédicaces relatives à la construction et à la réfection des synagogues. Paris: Gabalda. Maier, J. 1990. “Zu Kult und Liturgie der Qumrangemeinde”. RevQ 56: 543–586 [= García Martínez, F. (ed.), The Texts of Qumran and the History of the Community: Proceedings of the Groningen Congress on the Dead Sea Scrolls (20–23 August 1989), Vol. 3: The History of the Community]. Mann, J. 1971. “The Observance of the Sabbath and the Festivals in the First Two Centuries of the Current Era according to Philo, Josephus, the New Testament and the Rabbinic Sources”. In: Mann, J. (ed.), The Collected Articles, Vol. 1. Jerusalem: M. Shalom, 1–59. Manns, F. 1986. La prière d’Israël à l’heure de Jésus. ASBF 22. Jerusalem: Franciscan Printing Press. Martin, M. J. 2000a. “Philo’s Interest in the Synagogue”. ANES 37: 215–223. Martin, M. J. 2000b. “The School of Virtue and the Tent of Zion: An Investigation into the Relationship between the institutions of the Graeco-Roman Diaspora Synagogue and the Jerusalem Temple in Late Second Temple Judaism. Philo of Alexandria – a Case Study”. PhD thesis, University of Melbourne. McKay, H. A. 1994. Sabbath and Synagogue: The Question of Sabbath Worship in Ancient Judaism. RGRW 122. Leiden: Brill. Mélèze Modrzejewski, J. 1997. The Jews of Egypt: From Rameses II to Emperor Hadrian. Trans. R. Cornman, foreword by S. J. D. Cohen. 2nd ed. Princeton: Princeton University Press.

236

Jutta Leonhardt-Balzer

Mendelson, A. 1994. “‘Did Philo Say the Shema?’ and Other Reflections on E. P. Sanders’ Judaism: Practice and Belief”. SPhA 6: 160–170. Neyrey, J. H. 2005. “God, Benefactor and Patron: The Major Cultural Model for Interpreting the Deity in Greco-Roman Antiquity”. JSNT 27: 465–492. Nock, A. D. 1972. “Soter and Euergetes”. In: Stewart, Z. (ed.), Arthur Darby Nock: Essays on Religion and the Ancient World. 2 vols. Cambridge, MA: Harvard University Press, 720– 735. Oesterley, W. O. E. 1941. The Jews and Judaism during the Greek Period: The Background of Christianity. London: SPCK. Oster, R. E. 1993. “Supposed Anachronisms in Luke-Acts’ Use of συναγωγή: A Rejoinder to H. C. Kee”. NTS 39: 178–208. Perrot, C. 1984. “La lecture de la bible dans la diaspora hellénistique”. In : Kuntzmann, R. and Schlosser, J. (eds.), Études sur le judaïsme hellénistique: Congrès de Strasbourg (1983). LD 119. Paris: Cerf, 109–132. Pucci Ben Zeev, M. 1998. Jewish Rights in the Roman World: The Greek and Roman Documents Quoted by Flavius Josephus. TSAJ 74. Tübingen: Mohr Siebeck. Rajak, T. 1996. “Benefactors in the Greco-Roman Diaspora”. In: Cancik, H., Lichtenberger, H., and Schäfer, P. (eds.), Geschichte – Tradition – Reflexion: Festschrift für Martin Hengel zum 70. Geburtstag, Bd. I: Judentum. Tübingen: Mohr Siebeck, 305–319. Riesner, R. 1995. “Synagogues in Jerusalem”. In: Bauckham, R. (ed.), Palestinian Setting. BAFCS 4. Grand Rapids, MI: Eerdmans, 179–211. Runesson, A. 2001. The Origins of the Synagogue: A Socio-Historical Study. ConBNT 37. Stockholm: Almqvist & Wiksell. Runesson, A., Binder, D. D., and Olsson, B. 2008. The Ancient Synagogue from its Origins to 200 C.E.: A Source Book. AJEC 72. Leiden: Brill. Safrai, S. 1977. “The Synagogue and Its Worship”. In: Avi-Yonah, M. and Baras, Z. (eds.), Society and Religion in the Second Temple Period. WHJP: First Series: Ancient Times 8. Jerusalem: Jewish History Publications and Massada Publications, 65–98. Salzmann, J. C. 1994. Lehren und Ermahnen: Zur Geschichte des christlichen Wortgottesdienstes in den ersten drei Jahrhunderten. WUNT 2/59. Tübingen: Mohr Siebeck. Sanders, E. P. 1990. Jewish Law from Jesus to the Mishnah. London: SCM. Sanders, E. P. 1992. Judaism: Practice and Belief, 63 BCE–55 CE. London: SCM. Sandmel, S. 1978. Judaism and Christian Beginnings. New York: Oxford University Press. Schrage, W. 1990. “συναγωγή”. TWNT 7: 798–840. Sarason, R. S. 2003. “Communal Prayer at Qumran and Among the Rabbis: Certainties and Uncertainties”. In: Chazon, E. G., Clements, R. A., and Pinnick, A. (eds.), Liturgical Perspectives: Prayer and Poetry in Light of the Dead Sea Scrolls. Proceedings of the Fifth International Symposium of the Orion Center for the Study of the Dead Sea Scrolls and Associated Literature, 19–23 January, 2000. STDJ 48. Leiden: Brill, 151–172. Schröder, B. 1996. Die “väterlichen Gesetze”: Flavius Josephus als Vermittler von Halachah an Griechen und Römer. TSAJ 53. Tübingen: Mohr Siebeck. Schürer, E. 1973–1987. The History of the Jewish People in the Age of Jesus Christ (175 BC– AD 135). Rev. and ed. by G. Vermes et al. 3 vols. Edinburgh: T&T Clark. Schwartz, D. R. 1996. “Temple or City: What Did Hellenistic Jews See in Jerusalem?” In: Poorthuis, M. and Safrai, C. (eds.), The Centrality of Jerusalem: Historical Perspectives. Kampen: Peeters, 114–127. Taylor, J. E. 2003. Jewish Women Philosophers of First-Century Alexandria – Philo’s ‘Therapeutae’ Reconsidered. Oxford: Oxford University Press. Taylor, J. E. 2012. The Essenes, the Scrolls, and the Dead Sea. Oxford: Oxford University Press.

Philo and the προσευχή

237

Thyen, H. 1955. Der Stil der Jüdisch-Hellenistischen Homilie. FRLANT 65, N.S. 47. Göttingen: Vandenhoeck & Ruprecht. Van der Horst, P. 1999. “Was the Synagogue a Place of Sabbath Worship Before 70 CE?”. In: Fine, S. (ed.), Jews, Christians and Polytheists in the Ancient Synagogue: Cultural Interaction during the Greco-Roman Period. London: Routledge, 18–43. Veyne, P. 1990. Bread and Circuses: Historical Sociology and Political Pluralism. Abridged English trans. by B. Pearce. London: Penguin. Winston, D. 1990. “The Sage as Mystic in the Wisdom of Solomon”. In: Gammie, J. G. and Perdue, L. G. (eds.), The Sage in Israel and the Ancient Near East. Winona Lake, IN: Eisenbrauns, 383–397. Winter, B. W. 1994. Seek the Welfare of the City: Christians as Benefactors and Citizens. Grand Rapids, MI: Eerdmans.

In Search of the Petichah: Some Thoughts on the Torah, the Prophets, and the Scriptures in the Synagogues and Beyond

Hermut Löhr

I. Introduction The observations and suggestions made in this article are offered in order to instigate further reflection on two distinct, enduring questions which have been discussed in scholarship relating to ancient Judaism and emergent Christianity: a) What was the regular, ritual or other, function of the early Jewish synagogues, especially in the period before 70 CE? b) What can we say about the emergence of the twofold or threefold collection (or: canon)1 of sacred scriptures in ancient Judaism and emerging Christianity? The combination of these two questions can be collapsed into a single, more directed question: c) Has the development of the twofold or threefold structure of the Hebrew Scriptures, as we know it from tradition, anything to do with the use of sacred scriptures in the synagogues, which may be understood both as an institution and a space—viz., a building? While it might seem plausible to assume such a connection from the outset, we should hesitate to consider such a link as a matter of fact. Firstly, the assumption that textual features and structures can be traced back to practical and repeated use in the life of an institution or a community, while flourishing in the high times of classical form criticism, is not so well accepted in current scholarship. In this regard, I would opt for a middle position: While both the inductive and the deductive approaches of classical form criticism, especially in biblical studies, were far too optimistic in isolating oral and/or ritual textual traditions even in literary texts, and in linking them to quite specific social settings—the famous Sitz im Leben—a complete denial of a connection between written texts and ritual practice would be an exaggeration, too. This is because  1

It is beyond the aims of this paper to probe the complex literary, cultural, historical, and philosophical, implications and explications of the notion or concept of canon in antiquity. The literature on this field of research is rich, cf., e.g., Auwers and de Jonge 2003; Finkelberg and Stroumsa 2003; McDonald and Sanders 2002; Norelli 2004; Theobald 1990.

240

Hermut Löhr

such a position would not take into account the little we know of the ritual life of Judaism and emergent Christianity in antiquity. And, more generally, such a denial would pass over the obvious: the importance of the performative dimension of texts from antiquity in general. Secondly, the whence and when of the development of the structure of the Tanakh is a matter of longstanding dispute, with advocates for both early and later datings. The difference between these datings span two centuries or so. Part of the dispute relates to a few passages from Jewish literature from the period in question and their implications for the issue in focus. Two important questions have to be asked: Do some of those texts, which can be attributed to quite precise periods from the second century BCE to the first half of the second century CE, really refer to the threefold structure of the canon as we know it from rabbinic texts?2 And, equally important but perhaps less discussed than the first question: What do the passages in view reveal of the use either of the twofold or the threefold canon? Do they help us to understand how, when, and where the Tanakh was read in the Second Temple period and beyond? I will come back to this issue below. Thirdly, the function and practice of the synagogue of the first centuries and before the year 70 CE or so is far from clear in the relevant sources; its description demands detailed scrutiny and cautious conclusions. We cannot assume from the outset that the synagogue was primarily a place of worship, in analogy to later Christian churches, with singing, prayer, reading from the Holy Scriptures, preaching, and blessing (minus a sacrificial ritual like the early Christian eucharist, of course). It could be entirely different, and it could very well be variable according to time and location of the synagogues of which we are speaking. So while it would be a (primarily Christian) prejudice to understand the rituals of Judaism and emergent Christianity in antiquity uncritically against the background of the liturgical order of the later Christian Church, it is nonetheless true that, at least in its beginnings, the ritual tradition of Christianity was primarily moulded on the model of Jewish rituals of worship. This implies, I am convinced, that we are entitled to look for analogies, and even to use categories from later Christian tradition heuristically to elucidate what they did in the synagogues, provided we are as sensitive with regard to discrepancies and discontinuities as with regard to analogies and filiation. Evidence for the function of and practice in the synagogues is, as it has been stressed many times in scholarship, rare in pre-70 CE Judaism, and it is more indirect than direct in character. Thus, it is certainly relevant to pay attention to the designations used for what we may call a synagogue,3 that is, most prominently, προσευχή and συναγωγή, but also other terms such as the σαββατεῖον  2 For rabbinic attestation see Hengel 1994: 10 n. 37. The threefold structure is explicitly mentioned for the first time in rabbinic literature in b. Sanh. 90b (baraita). 3 A useful overview is provided by Claußen 2002: 113–150.

In Search of the Petichah

241

mentioned once in a decree by Octavian Augustus and cited by Flavius Josephus.4 But do they refer to the unique function of the synagogue? Or are they more metonymic in character? Not to mention the fact that these designations are, at their best, a short-hand expression of functions and rituals while not revealing any details. It should be noted, in my opinion, that among the terminology known so far from inscriptions or literary texts, only the designation as διδασκαλεῖον, which can be found in the works of Philo,5 may relate to reading and explaining the sacred scriptures. Thus, Philo, Decal. 40 speaks of the ‘school of the sacred laws’, which potentially has the synagogue in view. And in Mos. 2.215–216, he equates προσευκτήρια and διδασκαλεῖα. There are only a few texts which may be understood as direct descriptions— more precisely, as random sketches—of what people did in the synagogues of the first centuries of their existence. For the present purpose I would like to mention only two prominent and quite distinct examples: a) In the well-known and often cited Theodotus inscription (CIJ II 1404 = CIIP I 9) which was found more than a hundred years ago in Jerusalem and which may very well go back to the Second Temple period, the purpose of the building which was financed by the priest and ἀρχισυνάγωγος Theodotus is said to function “for the reading of the Law and the teaching of the commandments”.6 It cannot be stated with certainty whether the context of reading and teaching is regular worship or not. Thus, it is very well possible that the building was meant for both study and worship, and both of them communal and individual. In other words: The later functional distinction between beit ha-midrash and beit ha-kneset may not be pertinent at all for Second Temple Judaism. We are reminded here, inter alia, of passages in Philo’s works according to which the purpose of the (Sabbath) gatherings in the synagogue is primarily communal philosophical instruction,7 or reading and explaining the sacred laws,8 or even those of Ben Sira. b) The writings that would later comprise the New Testament canon are another vital source for our knowledge of Second Temple Judaism. While their overall stance towards Jews who were not Jesus-followers should guard us against hastily accepting the picture they provide as historical fact, nevertheless the texts do contain important information about Jewish life and institutions of the time. This is true especially for the history of the Pharisaic movement, but also with regard to the synagogues in the Land of Israel and in the Jewish Diaspora before 70 CE. In this perspective it is interesting to note that our earliest extant wit  4

A.J. 16.164. Cf. Spec. 2.61–62; Mos. 2.215–216; s. also Decal. 40; Praem. 66. 6 εἰς ἀν[άγν]ω||σ[ιν] νόμου καὶ εἰς [δ]ιδαχ[ὴ]ν ἐντολῶν. – For an extensive treatment of the inscription, including transcription, English translation, and photograph, cf. Hachlili 2013: 523–526. 7 Legat. 156. 8 Hypoth. [= Eusebius, Praep. ev. 8.] 7.12–13). Other passages include Prob. 82; Somn. 2.127; Spec. 2.62. 5

242

Hermut Löhr

nesses, the surviving letters of Paul of Tarsus, do not explicitly mention the synagogue as an institution, or synagogue buildings, for that matter. Nevertheless, they may very well reveal some elements of synagogue practice, such as the forty strikes minus one (2 Cor 11:24; Deut 25:3) which we know from Mishna Makkot to have been part of synagogue discipline. And the reference to (communal) reading (ἀνάγνωσις) of the “old covenant” (probably the five books of the Torah are in view) in 2 Cor 3:14 is best understood, I would argue, as a reference to a contemporaneous synagogue practice well known to Paul. If this is indeed the case, then other aspects of the ritual and life of the Second Temple synagogue inside or outside the Land of Israel may also be echoed or alluded to in the Pauline letters. Synagogues and synagogue life are then explicitly and repeatedly mentioned in the two writings attributed to one of Paul’s disciples, namely the Gospel of Luke and the Acts of the Apostles. For example, Luke 4:16–30 narrates Jesus’ visit to the synagogue in his hometown Nazareth and his reading from the book of Isaiah, followed by a short and spontaneous address to the other participants in the Sabbath service. Interestingly, this pericope states that visiting the synagogue on the Sabbath day was Jesus’ custom, thus presupposing (but not necessarily proving, of course) a regular Sabbath service in the synagogues and probably also the existence of synagogues throughout the region of Galilee in the first half of the first century CE. While Luke 4 presents Jesus reading from a prophet scroll, thus testifying to the custom of haftarah (but without mentioning a Torah reading),9 and perhaps a short ritual connected to it, according to Acts 13:15, in the Diaspora synagogue of Antioch in Pisidia, “reading from the Law and the Prophets” on the Sabbath day is a regular custom. The synagogue leaders (οἱ ἀρχισυνάγωγοι) then invite the visitors from abroad, with Paul among them, to deliver a “word of consolation” (λόγος παρακλήσεως).10 Is that meant as a usual practice? Or is this invitation intended to honour a highly regarded guest? We can only speculate. Of course, these passages were composed in the last decades of the first or even the first decades of the second century CE, and their primary historical value is to testify for the first century synagogue in the Land of Israel and the Diaspora as imagined in the time after 70 CE. It would be difficult, however, to make plausible the assumption that this historical colouring is a mere anachronism. An interpretatio in optimam partem would favour the historical value of the information given for the time of Jesus. c) Other, more indirect indications for the practice of the early synagogues include textual witnesses, features, and artifacts, which, inter alia, suggest Sabbath or daily prayers and lectionary cycles or which could be understood as reading marks for ritual pericope readings. As the manuscripts found in the caves  9

For a possible interpretation, see Perrot 1973. Cf. Heb 13:22.

10

In Search of the Petichah

243

near Qumran are especially rich in this regard, the question arises whether the community or group which produced or copied these texts practiced regular synagogue meetings. d) The archaeology and architecture of early synagogues can also help us to shed light on the synagogue practice, be it in the Land of Israel or in the Diaspora. Features like the location or the orientation of the building, the existence of side rooms or—possibly ritual—baths, decoration, steps, benches etc. ask for functional interpretation. The discovery of manuscript fragments in rooms possibly pertaining to a synagogue is even more revealing, although rather exceptional. Thus, the fact that under the floor of a small side room of the synagogue of Masada scroll fragments with texts from the last chapters of Deuteronomy and from Ezekiel 37 were found, may be indicative of the function of the building. Yigael Yadin thus identified the respective location as the genizah of the Masada synagogue. If this is accepted, then the use of—that is, probably, the public reading from—the Torah and the Prophets might have been part of the practice applied (i.e., the ritual performed) in that synagogue in the time before 73 CE, when Masada was taken by Roman troops.11 Others, like Fine, have suggested that the room functioned not only as a genizah but also a storage room for scrolls in use.12

II. Structures of Scripture We have already mentioned allusions to a threefold structure of the collection of authoritative scriptures in ancient Judaism which clearly antedate the canonization of the New Testament writings. The earliest example known so far is, as is well known, the Greek prologue to the Book of Ben Sira, a passage which was composed in the second half of the second century BCE. Its first line refers to the “Law and the Prophets and the others following them”,13 with “the others” (Greek: τῶν ἄλλων) probably being an abbreviated expression for “the other writings” (or: authors) in analogy to the Law and the Prophets. Understood in this way, the passage is an attestation of a threefold structure of the Scriptures. This view is corroborated by a second passage in the prologue, lines 7–10, which speak of the “reading of Law, the Prophets, and the other books of the Fathers.”14 While “the Law and the Prophets” are quite precise designations, the terminology is more vague with regard to character and contents of the other writings. One may ask whether this is an indication of an  11

Cf. Yadin 1966: 180–192. Cf. Fine 1997: 30. 13 Sir. Prol. l. 1: διὰ τοῦ νόμου καὶ τῶν προφητῶν καὶ τῶν ἄλλων τῶν κατ᾿ αὐτοὺς ἠκολουθηκότων. 14 Sir. Prol. ll. 7–10: εἴς τε τὴν τοῦ νόμου καὶ τῶν προφητῶν καὶ τῶν ἄλλων πατρίων βιβλίων ἀνάγνωσιν. 12

244

Hermut Löhr

open canon and of a gradual development of the threefold structure which is not yet fixed by the time of Ben Sira’s grandson. For our purpose, it is more relevant to notice that the scriptures mentioned come in focus both as a collective heritage and as the object of intensive private lecture and study. In a third passage,15 the prologue deals with the issue of original language and translation, and it should be noticed that here, in line 26, the text does not use ἀνάγνωσις (as in line 10), which probably means public reading, but the verb λέγειν. One may ask whether this small difference is indicative of a variety of ways of performance or reception of the scriptures. At the same time the phrase τὰ λοιπὰ τῶν βιβλίων used here is, I would suggest, more in favour of the notion of a fixed collection of writings than of an open and fluid one. Another possible, albeit contested16 early example of the threefold structure of the collection of biblical writings was identified in 4QMMT, a text, which may originate in the same period as the prologue to Ben Sira, but which was also attributed to later Hasmonean times. The exact terminology differs from that used in Ben Sira, but again it potentially refers to a threefold structure of authoritative writings, namely “the book of Moses, the books of the Prophets, and the psalms of David” (4Q397 = 4QMMTd col. iv frag. 14–21, line 10).17 In line 15 of the same fragment, however, only the Book of Moses and the Books of the Prophets seem to be mentioned (the line is preserved only in parts). The context in line 10 refers to the understanding of these scriptures, an understanding which is provided by the argument of the text in hand and its explicit author himself. At the same time these and other fragments of 4QMMT adduce biblical quotations with the formula “it is written” (‫)כתוב‬. Other indications concerning the performance or reception of these writings cannot be identified in the text. The possible juxtaposition of the threefold structure with a twofold one, which is paralleled in later texts, within a few lines of 4QMMT asks for an explanation. Certainly, speaking just of “the Law and the Prophets” could be an abbreviated expression, but this explanation is, in my eyes, not satisfactory. The fact that the third part of the threefold structure, the “writings”, is not mentioned, cannot be explained either by assuming that a third part of scripture was not yet heard of. Thus, one might hypothesize that the parallel or contemporaneous mentioning of threefold and twofold structures of the scriptures are not an indication of stages of the development in the history of canonization, but more of different accents or perspectives on the Scriptures which can co-exist, the one juxtaposing the Torah and the Prophets (and possibly also other writings), while the other makes a distinction within—but at the same time stressing all three parts of—the collection. While it would be tempting to go further and attribute these different ways of  Sir. Prol. ll. 24–26: ἀλλὰ καὶ αὐτὸς ὁ νόμος καὶ αἱ προφητεῖαι καὶ τὰ λοιπὰ τῶν βιβλίων οὐ μικρὰν ἔχει τὴν διαφορὰν ἐν ἑαυτοῖς λεγόμενα. 16 See Campbell 2000; Kratz 2006; Ulrich 2003. 17 For a transliteration and reconstruction of the passage, see the editio princeps, DJD 10: 27–28. 15

In Search of the Petichah

245

conceptualizing the canon to different social settings and practices of reading or studying, this would be too speculative. Other references to the threefold structure are to be dated to the first century CE. They include Philo’s account of the life of the Therapeutae in Contempl. 25. The context is not about communal worship, but it mentions individual study in private houses. A more extensive reference to the threefold canon, combined with an enumeration of the individual books pertaining to each section, can then be found in Josephus, C. Ap. 1.38–41. The rationale explicitly given here for the structure is a combination of chronological or historical and thematic aspects. The end of the formative epoch is marked by the reign of Artaxerxes. The passage makes no reference to public reading or study of the Bible. The twofold structure of the Law (of Moses) and the Prophets is more often alluded to in Second Temple Literature as well as in the New Testament writings. If the interpretations of the prologue of Ben Sira and of 4QMMT suggested above were wrong, then these passages would figure here as the earliest explicit witnesses. Other examples include 2 Macc 15:9 from the second half of the first century CE (Judas encouraging his comrades with words from the Law and the Prophets), 4 Macc 18:10 (the mother of the seven sons reminding them of Bible lessons with their father), Rom 3:21 as the earliest New Testament passage, as well as several verses from the Gospels of Matthew,18 Luke,19 and John.20 It is also mentioned several times in the Acts of the Apostles.21 The reference to the twofold Bible serves the aims of moral or historical orientation or else it gives authoritative proof for teaching. Only in Acts 13, mentioned above, as well as in Acts 28:23, reference is made to the setting of communal service or instruction. The probable date of some of these texts—as well as the fact that in the Gospel of Luke the threefold and the twofold structure are mentioned alongside each other in chapter 24—corroborate our view that reference to the twofold structure of Law and Prophets cannot be taken as proof for an earlier stage of the history of the Hebrew canon. If these different ways of referring to the Bible are meaningful at all, then they are more plausibly understood as the expression of different perspectives or aspects of Scripture. It is important to note that reference to a twofold or a threefold structure of the Bible should not be confused with a fixation of the exact number and names of texts, contents, or wordings of these parts. According to the data we currently possess, Second Temple Judaism was flexible in these regards. Vice versa, it is a non sequitur to conclude from these variations and differences that the notion of a structured collection of sacred writings must be late. Furthermore, the notion  18

Matt 5:17; 7:12; 22:40. Luke 16:16, 29, 31; 24:27. 20 John 1:45. 21 Acts 13:15; 24:14; 28:23. 19

246

Hermut Löhr

of such a twofold or threefold structure is not necessarily generally accepted and exploited.

III. A Pattern of Interpretation Further insights into the practice of reading the scriptures in early Judaism and emergent Christianity may be gained through a closer look at the different methods of biblical interpretation manifest in the extant sources. While this is a vast area of scholarship past and present, for the time being we are specifically interested in approaches that are based on the twofold or the threefold structure of the sacred scriptures. The most prominent example of this type of interpretation known from Jewish literature from antiquity is certainly the so-called petichah, or “proem”, the artful and playful combination of verses from different parts of the Tanakh, mostly, but not exclusively to be found in haggadic midrashim from the Amoraic Period onward. Scholarship is divided over the question whether the petichah (a modern descriptive designation not found in the sources) is to be understood as the introduction to a more extended homily, or whether the extant petichot may be seen as complete homilies. From a hermeneutical perspective, the petichah constructs a dialogue between a verse taken from the Prophets or, more often, as it seems, from the ketuvim and a verse from the Torah (the so-called seder verse), which is chosen from the pericope which is in the focus of the midrash, or the homily. It should be stressed that petichot do also occur in non-homiletic midrashim, and thus it cannot be corroborated with complete certainty that the petichah originated in the homiletical tradition of the synagogue.22 Unfortunately, the emergence and prehistory of the petichah are unknown. While some scholars would argue in favour of an origin already in tannaitic times, there is no consensus on such a dating. Focusing on literature from the Second Temple period and from emergent Christianity, the general conviction in scholarship is that no analogies or models for the petichah or some of its characteristic features can be identified in the extant sources. Prominent types and examples of exegesis like the series of Torah commentaries by Philo of Alexandria, the pesharim or the thematic commentaries and florilegia from Qumran, or the extant examples of “rewritten Bible” from Second Temple Judaism, leave no place, so it seems, for the systematic application of the interplay between verses or passages from different parts of the canon. Hartwig Thyen, a New Testament scholar who, some decades ago, wrote a small but influential monograph on the “Jewish-Hellenistic homily”, while acknowledging the existence of combinations of scriptural quotations as a feature  22

See, e.g., Stemberger 2008.

In Search of the Petichah

247

for the period in question, explicitly denied the existence of a homiletic technique comparable to the petichah in the literature under scrutiny.23 Texts identified by Folker Siegert as rare examples of Jewish-Hellenistic homilies are not constructed, so it appears, in a way reminiscent of the petichah.24 The picture may have to be nuanced somewhat, e.g., by the analysis of quotations from the latter Prophets in the writings of Philo of Alexandria, as proposed by Naomi Cohen in an article published in 1997 and later expanded in her book, Philo’s Scriptures.25 Cohen concluded from her findings that in the exegetical works of Philo we find the earliest extant witness of the string of haftarot recited later between the 17th of Tammuz and Sukkot. Already in 1923, Henry St. John Thackeray had advocated the somewhat speculative thesis of a haftarah cycle being present even earlier in the Book of Baruch.26 And the combination of quotations and comments on Psalms 1 and 2, Deuteronomy 33 and 2 Samuel 7 in 4QFlorilegium led George Brooke, in his seminal study on Exegesis in Qumran, to argue for a liturgical background of the text, namely the Feast of Tabernacles.27 I am optimistic that future research will bring to light still other examples from Second Temple literature which do not just put together quotations from different texts or parts of the Bible, but which try to stage a dialogue between them to make the scriptures speak in a new way. This is, almost needless to say, not the petichah as we know it from the later midrashim. Nevertheless, the notion of different parts of Scripture—and not just different writings—, as well as the possibility of a dialogue resulting from it, are essential and characteristic preconditions for the development of the petichah form. In fact, clear examples of this technique can be identified in well-known texts from the first century CE, which however, possibly due to their acceptance into the canon of the New Testament, are arguably passed over mostly unnoticed in discussions on the exegesis and use of Scripture or its possible social and ritual setting in Judaism in antiquity. The first two examples are to be found in the writings of the Pharisee and (possibly so) scribe Paul the Apostle. Thus, the interpretation of the figure of Abraham in Gal 3 is built primarily on a dialogue of the passage Gen 15:6 and the prophetic verse Hab 2:4, enriched by repeated reference to Deut 27:26, but also to Lev 18:5, Deut 21:23, Gen 13:15, and 17:8. It is important to note that Paul does not just use a chain of quotations to prove his case, but that he selects and arranges scriptural verses and links them with the help of comments which direct  23

Thyen 1955: 75. Nevertheless Siegert 1992: 29–30 identifies Ps.-Philo, De Jona as a homily on the haftarah on the afternoon of Yom Kippur, but without further proof. 25 Cohen 1997; 2007. – But see the sceptical view of Cohen’s thesis in the contribution by Jutta Leonhardt-Baltzer to the present volume. 26 Thackeray 1921: 80–111. 27 Brooke 1985. 24

248

Hermut Löhr

the argument. A double verbal link exists between the Torah verse and the quotation from the prophets saying, with πιστεύειν and δικαιοσύνη in the Genesis verse, and δίκαιος and πίστις in Habakkuk. This model is taken up and reworked in Romans 4, which starts again from the Torah verse Gen 15:6 and then quotes Ps 31:1+2 LXX. Here the verbal link is the use of the verb λογίζεσθαι in both passages. Other biblical passages may be alluded to in the same chapter, but they are not marked by explicit introductory formula. As I suggested in an article published several years ago,28 more sophisticated examples of this exegesis built on the dialogue of different parts of scripture can be found in the so-called Epistle to the Hebrews, an early Christian text from the end of the first or the beginning of the second century impregnated with quotations and interpretations of texts from the Bible. In fact, the text witnesses a variety of exegetical techniques; I have no doubt that its unknown author must have had some training and experience in the field—even more so than Paul, it appears. My article analysed in detail the method applied in Hebrews 3 and 4, two chapters which contain an argument built on the dialogue between a lengthy quotation from Psalm 95 and the Torah verse Gen 2:2. What I described in exemplary detail for Hebrews 3–4, was identified in fact as a repeated pattern of the Epistle to the Hebrews some years later in Gert Steyn’s important monograph on the Septuagint “Vorlage” of the quotations in Hebrews.29 Thus, the figure of Melchizedek in chapter 7 is interpreted by means of a combination of Gen 14:17–20 and Ps 110:4 and prepared before by the juxtaposition of Ps 110:4 and Ps 2:4, but also by the quotation of Gen 22:17. Chapters 8–10, focusing on the heavenly sanctuary and its sacrificial cult, but also on the new covenant announced in Jer 31, come out as a rather complex combination of different verses from Exod 24 and 25 (24:8; 25:40), Jer 31 and Ps 40. In Heb 10, quotations from Deut 32:35–6 and Isa 26:20 combined with Hab 2:3–4 are interpreted in the perspective of final divine judgement. The warning passage Heb 12:18–29 adduces Deut 9:19 together with Hag 2:6, 21. Even in the paraenetical coda of the text, in Heb 13:1–6, such a combination (here, of Deut 31:6 and Ps 118:6) can be identified. This can be no accident, I think. The pattern is not always as clearly exposed as in Heb 3–4. In some cases, the argument, while not beginning with the Torah verse, seems to aim at its interpretation like in the amoraic petichah. But we cannot say from the textual evidence that stress is always primarily laid on the interpretation of the Torah in comparison to that of passages from other parts of Scripture. In some cases, the exegesis starts with the Torah verse itself, as in the examples given from the Pauline writings. And there are numerous examples of exegesis in Hebrews which do not fit  28 29

Löhr 1994. Steyn 2011.

In Search of the Petichah

249

at all into the pattern in view. Thus, while the exegesis in Hebrews does show some characteristic features used later in the petichah, as the other examples adduced, it does not provide precise analogies or models. The variety of techniques present in Hebrews, however, is illuminating if we understand it as an integral part of the history of reading and interpreting the Tanakh in the first century CE and before. The pattern is even more discernible as a distinct technique if we compare these examples with the combination and interpretation of biblical quotations, e.g., in some of the speeches of Acts, or in the Epistle of Barnabas. In these particular texts, the argument is also built on different quotations from Scripture, but it does not exploit the twofold structure of the canon systematically. Thus, we are entitled to make a distinction between two ways of scriptural exegesis built on the combination of different verses from the Bible, the one relying on the dialogue between the different parts of the canon, the other one not using this technique. Whether the pattern sketched here originates with the emergence of reading from different parts of the Tanakh in the ritual of the early synagogue, or whether it is the outcome of specialized study and training in interpretation, cannot be said from the evidence we have. And again, as suggested above with regard to the function of the synagogues in the Second Temple period, this alternative may not be mutually exclusive.

*** In conclusion, it has been suggested here that in order to enrich our knowledge on the function and the ritual of the early synagogue, we may take into account early evidence for the twofold or even threefold structure of the Hebrew Bible, as well as exegetical methods in contemporaneous texts which seem to be referring to this structure. I suggested to understand reference to the “Law and the Prophets” (with omission of the writings) not so much in terms of an earlier stage of the development of the Tanakh, but as a shorthand expression of a specific aspect of the Holy Scriptures. To assume a direct link to the practice of ritual readings from the Torah and the Prophets is a plausible option, but not more. With more certainty, we can identify a distinct type of exegesis which builds on the twofold structure of the canon in the first century CE at the latest.

250

Hermut Löhr

Bibliography Auwers, J.-M. and de Jonge, H. J. (eds.) 2003. The Biblical Canons. BETL 163. Leuven: Peeters Publishers. Avemarie, F. 1994. “Schriftgebrauch in der haggadischen Exegese der Amoräer: Am Beispiel der Peticha WaR 27,3”. In: Hengel, M. and Löhr, H. (eds.), Schriftauslegung im antiken Judentum und im Urchristentum. WUNT 73. Tübingen: Mohr Siebeck, 133–152. Bacher, W. 1913. Die Prooemien der alten jüdischen Homilie. Beitrag zur Geschichte der jüdischen Schriftauslegung und Homiletik. Leipzig: J. C. Hinrichs. Brooke, G. J. 1985. Exegesis at Qumran. 4QFlorilegium in its Jewish Context. JSOTSup 29. Sheffield: Sheffield Academic Press. Campbell, J. G. 2000. “4QMMTd and the Tripartite Canon”. JJS 51: 181–190. Claußen, C. 2002. Versammlung, Gemeinde, Synagoge: Das hellenistisch-jüdische Umfeld der frühchristlichen Gemeinden. SUNT 27. Göttingen: Vandenhoeck & Ruprecht. Cohen, N. G. 1997. “Earliest Evidence of the Haftarah Cycle for the Sabbaths between   and  in Philo”. JJS 48: 225–249. Cohen, N. G. 2007. Philo’s Scriptures: Citations from the Prophets and Writings. Evidence for a Haftarah Cycle in Second Temple Judaism. JSJSup 123. Leiden: Brill. Fine, S. 1997. This Holy Place: On the Sanctity of the Synagogue during the Greco-Roman Period. Christianity and Judaism in Antiquity Series 11. Notre Dame, IN: The University of Notre Dame Press. Finkelberg, M. and Stroumsa, G. G. (eds.) 2003. Homer, the Bible, and Beyond: Literary and Religious Canons in the Ancient World. JSRC 2. Leiden: Brill. Goldberg, A. 1979. “Peti#a und "ariza: Zur Korrektur eines Mißverständnisses”. JSJ 10: 213–218. Goldberg, A. 1980. “Versuch über die hermeneutische Präsupposition der Peti#a”. FJB 8: 1–59. Hachlili, R. 2013. Ancient Synagogues—Archaeology and Art: New Discoveries and Current Research. HdO I 105. Leiden: Brill. Hengel, M. 1994. “‘Schriftauslegung’ und ‘Schriftwerdung’ in der Zeit des Zweiten Tempels”. In: Hengel, M. and Löhr, H. (eds.), Schriftauslegung im antiken Judentum und im Urchristentum. WUNT 73. Tübingen: Mohr Siebeck, 1–71. Kratz, R. G. 2006. “Mose und die Propheten: Zur Interpretation von 4QMMT C”. In: García Martínez, F. et al. (eds.), From 4QMMT to Resurrection: Mélanges qumraniens en hommage à E. Puech. STDJ 61. Leiden: Brill, 151–176. Lim, T. H. 2001. “The Alleged Reference to the Tripartite Division of the Hebrew Bible”. RevQ 20: 23–37. Löhr, H. 1994. “‘Heute, wenn ihr seine Stimme hört …’: Zur Kunst der Schriftanwendung im Hebräerbrief und in 1 Kor 10”. In: Hengel, M. and Löhr, H. (eds.), Schriftauslegung im antiken Judentum und im Urchristentum. WUNT 73. Tübingen: Mohr Siebeck, 226– 248. McDonald, L. M. and Sanders, J. A. (eds.) 2002. The Canon Debate. Peabody, MA: Hendrickson. Norelli, E. (ed.) 2004. Recueils normatifs et canons dans l’Antiquité: Perspectives nouvelles sur la formation des canons juif et chrétien dans leur contexte culturel. PIRSB 3. Lausanne: Zèbre. Perrot, C. 1973. “Luc 4, 16-30 et la lecture biblique de l’ancienne synagogue”. RSR 47: 324– 340.

In Search of the Petichah

251

Siegert, F. 1992. Drei hellenistisch-jüdische Predigten: Ps.-Philon, “Über Jona”, “Über Jona” (Fragment) und “Über Simson”, Vol. II: Kommentar nebst Beobachtungen zur hellenistischen Vorgeschichte der Bibelhermeneutik. WUNT 61. Tübingen: Mohr Siebeck. Stemberger, G. 2008. “The Derashah in Rabbinic Times”. In: Deeg, A. et al. (eds.), Preaching in Judaism and Christianity: Encounters and Developments from Biblical Times to Modernity. SJ 41. Berlin: De Gruyter, 7–21. Stewart-Sykes, A. 2001. From Prophecy to Preaching: A Search for the Origins of the Christian Homily. VCSup 59. Leiden: Brill. Steyn, G. J. 2001. A Quest for the Assumed LXX Vorlage of the Explicit Quotations in Hebrews. FRLANT 235. Göttingen: Vandenhoeck & Ruprecht. Thackeray, H. St. J. 1921. The Septuagint and Jewish Worship: A Study in Origins. The Schweich Lectures 1920. London, published for the British Academy by H. Milford: Oxford University Press. Theobald, C. (ed.) 1990. Le Canon des Écritures. LD 140. Paris: Cerf. Thyen, H. 1955. Der Stil der Jüdisch-Hellenistischen Homilie. FRLANT 65. N.S. 47. Göttingen: Vandenhoeck & Ruprecht. Ulrich, E. 2003. “The Non-attestation of a Tripartite Canon in 4QMMT”. CBQ 65: 202–214. Yadin, Y. 1966. Masada: Herod’s Fortress and the Zealots’ Last Stand. New York: Random House.

Rabbis, Nonrabbis, and Synagogues in Roman Palestine: Theory and Reality

Ruth Langer

I. Introduction and a Discussion of Method When and how did the communal synagogue become ‘rabbinic’ and rabbinic liturgy become Jewish liturgy? Related of course is the question of when rabbinic aspirations to leadership succeeded.1 The collected evidence demonstrates that a non-sacrificial Jewish liturgical system had emerged by Late Antiquity; its prayer texts, though, have been preserved only from the late first millennium CE. Until the 1980s, all historians of Jewish liturgy presumed communal prayer to be among the functions of the Second Temple synagogue and hence located the liturgy’s origins there. By 1986, Shmuel Safrai opened these questions anew with his now widely accepted recognition that, while there is good evidence for Torah reading, there simply is no evidence for rabbinic-style prayer in the Second Temple-era synagogue.2 The only fairly comprehensive theory of liturgical development to emerge in the wake of Safrai’s realization has been that of Ezra Fleischer, published in a book’s worth of articles beginning in 1990.3 Fleischer claims that rabbinic liturgy emerged de novo as a response to the covenantal and spiritual void left by the loss of the Jerusalem Temple. Under Rabban Gamliel, the rabbis developed an alternative, universally obligatory, form of verbal worship. In order to spread this worship, they immediately co-opted the pre-existent synagogue.4 While in the last quarter-century, various scholars have probed elements of Fleischer’s theory, no international consensus has emerged around it. Fleischer’s theory is built upon a  1 Others who have asked these questions explicitly in liturgy or synagogue-related discussions include: Fleischer 2012, with relevant essays dating from 1990; Levine 2005: especially chaps. 13, “The Sages and the Synagogue”, and 16, “Liturgy”; and Miller 1999: 64. 2 In an article first published in 1981, Shmuel Safrai posited prayer as a function of the Second Temple synagogue, at least towards the end of that period. However, he taught otherwise in a graduate seminar on the synagogue I attended at Hebrew University in 1986/87. That spring he presented his new understanding at a conference on the synagogue in Haifa, Safrai 1989. 3 For the Hebrew articles themselves, see Fleischer 2012; for a review essay on most of them, see Langer 1999 and the published correspondence between us, Fleischer and Langer 2000. 4 This last point appears in “Annual and Triennial Reading of the Bible in the Old Synagogue” (in Hebrew) in Fleischer 2012: 37–38 (section 7).

254

Ruth Langer

series of logical deductions, at the base of which often stands a problematic reading of the evidence. It thus builds a very neat picture, but not one that reflects a verifiable historical reality. There is a deep ongoing need for scholars carefully to differentiate between conjecture and fact in their reconstructions of the history of Jewish liturgy. However, rabbinic texts are essentially our only detailed sources for Jewish liturgical history in Late Antiquity after the demise of the Qumran community. These texts embed within them the rabbis’ own story of their history, one that presumes that the rabbis had authority and that their leadership was immediately accepted, at least from the early first century CE if not before. Most liturgical scholars have accepted the rabbinic self-portrayal as historical truth and reconstruct the early history of rabbinic prayer accordingly. However, there are many who today question this method of reading rabbinic texts, labeling it “historical positivism”, and calling instead for a cautious and critical approach to the sources, one that is conscious (to the extent possible) of the editorial layers of the documents at hand, of these editors’ agendas in reworking their received traditions, and of our own agendas as interpreters of the texts.5 Critical readings of texts based on manuscript variants and parallel versions, while certainly necessary, are insufficient. Extreme critics of positivist readings exclude the possibility of learning any history from rabbinic literature.6 However, the argument that one should “derive historical information” about this period only on the basis of extra-rabbinic— preferably, material—evidence is simply overly limiting. In many areas, like liturgy, we simply lack alternative sources. In this situation, a variety of scholars have sought to discern credible methods for applying what David Levine (2009: 64) calls a “skeptical positivism” or “realistic positivism”. Such an approach does require us to apply all available critical methods to the texts about the synagogue and its liturgy, and read this along with what material evidence exists. Stefan Reif (1993: 91–93) was the first liturgist to call for such an approach, but he did not implement it fully because the critical literary tools were not yet at hand. Lee Levine’s (2005) chapters on the rabbis and the liturgy come much closer, though his discussions are still open to some refinement. Such a method helps us to rein in the imaginative leaps that have characterized most liturgical histories while also accepting that we will probably never know the full picture. While these issues of method generate questions about all aspects of early rabbinic liturgy, this essay focuses on applying them to the question of when, how, and to what extent the rabbis associated their liturgy with the institution of the synagogue in the Land of Israel in the Roman era, i.e., in the time of the tannaitic  5

For example, Levine 2009: 64; and Alexander 2010. See, for instance, the responses of Isaiah Gafni 2010; 2011 to the undermining of his own historical work. See also Newman 2006, especially his closing paragraph, which begins by acknowledging, “Historicizing readings are not all equally valid and some are claptrap. There are occasions, however, when a historicist interpretation is plausible, though not demonstrable with anything approaching mathematical certainty”. 6

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

255

and early amoraic rabbis.7 The results nuance Levine’s (2005) discussions, being generally more cautious about placing prayer in the synagogue and about Babylonian reworkings or inventions of earlier traditions. The conclusions also aim to be appropriately skeptical and hence cautious about introducing assumptions beyond what the evidence itself actually supports.

II. Tannaitic Evidence A. Where was Rabbinic Liturgy Located? 1. Mishnah Megillah The earliest rabbinic text, redacted ca. 220 CE, the Mishnah, explicitly discusses in Tractate Megillah (The Scroll [of Esther]) the synagogue as the locus of the Torah scroll, its associated holiness, and presumably its reading. Chapter Three of Tractate Megillah begins with an explicit recognition that the synagogue itself belongs to the community; they, not the rabbis, are the people who may buy and sell it. However, the rabbis, not named, do not fully approve of the accepted protocols governing this. Hence, the Mishnah teaches (m. Meg 3:1): ‫ תיבה‬.‫ בית הכנסת לוקחין תיבה‬.‫בני העיר שמכרו רחובה של עיר לוקחין בדמיו בית הכנסת‬ .‫ ספרים לוקחים תורה‬.‫ מטפחות לוקחין ספרים‬.‫לוקחין מטפחות‬ .‫ מטפחות לא יקחו תיבה‬.‫ ספרים לא יקחו מטפחות‬.‫אבל אם מכרו תורה לא יקחו ספרים‬ ‫ וכן במותריהן‬.‫ בית הכנסת לא יקחו את הרחוב‬.‫תיבה לא יקחו בית הכנסת‬ .‫אין מוכרין את של רבים ליחיד מפני שמורידין אותו מקדושתו דברי רבי מאיר‬ .‫ אם כן אף לא מעיר גדולה לעיר קטנה‬:‫אמרו לו‬ City residents who sold a public square may use the funds for a synagogue. If they sold a synagogue, they may buy a chest [for Torah scrolls]; a chest—they may buy wrappings [for the scrolls]; wrappings—books; books—Torah.  But if they sold a Torah, they may not purchase other books; books—not wrappings; wrappings—not a chest; a chest—not a synagogue; a synagogue—not a public square; and so too with the rest of the funds.  One may not sell public property to an individual because one lessens its level of holiness, according to Rabbi Meir.  They said to him: If this is the case, then one also may not sell from a large city to a small city.8

 7 The tannaim are the early rabbis, up to the generation of Rabbi Yehuda HaNasi, who is credited with redacting the Mishnah in the early third century. The amoraim are the rabbis of the centuries following whose traditions shape the Jerusalem and Babylonian Talmuds, i.e., up to about the sixth century. 8 Compare m. Ned. 5:5 which also explicitly identifies the synagogue, chest, and books as belonging collectively to the inhabitants of the city. The reading “Rabbi Meir” follows the manuscripts of the Mishnah and the Talmuds (as they are listed for b. Meg. 27b at bavli.geniza.org on 26 December 2017) as well as their printed editions, but not the printed editions of the Mishnah.

256

Ruth Langer

The rabbis here express their hierarchy of symbolic values and seek to impose them on those who actually own the synagogues. Funds from a communal sale may only be applied to something of greater sanctity. The synagogue, they teach, is holier than other communally owned property, but ultimately less holy than even the accoutrements for the Torah scroll, let alone that scroll itself. Are these rabbis simply recording generally held communal values or are they seeking to impose their own values on the broader community, to legislate against behavior they perceive as inappropriate? The continuation of the Mishnah and the related Tosefta ban very specific behaviors, indicating that failure to respect the synagogue’s inherent sanctity deeply disturbs the rabbis. In m. Meg. 3:3, the second-century Rabbi Yehuda restricts the use of even a ruined synagogue as a convenient open space for eulogizing the dead, making ropes, drying fruits, spreading out hunting nets, or taking a shortcut. The Tosefta’s parallel list (t. Meg. 2:18) applies to active synagogues, forbidding acting lightheartedly there, taking shelter from the weather, eating, drinking, and sleeping. The last three, we know, were the actual purpose of at least one first-century synagogue in Jerusalem, as the Theodotus inscription (CIJ II 1404 = CIIP I 9) documents.9 Thus, this rabbinic discussion of synagogue sanctity seems to be aspirational, a reflection of rabbinic values, and not an expression of universally accepted Jewish understandings of the sanctity of the synagogue. The tannaitic rabbis apparently do not understand Torah reading itself to be dependent on this community synagogue. A literary reading of this Mishnah, as we have received it, almost suggests that the rabbis more or less “demolish” the lay-owned institution in favor of the locus of their own liturgical life. M. Meg. 3:2, like its predecessor, discusses ways of selling the synagogue; m. Meg. 3:3 discusses destroyed ones. The remainder of this chapter and the parts of the next (m. Meg. 3:4–6; 4:1–2, 4, 10) discuss details of reading scripture; m. Meg. 4:3 and then 4:5– 9 broaden the scope to include the rest of rabbinic liturgy.10 The synagogue itself, as a context, fully disappears from the discussion.11 In other words, this thirdcentury CE composition moves seamlessly from, so to speak, selling off or destroying synagogues, to legislating about their pre-rabbinic functions as a place of Torah reading, now removed from that location, to prescribing elements of the larger rabbinic liturgical system. This larger rabbinic system is undocumented from outside of rabbinic sources, but the rabbis present this synthesis as a seamless, universally recognized whole.  9 Fleischer 2012: 12–13 points out the absence of prayer from this list. Reif 1991: 679–680 questions the historical significance of this argument from silence. 10 This reading suggests that the Babylonian decision to switch the order of these two chapters was not sensitive to the internal flow of the Mishnah’s discussion. This transition from the topic of Scripture reading to the rest of the liturgy, begun in 4:3, might explain why 4:5 and 4:6 begin with issues of Scripture reading and then move to the other elements which chronologically precede them in the service. 11 Levine 2005: 492 offers a different interpretation of this juxtaposition of themes, suggesting that these passages and their Toseftan equivalents (with different details) may reflect either rabbinic choices among existing practices or attempts to create something new.

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

257

We need to ask: is it only our presuppositions, based on later realities and this literary juxtaposition, that lead us to imagine that rabbinic liturgy was located in the same synagogue that the inhabitants of the town had been buying, selling, or misusing? Or is it that the Mishnah’s editors wish us to understand this? Was this the rabbinic ideal or the rabbinic reality, and if so, from when? To answer these questions, we need to look beyond this tractate and its concerns.12 2. Prayer and the Tevah The tannaim refer to the person who leads the communal recitation of their central prayer, the ‘amidah,13 as ‫העובר )היורד( לפני התיבה‬, “the one who passes (or: goes down) before the tevah”, i.e., the freestanding chest containing the Torah scroll(s).14 The hierarchy of holiness in m. Meg. 3:1 places this chest in the synagogue, but were such chests housed only in synagogues? The vast majority of tannaitic references to a person leading the ‘amidah from before the tevah do not otherwise indicate the location of the liturgy. Going down before the tevah is among the rituals m. Meg. 4:315 lists as requiring a quorum of ten, i.e., a situation of public communal worship. Thus, the space for the ritual requires only the tevah and the quorum, but that tevah was not a fixed furnishing initially; this language therefore does not necessarily indicate any dedicated place where that gathering occurs.16 That Mishnah Megillah and other sources list  12

The transfer of select non-sacrifical Temple rituals to the synagogue is also relevant here. Mishnah Rosh Hashanah is ambiguous about the relationship of the shofar to the synagogue and to rabbinic prayer. See m. Roš Haš. 3:7 (outside and inside synagogue); 4:1–2 (neither); 4:5–6, 8 (in rabbinic prayer but no synagogue); 4:7 (shofar in synagogue). The transfer of daily waving the lulav during the festival of Tabernacles from the Temple to the synagogue appears in m. Sukkah 3:12–13 and t. Sukkah 2:10. This waving occurs in the context of recitation of the Hallel Psalms (Pss 113–118), also a transferred Temple ritual. The Tosefta passage mentions more details of the synagogue rituals, the only one of which suggests a specifically rabbinic context being “to pass before the chest”. See too t. Pisa 10:8 which transfers these Psalms to the synagogue on Passover for the “city residents” who have no one to recite them in their homes. 13 M. Ber. chs. 4–5 presumes the existence of this prayer and places rabbinic discussions requiring public participation in it ca. 100 CE at Yavneh. However, stories of its formulation there appear only in the Babylonian Talmud, see b. Ber. 28b (with a partial parallel in b. Meg. 17b–18a), and may be more mythological than factual. For a brief introduction and further bibliography, see Langer 2015: chap. 5.3. 14 On the tevah, see Levine 2005: 351–356. On the verbs used, see Weiss 1990; and Hakohen, Rosenson, and Weiss 1992. 15 M. Meg. 4:3: ‫אין פורסין את שמע ואין עוברין לפני התיבה ואין נושאין את כפיהם ואין קורין בתורה ואין מפטירין בנביא ואין‬ .‫עושין מעמד ומושב ואין אומרים ברכת אבלים וחתנים ואין מזמנין בשם פחות מעשרה‬ They do not perform the public recitation of shema‘, or pass before the chest [to lead the ‘amidah], or [have priests] raise their hands [in blessing], or read Torah [liturgically], or close with the prophetic reading, or perform the [funeral ritual of] standing and sitting, or recite the blessing for the mourners or for the bridegrooms, or invite participation in the Grace after Meals with the Divine Name with less than ten [adult men]. 16 This is the opposite of Neusner’s argument, 2001: 168–169. He asserts that the synagogue itself is this gathering of ten and not a particular place. Part of his argument grew from the dearth of early synagogues then uncovered archaeologically; more have been found subsequently.

258

Ruth Langer

“passing before the tevah” among the other elements of rabbinic liturgy that include Torah reading only suggests a possible synagogue context but does not establish one.17 T. Meg. 3:21 indicates clearly that the tevah was very mobile for it discusses what direction the tevah should face when it is set down, and how the people of various ranks should orient themselves in relationship to one another and to it. However, nothing in this particular text hints at rabbinic prayers, though this seems to describe a synagogue.18 In contrast, t. Sukkah 2:10 lists “passing before the tevah” among the rituals one may perform in the synagogue while holding one’s lulav (palm branch, bound with willow, myrtle and citron; Lev 23:4). Is this a counter source? It seems to be a tannaitic tradition that developed along with m. Sukkah 3:12–13. The Mishnah teaches that in Temple times and after, outside Jerusalem, the ritual waving of the lulav—probably during the recitation of the Hallel Psalms (Pss 113– 118)—took place in the synagogue; however, the Mishnah names no other ritual elements. In contrast, the Tosefta relates the former custom of Jerusalem Jews to manage the various elements of rabbinic liturgy while holding a lulav. They could hold it while translating scripture or passing before the tevah, but not when reading scripture or raising their hands in the priestly benediction, rituals for which their hands were otherwise needed. A positivist reading takes this at face value. We need to ask whether this reflects a genuine Second Temple-era custom, the custom of later Jerusalemite refugees, or a retrojection of a later reality? The text seems confused, as pre-destruction Jerusalemites would have celebrated in the Temple, not the synagogue; consequently the Mishnah specifies that its discussion applies only outside of Jerusalem.19 We cannot rely on the Tosefta text as a historical witness to rabbinic-style prayer in the presence of the tevah in the synagogue before the era of the Tosefta’s redaction.20 Mishnah and Tosefta Taanit chap. 2 also record that, at least in times of drought, on public fast days, the tevah was brought into the city square and an elaborate ritual was performed there in its presence. After placing ashes on it and on the heads of the community’s leaders as a sign of debasement before God, the community in need of rain recited seven supplicatory prayers accompanied by  17

See also t. Meg. 3:10. The only parallel text (y. Bik. 3:3, 65c) lacks even the possible synagogue context. 19 Both m. Sukkah 3:12 and m. Roš Haš. 4:3 list the lulav as one of the Temple rituals that Rabbi Yoanan ben Zakkai adjusted to the post-Temple reality. B. Sukkah 43a and 44a try to reconcile these sources, suggesting that one refers to the Temple and the other to the synagogue after the Temple’s destruction, or alternatively, one refers to the Temple and the other to observance by those not present in the Temple, but while it still stood. The Tosefta and b. Sukkah 41b transmit the baraita in the name of Rabbi Lazar bei Rabbi Tzaddok; there are three tannaim by this name, generations II, III, and IV. Y. Sukkah 3:14, 54a introduces this baraita with ‫“( ותניי תמן‬it is taught there”) without specific attribution. While Lieberman 1992b: IV 865; 1937: I 199 and others suggest that this points to a Babylonian chain of tradition or origin, Cohen 2017: 166–170, 189 disagrees. 20 That the talmudic citations of this baraita (see previous note) add the recitation of shema‘ to the list of rituals suggests that the Tosefta’s omission of this may be significant and can help us date this tradition to before the full merger of these rabbinic liturgical elements, probably in the third century. 18

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

259

appropriate biblical texts, punctuated by priests blowing horns. For our purposes here, what is important is that this prayer seems not to have conformed to rabbinic prayer patterns and did not take place in the synagogue. Even when the rabbis domesticated this ritual into their liturgical system, creating an ‘amidah of twenty-four benedictions, they did not change its location. Therefore, prayer before the tevah did not necessarily mean prayer in the synagogue or even rabbinic-style prayer.21 If this is the case, where else might the tevah have been located? There are few textual clues from this era, but this movable container of sacred texts could have been found among those wealthy enough to own books, often but not exclusively rabbis, perhaps in their homes or where they studied. However, there is no material evidence for this (Hezser 2001: 147–149, 165). There may also have been spaces functioning as places for prayer gatherings, perhaps not labelled as such, in the homes of those rich enough to own the books. The only tannaitic text suggesting such a reality records occasions when two different students of Rabbi Eliezer went down before the tevah in his presence and offered creative versions of prayers (Mek. de-R. Ishmael Wa-yassa 1, 155; Mek. de-R. Shimon bar Yoai 15:25, 103; b. Ber. 34a).

B. Tannaitic Evidence for the Location of Prayer: Was it in the Synagogue? It is also fruitful to reverse the question and begin with the liturgy itself. As demonstrated by the Mishnah’s editorially separating its discussion in Megillah of the synagogue from its discussion in Berakhot of the main elements of rabbinic liturgy, the tannaitic tradition did not define “liturgy” as a synagogue-centered entity. Instead, Mishnah Berakhot, in chaps. 6–9, implicitly includes under “liturgy” both home and community-based food-centered rituals and private responses to God’s world. How then did the rabbinic prayers discussed in m. Ber. 1–5 function in the community? When the Temple stood, rotating courses of priests and Levites took care of Israel’s covenantal obligations to worship God through sacrificial offerings (Mishnah Tamid). A corresponding communal group (ma‘amad) gathered in their town—but in an undisclosed location—, fasted, and read Genesis’ creation narrative (m. Ta‘an. 4:1–2; t. Ta‘an. 3:3; Tabory 2003). This situation changed dramatically with the Temple’s destruction in 70 CE. The Mishnah attributes to late first century Rabban Gamliel II the expectation that everyone would pray  21

T. Ta‘an. 2:4 explicitly indicates that this outdoor ritual followed a Torah reading in the synagogue. The formulations of this passage made it very susceptible to scribal errors. The London manuscript skips these details entirely, the Erfurt manuscript skips the taking of the tevah out to the city square. See Lieberman 1992a: Mo‘ed 330. This casts some doubt on this source, but not enough to discard it.

260

Ruth Langer

daily. The Mishnah records debates about the nature and extent of this requirement, but not its details (m. Ber. 4:3–4:4a); we find Jewish precedent for such verbal prayer only in the sectarian community at Qumran that denied the validity of the Temple. How did Rabban Gamliel communicate and educate about this new commandment? Did the rabbis actively convene people for prayer, and if so, how often? Do we have indication that they employed the synagogue context? The tannaitic texts extend the requirement of prayer to non-rabbis wherever they may be. Travelers must pray at appropriate times while on their journeys, making compromises only, if necessary, in their posture, orientation to Jerusalem, or, in cases of danger, in the length of their prayer (m. Ber. 4:4b-6; t. Ber. 3:18). Laborers might have to descend to ground level to pray, but they remain in their workplace (m. Ber. 2:4). The picture in the Tosefta (t. Ber. 1:3; 2:6–8) is coherent with this, though it exempts from praying those engaged in the performance of other commandments, like the bridegroom on his wedding night or those writing sacred texts, and the porter physically bearing a load, because they lack the ability to concentrate properly on their prayers. It allows prayer by workers while in olive or fig trees, but not others, presumably because the trees are not so large as to lead them to fear falling. Curiously, though, these workers, themselves fully obligated, may not lead the prayers for others before the tevah (t. Ber. 2:9), suggesting a gap between rabbinic requirements and reality, as the rabbis, at least in the Tosefta, seem to be either questioning the workers’ actual competence in prayer or limiting the backlash that might hinder the success of their innovative form of worship were it to require significant loss of work time. Thus, these texts suggest no association between rabbinic prayer and the synagogue. The Mishnah does contain two passages that apparently suggest that rabbinicstyle prayer is taking place in the synagogue. One we can dismiss easily. M. Ber. 7:3 models a form of the invocation for the Grace after Meals on the synagogue call to the congregation using the imperative ‫“( ברכו‬Praise!”). Most liturgical historians, following Elbogen (1993: 17), read this as the origins of the call to worship (barekhu) preceding shema‘. However, this Mishnah almost certainly refers to the blessings before the Torah reading that begin with precisely the same language, themselves mentioned in m. Meg. 4:1. The call to worship before shema‘ cannot be documented before the geonic period.22 M. Bik. 1:4 is more complicated. It requires proselytes to differentiate themselves ritually from born Jews, including in the reference to the biblical ancestors in the first blessing of the ‘amidah, a text that addresses God, paraphrasing Exod 3:15.23 Where born Jews refer to God as the God of “our ancestors”, proselytes  22

See Soferim 10:6; 18:10. The discussions of this call in the Talmuds fit this explanation. See y. Ber. 7:3, 11c; b. Ber. 50a. Sifre Deut § 306 is less explicit but still suggests this reading. 23 This is one of the few examples of tannaitic evidence for fixed language within the body of a rabbinic prayer. This reference to God has no known Second Temple precedent.

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

261

praying privately should substitute “Israel’s ancestors”. In contrast, in the synagogue, i.e., in communal prayer, they quote Exod 3:15 precisely and say “your ancestors”.24 The specification of “synagogue” here might well be an interpolation into the text as it is not necessary; however the word appears in all known manuscripts, meaning that we cannot exclude this witness.25 Thus, at this earliest layer, with this one exception, the rabbis never associate the obligation to participate in their liturgical system with any particular location. Indeed, there are numerous sources, mostly later, that record the preference of particular rabbis to pray outside of the communal synagogue context, even when services were being held there. Earliest clues include t. Ber. 3:5, recalling that “when Rabbi Akiva prayed with a community, he would shorten his prayer…,26 but when he prayed privately, [he not only would pray at such length that] one could leave, return, [and still find him in mid-course] but his bowings and prostrations [i.e., expressions of religious enthusiasm] would move him from one corner of the room to the other”. In other words, he exercised spiritual restraint from consideration for others, but presumably preferred praying alone. A similar preference emerges from a teaching in the name of the first-century Rabbi Dosa ben Harkinas that sitting in the synagogue of the commoners shortens one’s life (m. ’Abot 3:10). A comparison between the references to prayer in the synagogue in the Mishnah and in the Tosefta suggests the possibility that a transformation was occurring gradually in the early third century between the final redactions of these texts.27 Mishnah Bikkurim might be an early manifestation of this change. Where the Mishnah includes only this one explicit mention of prayer in the synagogue, the Tosefta includes several. In addition to those already mentioned, t. Ber. 2:4 (cf. t. Meg. 2:3) contrasts the way an individual should correct a mistake in his private recitation of rabbinic prayers with the protocol for when he comes late to the synagogue and the community is already half way through them. Similarly, m. Meg. 4:5–6 only mentions a public liturgy named pores et (or: ‘al) shema‘, but  24

It is unclear whether this presumes that the proselyte is leading the prayers. According to the materials posted as of May 14, 2017 on the Friedberg Genizah Project website, this language appears all the in Genizah manuscripts containing this particular text: CUL: T-S E1.38, Fragment 1v, FGP No.: C166041; Oxford: MS heb. c.8/2, Fragment 1r, FGP No.: C459898; New York, JTS: MS R1622, fol. 17 (Alt: MS 8451, fol. 17), Fragment 1r, FGP No.: C89150; Cambridge, CUL: T-S AS 75.180, Fragment 1v, FGP No.: C99641; AIU: III.A.19, Fragment 3r, FGP No.: C76900 (only ‫ בית‬appears at the edge of the fragment). Y. Bik. 1:4, 64a does not discuss the synagogue context. For an extended discussion of this Mishnah, though without a liturgical focus, see Cohen 1999: Chap. 10. 26 I have skipped the words ‫ בפני כולם‬because Lieberman 1992b: I 29 rejects all possible translations of them. He notes that b. Ber. 31a revises the text to read: ‫( מקצר ועולה מפני טורח צבור‬abbreviates and comes up [from before the ark] because of causing aggravation to the community). The concepts of “causing aggravation to the community”, or adjusting liturgy because of the “honor of the community” appear only in the Babylonian Talmud and thus should not inform this translation. 27 There is no agreement on the question of how to characterize the relationship of the Tosefta to the Mishnah since Friedman 1999 and Hauptman 2000 argued for the Tosefta’s primacy and a dynamic relationship. See Mandel 2006. 25

262

Ruth Langer

t. Soah 6:3 describes this as a responsive performance located explicitly in the synagogue. It is the Tosefta that restricts workers from leading communal prayers before the tevah (not necessarily in a synagogue), suggesting the possibility that the Tosefta knows an elitist rabbinic prayer context. Zeev Weiss’ (1990) suggestion that the Tosefta’s greater use of “going down before the tevah” in comparison with the Mishnah’s almost exclusive “pass before the tevah” reflects a real shift is also useful here. He argues that this reflects a fourth-century shift in synagogue architecture; more likely, it reflects a third-century shift towards rabbinic prayer in synagogues.

III. Talmudic Developments and Assessing Baraitot While in the Mishnah and Tosefta, we found only isolated incidental references to prayer in the synagogue, that is not the case in the Talmuds, including in materials ascribed there to the tannaim. Essentially all liturgical scholars, including Levine (2005: 532), have presumed that tannaitic traditions in the Talmuds do provide historically reliable evidence. But it is precisely here that historical positivism becomes a significant danger. Comparisons of parallels demonstrate that the talmudic editors routinely revised received materials to reflect their own reality and values; they also seem to have pseudepigraphically constructed new materials. How then, should we read the tannaitic materials in the Talmuds, especially those that suggest wide participation in rabbinic-style prayer located in the synagogue? We need to consider seriously the possibility that some or even many Babylonian baraitot (content the Talmuds identify as non-Mishnaic tannaitic traditions), especially ones that lack parallels in sources from the Land of Israel,28 have the potential of being pseudo-baraitot,29 texts that have no authentic early tradition behind them. However, other unparalleled baraitot may be authentically tannaitic, but preserved only in this later source. Therefore, other characteristics, such as the baraita’s fitting an obviously Babylonian argument or scenario, or obvious reworking of the tannaitic original, must verify our suspicions. A crucial prerequisite for assessing these baraitot, then, is to understand the cultural realities of the worlds through which the traditions might have passed. What editorial constructions might have shaped them? As Shamma Friedman (2000: 46–47) writes, failing to identify pseudo-baraitot can lead the scholar “to enticing pieces  28 Most importantly, the Tosefta and halakhic midrashim, but also the Jerusalem Talmud and early aggadic midrashim. 29 For this term and a discussion of it see Rubenstein 1999: 261–262. The concept itself is not original with him, as he acknowledges. Much of the most detailed work on the question has been done by Friedman 2010. Jacobs 1971 takes a somewhat more conservative approach to what he calls “fictitious baraitot” in dialogue with yet earlier discussions. Cohen 2017, though, argues that the Babylonian amoraim transmitted authentic content of tannaitic traditions, albeit with some editing.

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

263

of information about the tannaim, which he or she would prefer not to discount. As a result, some of scholarship’s essential historical conclusions are based upon such baraitot in the Bavli”. Indeed, key elements of existing histories of Jewish liturgy are built upon such shaky evidence.30 While tannaitic evidence leaves open the possibility that communal participation in the rabbinic liturgical system occurred in the presence of the Torah scroll but not otherwise in a dedicated liturgical space, both Talmuds suggest a different situation was emerging in the third century CE. They attribute many traditions about prayer, explicitly in the synagogue, to the first generations of amoraim.31 It is against the backdrop of this demonstrable development from the tannaitic discourse that we need to assess the historicity of the tannaitic materials cited only in the Talmuds and especially the Bavli.32 Both Talmuds present us with arguments constructed to convince both nonrabbis and the rabbis themselves that there are advantages to participating in public prayer in the synagogue. This reflects the reality that by the time of the redaction of these texts, the synagogue was the official institutional locus of rabbinic prayer, both in the Land of Israel and in Babylonia. However, the need for persuasion suggests an underlying reality of non-participation in the rabbinic system. It therefore provides a key to the emergence of the successful combination of the institutions of the popular synagogue and rabbinic prayer.33

A. The Jerusalem Talmud One cannot analyze the Bavli without examining first its roots in the Yerushalmi. This earlier text contains a few traditions that seek to persuade, both rabbis and other Jews, that the synagogue is the ideal locus for the statutory prayers. Two discussions (see Appendix 1 below) record slightly different traditions in the name of the mid-third century Rabbi Yoanan: that one must pray in a place specifically designated for prayer (1a, y. Ber. 4:4, 8b); and that when one prays within one’s own home, it is as if one were encircled by an iron wall (2c, y. Ber. 5:1, 8d– 9a). Standing alone, both teachings contain inherent ambiguities: the place one  30

For instance, the baraita narrating the addition of the birkat haminim to the ‘amidah at Yavneh in b. Ber. 28b is presented as an answer to a clearly Babylonian question. At the very minimum, the meaning of the baraita has been transformed by its context. (Conversation with Judith Hauptman, 17 December 2017). For a detailed discussion, see Langer and Sarason 2019: 223–229. 31 Some—like Neusner 2004—question any reliance on attributions to date rabbinic traditions and limit dating texts to the redaction of their larger contexts. As a consequence, one cannot “do history” based on this literature. Pomerantz 2016: 205–209 argues for a method by which to derive historical insights while acknowledging that actual attributions may be unreliable. 32 This text was redacted by anonymous rabbis in Babylonia (Bavel), probably in circa the 7th century. It knows and uses materials found in the Yerushalmi (Jerusalem or Palestinian Talmud), itself redacted in less polished form in the fifth century in the Land of Israel. 33 For other sorts of evidence for rabbinic prayer in the synagogue, see Levine 2005: ch. 16.

264

Ruth Langer

designates for prayer could be in any number of locations; and a wall of iron could be protective rather than preventing prayer’s communication. However, the Jerusalem Talmud juxtaposes these traditions with others naming the synagogue explicitly and thus presents its readers with a call for locating prayer in the synagogue. Some of these are attributed to Rabbi Yoanan’s predecessors. His teacher, Rabbi Hoshaya, teaches that when one prays in the synagogue, it is as if one offered a pure flour sacrifice in the Temple (2e). In other words, the synagogue is an especially efficacious location for verbal worship. Rav una (fourth century) associates those failing to enter the synagogue with the wicked and threatens them with being excluded from the synagogue in the afterlife (2a–b). However, these traditions nevertheless leave us with a degree of ambiguity about whether prayer offered elsewhere is also efficacious, as long as it is in a place one has designated for prayer. 2d offers a resolution to the apparent contradiction between the two statements of Rabbi Yoanan, suggesting that while the individual may pray elsewhere, the community needs to pray in the synagogue. 1a itself suggests that prayer may be offered anywhere that God so desires, and that this is not dependent on human initiative. But does this include private homes? In 2f, Rabbi Abbahu, Rabbi Yoanan’s student, preserves some of his teacher’s ambiguity about where God is to be found, naming both the synagogue and the study house as efficacious loci for prayer.34 Bottom line, both the ambiguity and the voices of exhortation suggest that the late-Roman period rabbis were still short of their goal of extending their liturgical system broadly and inserting it into the popular synagogue.35

B. The Babylonian Talmud The Babylonian Talmud’s discussion in b. Ber. 5b–8b (Appendix 2) is much longer and more complex. It incorporates most of the Yerushalmi’s traditions and similarly encourages prayer in the synagogue. However, it adds some materials for which we have no parallels that it identifies as tannaitic that require attention. We also see evidence here of a disjunct between the expectations about the location of prayer, at least for rabbis, in the two communities, a disjunct that has the potential of influencing the editing of the texts and Babylonian revisions to them.  34 See also y. Ber. 9:1, 13a, in the name of the third generation (290–320 CE) amora Rabbi Abba Samuqa, that even though God is enormous, when a person enters the synagogue, stands behind a column, and whispers his prayers (like Hannah did), God hears them as if one were whispering into God’s ear. 35 Levine 2005: 494 points to the story of Rabbis Yoanan and Yonatan encountering a service in the cities of the south (y. Ber. 9:3, 12d) as evidence that they were involved in synagogues. However, this text just says “place” (‫ )אתר‬and does not otherwise name the location of this public prayer.

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

265

This discussion begins within a collection of four baraitot ascribed to one Abba Binyamin, a sage who appears in rabbinic tradition only here, so he cannot be otherwise identified or dated. Abba Binyamin’s two teachings about prayer in the synagogue provide the segue to the Bavli’s long discussion of that institution. They seem inconsistent with the general tenor of tannaitic discussions. The first (A), threatens dire consequences for bad manners: if one enters the synagogue36 with another to pray but does not wait upon finishing, then one’s own prayers are ruined and one causes the indwelling Divine presence, the Shekhinah, to depart the world.37 Although Abba Binyamin speaks only of two people here, this teaching reinforces a conception of prayer as a communal endeavor whose locus is the synagogue. His second relevant baraita (B), is similar to the teachings of the Yerushalmi, but goes beyond them. Abba Binyamin also teaches that God only hears prayers offered in the synagogue, but his prooftext is Solomon’s thanking God for heeding his rinnah and tefillah (1 Kgs 8:28). Rabbinic sources regularly rely on this verse to explain the organization of the parts of the amidah: praise of God, rinnah, precedes petition, tefillah (cf. t. Ber. 3:6; y. Ber. 4:4, 8b; b. Ber. 31a). This baraita uniquely applies this verse instead to the synagogue, explaining, “In the location where there is rinnah (praise or song), that is where there should be tefillah (prayer, or petition)”.38 This reflects a presumption of elaborate prayer in the synagogue, not something otherwise well-established from tannaitic texts. Thus we must question whether these traditions ascribed to Abba Binyamin are genuinely early or whether they are either pseudepigraphic or reworked texts to fit the Babylonian reality. To buttress Abba Binyamin’s reading, the Bavli cites a tradition attributed to the third-generation amora Rabbi Yiaq39—i.e., taking us perhaps into the early fourth century—who teaches first that God is found in the synagogue and second, that when ten pray together, God’s indwelling presence, the shekhinah, is with them. A similar set of teachings, though without explicit mention of prayer, appears with the same attribution in (C). These later traditions are indeed coherent with Abba Binyamin’s teachings, including his presumption that the shekhinah is involved in prayer—why else would error in prayer etiquette cause  36

While the printed versions of the text do not specify “synagogue”, it does appear explicitly in Ms. Paris 671, Ms. Munich 95, and most of the Genizah manuscripts. However, even without this term, the verb “enters” implies some structure. 37 The Tosafot explain this as referring to a synagogue out in the fields, where leaving an individual to return alone would endanger him. This is not a necessary explanation. 38 B. Ber. 6a, citing 1 Kgs 8:28 (= 2 Chr 6:19). 39 This presumes that this Rabbi Yiaq is Rabbi Yiaq Nappaa, a second generation amora from the Land of Israel, who often appears without the second name. It is also plausible that this is a fifth generation Babylonian tanna known just as Rabbi Yiaq. However, this identification means that the Bavli’s editors have elided these men together. Passages about the synagogue ascribed to Rabbi Yiaq are one of the organizing principles of this section, and (D) clearly describes the amora. The rabbi transmitting Rabbi Yiaq’s traditions is a third generation Babylonian amora.

266

Ruth Langer

it to leave? This too suggests that his teachings fit better with this later period’s understanding of the relationship between the synagogue and rabbinic prayer. How might we understand the intersection between the Talmud’s exhortations to people to pray in community and its exhortations to do so specifically within the synagogue? Are these now the same, or do they still represent (potentially) separate discourses? In (D), our same Rabbi Yiaq who insisted on the benefits of prayer in the synagogue, visits Babylonia and encounters the head of the Nehardea academy, Rav Naman, who sets a different example.40 Rabbi Yiaq seems shocked that Rav Naman chooses not to attend public prayers in the synagogue or even to coordinate his prayers with theirs. He pesters Rav Naman with a series of questions designed to open a path towards some conformity with the emerging norm of communal prayer in the synagogue among rabbis in the Land of Israel that was evident in the Yerushalmi traditions. Rav Naman resists all his importuning and even voices annoyance at his guest; we are not told that he changes his ways. The talmudic editor turns to marshalling alternative scriptural support for the efficacy of communal prayer with which this vignette ends. Its third suggestion (E) is a baraita in the name of Rabbi Natan that offers a midrash on Job 36:5. Sifre Num § 135, from a tannaitic midrash, cites this tradition in contrast to Moses’ complaint in Deut 3:26 that God did not accept the prayer he offered as an individual. However, by juxtaposing Job 36:5 with a second prooftext from Ps 55:19, the amoraim functionally rework the tannaitic midrash to make the synagogue its venue. The talmudic traditions interpret this second verse rather audaciously and against its simple meaning, to suggest that the community praying together redeems God. It comments, “The Holy One, blessed be He, said: All who engage in Torah [study] and in acts of loving kindness and who pray with the community, I consider it as if they have redeemed Me, Me, and my children, from among the nations of the world”. This conjunction of “Torah and acts of loving kindness” is fairly common in rabbinic literature, but “praying with the community” appears with them only here and must be considered a Babylonian addition, not a genuine part of this baraita. The Bavli then reinforces this assertion with a complementary tradition (F), attributed to Rabbi Yoanan’s colleague, Resh Lakish. He teaches that one who has a synagogue in his city but does not enter it to pray is called a bad neighbor and additionally causes exile for himself and for his children. This rhetoric continues in (G), now in Aramaic teachings ascribed to third century rabbis in the Land of Israel. Here, Rabbi Yoanan is taught that, in spite of their dwelling outside the Land of Israel, Babylonian Jews live long lives because they attend synagogue early and late, and Rabbi Yehoshua ben Levi, even  40 B. Ta‘an. 5a–6a preserves a series of conversations between the two, suggesting that they did speak face to face. The nature of the challenge in b. Ber. suggests that this meeting, at least, occurred in Rav Naman’s home territory.

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

267

earlier, taught his children to ensure long lives by doing so. These very possibly reflect the specific exhortation of the later Babylonian Talmud and not the historical reality of the third century rabbis in the Land of Israel. It definitely contrasts with m. ’Abot 3:10, where Rabbi Dosa ben Harkinas warns that “sitting in the synagogues of commoners removes a person from the world”, i.e., shortens one’s life. The Bavli has definitely reworked (H), that one should always enter a minimum of two door-widths into the synagogue, based on the plural reference to “doors” in Prov 8:34. The parallel in y. Ber. 5:1, 9a leaves unspecified the purpose of entering the building—it could be just to hear the Torah reading or a sermon—and here, consistent with this discussion, it is explicitly in order to pray. Multivocality and a degree of ambiguity characterizes the Babylonian Talmud’s own discussion as well. While it argues for the value of prayer in the synagogue and criticizes Rav Naman’s decision to avoid that institution and any association with its prayers, it also contains ample evidence that the rabbis preferred their own study contexts, both in general and as places for prayer. Many of these traditions are attributed to amoraim from the Land of Israel in the late Roman period—but they appear only in the Babylonian Talmud. We need to be careful about taking these traditions and their attributions at face value, but rabbinic self-segregation from the popular synagogue seems to have been an ongoing tension. For instance, b. Soah 22a records that Rabbi Yoanan challenged the decision of a widow who daily passed by closer local synagogues in order to pray in his study house. He seems to be trying to encourage her to participate in the community’s place of prayer. This suggests that Rabbi Yoanan’s exhortations to people to establish a fixed place for their prayers, in public, did apply to himself, but he did not consider it necessary for rabbis themselves to gather in the popular communal synagogue. (I) reinforces this observation as it cites rabbis from the generation following Rabbi Yoanan, at the very end of the Roman period, who chose to pray among rabbis only, i.e., not in private but in an elite community. Babylonians following this lead, like Rav Naman above (perhaps) and Abaye in this passage, seem less concerned about the communal aspect. Am I falling into the trap of historical positivism in my reading of these texts? I hope not. What is marked here is the consistent attribution of these traditions exhorting Jews to pray in the synagogue to a small group of leading rabbis from the Land of Israel in the mid- to late-third century. Either rabbis like Rabbis Yoanan and Yiaq are constructed characters in a wide swath of sources, throughout both Talmuds, or they were genuinely concerned about the rabbinization of the synagogue and making it the primary and ideal place for their liturgical system. Even if these are not historically accurate attributions, we somehow need to account for the literary placement of this dynamic in this set of decades. These attributions also roughly correlate with the apparent rabbinization in various ways of the synagogue buildings discovered from this period.

268

Ruth Langer

One last baraita (Appendix 3; b. Ber. 4b) requires discussion here. If it is genuinely tannaitic, then early rabbis argued strongly that prayer’s locus, for everyone, is in the synagogue. It therefore undercuts my arguments. However, there is good reason to consider it a pseudo-baraita. This baraita elaborates on m. Ber. 1:1, which teaches that the actual latest time for reciting the evening shema‘ is dawn, but that the sages taught that it needs to be recited before midnight “in order to distance a person from transgression”. Translating this into real terms, the baraita teaches that common laborers have obligations to pray rabbinic prayers. However, the rabbis recognize that in their exhausted haze at the end of the day their natural focus is on themselves instead of on God. This text responds with an extreme threat: anyone who gives in to these human needs and fails to meet God’s demands deserves death. The antidote? Common laborers need to act like rabbis, and coming directly from work, sweaty, smelly, and dirty, enter the synagogue. There, they should first study—including the rabbinic Oral Torah if they are able—and then pray. They should even eat there so that the rabbis can be sure that they bless God properly for their food. If this is a genuine baraita, it is evidence for prayer in the synagogue in the tannaitic period, at least as a rabbinic expectation. The first test is to look for parallels—and here we do find one in ’Abot de-Rabbi Nathan, especially its version A.41 However, how do we date that text? Menahem Kister (1997) suggests that an early common tannaitic layer lies deep behind its two distinct versions, but that the two versions evolved independently and probably reached their final forms only in the eighth or ninth century. Version A had clear interactions with Babylonian traditions, though he is unsure which text influenced the other. There are major differences between the two ’Abot de-Rabbi Nathan versions of our text. Version B only discusses the requirement to recite shema‘ as one is getting into bed and indicates that failure to recite shema‘ is a capital offense; Version A, in contrast, includes the requirement of going to the synagogue and defines the capital offense specifically as being a transgression against the words of the sages (or the words of Torah, in some manuscripts). Version B only expects serious study from the sages themselves, while version A, like our baraita, invites everyone to participate. This suggests to Kister (1998: 131–132) that the Babylonian Talmud may be citing version A. Therefore, the Talmud is not quoting the shared early tannaitic layer of this text; Version B does not locate the evening shema‘ in the synagogue or even in conjunction with the ‘amidah, suggesting that these elements were not part of the underlying tannaitic scenario and discourse. Version A and our baraita represent much later stages of evolution, one reflecting a different set of values. We can therefore justifiably call this a pseudo-baraita and exclude it from evidence for the Roman-era synagogue.  41

Version A at the end of Chapter 2, and Version B in Chapter 3. Becker 2006: A2: 69–71, 52; B3:1416, 324–325.

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

269

Conclusion When did locating rabbinic prayer in the synagogue become a norm? As Lee Levine (2005: 472, 486–487, 494, 562, 566) correctly notes, the situation is complicated. His reading of the evidence and mine support a finding that rabbinic involvement in the synagogue increased markedly in the third to fourth centuries, and he suggests that this indicates that the synagogue’s rituals then generally met with rabbinic approval. However, both the need to constantly reiterate this approval and occasional specific criticisms also suggest that the process was incomplete. Methodologically, Levine takes us a long way in his careful division of the textual evidence by its layers and his combination of it, where possible, with material evidence. I do not dispute his conclusions. However, he accepts at face value the attributions of texts, considers all baraitot to be tannaitic sources, and is generally not alert to editorial reworkings of the texts. I suggest that there are certain baraitot that must be considered pseudo-baraitot, constructions or reworkings of the Babylonian Talmud (especially) to fit its own narrative. Thus, the traditions in the names of Abba Binyamin and the last baraita about the workers do not fit the pattern of tannaitic teachings, but rather seem to represent understandings of the role of synagogue prayer usually attributed to amoraim of the late third century or even later. They may even reflect the understanding of the talmudic editors themselves, but neither Levine nor I have yet isolated that layer of discourse to determine its agenda on this question. In a future study I hope to pursue such questions past the end of the Roman period and more rigorously outside the borders of the Land of Israel.

Appendices Appendix 1: Jerusalem Talmud Texts42 39 '‫ עמ‬,‫ דף ח טור ב‬,‫תלמוד ירושלמי )ליידן( מסכת ברכות פרק ד ה"ד‬.1 ‫ "בכל‬.‫ ומה טעם‬.‫ צריך אדם להתפלל במקום שהוא מיוחד לתפילה‬.‫ר' אבא ר' חייא בשם ר' יוחנן‬ ‫ אלא "בכל מקום אשר‬.‫ "אשר תזכיר את שמי" אין כת' כאן‬."‫המקום אשר אזכיר את שמי‬ ."‫אזכיר‬ ‫ "ויהי דוד‬.‫ ומה טעם‬.‫ צריך אדם לייחד לו מקום בבית הכנסת להתפלל‬.‫אמ' ר' תנחום בר חנינא‬ ."‫ "אשר השתחוה שם לאלהים" אין כת' כאן אלא "אשר ישתחוה שם לאלהים‬."‫בא עד הראש‬

  42

According to Talmud Yerushalmi 2001.

270

Ruth Langer

1. Jerusalem Talmud Berakhot 4:4, 8b a. Rabbi Abba, Rabbi "iyya in the name of Rabbi Yo#anan: A person must pray in a place that is designated for prayer. What is the reason? “In every place where I mention My name [I will come to you and bless you]” (Exod 20:21). “Where you mention my name” is not what is written here, but “where I mention”. b. Rabbi Tan#um bar "anina said: One must designate for oneself a place in the synagogue for praying. And what is the reason? “When David came to the summit… (2 Sam 15:32). Where God was worshiped is not written here but, where God will be worshiped.” 43 '‫ עמ‬,‫ ה"א‬/ ‫תלמוד ירושלמי )ליידן( מסכת ברכות פרק ה דף ח טור ד‬.2 ."‫ שנ' "סביב רשעים יתהלכון‬.‫חונה אמ' המתפלל אחורי בית הכנסת נקרא רשע‬ ‫ בבית הכנסת בעולם הזה אינו נכנס לבית הכנסת‬43נכנס‬.'‫רב חונה אמ‬ ."‫ "סביב רשעים יתהלכון‬.‫ מה טעם‬.‫לעתיד לבוא‬ .‫ המתפלל בתוך ביתו כאילו מקיפו חומה שלברזל‬.‫אמ' ר' יוחנן‬ ‫ צריך לאדם להתפלל‬.‫ תמן אמ' ר' אבא אמר ר' חייא בשם ר' יוחנן‬.‫מחלפה שיטתיה דר' יוחנן‬ .‫ כאן ביחיד וכאן בציבור‬.‫ וכה אמ' אכן‬.‫במקום שהוא מיוחד לתפילה‬ .‫ מה טעם‬.‫ המתפלל בבית הכנסת כאילו מקריב מנחה טהורה‬.‫ר' פינחס בשם רבי הושעיא‬ ."‫"כאשר יביאו בני יש' את המנחה בכלי טהור בית י'י‬ ‫ בבתי כניסיות ובבתי‬.‫ איכן הוא מצוי‬."‫ "דרשו את י'י בהמצאו‬.‫ בשם ר' אבהו‬44‫ר' אבהו‬ ...‫מדרשות‬ 2. Jerusalem Talmud Berakhot 5:1, 8d a. "una said: One who prays behind a synagogue is called evil, for it says: “On every side the wicked roam”(Ps 12:9). b. Rav "una said: Anyone [who does not enter] the synagogue in this world will not enter into the synagogue in the eschatological future. And what is the reason? “On every side the wicked roam” (Ps 12:9). c. Rabbi Yo#anan said: One who prays in his home, it is as if an iron wall encircles him. d. This reverses the method of Rabbi Yo#anan There (in 1.a) Rabbi Abba cited Rabbi "iyya in the name of Rabbi Yo#anan: A person must pray in a place designated for prayer. Yet here he says this! One refers to prayer as an individual, one to prayer in community. e. Rabbi Pin#as in the name of Rabbi Hoshaya: One who prays in the synagogue, it is as if he offered a pure grain offering. What is the reason? “Just as the Israelites bring an offering in a pure vessel to the House of the Eternal” (Isa 66:20 end). f. Rabbi Abbahu cited Rabbi Abbahu: “Seek the Eternal where He can be found” (Isa 55:6)—where may He be found? In synagogues and study houses.

 43 44

Added based on Ms. Vatican 133. Ms. Leiden editors mark this as a likely error in the tradition.

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

271

Appendix 2: Babylonian Talmud Berakhot 5b–8b (selected passages)45 ‫ וקדם אחד מהם להתפלל‬,‫ להתפלל‬46[‫ לבית הכנסת‬+] ‫ שנים שנכנסו‬:‫ אבא בנימין אומר‬,‫ תניא‬ .‫ טרף נפשו באפו הלמענך תעזב ארץ‬:‫ שנאמר‬,‫ טורפין לו תפלתו בפניו‬- ‫ולא המתין את חברו ויצא‬

47

;‫ ויעתק צור ממקמו‬:‫ שנאמר‬,‫ולא עוד אלא שגורם לשכינה שתסתלק מישראל‬ .‫ שנאמר צור ילדך תשי‬,‫ואין צור אלא הקדוש ברוך הוא‬ A. It is taught, Abba Binyamin says: When two enter [mss. + the synagogue] to pray, and one finishes more quickly and does not wait for his fellow, his prayer is torn up in his face, for it says, “You who tear yourself to pieces in anger—earth will be forsaken because of you?” (Job 18:4a). And not only this, but one causes God’s indwelling Presence to leave Israel, for it says, “Will rocks be dislodged from their place?” (Job 18:4b).48 And there is no Rock except the Holy One, blessed be He, for it says, “You neglected the Rock that begot you” (Deut 32:18a). ‫ לשמוע אל‬:‫ שנאמר‬,‫ אין תפלה של אדם נשמעת אלא בבית הכנסת‬:‫ אבא בנימין אומר‬,‫תניא‬B .‫ במקום רנה שם תהא תפלה‬,‫הרנה ואל התפלה‬ :‫ מנין שהקדוש ברוך הוא מצוי בבית הכנסת שנאמר‬:‫אמר רבין בר רב אדא אמר רבי יצחק‬ .‫ אלהים נצב בעדת אל‬:‫ שנאמר‬- ‫אלהים נצב בעדת אל; ומנין לעשרה שמתפללין ששכינה עמהם‬ B. …It is taught, Abba Binyamin says: a person’s prayer is heard only in the synagogue, for it says, “To hear the praise and the prayer” (1 Kgs 8:28). In a place where there is praise, there may also be prayer. Rabin bar Rav Ada cited Rabbi Yi%#aq: From whence do we know that the Holy One, blessed be He, is found in the synagogue? For it is said, “God stands in the Divine assembly” (Ps 82:1); and whence do we know that when ten pray together, the divine Indwelling Presence is with them? For it is said, “God stands in the Divine assembly” (Ps 82:1). - ‫ כל הרגיל לבא לבית הכנסת ולא בא יום אחד‬:‫ אמר רבין בר רב אדא אמר רבי יצחק‬...C ‫ מי בכם ירא ה' שמע בקול עבדו אשר הלך חשכים ואין‬:‫ שנאמר‬,‫הקדוש ברוך הוא משאיל בו‬ .‫ ואם לדבר הרשות הלך אין נוגה לו‬,‫ נוגה לו‬- ‫י(; אם לדבר מצוה הלך‬:‫נגה לו )ישע' נ‬ .‫ משום דהוה ליה לבטוח בשם ה' ולא בטח‬- ‫ מאי טעמא‬,'‫יבטח בשם ה‬ ,‫ מיד הוא כועס‬- ‫ בשעה שהקדוש ברוך הוא בא בבית הכנסת ולא מצא בה עשרה‬:‫אמר רבי יוחנן‬ .[‫ב‬:‫ מדוע באתי ואין איש קראתי ואין עונה ]ישע' נ‬:‫שנאמר‬ C. …Rabin bar Rav Ada cited Rabbi Yi%#aq: Anyone who habitually comes to the synagogue and does not come one day, the Holy One, blessed be He, inquires about him, as it is said, “Who among you fears the Eternal and obeys the voice of his servant, who

 45

Critical notes are according to variants on the Hachi Garsinan – Bavli site of the Friedberg Jewish Manuscript Society, May 2017 (https://fjms.genizah.org/). ‫ = מ‬Ms. Munich 95; ‫ = פ‬Ms. Paris 671. For specifics on Genizah fragments, consult this site. 46 .‫ ורוב קטעי הגניזה‬,‫ מ‬,‫פ‬ 47 .‫ גניזה כדומה‬,‫ מ וסיים ויצא‬,‫פ וסיים תפילתו ויצא‬ 48 NJPS and NRSV translations of Job 18:4 do not fit the use of it made here.

272

Ruth Langer

walks in darkness and has no light” (Isa 50:10a). If he has “walked” for the sake of performing a commandment, it is light for him, but if for something simply permitted, he “has no light”. What is the reason for “yet trusts in the name of the Lord” (Isa 50:10b)? Because he should have trusted in the name of God and did not. Rabbi Yo#anan said: When the Holy One comes to the synagogue and does not find [a quorum of] ten, immediately He grows angry, for it is said, “Why was no one there when I came? Why did no one answer when I called?” (Isa 50:2). ‫ לא‬:‫ מאי טעמא לא אתי מר לבי כנישתא לצלויי? אמר ליה‬:‫ אמר ליה רבי יצחק לרב נחמן‬...D ‫ ולימא ליה מר לשלוחא‬- .‫ טריחא לי מלתא‬:‫ אמר ליה‬.‫ לכנפי למר עשרה וליצלי‬:‫ אמר ליה‬.‫יכילנא‬ ‫ דאמר‬:‫ מאי כולי האי? אמר ליה‬:‫ אמר ליה‬- .‫ בעידנא דמצלי צבורא ליתי ולודעיה למר‬,‫דצבורא‬ ‫ )מאי דכתיב( ]אין תפלתו של אדם נשמעת אלא בשעה‬,‫רבי יוחנן משום רבי שמעון בן יוחי‬ ‫ בשעה‬- ‫ אימתי עת רצון‬- (‫יד‬:‫ ואני תפלתי לך ה' עת רצון )תה' סט‬49['‫שהצבור מתפללים שנ‬ .‫שהצבור מתפללין‬ D. …Rabbi Yi%#aq asked Rav Na#man, “What is the reason that you, sir, do not come to the synagogue to pray?” He answered, “I am not able to.” He (Rabbi Yi%#aq) suggested, “Then, sir, gather together ten and pray.” He answered, “That is too much bother.” “Then, sir, tell the communal prayer leader that when the community is praying, he should come and inform you.” He responded, “What is all this about?” [Rabbi Yi%#aq] said to him that Rabbi Yo#anan taught in the name of Rabbi Shimon ben Yo#ai: [One’s prayer is only heard at the time when the community is praying, for it is written,]50 “And I will pray to You, Eternal, at a favorable time” (Ps 69:14). When is “a favorable time”? At the time when the community is praying.51 :‫ שנאמר‬,‫ מנין שאין הקדוש ברוך הוא מואס בתפלתן של רבים‬:‫ רבי נתן אומר‬,‫ תניא נמי הכי‬...E ‫ אמר הקדוש ברוך הוא כל העוסק‬.'‫ לי וגו‬- ‫ פדה בשלום נפשי מקרב‬:‫ וכתיב‬,‫הן אל כביר ולא ימאס‬ ‫ מבין אומות‬,‫ לי ולבני‬,‫ מעלה אני עליו כאילו פדאני‬- ‫בתורה ובגמילות חסדים ומתפלל עם הצבור‬ .‫העולם‬ E. …It is also taught thus: Rabbi Natan says: From whence that the Holy One, blessed be He, does not despise the prayers of the multitude? For it is said, “Surely God is mighty and does not despise any” (Job 36:5a); and it is written, “He redeems me unharmed from the battle against me; [it is as though many are on my side] (He redeems Me…, for they were with me in large number)” (Ps 55:19). The Holy One, blessed be He said, “All who engage in Torah study and deeds of lovingkindness and pray with the community, I consider it as if they had redeemed Me, Me and my children from among the nations of the world.” ,‫ נקרא שכן רע‬- ‫ כל מי שיש לו בית הכנסת בעירו ואינו נכנס שם להתפלל‬:‫ אמר ריש לקיש‬F ‫ כה אמר ה' על כל שכני הרעים הנגעים בנחלה אשר הנחלתי את עמי את ישראל; ולא‬:‫שנאמר‬ .‫ הנני נתשם מעל אדמתם ואת בית יהודה אתוש מתוכם‬:‫ שנאמר‬,‫עוד אלא שגורם גלות לו ולבניו‬

 49

.‫פ מ‬ The printed editions of the Talmud have at this point only the question, “What is written?” This sentence, found in Mss. Paris 671 and Munich 95 is stylistically a better reading, and could easily have been eliminated by scribal error. 51 B. Ber. 7b cites this as an elaboration on a tradition attributed to Rabbi Yoanan in the name of the mid-second century tanna, Rabbi Shimon bar Yoai, a tradition not attributed to him in tannaitic sources. It teaches that “anyone who fixes a place in which to pray, his enemies fall before him.” This has obvious similarities to Rabbi Yoanan’s own teaching in the Jerusalem Talmud, Appendix 1, Text 1.a. 50

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

273

F. Resh Lakish said: Anyone who has a synagogue in his city and does not enter there to pray is called a bad neighbor, for it is said, “Thus said the Eternal: As for My wicked neighbors who encroach on the heritage that I gave to My people Israel.” Not only this but he causes exile for Him and his children, for the verse continues, “I am going to uproot them from their soil, and I will uproot the House of Judah out of the midst of them” (Jer 12:14). ‫ למען ירבו ימיכם וימי בניכם על האדמה‬:‫ תמה ואמר‬.‫ איכא סבי בבבל‬:‫אמרו ליה לרבי יוחנן‬G ‫ היינו דאהני‬:‫ אמר‬,‫ מקדמי ומחשכי לבי כנישתא‬:‫ אבל בחוצה לארץ לא! כיון דאמרי ליה‬,‫כתיב‬ .‫ כי היכי דתורכו חיי‬,‫ קדימו וחשיכו ועיילו לבי כנישתא‬:‫ כדאמר רבי יהושע בן לוי לבניה‬.‫להו‬ G. They said to Rabbi Yo#anan, “There are old men in Babylonia!” He was surprised and quoted, “So that your days and the days of your children may be long upon the earth” (Deut 11:21), meaning that this applies only in the Land of Israel but not outside it. When they said to him, “They go early and late to the synagogue”, he replied, “That is what benefits them.” As Rabbi Yehoshua ben Levi said to his children, “Go early and late and enter the synagogue, for this is how you will live long lives.” ‫ אשרי אדם שמע לי לשקד על דלתתי יום יום לשמור‬- ‫ מאי קרא‬:‫אמר רבי אחא ברבי חנינא‬H .‫ כי מצאי מצא חיים‬:‫ כתיב בתריה‬,‫מזוזת פתחי‬ .‫ לעולם יכנס אדם שני פתחים בבית הכנסת‬:‫אמר רב חסדא‬ .‫ ואחר כך יתפלל‬,‫ שיעור שני פתחים‬:‫ אימא‬,‫ אלא‬- ?‫שני פתחים סלקא דעתך‬ H. Rabbi A#a son of Rabbi "anina said: What is the meaning of the verse, “Happy is the one who listens to me, coming early daily to my gates, guarding the posts of my doors?” It is written immediately following, “For the one who finds Me finds life” (Prov 8:34– 35). Rav "isda said: One should always enter two doors into the synagogue. Do you think this means “two doors”? Rather, say, “the width of two doors”, and then pray. ‫ כיון דשמענא להא דאמר רבי‬,‫ מריש הוה גריסנא בגו ביתא ומצלינא בבי כנישתא‬:‫ ואמר אביי‬.I ‫ מיום שחרב בית המקדש אין לו להקדוש ברוך הוא בעולמו אלא‬:‫חייא בר אמי משמיה דעולא‬ .‫ לא הוה מצלינא אלא היכא דגריסנא‬- ‫ארבע אמות של הלכה בלבד‬ ,‫רבי אמי ורבי אסי אף על גב דהוו להו תליסר בי כנישתא בטבריא לא מצלו אלא ביני עמודי‬ .‫היכא דהוו גרסי‬ I. Abaye said: Originally I studied in my house and prayed in the synagogue, but when I heard that Rabbi "iyya bar Abba taught in the name of Ulla, “From the day that the Temple was destroyed, the Holy One, blessed be He, has in His world only the four ells of halakhah”, I have only prayed where I study. Rabbi Ami and Rabbi Assi, even though they had thirteen synagogues in Tiberias, only prayed among the pillars of where they studied.

   

274

Ruth Langer

Appendix 3: Babylonian Talmud Berakhot 4b ‫ אלך‬:‫ בערב ואומר‬53‫ כדי שלא יהא אדם בא מן השדה‬,‫ סייג לדבריהם‬52‫ חכמים עשו‬:‫כדתניא‬ .‫ ואחר כך אקרא קריאת שמע ואתפלל‬,‫לביתי ואוכל קימעא ואשתה קימעא ואישן קימעא‬ ‫ נכנס לבית‬,57‫ אדם בא מן השדה בערב‬56‫; אבל‬55‫ שינה ונמצא ישן כל הלילה‬54‫וחוטפתו‬ ‫ ואוכל‬,‫קורא קריאת שמע ומתפלל‬60‫ ו‬,‫ ואם רגיל לשנות שונה‬,‫ רגיל לקרות קורא‬59‫ אם‬,58‫הכנסת‬ .‫; וכל העובר על דברי חכמים חייב מיתה‬61‫פתו ומברך‬ The sages created a fence62 for their words, so that a person would not come from the field63 in the evening and say, “I’ll go home, eat a bit and drink a bit and take a nap, and afterwards recite shema‘ and pray [the ‘amidah]”—lest sleep overcome him64 and he finds that he has slept through the entire65 night. Instead, a person should come from the field in the evening, enter the synagogue,66 and if he is accustomed to recite, he should recite [scripture]; if he is accustomed to repeat, he should repeat [rabbinic teachings], and recite shema‘ and pray [the ‘amidah], and eat his bread and bless.67 Anyone who transgresses the words of the sages should be sentenced to death.

Bibliography Alexander, P. 2010. “Using Rabbinic Literature as a Source for the History of Late-Roman Palestine: Problems and Issues”. In: Goodman, M. and Alexander, P. (eds.), Rabbinic Texts and the History of Late-Roman Palestine. Proceedings of the British Academy 165. Oxford: Oxford University Press for the British Academy, 7–24. Becker, H.-J. 2006. Avot de-Rabbi Natan: Synoptische Edition beider Versionen. TSAJ 116. Tübingen: Mohr Siebeck. Cohen, B. S. 2017. For out of Babylonia Shall Come Torah and the Word of the Lord from Nehar Peqod: The Quest for Babylonian Tannaitic Traditions. BRLJ 55. Leiden: Brill.

 52

.(‫חיזוק )כמה קטעי גניזה‬ .(‫ממלאכתו )כמה קטעי הגניזה‬ 54 .‫פ שמא יאניסהו; מ ואם הט?פ?תו; ג ואם הטיבתו; גניזה – חסר לגמרי ברובם‬ 55 .‫כולו מ ורוב קטעי הגניזה‬ 56 .‫אלא פ מ כמה קטעי הגניזה‬ 57 .(‫ממלאכתו )כמה קטעי הגניזה‬ 58 .‫או לבית המדרש – כמה קטעי הגניזה‬ 59 .‫היה מ‬ 60 .‫אחר כך פ‬ 61 .‫תחילה וסוף מ‬ 62 Many Genizah fragments read ‫ חיזוק‬instead of ‫סייג‬, i.e., that the rabbis reinforced their words. In some ways, this fits our reading of this as an exhortation better. The use of ‫ סייג‬here may be influenced by m. ’Abot 1:1’s language: ‫( ועשו סייג לתורה‬construct a fence for the Torah). 63 Some Genizah fragments read ‫“ ממלאכתו‬from his work/labor” both here and below. This is the language of ’Abot de-Rabbi Nathan A. 64 Reading with Mss. Paris 671 and Munich 95, whose sense is similar, though the wording differs. Some Genizah fragments lack this phrase. 65 In Ms. Munich 95 and most Genizah manuscripts. 66 Several Genizah manuscripts add: or the study house. This language is also universal in the versions of ’Abot de-Rabbi Nathan A. 67 I.e., recite the Grace after Meals. Ms. Munich 95 adds “before and after”, therefore adding the blessing over the specific food as well. 53

Rabbis, Nonrabbis, and Synagogues in Roman Palestine

275

Cohen, S. J. D. 1999. The Beginnings of Jewishness: Boundaries, Varieties, Uncertainties. Hellenistic Culture and Society. Berkeley: University of California Press. Elbogen, I. 1993. Jewish Liturgy: A Comprehensive History. Trans. R. P. Scheindlin. Philadelphia, PA: Jewish Publication Society. Fleischer, E. and Langer, R. 2000. “Controversy”. Prooftexts 20, 3: 380–387. Fleischer, E. 2012. Statutory Prayers: Their Emergence and Development. Ed. by S. Elizur and T. Beeri. 2 vols. Jerusalem: Magnes Press (in Hebrew). Friedman, S. 1999. “The Primacy of Tosefta to Mishnah in Synoptic Parallels”. In: Fox, H. and Meacham, T. (eds.), Introducing Tosefta: Textual, Intratextual and Intertextual Studies. [Hoboken, NJ:] Ktav, 99–121. Friedman, S. 2000. “Uncovering Literary Dependencies in the Talmudic Corpus”. In: Cohen, S. J. (ed.), The Synoptic Problem in Rabbinic Literature. BJS 326. Providence, RI: Brown Judaic Studies, 35–57. Friedman, S. 2010. Talmudic Studies: Investigating the Sugya, Variant Readings, and Aggadah. New York: Jewish Theological Seminary of America. Gafni, I. 2010. “The Modern Study of Rabbinics and Historical Questions: The Tale of the Text”. In: Bieringer, R., García Martínez, F., Pollefeyt, D., and Tomson, P. J. (eds.), The New Testament and Rabbinic Literature. JSJSup 136. Leiden: Brill, 43–61. Gafni, I. 2011. Rethinking Talmudic History: The Challenge of Literary and Redaction Criticism. Jewish History 25 (3–4): 355–375. Hakohen, A., Rosenson, I., and Weiss, Z. 1992. “Discussion: Goes down before the Ark”. Cathedra 64: 172–179 (in Hebrew). Hauptman, J. 2000. “Mishnah as a Response to “Tosefta””. In: Cohen, S. J. (ed.), The Synoptic Problem in Rabbinic Literature. BJS 326. Providence, RI: Brown Judaic Studies, 13–34. Hezser, C. 2001. Jewish Literacy in Roman Palestine. TSAJ 81. Tübingen: Mohr Siebeck. Jacobs, L. 1971. “Are there Fictitious Baraitot in the Babylonian Talmud?”. HUCA 42: 185– 196. Kister, M. 1997. “Introduction”. In: Avot de-Rabbi Natan, Mahadurat Sh. Z. Shekher: ‘Im Tziyunim la-Mabilot ben ha-Nusaim ela-Tosafot shebe-Mahadurat Shekher. New York: Jewish Theological Seminary of America, 7–41 (in Hebrew). Kister, M. 1998. Studies in Avot D’Rabbi Natan: Text, Editing, and Commentary. Jerusalem: Hebrew University Talmud Department and Yad Ben-Zvi (in Hebrew). Kraemer, D. 1989. “On the Reliability of Attributions in the Babylonian Talmud”. HUCA 60: 175–190. Langer, R. 1999. “Revisiting Early Rabbinic Liturgy: The Recent Contributions of Ezra Fleischer”. Prooftexts, 19, 2: 179–194. Langer, R. 2015. Jewish Liturgy: A Guide to Research. Lanham, MD: Rowman and Littlefield. Langer, R. and Sarason, R. 2019. “Re-Examining the Early Evidence for Rabbinic Liturgy: How Fixed Were Its Prayer Texts?”. In: Calduch-Benages, N., Duggan, M., and Marx, D. (eds.), On Wings of Prayer: Sources of Jewish Worship. Essays in Honor of Professor Stefan C. Reif on the Occasion of his Seventy-fifth Birthday. DCSL 44. Berlin: De Gruyter, 203– 231. Levine, D. 2003. “A Temple Prayer for Fast-Days”. In: Chazon. E. (ed.), Liturgical Perspectives: Prayer and Poetry in Light of the Dead Sea Scrolls. STDJ 48. Leiden: Brill, 95–112. Levine, D. 2009. “Is Talmudic Biography Still Possible?”. Jewish Studies 46: 41–64 (in Hebrew). Levine, L. I. 2005. The Ancient Synagogue: The First Thousand Years. 2nd ed. New Haven, CT: Yale University Press. Lieberman, S. 1937. Tosefeth Rishonim. Jerusalem: Bamberger and Wahrmann (in Hebrew). Lieberman, S. 1992a. The Tosefta. 5 vols. 2nd ed. Jerusalem: The Jewish Theological Seminary of America (in Hebrew).

276

Ruth Langer

Lieberman, S. 1992b. Tosefta Ki-Fshuah: A Comprehensive Commentary on the Tosefta. 8 vols. 2nd ed. Jerusalem: The Jewish Theological Seminary of America (in Hebrew). Mandel, P. 2006. “The Tosefta”. In: Katz, S. T. (ed.), The Cambridge History of Judaism: Volume IV, The Late Roman-Rabbinic Period. Cambridge: Cambridge University Press, 316–335. Miller, S. S. 1999. “The Rabbis and the Non-Existent Monolithic Synagogue”. In: Fine, S. (ed.), Jews, Christians, and Polytheists in the Ancient Synagogue: Cultural Interaction during the Greco-Roman Period. London: Routledge, 57–70. Neusner, J. 2001. “The Synagogue in Law: What the Texts Lead Us to Expect to Find”. In: Neusner, J. and Strange, J. F. (eds.), Religious Texts and Material Contexts. Lanham, MD: University Press of America, 151–173. Neusner, J. 2004. “What Use Attributions? An Open Question in the Study of Rabbinic Literature”. In: Harrington, D., Avery-Peck, A. J., and Neusner, J. (eds.), When Judaism and Christianity Began: Essays in Memory of Anthony J. Saldarini. JSJSup 85. 2 vols. Leiden: Brill, 2:441–460. Newman, H. 2006. “Closing the Circle: Yonah Fraenkel, the Talmudic Story, and Rabbinic History”. In: Kraus, M. A. (ed.), How Should Rabbinic Literature Be Read in the Modern World? Judaism in Context 4. Piscataway, NJ: Gorgias Press, 105–113. Pomeranz, J. 2016. “Ordinary Jews in the Babylonian Talmud: Rabbinic Representations and Historical Interpretations”. PhD thesis, Yale University. Reif, S. C. 1991. “On the Earliest Development of Jewish Prayer (in Response to Ezra Fleischer’s Article)”. Tarbiz 60: 677–681 (in Hebrew). Reif, S. C. 1993. Judaism and Hebrew Prayer: New Perspectives on Jewish Liturgical History. Cambridge: Cambridge University Press. Rubenstein, J. L. 1999. Talmudic Stories: Narrative Art, Composition, and Culture. Baltimore: Johns Hopkins University Press. Safrai, S. 1981. Beshilhei Habayit Hasheini uveTekufat HaMishnah: Perakim BeToldot Ha evrah vehaTarbut (In the Late Second Temple and Mishnaic Periods: Chapters in Social and Cultural History). Ed. by Y. Kohen. Jerusalem: Department of Education, Religious Cultural Branch. Safrai, S. 1989. “Gathering in the Synagogues on Festivals, Sabbaths and Weekdays”. In: Hachlili, R. (ed.), Ancient Synagogues in Israel. Oxford: BAR, 7–15. Tabory, J. 2003. “Ma‘amadot: A Second Temple Non-Temple Liturgy”. In: Chazon, E. G. (ed.), Liturgical Perspectives: Prayer and Poetry in Light of the Dead Sea Scrolls. STDJ 48. Leiden: Brill, 235–262. Talmud Yerushalmi: According to Ms. Or. 4720 (Scal. 3) of the Leiden University Library with Restorations and Corrections [Introduction by Y. Sussman]. 2001. Jerusalem: The Academy of the Hebrew Language. Weiss, Z. 1990. “The Location of the Sheliah Tzibbur during Prayer”. Cathedra 55: 8–21 (in Hebrew). Zunz, L. 1832. Die gottesdienstlichen Vorträge der Juden, historisch entwickelt: Ein Beitrag zur Alterthumskunde und biblischen Kritik, zur Literatur- und Religionsgeschichte. Berlin: A. Asher.

The Origins of Torah Reading as a Ritual and its Social Context

Clemens Leonhard

Introduction The Torah was written in order that somebody—at least the king, who was supposed to possess his own copy, as per Deut 17:17–18—may read it. The book of Deuteronomy contains another explicit self-referential remark in that regard. It must be read every seventh year at the feast of Tabernacles in front of the whole people of Israel (Deut 31:10–12). Normal Israelites would not be in a position to read it. The ideal members of the people of Israel know it (or certain parts of it) by heart. The Torah itself does not mandate anybody to read it every Sabbath, let alone on Mondays and Thursdays. Nevertheless, the people felt compelled to put the Law—or rather parts of it—to uses other than recitation or reading. They inserted (snippets of) Torah text into amulets. They used its texts in other ostensibly apotropaic ways, such as writing it on doorposts and gates (Deut 6:9). Another precept, to talk about “these words” constantly—apparently, not to read them—is ambiguous. A modern, enlightened exegesis may regard this practice as a form of study or as a means to internalize the ethical contents of the text. However, a text that one carries around in amulets may also suggest itself as a powerful spell when uttered. However, is still a far cry from Rabbinic Torah reading practices. Post-biblical (pre-rabbinic and non-rabbinic) sources hardly ever describe any details of the procedure of Torah reading. It is the nature of things that the most explicit sources were written by and addressed to the upper echelons of society or even of the Roman Empire (in the case of Josephus). Observations of high society reading habits establish links between Roman and Jewish intellectual elites’ customs of reading and reciting texts. For an assessment of the less affluent people’s interest in an occupation with Torah reading, one may tap other sources. Thus, one may try to assess who actually read which texts in what kind of social setting. Furthermore, the question arises, which kind of Second Temple reading habits generated the ritualized performance of Torah reading of the synagogue. Since when and under which circumstances was it customary to call up Jewish men to read (certainly only later to watch someone else reading) portions of the

278

Clemens Leonhard

Torah to their congregations (on Sabbaths and on certain other weekdays)? One can answer both questions by extrapolating rabbinic and post-rabbinic reading rituals into Second Temple situations. Hence, any room containing stone benches not only elicits associations of domestic, political, religious, or academic meetings, but also the rabbinic way to celebrate Torah readings. However, the rabbinic and hence medieval and modern ritual of Torah reading may not have been conceived as a ritual from the beginning. More plausibly, it evolved from other forms of social behaviour. This discussion is connected with the study of literacy in Ancient Palestine. Ritualized Torah reading requires a certain level and character of literacy of performers and/or audiences. Thus, Michael Wise has severely criticized earlier approaches, especially Catherine Hezser’s Jewish Literacy in Roman Palestine.1 Wise claims that Hezser misinterprets many sources from the pre-Bar Kokhba period. His suggestion to refrain from reading rabbinic data into earlier times and to reconstruct ancient literacy based on extant data is reasonable. However, in several parts of his argument, Wise himself presupposes a rabbinic, almost modern approach to the reconstruction of ways to read, expound, and memorize the text of the Torah in pre-Bar Kokhba Palestine.2 Wise claims that his ancient protagonists met regularly, read the Torah in an archaizing way, and taught it to the members of their households.3 The contribution of his analyses to the history of literacy in pre-Bar Kokhba Palestine are important because of his use of non-literary sources. However, their explanatory force with regard to the pre-history of the rabbinic way to read and recite the Torah is debatable. The following essay presupposes Wise’s (2015) analysis of literacy regarding the first centuries BCE and CE. Even so, it suggests revisiting his great optimism about the possibility to reconstruct the procedures of congregational Torah reading in ancient Palestine. Additional observations to Wise’s approach make recourse to William A. Johnson.4 The following essay suggests, in comparison with Roman elitist reading habits, how the reading and recitation of the Torah might have evolved. It suggests that the ritualization of the reading of the Torah in Judaism inherits Roman practices regarding the study and discussion of books. In terms of chronology, the present approach draws upon Wise’s (2015: 29) contention that the destruction of Jewish culture in the wake of the Bar Kokhba revolt led to a broader acceptance of Roman ways of life. Thus, the following observations describe the movement of the rabbis in Roman terms. It assumes  1 2

Wise 2015; Hezser 2001. Wise 2015: 49, 58 [following David Carr], 304, 316, 336 [reading the Law in Greek to one’s family],

354. 3 Some of Wise’s 2015 conclusions on ancient Jewish customs of Torah reading are based on Mishnah ’Abot (309, cf. 275, 287). In all cases where Mishnah ’Abot is his only proof text for a phenomenon of rabbinic or even pre-rabbinic times, the argument is spurious. Günter Stemberger (2010b; 2010c; 2010d; 2010e) has shown that Mishnah ’Abot is a post-Talmudic, perhaps anti-Karaite tractate. 4 Cf. Wise 2015: 303–304, referring to Johnson 2010.

Origins of Torah Reading as a Ritual

279

that the later rabbis changed their forbears’ Roman elitist approach with regard to congregational Torah reading. The level of ritualization in the performance of Torah increased when it spread beyond the circles of rabbinic specialists in what one may understand as the age of the Talmud. The following essay claims that the democratization of elitist reading customs eventually led to their ritualization and changed their social function. Nevertheless, it must also briefly address the assumption that congregational Torah reading was at least to some extent always typical for Jewish life in any of its forms and on all of its social strata. If this were true, it would be a survival and not an innovation in Talmudic times. In this respect, it seems rather appropriate to follow the lead of Catherine Hezser rather than Michael Wise. Jewish scholars, intellectuals, and writers composed an abundance of outstanding literature before and after the age of Bar Kokhba, and many people could read and write to some extent. The question is, however, when, why, and how did many of them start to participate in liturgies of Torah reading.

Philo and Josephus Philo refers to the congregational reading of the Torah and its explanation throughout the Sabbath in an apologetic text that has been partly preserved by Eusebius.5 This text emphasizes the accurate tradition and the total devotion of the whole nation of Israel to the Law.6 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”, which is observed among the people … 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 (anaginōskein) 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. … any one of them whom you attack with inquiries about their ancestral institutions can answer you readily and easily.7

The hearers spread the knowledge among the members of their households. Eusebius cuts the tractate at the remark: “This is what I have to say about the seventh day.” In Probus, Philo describes the Essenes as “athletes of virtue produced by a philosophy free from the pedantry of Greek wordiness, a philosophy which sets its pupils to practise themselves in laudable actions …”.8 This group assembles in  5

Cf. Eusebius, Praep. ev. 8 = Philo, Hypoth. 7.13–15 (Colson 1941: 430–437). Cf. Philo, Hypoth. 6.9 (Colson 1941: 420–421). 7 Quotations: Hypoth. 7.12–20 (Colson 1941: 430–437). Cf. for the works of Philo, Stemberger 2010a: 6

30. 8

Philo, Prob. 75–89 (Colson 1941: 52–61), quotation: Prob. 88 (Colson 1941: 60–61).

280

Clemens Leonhard

“holy” places, called synagogues, and has one of them read portions of the Law that are then expounded by “another of especial proficiency”.9 The seventh day is also the topic of a section in De Vita Mosis.10 Moses gave the Sabbath a raison d’être, the “pursuit of wisdom” (philosophein), which includes “thoughts, words, and deeds”. In an aside to a story of Moses’ life, Philo remarks that in his own time, 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 rightly performed?11

Apart from the usual hyperbole, the text does not mention Torah reading, let alone a ritualized form of it. Philo makes use of the same trope of a universal knowledge of the “sacred laws and also the unwritten customs” which are taught to Jews “from the cradle” in Legatio ad Gaium.12 In general, Jews are said to know the Law and more than that. This text does not claim that everybody participates in ritualized reading performances on Sabbaths. Similarly, the Therapeutai and Therapeutrides celebrate their festival banquets on every fiftieth day praying, listening to learned, allegorical explanations of topics “arising in the Holy Scriptures” or other material13 and continue singing and dancing. Each of them possesses books in their houses. However, they study them in solitude. In De Specialibus Legibus, Philo mentions the same dedication of Jews to philosophy and ethics. The Sabbath is the day, 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 teacher’s words supply, while one of special experience rises and sets forth what is the best and sure to be profitable and will make the whole of life grow to something better.14

Apparently following Philo’s apology in Hypothetica,15 Josephus states that Moses decreed “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”, as a uniquely Jewish practice. He continues to claim:

 Prob. 82: ἕτερος δὲ τῶν ἐμπειροτάτων (Colson 1941: 56). Philo, Mos. 2.209–216 (Colson 1935: 552–557). 11 Philo, Mos. 2.216 (Colson 1935: 556–557); similarly, Philo, Opif. 128 (Colson 1929: 100–101). 12 Philo, Legat. 115 (Colson 1962: 56–57) and cf. 210 (Colson 1962: 108–109). 13 Philo, Contempl. 75–78 (Colson 1941: 158–161); similarly, on Sabbaths 30–33 (Colson 1941: 130– 133). 14 Philo, Spec. 2.62 (Colson 1937: 346–347). 15 Cf. Stemberger 2010a: 30. 9

10

Origins of Torah Reading as a Ritual

281

But, should anyone of our nation be questioned about the laws, he would repeat them all more readily than his own name. The result, then, of our thorough grounding in the laws from the first dawn of intelligence is that we have them, as it were, engraven on our souls. A transgressor is a rarity; evasion of punishment by excuses an impossibility.16

Like Philo, Josephus presents Judaism as an outstanding exception with regard to a tremendously broad dissemination of knowledge and corresponding behaviour: “Indeed, most [non-Jewish] men, so far from living in accordance with their own laws, hardly know what they are”.17 The assertion that every Jew knows the (biblical) Law by heart is as trustworthy as the idea that “a transgressor is a rarity” in Judaism—especially in comparison with Greeks and Romans, who do not, allegedly, know any law and who have a higher incidence of crime as a nation. In the rare cases when Josephus describes meetings in synagogues, he does not mention the reading of the Torah (or other biblical texts).18 However, the synagogue is a place where the (books of the) Law is (are) stored.19 Philo’s and less so Josephus’ works offer a wide range of discretion to modern interpreters. On the one hand, one may foreground Philo’s hint at scriptural readings in his Hypothetica. One may also claim that all other remarks about sermons, discourses, and the Jews’ general devotion to the learning of their ancestral philosophy attest to congregational readings of biblical texts. On the other hand, remarks on actual reading are rare and even the virtuosi (the Therapeutai and Therapeutrides) are portrayed as listening to fine rhetoric on scripture instead of reading texts. This latter approach gains credence from studies on roughly contemporary ways of reading scientific texts and belles-lettres among Greeks and Romans (to be discussed below). Both Philo and Josephus blur their own pictures of Judaism considerably through their excessive hyperbole.20 Of course, their assertions are highly plausible as glimpses into modes of behaviour among their peers—i.e., the highest echelons not only of the Jewish but also the Roman society.21 Both Philo and Josephus were in touch with emperors and belonged to the leading families of their people. Apart from all clichés of the ideal philosophers that are claimed for Judaism and its sects and branches, it is obvious that Philo and Josephus talk about their own circles of high standing. As a result, this observation does not elucidate the knowledge of Jews of the Land of Israel, Rome, or of Alexandria with regard to the literary heritage of Judaism. It may even remove Philo and  16

Josephus Flavius, C. Ap. 2.175–178 (Thackeray 1926: 362–365). Josephus Flavius, C. Ap. 2.176 (Thackeray 1926: 362–363). 18 Josephus Flavius, Vita 277–279 (Thackeray 1926: 102–105); cf. Stemberger 2010a: 31. 19 B.J. 2.289–292 (Thackeray 1927: 434–437). 20 Wise 2015: 354 takes Josephus’ claims at face value. 21 Wise 2015: 329 characterizes the Dead Sea scroll deposits as “holdings of serious Judaean intellectuals”. 17

282

Clemens Leonhard

Josephus from the list of witnesses for the reading of Torah among the populace on Sabbaths.

Roman and Jewish Readers It was unfair to Jews that Seneca and Tacitus chided them for wasting a seventh of their lifetime in idleness.22 Tacitus speaks about otium—one’s free time—in this context. If one asked Tacitus and intellectuals of the second century like Gellius (the extremely learned philologist and author of the Attic Nights), Galenus (the polymath, but especially physician and philosopher), Pliny the Elder (natural scientist), Pliny the Younger (a lover of poetry and a statesman), Fronto (the teacher of Marcus Aurelius, philosopher, and statesman), and the emperor Marcus Aurelius, how one should spend one’s otium, all of them would answer: by reading, of course. Any rich and decent man must ideally spend at least a certain part of every day with reading. This activity of reading could be done alone in one’s private chambers, silently and undisturbed by worldly affairs with a view over one’s vineyards and the Mediterranean Sea in the background. Even though one’s negotium—public duties, obligations as a politician or advocate and the like—take absolute priority,23 it was among others a question of morality—not only of good taste—to devote time to reading as well as writing (poetry, for example) according to Pliny the Younger (Johnson 2010: 44). Before becoming emperor, Marcus Aurelius was accustomed to rising early in the morning in order to spend several hours reading before other activities of his day (Johnson 2010: 151–152). However, it is preferable to others to read in the company of one’s friends. Fronto criticizes Marcus Aurelius for his obsessive habit of reading alone and even eschewing the company of his friends when he retreated to a lonely place in the house in order to read a scroll (Johnson 2010: 148–153). Ideally, reading should lead to serious as well as pleasant discussions of philological, historical, medical, etc. issues inspired by the texts. The Roman intellectuals and political leaders were accustomed to bragging about spending several hours each day reading, discussing, and writing material that was not part of their political and professional duties. They did not spend less time of their otium than the Jews who allegedly used the whole Sabbath for that activity. If one of them had refrained from polemics, he could theoretically have asked himself why and which kind of Jews were reading the Torah on the Sabbaths in congregations. The answer would have been easy. Well-off Jews read the Torah, because every intellectual is supposed to spend his free time reading.  22 Seneca’s statement is preserved by Augustine, De Civitate Dei 6.11 (Stern 1974: 431–432 [= no. 186]); Tacitus, Historiae 5.4.3 (Stern 1980: 18, 25, and cf. n. 38 [= no. 281]). 23 Cf. Johnson 2010: 116 pointing to Gellius.

Origins of Torah Reading as a Ritual

283

Intellectual Jews should not have constituted an exception. The accusation that Jews actually waste the time of the Sabbath was nothing but an ethnic slur. Even though silent reading of all genres—oratory, poetry, sciences, history— is the rule (Johnson 2010: 3–16 [introduction]), virtuous intellectuals and leading philosophers are expected all the more so to share their books (as physical objects), their insights, their wisdom, and their reading (the text of their books) with their friends. As rich men, they hired lectors who read books to groups of visitors (as Galen remarks that it is the purpose of such accumulated wealth to be shared among one’s peers; see Johnson 2010: 93–94). This is the reason for the astonishment of Augustine to watch Ambrose sitting by himself and reading silently but surrounded by his adherents including young Augustine (Johnson 2010: 4, 8–9). Nobody is astonished about the mere fact that Ambrose is reading silently. Ambrose offends his entourage, because he would be expected to share not only his insights about a text but also the text itself. The master was supposed to read aloud and engage in conversation with the friends who were present. The difference to modern reading habits is not that the ancients would normally read aloud, let alone that they could only read aloud, but that they would share their reading if their friends were present. Intellectuals read their own poetry to each other at dinner parties or other occasions. Those readings are, however, concentrated on the production and amelioration of texts. The purpose of readings and discussions was the improvement of the written text (Johnson 2010: 51–52). The product was a book. Its author would not read it aloud again in the circle of the friends as soon as he regarded it as finished. A Roman observer of Philo, Josephus, and the rabbis would see a group of intellectuals meeting on the Sabbath—as well as at any other time that permitted them—to divert their attention from their negotium. They are absorbed in the reading and discussion of issues of importance and beauty, like the Law, its exegesis, or poetry. They do this because this is normal for sufficiently wealthy intellectuals. In other words, reading the Torah does not distinguish intellectual Jews from intellectual Greeks and Romans. They share texts and their meanings. Ancient people do not read the Torah aloud because they could not read silently.24 Even the mode of the rabbis’ Torah reading would not unsettle Roman observers. The procedure was mostly recognizable and acceptable for them. They saw a group of rabbis with a highly educated slave—a lector or anagnōstēs—reading in perfect vocalization, intonation, etc., a piece of text until somebody would interrupt him. Then they would discuss a question of interest, a philological problem, or an ethical issue. Instead of a slave, one of the intellectuals themselves could read a portion of a text or quote passages from memory. Nicholas Horsfall (1995) observes that in epochs before the (medieval) invention of eyeglasses,  24

Cf. Wise (2015: 304–305) who refers to sources about Roman ways of the recitation of texts.

284

Clemens Leonhard

especially elder scholars were, in any case, dependent upon a lector (or a friend) for the speedy and accurate consumption of a text. The rabbis may translate passages of the text. Gellius compares Greek texts with their Latin translations (Johnson 2010: 101). The rabbis or the Greeks may continue procedures of their former schooldays, where one would read a certain passage from Homer or from the Torah, explain its philological and ethical implications and conclude the discussion with a paraphrase. They may replace reading this paraphrase with a spontaneous translation into Greek or Aramaic (in the case of the Torah). Like Gellius, they would not translate the Torah into another language, because they did not understand it. Translation was a method to refine and express how one understands a passage. Some of the rabbinic principles regarding the performance of more or less public readings can also be found in the mindset of Roman intellectuals. Thus, the reader does not place a scroll in front of himself, because he does not know the text. Holt N. Parker (2009: 188) emphasizes that ancient readers could avoid reciting a text from memory and hence without a written text in front of them in order to distinguish between an actor in a play (who pretended to be someone, who he is not) and a reader of a text. The lector was supposed to read sitting and to avoid facial expressions (Parker 2009: 203). In this context, one may recall the famous story about Rabbi Meir, who must write down the scroll of Esther from memory in order to be able to perform its reading from a licit scroll.25 Because a licit scroll has been copied from a scroll, Rabbi Meir must copy the scroll that he just wrote from memory in order to read it. Even though he knows the text by heart, the procedure requires that he should read it. The presence of the scroll helps to distinguish reading from acting and establishes the source of the reading in a visual way. Ancient intellectuals quoted texts from memory, too. They knew exactly where to look up a passage in a book. They would produce on the spot and out of their memory concordances of the occurrences of terms with a certain meaning over a wide range of literary sources (Johnson 2010: 105 and its context for Gellius and cf. 114). Roman poets, masters of rhetoric, philosophers, scientists, and philologists would know immense amounts of texts by heart and were well able to handle the comparison of different texts (also in different languages) in an oral discussion. Johnson argues that “bookrolls were not, in gross terms, conceptualized as static repositories of information (or of pleasure), but rather as vehicles for performative reading in high social contexts. Central to the performative reading were questions of status” (Johnson 2000: 616). To some extent, that procedure manifests or establishes the group. For Romans as well as for rabbis, the performance of the discussion was of prime significance, the holdings of libraries were important but secondary.  25

See y. Meg. 4:1, 74d; t. Meg. 2:5 (Lieberman 1962: 349).

Origins of Torah Reading as a Ritual

285

Nevertheless, Gellius and his peers were proud of their access to private and public libraries, and their knowledge of the holdings of those libraries. They mention that a certain text is only extant in a single manuscript in the city of Rome. Galen visits several libraries in order to establish the correct reading of a text by collating all available witnesses (Johnson 2010: 93). The world of the rabbis’ literacy was not essentially different from that of the wealthiest of Roman intellectuals. The rabbis thought and worked in a world where orality and scriptural literacy dovetailed in the activities of scholars. Returning to the culture of reading, the writing of Greek deluxe manuscripts in narrow columns but without spacing between words required professional readers and intellectuals-in-the-making to pass through a long and painstaking period of training (cf. Johnson 2010: 17–31). It would have been easy to write more reader-friendly and likewise beautiful manuscripts. Yet, the difficulty of reading set a higher threshold for the access to the community of readers and was hence welcome as a barrier to ban the uneducated—or rather the uninitiated— from one’s circles. These elites were not interested in a wide dissemination of their habits of reading, understanding, and discussion (Johnson 2000: 615 n. 31). This may be a point of difference to the rabbis’ ideals. The rabbis are aware of this ambiguity of reading as a performance. On the one hand, reading may convey wisdom and competence. On the other hand, it presupposes wisdom and competence. On the one hand, the rabbis’ own prestige was dependent upon a certain degree of wisdom in the populace. They could only act as teachers and exegetes of the Law if many people regarded the knowledge of the Law as desirable. On the other hand, knowledge of the Law is the basis for the delimitation, prestige, and power of their group. Some passages of rabbinic texts express this ambiguity. Thus, the Babylonian Talmud remarks that a gentile who reads the Torah is liable to death. The ensuing statement compares this gentile with a high priest (b. Sanh. 59a). Gentiles may read the Torah, especially in its Greek translation. However, they would not understand it, because they are lacking the necessary supplement: the oral Torah. Furthermore, they lack access to the group of people who possess this supplement. Belonging to the right group is rather a prerequisite for the reading of the Torah than a consequence of this practice. At least one Christian intellectual shared this approach in general terms. Thus, the Ethiopian court official of the narrative in Acts 8 reads a scroll of the prophet Isaiah without understanding it. The text emphasizes that not the reading, but the encounter with the apostle makes him an insider. Baptism is a mere formality in the wake of the encounter. The outsider needs the explanation of the book given by someone from the right circle in order to understand anything. Furthermore, the Ethiopian is by all standards a rich man. He reads and owns a book. He may even have commissioned its writing. In terms of intellectual hierarchy, the Ethiopian is the apostle’s peer.

286

Clemens Leonhard

Lucian’s Insufficient Intellectual buys expensive scrolls (Johnson 2010: 157– 178). He even learned to scan the text with his eyes before pronouncing it orally. Yet, he does not reach a level of sophistication and nuance in his way of reading that would be sufficient for his acceptance by the true intellectuals. He does not understand the text. Only the knowledgeable and learned member of the right social stratum and the appropriate group is able to appreciate the display of wealth and the conveyance of social codes in its context with all subtleties. The ignorant book collector challenges the impermeability of the borders of the group. Lucian collects features of the others’ reading that reveal his inferior status. The possession of books and even the more or less correct performance of the reading cannot replace one’s having been educated in the small world of intellectuals. The level of interest bestowed on a certain text depends upon the intention of the group in the concrete situation. There was, by definition, no rule what to read, let alone how much to read. This recalls the sympotic setting of the rabbinic Seder of Pesach. The Mishnah rules that one should read and expound a certain text from the Torah “until he [apparently, the reader] finished the whole passage.” It is absurd to ask what should have been the end of the passage of the “laws of the Pesach” (sacrifice) or of miqra bikkurim (Deut 26:8–10) in the case of the performance of the Seder (not in the case of the preservation and transmission of the text). The composers of the Mishnah could not know in advance when the host or a learned guest might interrupt the reader or when the host (if he recited the text himself) would want to start a discussion. The same is true for Justin’s description of the Sunday Eucharist in the First Apology where the traditional texts are read “as long as time permits”.26 The group establishes when it is time to stop the recitation according to subtle rules of courtesy and hierarchy. Thus, one must ask to what extent the literary self-description of members of the Roman uppermost class should be representative for the rabbis in a remote province. Three bits of evidence may be adduced here in order to show that ideals to organize one’s social life and spare time had percolated down to less wealthy strata and to other places of the Roman Empire (and in the case of the Babylonian Talmud also far beyond its borders). First, Johnson (2010: 170–175) not only describes and summarizes elite behaviour and aspirations. He also refers to Lucian’s satirical work On Salaried Posts in Great Houses, where Lucian offers a scathing criticism of the wealthy pseudo-intellectual who affords to hire a poor, but learned Greek man in order to pass himself off as a lover of wisdom and philosophy. The satirical work imagines wisdom and learning on the side of the philosopher, not on that of the wealthy member of the Roman upper class. Second, the Christian author Justin the philosopher surrounds himself with a group of adherents who admire him and hence establish him as a philosopher.  26

Justin, 1 Apol. 67.3 (Minns and Parvis 2009: 258, lines 7–8).

Origins of Torah Reading as a Ritual

287

Like their Roman counterparts, they assemble in order to listen to a reader and to learned commentaries upon the readings. Obviously, Justin has assembled a private library containing biblical books of the prophets and texts that resembled the Gospels.27 Apparently, Justin was by no means poor, but there is no reason to think that he could aspire after the status of a senator. Third, a letter from Oxyrhynchus allows a glimpse into the world of learned individuals who are in chase after secondary literature on Greek classical drama—living in Egypt, but not in Alexandria. Johnson (2010: 25) emphasizes that these scholars collect books and share their interests as groups, not just as learned individuals. One may easily imagine a network of circles of rabbis who are interested in the pursuit of halakhah, exegesis, philology, and law. They share information and keep in touch with each other. Furthermore, nineteen papyri (sixteen of which were found in Oxyrhynchus) contain marginal notes that attribute variant readings to named scholars, sometimes spanning generations (based on palaeographical evidence). Johnson observes papyri with different readers’ remarks and punctuation. He concludes that these remarks “may also indicate that members of these intellectual groups repeatedly made performative readings of the text, whether by way of entertainment or as a springboard for discussion” (Johnson 2010: 191). The gain of knowledge provided by this evidence cannot be overestimated. Reading, study, and interpretation of difficult texts are “constructed as a collective endeavor” (192). The group hoi peri Diodōron28—those around Diodorus—is reminiscent of rabbinic study groups.

Rituals, Texts, and Readers Several similarities and differences between the customs of reading among Jews and other inhabitants of the Roman Empire emerge against the backdrop of these observations. Roman intellectuals met in the house of a wealthy member of their circle. They were happy if their host possessed a large and well-selected library. Such libraries were not encyclopaedic, let alone containing unimportant material (Johnson 2013: 359 n. 39). The library of the Pisones in the so-called Villa of the Papyri in Herculaneum was a tightly focused collection (Johnson 2013: 359). The choices of book collectors established canons because they were interested in certain texts and chose to disregard others. Intellectuals also used public libraries. One would enter a public library counting on the service of a knowledgeable slave. He would fetch the desired books from the shelves or bring them from the stacks. Very influential individuals (like the young Marcus Aurelius) could borrow books from such libraries. Others could still bribe the librarian to allow them to take out books. Normally, books  27 28

Cf. Leonhard 2018. Johnson 2010: 181 n. 5 = P.Oxy. XVIII 2192, lines 43–44 (2nd cent. CE).

288

Clemens Leonhard

were supposed to stay in the library. Despite Johnson’s (2013: 363) contention that there is no evidence whatsoever for recitations in the reading rooms of Roman public libraries, they were not quiet places in the second century CE. People would share their reading or discuss questions in small groups. As soon as the furniture of synagogues appeared in images, Torah shrines were depicted as bookcases. One could regard the presence of the Torah scrolls as imparting a kind of holiness to the building (Fine 1997). Nevertheless, considerable time elapsed before congregations began to adorn their physical containers in special ways. This is borne out by ancient literary references to the reading of the Torah in the Temple in Jerusalem. Some sources may be read as reflecting the setting of Roman public libraries in their descriptions of Torah reading procedures. The Mishnah (m. Soah 7:1, 7–8) mentions the reading of the high priest at the Day of Atonement and the haqhel in the seventh year at Sukkot. According to the Hebrew Bible, Ezra brings the Torah in front of the congregation in order to read it. He is standing on a wooden stage during the reading (migdal, Neh 8:2). The high priest of Mishnah Soah receives the Torah from the azzan ha-kneset through the head of the kneset, and the sgan. This description may be intended to increase the pomp of the reading ceremony. Nevertheless, it may also read synagogue and library procedures of Mishnaic times into the history of the Temple. In Luke 4, someone (whose office is not denoted) brings a scroll of Isaiah to Jesus. Luke 4 does not specifically attest to the atmosphere and customs of a first (or rather second) century synagogue in the New Testament. On the contrary, Luke 4 and rabbinic texts reflect public spaces that combine Roman libraries or wealthy patrons’ homes with a meeting place. For example, before reading and sharing one’s insights into the interpretation of the text, one would have a slave bring the scrolls. Different situations prevailed in Second Temple times in Palestine. Following Lee Levine, Michael Wise (2015: 489 n. 64) observes that, with the exception of Masada, “all known Second Temple synagogues lack architectural provision for the storage of book rolls.” He questions the notion of community ownership of Torah scrolls for this period. Yet, this observation does not imply that wealthy owners of books (whose existence is of course not to be doubted) brought books to the synagogue and read them to the “illiterate” peasants.29 Wise assumes that many people were literate. He distinguishes different degrees of literacy. Those degrees include being able to sign documents, to write documents, to read literary texts, etc.30 The differentiation of degrees of literacy diversifies answers to the question of who would have been able to read and/or write in Second Temple Palestine. On the one hand, this diversification confounds the notion that most of the people were simply illiterate. On the other hand, it does not prove that  29 30

Wise 2015: 304–305 and cf. 316. Cf. Wise’s 2015: 59–61 conclusion.

Origins of Torah Reading as a Ritual

289

many of these—to a certain extent literate—people met for the performance of Torah readings on Sabbaths. Modern textbooks refer to a few sources as evidence for the synagogue as the typical place for the reading of the Torah on Sabbaths. The passages of Luke’s Gospel (4:16–22; see above) and Acts (13:15, 21, 27; 15:21) are the only New Testament evidence for the reading of biblical texts in synagogues. 2 Macc 2:13–15 remarks that Judas, following the model of Nehemiah, tried to re-establish the library of the Temple. The Jews of Egypt may make a request for books in Jerusalem. 2 Maccabees assumes that the important books were just not present everywhere in the Diaspora. Certainly, Theodotus may have been a wealthy intellectual. He may have owned a private library in Jerusalem apart from the Temple. This would account for his interest in the study of the Law in Greek (Wise 2015: 335). Otherwise, few people would have owned Torah scrolls. Even people who owned such books did not necessarily perform ritualized readings from the Torah. The proprietors of Torah scrolls read them, in order to understand, learn, teach, and expound the text. This is also true for the early rabbis. One can read many rabbinic texts as traces of the rabbis’ interest in the understanding, memorization, and interpretation of the Torah and other books of ancient Hebrew literature. As long as everybody wanted to understand, learn, teach, and expound the text, the level of ritualization in the reading procedures was likely low. Of course, the Greek and especially Roman readers availed themselves of ritualized acts in their treatment of precious books and performed readings involving a low level of ritualization. The display of wealth, the social games of inclusion and exclusion, and the fostering of an upper class (or philosophical or intellectual) habitus implied subtle ritualized acts. Such readers did not need reading cycles or rules for the handling of scrolls other than precautions in order not do damage them. They did not need a ritual for the performance of the reading. The rules pertaining to the congregational reading of the Torah begin to materialize in a tentative way in the Mishnah and increase in their ritualization and elaboration. Phenomena like reading cycles, literary sermons, and an increasing set of laws how to perform the reading itself evolve after the age of the tannaim. Vestiges of an increased ritualization of the reading culture may point to a fundamental shift that removed the rabbinic ways of reading from Roman elite intellectualism and led to the establishment of the ritual of Torah reading in the synagogue. One decisive step towards the ritualization of the reading of the Torah in the sources would be traces of an allegorical interpretation of this performance. The need to give meaning to the act of reading points to its performance for reasons other than the appreciation of the contents of its texts. The reading of poetry, philosophy, and science by Gellius, Fronto, Galen, Pliny, or Marcus Aurelius had many more or less apparent purposes and social functions. Reading a text may lead to the understanding and/or enjoyment of this text and its meaning(s). The

290

Clemens Leonhard

act of reading per se does not mean anything. It is meaningless. The creation of meaning for a formerly meaningless but purposeful performance points to a loss of straightforward purposes and functions. The loss of an intellectual function of the reading leads to an increase in ritualization and may encourage participants and observers to engage in interpreting the mere act of reading. Formerly, functions determined the performance. If the performance loses functions, celebrants must act according to formal rules in order to create and preserve a performative tradition. As long as people just want to understand a text, somebody may read it aloud, in order to enable them to participate in the discussion of its contents. As soon as the perception of the contents becomes less important or the language incomprehensible, the group must ritually standardize the formal procedure in order to preserve it. The democratization of Torah reading required its ritualization while its ongoing ritualization facilitated its democratization. Ruth Langer points to important steps of this process (Langer 2005; Langer 2006). She collects traces for the interpretation of the reading of the Torah as a representation of the giving of the Law at Sinai. Note that this tradition emerges in the Yerushalmi (y. Meg. 4:1, 74d). It does not belong to the most ancient strata of rabbinic texts. The Torah ceases to be read in order that its legal and historical ramifications, its ethical implications, and its philological intricacies would be taught to its readers and listeners (or just in order that its readers could enjoy its stylistic beauty and wisdom). Not the more or less well-informed participation in the liturgy but the extra-liturgical study makes someone a scholar. The performance of the liturgy acquires meaning. In order to hold a study session in terms of Gellius’ group of philologists one must convene a group of rich, dedicated, creative, and highly educated rabbis. Leading a ritualized Sabbath morning service requires different skills of the precentor. Moreover, one trained precentor is sufficient for the performance.

Concluding Remarks To sum up, surveys of sources about the reading of the Torah in the synagogue should include contemporary Greek and Roman sources about reading habits and customs in the Roman Empire. Comparisons suggest that older rabbinic testimonies (as well as remarks by Philo and Josephus) about customs of Torah reading depict a small elite of highly educated groups who read and discussed texts of the Hebrew Bible. Even if the literacy rate among the inhabitants of Palestine before the Bar Kokhba revolt was high (on different levels), this does not mean that congregational reading of the Torah or other texts of the Hebrew Bible and the Septuagint were widespread. Parallels between the rabbis and their forerunners in Second Temple times point to a general participation in Mediterranean cultures of reading rather than to timeless customs of congregational Torah

Origins of Torah Reading as a Ritual

291

reading. Much later (Talmudic) sources point to a rising dissemination of this elitist culture of reading the Torah in Judaism. When the custom spread among the populace, it changed considerably. Reading cycles and other norms together with the development of meanings of the performance both facilitated the propagation and fostered the ritualization of the reading. Reconstructions of medieval Torah reading liturgies in ancient buildings like the synagogue of Magdala are anachronistic. The furniture, equipment, appurtenances, and spatial organization of the synagogues of this age had functions and reasons. Its decorations may have been associated with meanings. However, one cannot decode these meanings with recourse to the rabbinic liturgies of Torah reading.

Bibliography Colson, F. H. 1929. Philo: On the Creation. Allegorical Interpretation of Genesis 2 and 3. Philo 1. LCL 226. London: Heinemann. Available as DOI 10.4159/DLCL.philo_judaeusaccount_worlds_creation_given_moses.1929 from: https://www.loebclassics.com (accessed 19 August 2019). Colson, F. H. 1935. Philo: On Abraham. On Joseph. Moses I and II. Philo 6. LCL 289. London: Heinemann. Available as DOI 10.4159/DLCL.philo_judaeus-moses_i_ii.1935 from: https://www.loebclassics.com (accessed 19 August 2019). Colson, F. H. 1937. Philo: On the Decalogue. On the Special Laws, Books 1–3. Philo 7. LCL 320. London: Heinemann. Available as DOI 10.4159/DLCL.philo_judaeusspecial_laws.1937 from: https://www.loebclassics.com (accessed 19 August 2019). Colson, F. H. 1941. Philo: Every Good Man is Free. On the Contemplative Life. On the Eternity of the World. Against Flaccus. Apology for the Jews. On Providence. Philo 9. LCL 363. London: Heinemann. Available as DOI 10.4159/DLCL.philo_judaeus-hypothetica.1941 from: https://www.loebclassics. com (accessed 19 August 2019). Colson, F. H. 1962. Philo: On the Embassy to Gaius. Indices to Volumes I–X. Philo 10. LCL 379. London: Heinemann. Available as DOI 10.4159/DLCL.philo_judaeusembassy_gaius_first_part_treatise_virtues.1962 from: https://www.loebclassics. com (accessed 19 August 2019). Fine, S. 1997. This Holy Place: On the Sanctity of the Synagogue during the Greco-Roman Period. Christianity and Judaism in Antiquity Series 11. Notre Dame, IN: The University of Notre Dame Press. Hezser, C. 2001. Jewish Literacy in Roman Palestine. TSAJ 81. Tübingen: Mohr Siebeck. Horsfall, N. 1995. “Rome without Spectacles”. Greece & Rome 42.1: 49–56. Johnson, W. A. 2013. “Libraries and Reading Culture in the High Empire”. In: König, J., Oikonomopoulou, K., and Woolf, G. (eds.), Ancient Libraries. New York: Cambridge University Press, 347–363. Johnson, W. A. 2010. Readers and Reading Culture in the High Roman Empire: A Study of Elite Communities. Classical Culture and Society. New York: Oxford University Press. Johnson, W. A. 2000. “Toward a Sociology of Reading in Classical Antiquity”. AJP 121, 4: 593–627. Langer, R. 2006. “From Study of Scripture to a Reenactment of Sinai: The Emergence of the Synagogue Torah Service”. Journal of Synagogue Music 31: 104–125.

292

Clemens Leonhard

Langer, R. 2005. “Sinai, Zion, and God in the Synagogue: Celebrating Torah in Ashkenaz”. In: Langer, R. and Fine, S. (eds.), Liturgy in the Life of the Synagogue. Studies in the History of Jewish Prayer. Duke Judaic Studies Series 2. Winona Lake: Eisenbrauns, 121–159. Leonhard, C. 2018. “No Liturgical Need for a Gospel in the Second Century”. In: Heilmann, J. and Klinghardt, M. (eds.), Das Neue Testament und sein Text im 2. Jahrhundert. TANZ 61. Tübingen: Narr Francke Attempto, 89–106. Lieberman, S. (ed.) 1962. Tosefta: According to Codex Vienna, with Variants from Codex Erfurt, Ms. Schocken and Edition Princeps (Venice 1521). The Order of Mo‘ed. New York: The Jewish Theological Seminary of America Press. Minns, D. and Parvis, P. 2009. Justin, Philosopher and Martyr: Apologies. Oxford Early Christian Texts. Oxford: University Press. Stemberger, G. 2010a [1996]. “Öffentlichkeit der Tora im Judentum – Anspruch und Wirklichkeit”. In: Stemberger, G. (ed.), Judaica Minora I: Biblische Traditionen im rabbinischen Judentum. TSAJ 133. Tübingen: Mohr Siebeck, 27–37. Stemberger, G. 2010b [2005]. “Mischna Avot – Frühe Weisheitsschrift, pharisäisches Erbe oder spätrabbinische Bildung?”. In: Stemberger, G. (ed.), Judaica Minora II: Geschichte und Literatur des rabbinischen Judentums. TSAJ 138. Tübingen: Mohr Siebeck, 317–330. Stemberger, G. 2010c [1996]. “Die innerrabbinische Überlieferung von Mischna Avot”. In: Stemberger, G. (ed.), Judaica Minora II: Geschichte und Literatur des rabbinischen Judentums. TSAJ 138. Tübingen: Mohr Siebeck, 331–346. Stemberger, G. 2010d [2003]. “‘Moses received Torah …’ (mAv 1,1) – Rabbinic Conceptions of Revelation”. In: Stemberger, G. (ed.), Judaica Minora II: Geschichte und Literatur des rabbinischen Judentums. TSAJ 138. Tübingen: Mohr Siebeck, 347–359. Stemberger, G. 2010e [1998]. “Verdienst und Lohn – Kernbegriffe rabbinischer Frömmigkeit? Überlegungen zu Mischna Avot”. In: Stemberger, G. (ed.), Judaica Minora II: Geschichte und Literatur des rabbinischen Judentums. TSAJ 138. Tübingen: Mohr Siebeck, 434–452. Stern, M. 1974. Greek and Latin Authors on Jews and Judaism, Vol. 1: From Herodotus to Plutarch. Publications of the Israel Academy of Sciences and Humanities. Section of Humanities. Fontes ad Res Judaicas Spectantes. Jerusalem: The Israel Academy of Sciences and Humanities. Stern, M. 1980. Greek and Latin Authors on Jews and Judaism, Vol. 2: From Tacitus to Simplicius. Publications of the Israel Academy of Sciences and Humanities. Section of Humanities. Fontes ad Res Judaicas Spectantes. Jerusalem: The Israel Academy of Sciences and Humanities. Thackeray, H. S. J. 1927. Josephus: The Jewish War, Volume I Books 1–2. Josephus Volume 2. LCL 203. London: Heinemann. Available as DOI 10.4159/DLCL.josephusjewish_war.1927 from: https://www.loebclassics.com (accessed 20 August 2019). Thackeray, H. St. J. 1926. Josephus: The Life. Against Apion. Josephus 1. LCL 186. London: Heinemann. Available as DOI 10.4159/DLCL.josephus-apion.1926 from: https:// www.loebclassics.com (accessed 20 August 2019). Wise, M. O. 2015. Language and Literacy in Roman Judaea: A Study of the Bar Kokhba Documents. ABRL. New Haven, CT: Yale University Press.

IV. Legal, Political, and Cultural Contexts of Ancient Synagogues



“Synagogues” in Ptolemaic and Early Roman Egypt

Kimberley Czajkowski

Introduction Egypt has for a long time occupied an important place in synagogue studies. Not only was it the centre of a large, longstanding diasporan community by the early Roman era,1 but Egypt is the heart of one of the origin theories of the synagogue as an institution.2 The two factors are, of course, not unconnected. A large diasporan community outside of the homeland would seem a natural place for the development of a communal institution. This is helped by evidence for what are typically labelled “synagogues” in a range of sources: literary, epigraphic and documentary, even if archaeological evidence for this region is absent. On the literary side, Philo has perhaps dominated certain aspects of scholarship on the subject of “synagogues” in Egypt, including, as he does, multiple references to Jewish groups and their buildings in Alexandria.3 And yet there are more problems with writing a history of “synagogues” in Egypt than is perhaps sometimes anticipated. First, Philo, though rich in source material, comes from a very specific Alexandrian background. This is a concern when we consider the immense difference in the lives of Alexandrians and the inhabitants of the chōra: the Egyptian countryside was in many ways another world entirely from the Greek city of the capital. Furthermore, in view of the  1 On Jewish immigration to and settlement in Egypt see Tcherikover 1957: 1–5; Barclay 1996: 22–34; Gruen 2002: 68 and Ritter 2015: 113–131 on the origins of the community in Alexandria specifically. The Elephantine papyri attest a Jewish garrison in Egypt in the 5th century BCE (see Cowley 1923; Kraeling 1953, cf. Aimé-Giron 1931: nos. 1, 33, and 78; also now Porten et al. 2011 for an English translation); cf. also Jer 44:1; 46:14 for earlier attestations. Josephus traces the origins of the community in Alexandria back to Alexander the Great at C. Ap. 2.35 and C. Ap. 2.42 and states that they received their rights from the earliest times at A.J. 19.281; this is contradicted at B.J. 2.487 (space is allotted by the successors), and A.J. 12.7–8 where Ptolemy leads them into Egypt and allots them equal rights (ποιήσας ἰσοπολίτας) to the Macedonians; cf. A.J. 12.45–50 on the other rights bestowed by Ptolemy. Let. Aris. 13–14 similarly states that Ptolemy I Soter brought up to 100,000 Jews from Judaea to Egypt (cf. Josephus, C. Ap. 1.186). While the numbers are unreliable, a sizeable community is still plausible: see Modrzejewski 2001: 73 on the numbers. 2 See, for example, Griffiths (1995) and Grabbe (1988), who date this to the third century BCE. 3 See Jutta Leonhardt-Balzer’s chapter in this volume for a detailed examination of Philo’s depiction of the activities that took place in “synagogues”.

296

Kimberley Czajkowski

changes that occurred with the Roman occupation,4 the potential transformation in the status and composition of Jewish groups from the Ptolemaic to the Roman period should also be factored in.5 When we move into the chōra, there are further questions to be raised about the prevalence of Greek culture compared with that of the Egyptian natives, which should in turn be incorporated into our historical construction of the context in which Jews and Jewish groups were operating. In short: Egypt was a land of contrasts, made up of inhabitants and settlements of very different cultural and linguistic backgrounds. Any historical analysis of the role of “synagogues” must take careful account of these variations. But there is also a more fundamental underlying problem, one that is on the surface purely terminological but has deeper implications. In essence: what, exactly, do we mean by “synagogues” in an ancient context in which the term (synagōgē) is very rarely used? While it may seem a banal point to quibble about the use of modern terminology, in fact, this contrast between the ancient and the modern raises the question of our very object of study. If we wish to profile the kinds of Jewish organizations we find in Egypt, then we must make some key decisions about what we are looking for and likely broaden our scope of study to take into account all types of Jewish organizations before determining whether they fit what we think we are discussing under the heading “synagogue”. A further complication is raised by the limited attestation of such organizations in and of themselves: evidence of buildings, rather than for the organizations themselves, are more commonly found, and these are again not called synagōgai. This chapter will trace the evidence for “synagogues”—in the sense of both buildings and organizations—in Egypt in the Ptolemaic and early Roman era, whence most of our evidence derives.6 It will be argued that a rather differentiated picture is needed that takes into account a range of historical constraints: the difference between Alexandria and the rest of Egypt, and the limits this places on the value of Philo’s evidence;7 the limited amount of the evidence for Jewish communities and communal buildings outside of Alexandria; chronological specificities, in particular the marked difference between the Ptolemaic and  4 Bowman and Rathbone 1992 is still a good, emphatic argument for the case that there were extensive changes to the administration of Egypt upon the Roman annexation. For more recent contributions to the debate of the “Romanity” of Roman Egypt, see Capponi 2005; Jördens 2009; Monson 2012 and Rathbone 2013. 5 See Honigman 2016 with regards to the so-called politeuma of the Jews in Alexandria, and the changes in social categorization as a result of the annexation; Czajkowski 2019 looks more specifically at its legal implications. 6 The activities that took place within “synagogues” (often profiled for this region in relation to Philo’s evidence) are not the primary concern of this chapter, in part because they are dealt with elsewhere in this volume: see Leonhardt-Balzer’s chapter. Some passages in Philo sketch what are taken as typical synagogue activities, though these are quite often theoretical or idealizing in nature and the same cautions as those laid out in this paper should be applied if we wish to utilize them as fully representative of activities that occurred in areas beyond Philo’s very immediate, specific locale. 7 Levine 2005: 90 makes brief comments to this effect in limiting the applicability of Philo’s evidence; this should be applied more generally.

“Synagogues” in Ptolemaic and Early Roman Egypt

297

Roman contexts. The overall impression is that we should not be tempted to synchronize our evidence base, and should allow for a rather varied picture in modelling Jewish communal organizations across the Ptolemaic and Roman era.

“Synagogues” in Egypt: Terminology and Evidence First, some definition of the subject matter and terminology is necessary.8 “Synagogue” may have an ancient derivation, but the term is loaded with modern associations. This is a particular problem for Egypt, as synagōgē is extremely rarely used as a terminological designator for a Jewish community, as opposed to a general term for the gathering (i.e., assembly) of a group that might occur within their communal building. In fact, the only such clearly confirmed case is rather late: a translation of a manumission document from Oxyrhynchus, 291 CE, in which an Aurelius and his sister manumit a forty-year-old slave woman and her children.9 Their freedom—costing 14 talents—is paid for by the synagōgē of the Jews through a person called Aurelius Dioskoros. This appears to be marked as a translation, presumably of a Latin original, so the terminology may be viewed in that light.10 It is the late date of the deed that is most notable: by this point, synagōgē had become a technical term, and as a designator was used solely by Jewish organizations.11 We would therefore never expect there to have been any ambiguity at this time: a synagōgē was Jewish. In fact, there are very few cases in Egypt in which synagōgē is ever used as a group’s designation and that group is confirmed to be Jewish. In addition to the late example (above), there is a doubtful case: an inscription on a statue base, dated to the first century BCE or CE, set up by an Artemon son of Nikon to the  8

See Markus Öhler’s chapter in this volume for more details on terminology, with some differences. P.Oxy. IX 1205 (= CPJ III 473), ll.1–8: [ -ca.?- ἑρμηνεία ἐ]λευθ[ερώσε]ως | [Αὐρήλιος - ca.27 - τῆς λα]μπρ[ᾶς καὶ] λαμπροτάτης Ὀξυρυγχειτῶν πόλεως καὶ ἡ ὁμομη-|[τρία ἀδελφὴ Αὐρηλία - ca.23 -]ος γεν[ο]μένου ἐξηγητ[οῦ] βουλευτοῦ τῆς αὐτῆς π[όλ]εως μετὰ κου- | [ράτορος - ca.33 -] παραδόξου Παραμόνην οἰκογενῆ δούλην ἑα[υ]τῶν ὡς (ἐτῶν) μ [κ]αὶ | [τὰ ταύτης τέκνα - ca.22 - οὐλὴ τ]ραχήλῳ ς (ἔτῶν) ι καὶ Ἰακ[ὼ]β ὡς (ἐτῶν) δ μεταξὺ φίλων ἠλευθε-|[ρώσαμεν καὶ ἀπελύσαμεν - ca.17 - ἀπὸ] παντὸς τοῦ πατρωνικοῦ δικαίου καὶ ἐξουσίας πάσης, ἀριθμη-| [θέντων ἡμῖν ὑπὲρ τῆς ἐλευθερώσεως καὶ ἀπολύσ]εως παρὰ τῆς συνα[γ]ωγῆς τῶν Ἰουδαίων διὰ Αὐρηλίων | [Διοσκόρου] (Translation of manumission. We, Aurelius … of the illustrious and most illustrious city of Oxyrhynchus, and his sister by the same mother Aurelia … daughter of … the former exegetes and bouleutēs of the same city, with her guardian … the admirable …, have manumitted and discharged inter amicos our house-born slave Paramone, aged 40 years, and her children … with a scar on the neck, aged 10 years, and Jakob, aged 4 years, … from all the rights and powers of the owner: fourteen talents of silver having been paid to us for the manumission and discharge by the synagōgē of the Jews through Aurelius Dioskorus [P.Oxy. edition translation, slightly adapted]). 10 The first line is fragmentary, meaning the “translation” (ἑρμηνεία) is restored, but this seems obvious in part due to the Latinisms in the text: see the comments of the editor of P.Oxy. Hunt 1912: 241. 11 See Eckhardt in Eckhardt and Leonhard 2018: 67–74 for a detailed examination of the terminology. 9

298

Kimberley Czajkowski

synagōgē (τῇ συναγωγῇ).12 Both names are attested in Jewish contexts. The case against classifying this as a Jewish group is the statue base, which is considered to conflict with aniconic norms, especially Philo’s accounts of opposition to statues in synagogues in the early Roman era.13 The evidence of the statue-base, however, is not at all conclusive: Jewish attitudes to the ban on images varied considerably dependent on location and period.14 The arguments about Philo’s representativeness (below) should also be taken into account here before imposing his interpretation. Ultimately there is simply nothing to judge either way in this case: this might be a Jewish group, it might not. That is the only attestation of the term as a designator for a group that might be Jewish; but furthermore, synagōgē is otherwise never used as a group designator in this era and refers instead simply to an assembly.15 There are attestations of synagōgoi and archisynagōgoi as leaders, though this is a different matter from the use of synagōgē as a group designator. Of these, synagōgos is found as a leader’s title fairly regularly in non-Jewish associations; archisynagōgos is quite rarely attested in Egypt, and in one case the beginning of the term is restored.16  12 CIJ II 1447 (= JIGRE 20): Ἀρτέμων | Νίκωνος π(ροστάτης) | τὸ ια’ (ἔτος) τῇ | συναγωγῇ | [..]ντηκηι. (Artemon, son of Nikon, p[rostatēs ?] for the eleventh year, to the association). The abbreviation π has been read as standing for prostatēs: if correct, this helps little in identifying this as Jewish or not. The participle form is attested in CIJ II 1441 (= JIGRE 24; the Xenephyris village inscription that will be discussed in due course). There is possibly an [ἀρ]χιπροστάτης in line 4 of IGRR I 1077 (JIGRE 18; see n. 46 for full text that follows JIGRE): de Ricci (1903: 430, no. 5) read ]ι πρὸς τὰ τῆς διοι[κήσεως], remarking that πρὸς τὰ τῆς was better than προστάτης; Breccia (1911: 31 no. 47), ]ρι προστάτης διοι[κήσεως εἶπεν]; IGRR has ]ρι προστάτης διο[….εἶπεν]. In general, προστάτης is confirmed, ἀρχιπροστάτης in fact seems unlikely. 13 More generally, on tensions between euergetistic practices and Jewish norms in Josephus (with wider implications), see Schwartz 2010: 80–109. On this inscription, see Lifshitz 1967: no. 98; Mélèze Modrzejewski 2001: 96 also raises the possibility this is a non-Jewish association; Levine 2005: 88–89 argues for identifying this as a Jewish synagōgē. 14 The JIGRE editors suggested Artemon was a Judaizer and the association not necessarily Jewish. The one does not necessarily follow the other. Artemon could have been a convert (rather than a “Judaizer” without converting), and thus engaged in non-typical practices for Jews (i.e., putting up statues) as part of his membership in a Jewish association. This is all speculative: the point remains, there is no way to decide on the final classification of this inscription. 15 See the term in OGIS 737 (= SB V 8929) (112–111 BCE, Memphis) to describe the assembly of the politeuma of Idumaeans; I.Delta I 446 (67 and 64 BCE, Psenamosis) for an assembly of a synodos of farmers; P.Lond. VII 2193 (69–58 BCE, Philadelphia) refers to the meetings, gatherings (συναγωγά) and outings of a synodos of Zeus Hypsistos; I.Delta I 11 (= IGRR I 1095) (29/28 BCE, Kanopos) uses the verb form to describe gathering the association together; SB XXII 15460 (5 BCE, Alexandria), also attested in BGU IV 1137 (6 BCE) to describe the assembly of a synodos of the god emperor Caesar. 16 See IGRR I 1077 (= JIGRE 18) for an archisynagōgos (3 CE, Alexandria): this inscription is typically labelled as Jewish based on the attestation of this official. I.Fayum I 9 (80–68 BCE) also has an archiereus who may also be an archisynagōgos (though only -γωγος is preserved). Synagōgoi in Egypt: I.Fayum II 204 (= SB I 4211) (79 BCE, Krokodilopolis) has a Helenos as the synagōgos of a synodos of Esenchebis; I.Fayum III 205 (51 BCE, Soknopaiou Nesos) attests a synagōgos of the synodos of Snonais; SB XXII 15460 (5 BCE, Alexandria; see n. 15 above) also includes a synagōgos called Primos who is also mentioned in BGU IV 1137; see also I.Delta II 28 (= JIGRE 26) for a synagōgos of a Sambathic synodos (30 BCE–14 CE, Naukratis) (the nature of this group is much debated; see below). See Eckhardt in Eckhardt and Leonhard 2018: 69 n. 29 for the attested terms beyond Egypt.

“Synagogues” in Ptolemaic and Early Roman Egypt

299

The other, a fragmentary honorific decree from Alexandria (3 CE), is another case that is completely unclassified and may or may not be Jewish—there is simply not enough information in the inscription as it stands to decide the issue. Elsewhere, the title is most normally used for Jewish officials, with only a couple of non-Jews using it:17 cautiously we might reason out from there to see these two as leaders of Jewish groups. It remains the case, however, that the group they lead is not preserved as a synagōgē. Put simply: the term synagōgē was not in common use in Egypt as a designation either for Jewish or non-Jewish groups in the late Ptolemaic and early Roman era. In the context of Jewish groups, use of the modern designator “synagogue” should therefore be approached with extreme caution in this region: modern terminology can of course be used, but, in this case, it should be divorced from any assumptions about what is being referred to in the ancient sources. Furthermore, since our evidence about what Jewish organizations did in this period is limited, it might be worth abandoning the term altogether:18 it does not add much of positive value to the scholarly debate. We should also be clear about what our object of study is. “Synagogue” is used in modern scholarship to refer both to Jewish communities and to their buildings. The latter could be taken to indicate the existence of the former: if a group had a communal building, one assumes they met in it. But the distinction should nonetheless be drawn, since the terms used in antiquity are different. Both will be considered below, but the distinction will be maintained as it has a bearing on the amount of information we may extract about Jewish communities. There are a couple of other possible attestations of Jewish groups, which employ different terms from the norm. First, there is the Sambathic synodos attested in Naukratis in 30 BCE–14 CE.19 Synodos is an extremely common designator for associations in general. The problem here is whether to see Sambathic as a Jewish designator: a proselyte context has also been suggested.20 If this is defined as a Jewish group, then they are employing a very common term for an association: there is nothing specifically Jewish about synodos. Politeumata of Jews are also attested at Alexandria and Herakleopolis in the Ptolemaic era. These are, however, unlikely to be the kind of privately organized communities modern scholarship usually understands to come under the “synagogue” label. Instead, under the Ptolemies, they were ethnic groups with a military purpose, strongly  17

See Eckhardt in Eckhardt and Leonhard 2018: 69. See the arguments by Last (2016) for this argument beyond Egypt; see also Eckhardt’s (2017) response. 19 I.Delta II 28 (= JIGRE 26): [ - - Ἀ]μμωνίου συναγωγὸς | [ - - σ]υνόδῳ Σαμβαθικῇ | [(ἔτους) .. Καί]σαρος, Φαμενὼθ ζ' . (. . . son of Ammonios, leader . . . to the Sambathic synodos, in the year of Caesar, Phamenoth 7). 20 See comments on this text in Binder, Runesson, and Olsson 2008: 197; cf. Kraus 1922: 26; see also Horbury and Noy 1992: 44. We might compare with an inscription from Cilicia (OGIS 573) which refers to a group of Sabbatistai with an officer called a synagōgos. 18

300

Kimberley Czajkowski

connected to the state.21 Changes to their status occurred in the Roman era: this will be discussed below for the case of Alexandria, though we have no evidence that the Herakleopolis politeuma survived that long. These politeumata represent state organizations, not privately organized groups;22 we should thus note that Jewish communal organizations could range across the public and private in the Ptolemaic period, and in the former cases, their internal structures were likely to have been state determined. These are of course the minority cases. It is well known that proseuchē is the most commonly attested term used by Egyptian Jewry, though not for gatherings but buildings.23 This implies a building for a specific purpose: praying, as the name suggests, would be a very obvious one. But when we shift our focus thus to buildings, and not group designators, there is also a corresponding alternative vocabulary that is opened up. These are not, however, all that varied. A later papyrus, dating to 113 CE, gives the accounts of a water system in the Arsinoite nome, and potentially includes a different designator for a building of a Jewish group. An archōn of the proseuchē of the Theban Jews is mentioned, but this is followed soon after by an eucheion:24 this has been taken to refer to a Jewish building. This is not specified as it is for the proseuchē but is plausible: the same rate is paid by the eucheion as by the proseuchē and this is calculated monthly, while every other reckoning in the papyrus is daily. This is also a comparatively high sum. Thus, we can also factor in the single attestation of eucheion as a designator for a Jewish building. Perhaps the most significant to note, however, are the references to hiera, particularly in the Heliopolite nome.25 Indeed, Josephus’ claimed record of Onias IV’s letter is as follows:26

 21 See Kruse 2015 and Honigman 2016. Sänger 2013 gives an alternative interpretation that is slightly less convincing. 22 Politeumata will be discussed in more detail on p. 307–308. 23 See Richardson 1996: 94–95 for a summary of terminology (cases are also catalogued in the paper); Levinskaya 1990 on proseuchē as a specifically Jewish term. 24 P.Lond. III 1177 (= CPJ II 432): ο ἀρχόντων Ἰ[ου]δαίων προσευχῆς Θηβαίων μηνιαίω(ν) (δραχμαὶ) ρκη | Παχὼν (δραχμαὶ) ρκ[η], Παῦνι (δραχμαὶ) ρκη, Ἐπεὶφ (δραχμαὶ) ρκη, Μεσορὴ (δραχμαὶ) ρκη, | ιζ (ἔτους) Θὼθ (δραχμαὶ) ρκη, Φαῶφι (δραχμαὶ) ρκη (γίνονται) (δραχμαὶ) ψ[ξη] | εὐχείου ὁμοίως Παχὼν (δραχμαὶ) ρκη, Παῦνι (δραχμαὶ) ρκη, Ἐπεὶφ (δραχμαὶ) ρκη, Μεσο(ρὴ) (δραχμαὶ) [ρκη] | ιζ (ἔτους) Θὼθ (δραχμαὶ) ρκ[η], Φαῶφι (δραχμαὶ) ρκη (γίνονται) (δραχμαὶ) ψξη (From the archontes of the proseuchē of the Theban Jews 128 dr. monthly: Pachon, Payni, Epeiph, Mesore, 17th year, Thoth and Phaophi. Total 768 dr. From the eucheion likewise 128 dr. monthly: Pachon 128 dr., Payni 128 dr. Mesore 128 dr. in the 17th year Thoth 128 dr. and Phaophi 128 dr. Total 768 dr. [CPJ translation, slightly adapted]). 25 Dion 1977: 54–55 argued that this designation was never used by Jews for their meeting places, and that the letter (below) was specific to the situation of military colonists; in particular the Jew writing the letter deliberately picks hiera (Dion 1977: 50–51) to make analogies to Egyptian temples. This is a fair observation, but even on Dion’s reading one Jew at least was prepared to (temporarily?) refer to these places as hiera, and Josephus was willing to reproduce this: they thus merit inclusion. 26 Josephus, A.J. 13.65–66. Translation slightly adapted from the LCL.

“Synagogues” in Ptolemaic and Early Roman Egypt

301

καὶ εἰς Λεόντων δὲ πόλιν τοῦ Ἡλιοπολίτου σὺν τοῖς Ἰουδαίοις καὶ εἰς ἄλλους τόπους ἀφικόμενος τοῦ ἔθνους, καὶ πλείστους εὑρὼν παρὰ τὸ καθῆκον ἔχοντας ἱερὰ καὶ διὰ τοῦτο δύσνους ἀλλήλοις. And when I came with the Jews to Leontopolis in the Heliopolite nome and to other places of our people, and I found that most of them have temples contrary to what is proper, and on account of this they are ill-disposed toward one another.

The nature of these hiera has provoked debate, mainly centred on whether we should understand sacrifice to have taken place therein. If not, the logic runs, they should be understood as “synagogues”.27 They are, it should be noted, compared with Egyptian temples in this passage, and the temple at Leontopolis is meant to replace them. Again, the “synagogue” terminology is at worst a red herring, at best irrelevant. It would be better to note that some kind of Jewish communal building is here claimed to have existed at the time that Onias founded the Leontopolis Temple: the implication of Josephus’ statement is that the new foundation was meant to have made these redundant.28 Thus, we are in the main dealing with references to buildings in Egypt, not the remains themselves, and not standalone attestations of Jewish organizations. The most common designation for the buildings is proseuchē; hiera are also attested in the Ptolemaic era around Leontopolis. None of these are helpfully described as “synagogues” and may, indeed, indicate a range of different buildings, used for different purposes.

Proseuchai in Egypt How widespread were such buildings throughout Egypt? Here, we are primarily dealing with the proseuchai: there is very limited other evidence for Jewish buildings with other names, or for Jewish organizations more broadly. The Herakleopolis politeuma of the second century BCE has already been mentioned as a state organization: Other than that, and the debated cases above, no Jewish organizations are attested.29 Furthermore, even the buildings are in themselves rather scantily attested.  27

Binder 1999: 234–236 argues against sacrifice occurring; also Rutgers 1998: 99 for the assumption these lacked altars; Runesson 2001: 412–413 sees them as temples (i.e., with sacrifices); see also recently Krause (2017) for the possibility these hiera involved sacrifices. 28 Most recently, see Krause 2017 on Josephus’ narrative purpose in this passage; cf. Dion 1977: 50– 51 for an argument for its authenticity. 3 Macc 2:28 also mentions hiera in the context of the persecution of Jews by Ptolemy IV Philopator, on which see Dion 1977: 48–49, who argues that this was a Greek perception; Philo, Deus 8–9 also refers to hiera, though this may be a general argument rather than a specific reference to Jewish buildings. 29 The debated cases are the Sambathic synodos at Naukratis (I.Delta II 28 [= JIGRE 26]; 30 BCE–14 CE, see p. 299 above) and the unspecified synagōgē (CIJ II 1447 [= JIGRE 20]; first century BCE or CE, see pp. 297–298 above).

302

Kimberley Czajkowski

The chōra was of course not uniform, consisting of small villages but also somewhat larger towns. To take the latter first: There are a few attestations of proseuchai or other Jewish buildings in the urban centres. Leontopolis was mentioned above: Onias’ temple of course survived until 73 CE, and the possible existence of Jewish hiera at the time of its foundation has already been raised.30 Athribis, from the archaeological context a fairly large town from the Ptolemaic period onwards, also had a proseuchē in the 3rd or 2nd century BCE. This is attested in a fairly typical honorific inscription, worth giving in full:31 ὑπὲρ βασιλέως Πτολεμαίου | καὶ Βασιλίσσης Κλεοπάτρας, | Πτολεμαῖος Ἐπικύδου, | ὁ ἐπιστάτης τῶν φυλακιτῶν, | καὶ οἱ ἐν Ἀθρίβει Ἰουδαῖοι, || τὴν προσευχὴν | θεῶι ὑψίστωι. On behalf of King Ptolemy and Queen Cleopatra, Ptolemy, son of Epikydes, the chief of police, and the Jews of Athribis dedicated the proseuchē to God Most High.

This strongly resembles typical honorific inscriptions, with the exception that any reference to the king and queen’s divinity has been omitted (see above). The chief of police is involved in the dedication, perhaps suggesting the group was not marginalized. Another proseuchē from Krokodilopolis from roughly the same era is near identical, suggesting that these particular Jewish groups were choosing to participate in a somewhat standard dedicatory practice.32 This is the sum total of our evidence for town contexts. Villages look similar. All securely located attestations are also from the Ptolemaic era, and all are also dedicatory inscriptions: Schedia near Alexandria (3rd century BCE), Xenephyris, again fairly near Alexandria in Lower Egypt (2nd century CE) and Nitriai, a village in the Delta closer to Naukratis (2nd century BCE).33 Two variations may be noted: in Xenephyris, the proseuchē has a pylōn, which has sometimes been taken to suggest an Egyptian influence;34 and the Nitriai inscription mentions that it has “appurtenances” (τὰ συνκύροντα). A papyrus from Alexandrou-Nesos in the

 30

See pp. 300–301 above. CIJ II 1443 (= JIGRE 27). Translation slightly adapted from JIGRE. 32 CPJ III 1532a (= JIGRE 117), 246–221 BCE: ὑπὲρ βασιλέως |Πτολεμαίου τοῦ | Πτολεμαίου καὶ | Βασιλίσσης | Βερενίκης τῆς || γυναικὸς καὶ | ἀδελφῆς καὶ τῶν | τέκνων οἱ ἐν Κροκ[ο]δίλων πόλει Ἰου[δαῖ]|οι τὴν προ[σ]ε[υχ]ή[ν] || κ[αὶ ---] (On behalf of King Ptolemy, son of Ptolemy, and Queen Berenice his wife, and his sister, and their children, the Jews of Crocodilopolis dedicated the proseuchē and …). Note that there is still a proseuchē in Krokodilopolis-Arsinoe in the second century CE (CPJ II 432): see discussion on p. 300 for the eucheion in that papyrus. 33 CIJ II 1440 (= JIGRE 22) (Schedia); CIJ II 1441 (= JIGRE 24) (Xenephyris); CIJ II 1442 (= JIGRE 25) (Nitriai). 34 Griffiths 1987. 31

“Synagogues” in Ptolemaic and Early Roman Egypt

303

Fayum (3rd century BCE) is a little different: this is a petition to the king complaining that a certain Jew, Doritheos, from the same village stole a cloak.35 The relevant part of the papyrus runs as follows:36 When I saw him, he … the cloak in the proseuchē of the Jews … And Lezelmis, a holder of 100 arourai, came to help … the cloak to Nikomachos the nakoros until judgement.

Though this differs from the honorific inscriptions, it in fact gives us no more information than they do. Because the papyrus is fragmentary, it is even unconfirmed whether the nakoros is actually a nakoros of the proseuchē or of somewhere else.37 Thus, this adds to our proseuchai attestations but nothing more. There are a few more references to proseuchai whose location cannot be fixed. The first is mentioned in a fragmentary papyrus, probably from the second half of the first century BCE: this would be a rare specific attestation of a Jewish association in the documentary record since, although the meeting is held in the proseuchē, a synodos is referred to throughout.38 A number of names of officials are also mentioned in this papyrus, suggesting a developed organization. There are limits to what we may do with this data since this is a unique piece of evidence in this region and we do not know where the proseuchē was located, but the attestation of a community with a specific term (synodos) is notable, and it further shows a Jewish community using a common association designator. Another dedicatory inscription, this time by a Papous, his wife and children and from an unknown location, comes from slightly later in the Roman era (1st or 2nd century CE), though it adds no new information.39 There is then another dedicatory inscription from the Ptolemaic era, though this one has the interesting feature of being bilingual (in Latin and Greek): it replaces a previous dedicatory inscription, and also declares the proseuchē inviolate (ἄσυλος).40  35

CPJ I 129 (Alexandrou-Nesos). CPJ I 129, ll. 4–7: αἰσθομένης δέ μου κατε[ -ca.?- ] | [ -ca.?- τὸ ἱμ]άτιον ἐν τῆι προσευχῆι τῶν Ἰουδαίων ἐπιλα- | [ -ca.?- ]ωπους· ἐπιπαραγίνετα[ι] δὲ Λήζελμις, (ἑκατοντάρουρος) | [ -ca.?- τὸ ἱμά]τιον Νικομάχωι τῶι νακόρωι ἕως κρίσεως. 37 Contra Levine 2005: 86; the translators in CPJ also take this as the nakoros of the synagogue in their translation: this cannot be confirmed. Dion 1977: 66 also cautiously notes that the state of the text does not make the relationship of the nakoros to the proseuchē explicit and see Dion 1977: 65–74 for discussion of officials, with extensive reference to other materials. In view of the state of the papyrus, we cannot assume the connection. 38 CPJ I 138. 39 JIGRE 126: Παποῦς οἰκο|δόμηση τὴν | προσευχὴν | ὑπερ αὑτου | καὶ τῆς υν||αικὸς καὶ τ|ῶν τέκνων | (ἔτους) δ' φαρμοῦθι ' (Papous built the proseuchē on behalf of himself and his wife and children. In the 4th year, Pharmouthi 7). 40 JIGRE 125 (replacement from 47–31 BCE): Βασιλίσσης καὶ Βασι|λέως προσταξάντων | ἀντὶ τῆς προσανακει|μένης περὶ τῆς ἀναθέσε|ως τῆς προσευχῆς πλα||κὸς ἡ ὑπογεγραμμένη | ἐπιγραφήτω· [vacat] | βασιλεὺς Πτολεμαῖος Εὐ|εργέτης τὴν προσευχὴν ἄσυλον. | Regina et || rex iusser(un)t (On the orders of the queen and king, in place of the previous plaque about the dedication of the proseuchē let what is 36

304

Kimberley Czajkowski

This is the sum total of our evidence for proseuchai outside of Alexandria. Excluding the politeuma of Herakleopolis, this amounts to six proseuchai in identified locations, and three in unidentified locations over a period of about five centuries (3rd century BCE to 2nd century CE): we might also add the Leontopolis hiera to this. Including Leontopolis, six of the seven located attestations are from the Ptolemaic era (either 3rd or 2nd century BCE); only one (Xenephyris) is Roman. Two of the three from unknown locations are likely Roman era, with the other late Ptolemaic. Thus, the majority of the documentary evidence is from the 3rd or 2nd century BCE: much earlier than Philo was operating. We cannot make much of the village-town divide with such a small sample. It suffices to say that there were proseuchai in some villages and they did erect dedicatory inscriptions in Greek. The language choice might be significant: these few communities were choosing to engage with the ruling dynasty in their own language and they wanted to participate in this kind of exchange. The sample is tiny, and in an area so well-documented as Egypt, this should give us severe pause in emphasizing the importance of “synagogues” throughout Egypt or making any statements about their nature. It is only in the late Ptolemaic or early Roman era that a fragmentary papyrus (CPJ I 138) informs us of anything more than the fact that there was a proseuchē, which might have had a pylōn or “appurtenances”: this rare attestation of a synodos within the proseuchē also gives us a scheme of officials that suggests a highly elaborate, well-structured organization that would fit with what we expect from an associational culture. But otherwise the evidence is notably scant.

Alexandria One objection to the emphasis on the scarcity of the evidence would be that Philo in fact does inform us about activities in proseuchai, and the above therefore neglects the use of a valuable source on the subject. In some senses, this is a valid objection: Philo does give us information about Jewish communal buildings, and some have extracted possible activities of Jewish communities from his philosophical treatises. This means, however, either explicitly or implicitly taking Philo as representative of a wider picture, either for Egypt, or indeed more generally for Diaspora Jews. This, I would suggest, is problematic. Philo is reacting to a quite specific, historical, geographical and most importantly chronological setting: namely, Alexandria in the early Roman era. This context differs greatly from the rest of Egypt, severely limiting the use of Philo’s evidence for a broader understanding of Jewish organizations in the early Roman period and beyond. Furthermore, the date when Philo was writing is key: there is reason to emphasize  written below be written up. King Ptolemy Euergetes (proclaimed) the proseuchē inviolate. The queen and the king gave the order [JIGRE translation]).

“Synagogues” in Ptolemaic and Early Roman Egypt

305

Philo’s description of these groups in a Roman imperial context. Nonetheless, Alexandria and Philo merit some discussion in their own right and certainly form a piece of the Egyptian picture. Philo does indeed constitute the bulk of our evidence for proseuchai and Jewish organizations within Alexandria, but not its entirety. Rabbinic literature laments the destruction of the great basilica at Alexandria,41 often aligned with a synagogue; despite some learned argument for the potential historicity of elements of this description, there is little reason to take this as anything other than an idealized construction in the rabbinic mindset.42 There is also a smattering of epigraphic material: one mostly restored reference to a proseuchē,43 a dedicatory inscription from the second century BCE; one fairly typically honorific inscription from 27 BCE, in which an Alypus made the proseuchē on behalf of the king and queen for the great god.44 The first refers to the appurtenances of the synagogue, suggesting either adjuncts to the building or perhaps land: this might mean a substantial structure. The honorific inscription is notable for not referring to the divinity of the king and queen—a formula that fits with earlier dedicatory inscriptions from elsewhere in Egypt (discussed above)—evidence that Jewish communities adapted the typical honorific formula to suit their own needs. Two further, more questionable inscriptions (mentioned above) might also be added to the mix: the synagōgē to which Artemon makes a dedication, dated to the first century BCE or CE,45 and possibly the archisynagōgos in a fragmentary inscription from Alexandria dated to 3 CE.46 Neither adds any further new detail to our impression of Jewish communal buildings or organizations in  41

See especially t. Sukkah 4:6. See Levine 2005: 91–96 for a more positivist take. The literary similarities to the Temple and the schematization of the picture seem conclusive, notwithstanding the general problems of reconstructing earlier history from later rabbinic testimony (on which see the excellent comments of Alexander 2010). 43 CIJ II 1433 (= JIGRE 9): [---] | [-- θε]ῶι ὑψίστωι | [--τ]ὸν ἱερὸν | περίβολον (?) και] τὴν προσ|[ευχὴν καὶ τὰ συγ]κύοντα (… to God, the Highest … the sacred precinct and the proseuchē and its appurtenances … [JIGRE translation]). 44 OGIS 742 (= CIJ II 1432 = JIGRE 13): [ὑπὲρ] βασ[ιλίσ]|[ση]ς καὶ β[ασι]|[λ]έως θεῶι [με]|γάλωι ἐ[πηκό]ωι Ἄλυπ[ος τὴν] | προσε[υχὴν] | ἐπόει. | (ἔτους) ιεʹ Με[χεὶρ —ʹ] (On behalf of the queen and king, for the great God who listens to prayer, Alypus made the proseuchē in the 15th year, Mecheir [JIGRE translation]). 45 CIJ II 1447 (= JIGRE 20). See n. 12 for the text. 46 IGRR I 1077 (= JIGRE 18): [ἀγαθῆι τύχηι]· | [ἔτους δλʹ(?) Καίσαρος, Ἁ]θὺρ ιηʹ ἐπὶ τῆς γ[ενη]|[θείσης συναγωγῆ]ς τῶν ἀπὸ τῆς Το[—] | [— — —]Σ ὧν ἀρχισυναγω[γος] | [—?— καὶ ἀρ]χιπροστάτης Διοκ[λῆς] |[— — — ἐπειδὴ Β]ρασίδας Ἡρακλε[ίδου] | [— — — τὸ] γλʹ (ἔτος) Καίσαρος ο[․․․] | [— — —] ἐν ἅπασι ἀναστ[ρεφό]|[μενος —ca.8–9—ω]ς καὶ ὑγιῶς ΕΠ[— —] | [— — —]Σ τὴν δαπάνην Π[— —] | [— — — νε]ομηνιακὰς ἡμέ[ρας] | [— — — ἐ]πισκευὰς ἀκολ[ούθως] | [— — —]ΟΥ λόγῳ ἐπὶ ΤΟ[— —] | [— — ․π]οδεξαμενοι[— —] | [— — —] στεφάνῳ ΕΙ[— —] | [— — — καὶ ἄλ]λοις δυσὶ [— —] | [— — —]ΠΙ․[— —] | — — — ([To good fortune!] [Year 34 (?) of Caesar, Ha]thys 18, in the … of those from the …archisynagōgos [Or: -oi ?] … archiprostatēs … since Brasidas son of Kerakleides … 33rd year of Caesar … in all … and soundly … the expense … days … repair …. By word in … crown … with two … [JIGRE translation]) See n. 12 for the problems in restoring archiprostatēs. 42

306

Kimberley Czajkowski

Alexandria, beyond perhaps the possible titles of officials. This is then the extent of what we may extract, that is, not much. Philo’s picture is richer. Proseuchai throughout refer to buildings, as we see in the epigraphic evidence, and are apparently in every section of the city.47 These are the focus of the attacks in 38 CE by the non-Jewish inhabitants of Alexandria, and it is in the proseuchai that Gaius’ statues and portraits were meant to be installed.48 Gaius’ actions are contrasted with the kings, who apparently left the proseuchai alone.49 There is no real reason to doubt Philo on this basic information: Jewish communal buildings, proseuchai, existed and seem to have been numerous in Alexandria. There is a possibility of extracting a little more information from Philo about Jewish organizations in Alexandria more broadly beyond the existence of their buildings. Indeed, the buildings are generally taken to indicate the existence of some sort of organized Jewish groups: they suggest that these groups met regularly at least.50 Such groups have often been compared with associations,51 and indeed the In Flaccum offers some rich pickings for this type of approach. In this work, Philo gives a picture of a thriving associational culture in Alexandria, one that accords with the epigraphic record:52 one of Flaccus’ first actions, of which Philo approved, was also to dissolve these clubs, which are described as centres of drunkenness and excess.53 In general, Philo sets Jewish groups in contrast to  47 Philo, Legat. 132. It should be noted that Philo also uses συναγωγία on one occasion (Legat. 311, when paraphrasing Augustus’ orders to the governors of Asia: this suggests a takeover of Roman imperial terminology rather than local designators); see also Somn. 2.127, where the term is also put in the mouth of an outsider. Συναγωγαί are also equated to τόπους among the Essenes (Prob. 81). 48 Philo, Legat. 132, 134, 137, 165, 191–192; Flacc. 41, 122. See also Legat. 148, where Philo’s opponents before Gaius accuse the Jews of never having made dedications to Augustus in their proseuchai. 49 Philo, Legat. 139. 50 See the comments of Tcherikover 1957: 7–8. 51 Indeed, the “Greco-Roman associations” model has proved especially popular in recent years: on these lines, Jewish groups fit in with a wider associational culture. This has certainly been useful in drawing comparisons, and in setting Jewish diasporan groups in particular firmly within their wider cultural setting, but there are also limits to the uses of this approach. See Benedikt Eckhardt in this volume for more detail. 52 See especially Philo, Flacc. 73: θίασοι κατὰ τὴν πόλιν εἰσὶ πολυάνθρωποι, ὧν κατάρχει τῆς κοινωνίας οὐδὲν ὑγιές, ἀλλ’ ἄκρατος καὶ μέθη καὶ παροινίαι καὶ ἡ τούτων ἔκγονος ὕβρις· σύνοδοι καὶ κλῖναι προσονομάζονται ὑπὸ τῶν ἐγχωρίων (In the city there are clubs with many members, whose common ideal is not something sound, but is neat wine, drunkenness, drunken behaviour and the offspring of these, violence. They are called synodoi and klinai by the natives); cf. Flacc. 137 for Isodorus’ portrayal as a dangerous leader of multiple clubs. The epigraphic evidence supports the idea that there were such clubs in Egypt, most of them engaging in fairly typical activities: dedication of statues, collections of monthly Jews, communal banquets and so on. See the appendices of Arnaoutoglou 2005 for a thorough collection of all collegia in the first centuries BCE and CE in Egypt; politeumata are excluded from the list, most likely as not being viewed as private associations (rather than state institutions in the Ptolemaic era at least). 53 Philo, Flacc. 4: τάς τε ἑταιρείας καὶ συνόδους, αἳ ἀεὶ ἐπὶ προφάσει θυσιῶν εἱστιῶντο τοῖς πράγμασιν ἐμπαροινοῦσαι, διέλυε τοῖς ἀφηνιάζουσιν ἐμβριθῶς καὶ εὐτόνως προσφερόμενος (He dissolved the associations and clubs which were always feasting under the pretext of sacrifices, making a drunken mockery of matters, dealing severely and vigorously with those who resisted.)

“Synagogues” in Ptolemaic and Early Roman Egypt

307

this associational culture: in the Legatio this is done very directly, where the dangerous drunken gatherings of the clubs is contrasted with the “temperance and justice” of the Jewish communities.54 But seeing Jewish groups in Alexandria as simply another kind of association also poses problems: for example, Philo needs the communities to be allowed to continue to be able to meet even after Flaccus’ edict banning the clubs.55 It is possible that Philo’s argument is a fairly sophisticated legal one here, in which the edict banned illegal associations, but legal ones (i.e., like those of the Jews) were allowed to continue.56 Thus, in Philo’s rendering at least, some Jewish groups may have had the status of associations in Alexandria, but he is also very careful to differentiate their activities: in making the comparison he is concerned with the legal status, not a cultural equivalence or with emphasizing integration (quite the opposite). It should furthermore be noted that this legal status is a very specific concern to the Roman period, when restrictions on collegia were tightened. Connected with the status question is the thorny issue of the so-called politeuma at Alexandria.57 This has been central to discussions of the legal status of the Jewish community in the city, though in some ways it is connected with the wider problem of the existence of Jewish organizations within the city. This is now particularly important due to recent arguments about the nature of Ptolemaic politeumata and the way this changed upon the Roman annexation: to keep this brief, politeumata are now typically seen as fundamentally military units under the Ptolemies. When the Romans annexed Egypt, they would have been discontinued in their previous form, and lost their military nature or been attached to the Roman army in some way. Indeed, only three politeumata associated with Egypt are attested for the Roman era, the Alexandrian Judaeans not being one of them (the term is never used for the Alexandrian community in this period).58 This would mean that any such Jewish politeuma in Alexandria would have seen a drastic reduction of status when the Romans took over: one plausible  54 Philo, Legat. 312: μὴ γὰρ εἶναι ταῦτα συνόδους ἐκ μέθης καὶ παροινίας ἐπι συστάσει, ὡς λυμαίνεσθαι τὰ τῆς εἰρήνης, ἀλλὰ διδασκαλεῖα σωφροσύνης καὶ δικαιοσύνης ἀνδρῶν ἐπιτηδευόντων μὲν ἀρετήν, ἀπαρχὰς δὲ ἐτησίους συμφερόντων, ἐξ ὧν ἀνάγουσι θυσίας στέλλοντες ἱεροπομποὺς εἰς τὸ ἐν Ἱεροσολύμοις ἱερόν (For (he said) that these assemblies were not based on drunkenness and intoxication to promote conspiracy, so as to disturb the cause of peace, but were schools of temperance and justice, where men practised virtue and collected the first fruits every year, from which they pay for sacrifices, sending commissioners to convey them to the temple in Jerusalem). 55 Philo, Flacc. 4. 56 See Eckhardt 2018 for this argument based on the language of the edict and comparison with Ulpian, Opinions 4 (Dig. 47.11.2); see also Czajkowski 2019 for the implications of this argument for understanding the status of Jewish groups within the city. 57 The term is only ever used in Let. Aris. 310, never in the Roman era, though modern scholarship tends to use it as a designator throughout. 58 I.Alex.Imp. 74 (3 BCE) attests a politeuma of Phrygians on a statue base found in Pompeii, though the material, language and dating suggest that it was originally set up in Egypt; I.Fayum II 121 (93 CE) attests the reconstruction of a topos for the politeuma set up by Hathotes; I.Alex.Imp. 24 (120 CE) is a politeuma of Lycians in Alexandria. See Zuckerman 1985–1988: 178 for comments.

308

Kimberley Czajkowski

hypothesis has been that it was reduced to the status of a private association.59 This would add a further Jewish organization into the mixture of groups within the city: by Philo’s time, it would then have been one private Jewish association, potentially with a high social status because of its history under the Ptolemies.60 Although there is another politeuma of Jews attested in the Ptolemaic period in Egypt,61 Jewish politeumata are rare and the problems outlined above are quite specific to Alexandria. This was a city that from the Ptolemaic era had been the capital of the country, and its importance correspondingly grew. It was also the centre of Greek culture, the place for education par excellence and indeed by the Roman era was the second city of the empire. The language of operation was Greek, which was not necessarily so in the countryside, where Demotic still operated outside the administration. From the Roman period, the contrast became even more extreme. Alexandrian citizens had immense status and fiscal benefits (immunity from poll tax being the key one), and indeed the separation of the city from the rest of the country is even embodied in the title of the Roman prefect: praefectus Alexandreae et Aegypti. The flourishing of Alexandria made its culture even more separate from much of the rest of its Egyptian territory, and there were few other Greek cities in Egypt.62 This should caution us further about generalizing on the basis of the picture Philo presents for the city. The specificity of the Alexandrian situation is also pertinent when we consider Philo’s treatment of associations and their buildings more broadly. Philo does, admittedly, make references to proseuchai beyond Alexandria, but these are extremely generalizing. These references to proseuchai beyond Alexandria are also typically attestations of their existence globally, not in the Egyptian countryside. So to take Flacc. 47:63 There was reason to fear that people all over the world would take their cue from there and treat their Jewish fellow-citizens outrageously by taking violent measures against their proseuchai and their ancestral customs.

It is quite clear here that Philo envisages proseuchai to be a worldwide phenomenon, though this is so general as not to be particularly useful: Philo attests the presence of Jewish buildings in many cities. He calls them proseuchai¸ probably because that is the name employed in his location. Legat. 346 is also specific on  59

See Honigman 2016 and Kruse 2015. See Czajkowski 2019 for more details. 61 Herakleopolis (P.Polit.Iud.) is well documented (ca. 144–132 BCE); see Cowey and Maresch 2001 for these. 62 See pp. 295–296 for comments on the difference between city and countryside. 63 Flacc. 47: καὶ δέος ἦν, μὴ οἱ πανταχοῦ τὴν ἀφορμὴν ἐκεῖθεν λαβόντες ἐπηρεάζωσι τοῖς πολίταις αὐτῶν Ἰουδαίοις εἰς τὰς προσευχὰς καὶ τὰ πάτρια νεωτερίζοντες. The rest of this part of the In Flaccum also builds on this: Philo states that the rumour of the destruction of the proseuchai at Alexandria will spread elsewhere (Flacc. 45), which does little more than suggest that there were Jews elsewhere and that Philo thought they cared about their fellows in Alexandria. 60

“Synagogues” in Ptolemaic and Early Roman Egypt

309

this matter that proseuchai were in every city, and Gaius started by appropriating those in Alexandria.64 It seems here that Philo is imposing the Alexandrian picture on the rest of the world. We should thus be extremely cautious about transferring our reconstruction of the set up and workings of Jewish communities in Alexandria to the landscape of the rest of Egypt: the evidence detailed above for proseuchai throughout the chōra is extremely limited, and there is little reason to suppose that they hosted the same kind of “association” groups Philo implies for the city. Given that his arguments are also very specific to the era in which he writes, there is good reason to keep Philo and the other Egyptian evidence rather separate.

Conclusion Proseuchai are indeed well attested in Alexandria in the early Roman era: Philo bears witness to a good number of such buildings scattered throughout the city, and it is possible that the buildings themselves could have continued to exist from the time of the Ptolemies. On the other hand, the construction of the status of these groups and the comparison with associations in Alexandria made by Philo seem very contingent on certain realities of the Roman context. Thus, when we move beyond Alexandria, and indeed earlier in history, we should precede with extreme caution. There are acute limits to our evidence that are not always sufficiently noted, and most of this tells us very little about the role of these buildings in Jewish life or how the communities who met in them were organized. In view of the vast amount of documentary evidence from Egypt, the dearth in attestations in the Roman era especially—where the bulk of the material falls until 117 CE—is remarkable: the lack of documentary evidence for Jewish organizations is often noted, but when coupled with the specificity of Philo’s testimony to the situation in Alexandria, more account should be taken of this. This could lead to our questioning just how common proseuchai or Jewish associations really were to Jews, or might make us question whether there were good reasons for such organizations to remain invisible in—for example—official records. We should also perhaps differentiate more greatly between the two periods considering the changes wrought by the Romans when they annexed the region: in Alexandria, this certainly led to changes in legal status for certain groups. We might also question whether the coming of a new imperial power also changes the ways that the Jewish people organized themselves and represented their communities. At the very least this had an impact on those Jewish organizations  Legat. 346: διαφερόντως δὲ πρὸς τὸ Ἰουδαίων γένος, ᾧ χαλεπῶς ἀπεθανόμενος τὰς μὲν ἐν ταῖς ἄλλαις πόλεσι προσευχὰς ἀπὸ τῶν Ἀλεξανδρειαν ἀρξαμενος σφετερίζεται (But it was particularly marked towards the Jewish people. Because of his bitter hatred for it he appropriated the proseuchai in every city, starting with those in Alexandria). 64

310

Kimberley Czajkowski

(politeumata) that had previously been connected to the state. In short, a more differentiated approach to the subject is needed which acknowledges the differences inherent in Egypt’s settlements and people, and faces the limits of our evidence head on. In view of these considerations, we might also abandon the modern term “synagogue” with regard to Egypt as imposing a rather too uniform perspective on a fragmentary picture: the evidence suggests that Jewish communal organization in Egypt was more variegated. We should thus allow for a richer, more complex picture of Jews’ relationship to their surroundings than the “synagogue” model provides.

Bibliography Alexander, P. S. 2010. “Using Rabbinic Literature as a Source for the History of Late-Roman Palestine: Problems and Issues”. In: Goodman, M. and Alexander, P. S. (eds.), Rabbinic Texts and the History of Late-Roman Palestine. Proceedings of the British Academy 165. Oxford: Oxford University Press, 7–24. Aimé-Giron, N. 1931. Textes araméens d’Égypte. Cairo: Impr. de l’Institut français d’archéologie orientale. Arnaoutoglou, I. 2005. “Collegia in the province of Egypt in the 1st century A.D”. Ancient Society 35: 197–216. Barclay, J. M. 1996. Jews in the Mediterranean Diaspora: From Alexander to Trajan. (323 BCE – 117 CE). Edinburgh: T&T Clark. Binder, D. D. 1999. Into the Temple Courts: The Place of the Synagogues in the Second Temple Period. SBLDS 169. Atlanta, GA: SBL. Breccia, E. 1911. Iscrizioni greche e latine. Cairo: Impr. de l’Institut français d’archéologie orientale. Bowman, A. K. and Rathbone, D. 1992. “Cities and Administration in Roman Egypt”. JRS 82: 107–127. Capponi, L. 2005. Augustan Egypt: The Creation of a Roman Province. Studies in Classics 13. New York: Routledge. Cowey, J. M. S. and Maresch, K. 2001. Urkunden des Politeuma der Juden von Herakleopolis (144/3–133/2 v. Chr.) (P. Polit. Iud.). Papyrologica Coloniensia 29. Wiesbaden: Westdeutscher Verlag. Cowley, A. 1923. Aramaic Papyri of the Fifth Century. Oxford: Clarendon Press. Czajkowski, K. 2019. “Jewish Associations in Alexandria?”. In: Eckhardt, B. (ed.), Private Associations and Jewish Communities in the Hellenistic and Roman Cities. JSJSup 191. Leiden: Brill, 76–96. De Ricci, S. 1903. “Bulletin épigraphique de l’Égypte romaine. Inscriptions Grecques (1896– 1902)”. APF 2: 427–452. Dion, P. E. 1977. “Synagogues et temples dans l’Égypte hellénistique”. Science et Esprit 29: 45–75. Eckhardt, B. 2017. “Craft Guilds as Synagogues? Further Thoughts on “Private JudeanDeity Associations””. JSJ 48:2: 246–260. Eckhardt, B. 2018. “Religionis causa? Zur rechtlichen Lage der Vereine ‘fremder’ Götter in der römischen Kaiserzeit”. In: Blömer, M. and Eckhardt, B. (eds.), Transformationen paganer Religion in der römischen Kaiserzeit: Rahmenbedingungen und Konzepte. RVV 72. Berlin: De Gruyter, 113–152.

“Synagogues” in Ptolemaic and Early Roman Egypt

311

Eckhardt, B. and Leonhard, C. 2018. Juden, Christen und Vereine im Römischen Reich. RVV 75. Berlin: De Gruyter. Grabbe, L. L. 1988. “Synagogues in pre-70 Palestine: A Reassessment”. JTS 39: 159–181. Griffiths, J. G. 1987. “Egypt and the Rise of the Synagogue”. JTS 38: 1–15. Gruen, E. 2002. Diaspora: Jews Amidst Greek and Romans. Cambridge, MA: Harvard University Press. Honigman, S. 2016. “The Ptolemaic and Roman Definitions of Social Categories and the Evolution of Judean Communal Identity in Egypt”. In: Furstenberg, Y. (ed.), Jewish and Christian Communal Identities in the Roman World. AJEC 94. Leiden: Brill, 25–74. Hunt, E. S. 1912. The Oxyrhynchus Papyri. Vol IX. Nos. 1166–1223. London: Egypt Exploration Fund Graeco-Roman Branch. Jördens, A. 2009. Statthalterliche Verwaltung in der römischen Kaiserzeit: Studien zum praefectus Aegypti. Historia Einzelschriften 175. Stuttgart: Steiner. Kraeling, E. G. 1953. The Brooklyn Museum Aramaic Papyri: New Documents of the Fifth Century B. C. from the Jewish Colony at Elephantine. New Haven, CT: Yale University Press, for The Brooklyn Museum. Krause, A. R. 2017. Synagogues in the Works of Josephus: Rhetoric, Spatiality, and FirstCentury Jewish Institutions. AJEC 97. Leiden: Brill. Krauss, S. 1922. Synagogale Altertümer. Berlin: Benjamin Harz. Kruse, T. 2015. “Ethnic Koina and Politeumata in Ptolemaic Egypt”. In: Gabrielsen, V. and Thomsen, C. A. (eds.), Private Associations and the Public Sphere. Copenhagen: The Royal Danish Academy of Sciences and Letters, 270–300. Last, R. 2016. “The Other Synagogues”. JSJ 47:3: 330–363. Levine, L. 2005. The Ancient Synagogue: The First Thousand Years. 2nd ed. New Haven, CT: Yale University Press. Levinskaya, I. 1990. “A Jewish Or Gentile Prayer House? The Meaning Of ΠΡΟΣΕΥΧΗ”. TynBul 41.1: 154–159. Lifshitz, B. 1967. Donateurs et fondateurs dans les synagogues juives. Paris: Gabalda. Mélèze Modrzejewski, J. 2001. The Jews of Egypt: From Rameses II to Emperor Hadrian. Trans. R. Cornman, foreword by S. J. D. Cohen. Ebook ed. Skokie, IL: Varda Books. Monson, A. 2012. From the Ptolemies to the Romans: Political and Economic Change in Egypt. Cambridge: Cambridge University Press. Porten, B. et al. 2011. The Elephantine Papyri in English: Three Millennia of Cross-Cultural Continuity and Change. 2nd rev. ed. Atlanta, GA: SBL. Rathbone, D. 2013. “The Romanity of Roman Egypt: A Faltering Consensus?”. JJP 43: 73– 91. Ritter, B. 2015. Judeans in the Greek Cities of the Roman Empire: Rights, Citizenship and Civil Discord. JSJSup 170. Leiden: Brill. Runesson, A. 2001. The Origins of the Synagogue: A Socio-Historical Study. ConBNT 37. Stockholm: Almqvist & Wiksell. Rutgers, L. V. 1998. The Hidden Heritage of Diaspora Judaism. CBET 20. Leuven: Peeters. Sänger, P. 2013. “The Politeuma in the Hellenistic World (Third to First Century B.C.): A Form of Organisation to Integrate Minorities”. In: Dahlvik, J., Reinprecht, C. and Sievers, W. (eds.), Migration und Integration – wissenschaftliche Perspektiven aus Österreich. Jahrbuch 2/2013. Migrations- und Integrationsforschung 5. Göttingen: V&R unipress, 51–68. Schwartz, S. 2010. Were the Jews a Mediterranean Society? Reciprocity and Solidarity in Ancient Judaism. Princeton: Princeton University Press. Tcherikover, V. 1957. “Prolegomena”. In: Tcherikover, V. and Fuks, A. (eds.), Corpus Papyrorum Judaicarum. Volume I. Cambridge, MA: Harvard University Press, 1–111.

312

Kimberley Czajkowski

Zuckerman, C. 1985–1988. “Hellenistic ‘politeumata’ and the Jews: A Reconsideration”. SCI 8–9: 171–185.

Synagogues as Associations in the Roman Empire

Benedikt Eckhardt

Introduction Alongside civic institutions like councils and priesthoods, private associations were important corporate actors in many cities of the Hellenistic and Roman Mediterranean. As there was no normative terminology, they had many different names and fulfilled various functions. What unites them is a set of characteristics defined by modern scholars: membership determined not by birth or wealth, but by a choice on both sides (the member and the association); durable corporate structures including in-group hierarchies and magistracies (unlike ephemeral gatherings); regular meetings for meals or other purposes (unlike mere status groups); and finally, autonomy in the sense that associations made their own decisions, without interference by the state (and without those decisions having an impact on a person’s status as, e.g., a citizen).1 The associational sphere can thus be defined as a distinct space alongside more traditional social systems like the family or civic subdivisions. At the same time, these different social spaces— and the senses of belonging they created—necessarily overlapped to some degree and interacted with each other. The nature of such interaction was shaped by many factors, including reconfigurations of political authority and the changing attitudes of states towards private networks. It can reasonably be asked how Jewish communities (for which συναγωγή is just one of several designations) fit in here.2 As private groups with structures of corporate organization, we should expect them to have shared some, perhaps many features with other associations of their time. Indeed, such similarities are easy enough to recognize. Like other associations, Jewish communities needed spaces for meetings, money to finance their activities and benefactors who invested their own resources in the group, and the number of ways to organize  1

For a similar set of criteria, see Gabrielsen and Thomsen 2015. I use “Jewish” and not “Judaean” in this article, thereby avoiding a debate that seems to be leading nowhere. Unfortunately, it has become acceptable behaviour to attribute sinister motives to “Jud(a)ean”users (Marginalia Review of Books June 24, 2014), or to not even read scholarly works anymore that use “Judaean” instead of “Jew” (TLZ 143 [2018]: 461)—a clear sign that whatever good may originally have come from a debate on terminological precision is now outweighed by its potential to distract from the actual arguments made. 2

314

Benedikt Eckhardt

such things in antiquity was altogether limited.3 The real work starts not with collecting these similarities, but with their interpretation. What follows from the fact that Jewish communities can be described as “associations”? Here, things become much less straightforward. Basic methodological decisions determine both the interpretative approach and the end result of the investigation. Any attempt to engage with the question should thus be prefaced with a justification of the methodology applied. As this is not always the case, it will be useful to demarcate the possible choices first.

On the Two Ways There are two ways to approach the problem of “synagogues” as associations in the Roman empire.4 The first approach focuses on law and status. Were Jewish communities treated according to administrative procedures or even legal precepts that were developed for associations in general? If so, did that happen because they were in fact regarded as the same thing, or because Roman administrators found it advantageous? And if there were differences, how do we explain them? The second approach focuses not on law, but on practice. Did Jewish communities do similar things to other associations, and if so, can we perhaps reconstruct little-known aspects of synagogue history on the basis of association inscriptions? The interpretation of Jewish communities as associations under Roman law is the older approach. It can be traced back at least to François Baudouin’s masterly commentary on Pliny’s Christian letter, made an appearance in the first treatment of Roman professional associations by Gotthilff Schumann, was famously applied by Emil Schürer to the Jews of Rome and has in fact dominated the debate until quite recently.5 This should not surprise us, for the history of scholarship on associations—and of the comparison with Jewish and Christian phenomena— always starts with law. When scholars of the eighteenth and nineteenth centuries started seriously entertaining the idea that Christian communities should be compared with associations of their time, they wanted to clarify the legal status of Christianity, particularly in the Pauline period.6 When serekh ha-yaad was first compared in detail to the statutes of Greek and Roman associations in 1961, the question this was supposed to solve was how the “Qumran community” could have acquired the right to possess land.7 As associations are today (and could be in the Roman period) legally defined and restricted institutions, a focus on legal  3

For these aspects, see Richardson 2004; Barclay 2006; Harland 2015; Last and Harland 2020. The failure of the “Jew/Judaean”-debate notwithstanding, I avoid the term “synagogue” as a synonym for “Jewish communities”, following some of the arguments advanced by Last (2016). 5 Balduinus 1557: 52–53; Schumann 1729: 33; Schürer 1879: 8–10. 6 Haentzschelius 1729; de Rossi 1864: 101–108; Heinrici 1876. 7 Bardtke 1961. 4

Synagogues as Associations in the Roman Empire



matters is one obvious entry point into the debate, not least because legal categories promise a clear-cut result: Jews, Christians, or the members of the yaad were associations according to the law, or they were not. Such arguments are no longer popular. While Jewish and Christian communities are still compared to other associations of their time, the debate has turned to entirely different questions. Much recent work focuses on social realities: groups of people eating in buildings they either rented or owned; always being in need of money and depending on wealthy donors; with largely similar administrative structures; with their own cultic practice of course, but all part of the same Mediterranean cultural koinē, “claiming a place in ancient Mediterranean society”.8 This image can contrast with some traditional assumptions, and is sometimes deliberately pushed to its limits to explicitly do so. Most recently, Richard Last has argued that the conception of “synagogues” as closely-knit ethnic groups preserving Jewish tradition is a scholarly construct rooted in anti-Semitic fantasies.9 Instead, the fully integrated Jews of antiquity formed associations like everyone else, were members in other associations like everyone else, and brought their rites with them to those associations, which makes every association with one or more Jewish members a “synagogue” of the same right as others. There is no room for legal categories here. Denying the relevance of Roman law is in fact one of the cornerstones of the “Greco-Roman Associations” model: while it may be nice to see Roman magistrates occasionally applying legal regulations on associations to Jewish groups, all that really matters is the “etic” perspective, i.e., a scholar looking at something and calling it an “association”.10 Clearly, much is at stake here. The decision for one or the other approach determines the way we think about the position of Jewish communities in the cities of the Roman empire. Recent exchanges of arguments, while highlighting some of the methodological differences, cannot be said to have developed into a fruitful debate.11 The reason seems to be that the different paradigms serve different aims, and prove to be useful in their respective contexts. Still, I believe that some historical conclusions can only be reached if we take the categories available to historical agents into account. The following remarks are thus not concerned with counting the similarities in behaviour (eating together etc.), but with the  8

This image as well as the quotation are taken from Harland 2003; cf. Ascough 2015; Last 2016. Last 2016. 10 At the same time, any attempt to narrow the question down to an “essentialist” or “ontological” approach is now regarded as outdated; the debate is reframed as being not about whether or not Jewish communities were associations, but whether or not they were like them in important respects. But when there are no legal or other categorical differences that mark the boundaries of the “association”, this distinction is of course more rhetorical than actually relevant. 11 Eckhardt 2017b in response to Last 2016, with the reaction in Last and Harland 2020: 6 n. 20; Gruen 2016 in response to Ascough 2015, with the final response by Ascough 2017; see also Kloppenborg in RBL 07/2020 (https://www.bookreviews.org/pdf/13335_14892.pdf, accessed 8 September 2020). Discussion is not facilitated by the regrettable tendency of some scholars to claim that their counterparts “misunderstand”, “fail to grasp” or “miss” even the most basic aspects of the topic. 9

316

Benedikt Eckhardt

question whether Jewish groups were regarded as associations in the Roman empire, and what precisely that would mean.

Jewish Communities as Associations under Rome The questions raised above touch upon many points worthy of detailed discussion that cannot be provided here. In this section, I instead summarize my own reconstruction in five steps. What emerges is an alternative argument on Jewish communities as associations. To demonstrate that our reading of the evidence is largely determined by preconceived assumptions of what to look for, I have relegated all discussion of case studies to the appendix. 1. In the Hellenistic period, Jewish associations were likely organized like others. Especially in trade centres like Piraeus, Rhodes, or Delos, foreigners worshipping their traditional deities had an important part in the spread of private associations. It is plausible to assume that Jews would have participated in this trend. However, there is little evidence to support this. The Samaritans on Delos are our best shot; we may also point to some other cases elsewhere that possibly date to the very late Hellenistic period.12 In Egypt, some Jewish organizations appear very much on a par with others, but there, the common ground seems to be their status as politeumata. As these were ethnic organizations created and ultimately controlled by the Ptolemaic bureaucracy, we cannot treat them as private associations like the ubiquitous groups of Dionysiastai, Sarapiastai vel sim. in Greek cities.13 Perhaps the lack of attestations for Jewish groups in Hellenistic cities can be explained by the lack of a sizeable Jewish population. 2. Caesar and Augustus acknowledged Jewish groups as “legitimate” associations under Roman law. The evidence for this comes from Josephus (Octavian’s Parian letter) and Philo, who seems to be particularly well informed about the Roman legal discourse on collegia licita. Octavian says that just like Caesar, he exempts the Jews from the general ban on associations included in the lex Iulia de collegiis.14 This must mean that Jewish groups, like some other associations that qualified as either very old or “useful”, received the right to “convene, assemble

 12 IJO I Ach66–67 for the Israelites sending first fruits to the temple on Garizim; IJO II 26 from Nysa could be from the first century BCE; CPJ I 138 (late first century BCE) is very fragmentary; see also Josephus, A.J. 14.235 on Sardis. An overview is provided by Baslez 2019. 13 On the πολιτεύματα, see Kayser 2013; Kruse 2015. 14 Josephus, A.J. 14.213–216.

Synagogues as Associations in the Roman Empire

 

and contribute (money)” (coire convenire conferre).15 Those are precisely the features Philo points out as well.16 This is strong evidence that cannot easily be dismissed (that Greek texts do not use the word collegium is obvious). However much Jewish groups may have fitted the Hellenistic paradigm of associations, their organization since 47 BCE took place in a different, decidedly Roman framework. 3. As “legitimate” associations, Jewish groups must have been “useful”. Inscriptions from the first three centuries CE as well as occasional literary sources give us a clear picture of the sort of groups that could receive permission: the large builders’ associations (fabri) that could also be involved in fire-fighting;17 the textile dealers supplying the military;18 specialists needed for certain aspects of civic cult (the dendrophori, “tree carriers” of the Mater Magna cult, but also musicians);19 other professional associations, especially those involved in transport and food supply;20 civic age classes;21 organizations that worshipped the emperor or, in one case, his deified lover;22 and finally, explicitly attested in only one inscription, veterans.23 Some of these were municipal institutions, not private associations.24 They may have received official recognition as a legitimate corporation either because the respective cities were eager to show their desire to be part of the Roman order or because their capacity to act as separate legal entities was seen as advantageous by the parties involved. The other groups were either created by Roman officials or started out as private associations that were later turned into semi-public institutions through the recognition procedure.25 The  15

As in CIL VI 4416 (Rome, Augustean?) regarding the collegium symphoniacorum. The formula used there is (ll. 4–7) quibus | senatus c(oire) c(onvenire) c(onferre) permisit e | lege Iulia ex auctoritate | Aug(usti) ludorum causa. On the abbreviations, see Groten 2015: 293–295. 16 Philo, Legat. 311–313: συνέρχεσθαι ... συνιοῦσι ... συνεισφέρουσι. On his possible citation of the lex Iulia in Flacc. 4, see van Bynkershoek 1719: 254. 17 Asconius Pedianus, Corn. 75; Pliny, Ep. 10.33–34 (where Trajan denies permission); Pan. 54.3; CIL IX 2213 (Telesia; possibly also 5568 from Tolentinum); CIL X 5198 (Casinum); CIL XIV 168 (Ostia), 169, 256 (Portus); AE 1955: 177 (Ostia); AE 1963: 17 (Minturnae). 18 CIL II 1167 (Hispalis; I do not think that AE 1987: 496 actually improves the text); AE 1983: 731 (Solva) also presupposes official recognition. On the centonarii and their work, see Liu 2009. 19 Dendrophori: CIL VI 29691 (Rome); CIL X 3699–3700 (Cumae). Fabri, centonarii and dendrophori could also be addressed together as the “legitimate” collegia of a city, as the designation tria collegia most likely refers to them: CIL V 7811 (Cemenelum). Musicians: CIL VI 4416 (Rome); CIL X 1642–1643, 1647 (Puteoli). 20 Ulpianus, frg. Vat. 233 (bakers in Rome); SEG LXIII 974 (ship owners in Miletus); AE 1955: 175 (grain measurers in Ostia); CIL VI 85 (grain measurers in Rome); AE 1955: 184 (a group of ship owners navigating the Tiber in Ostia); CIL XIV 10 (textile workers in Ostia); CIL XIV 4573 (fullers?); CIL VI 1872 (fishermen and divers in the Tiber, Rome). 21 CIL III 7060 (νέοι of Cyzicus); perhaps TAM II 175 (γερουσία of Sidyma). 22 InscrIt X.5 223 (Brixia, Augustales); AE 2001: 854 (Liternum, Augustales); AE 2011: 203 (cultores Dianae et Antinoi; on this interpretation, see Bendlin 2011). 23 ILOP 129 (Portus). 24 This is clear for the Augustales and the civic age classes. 25 “Semi-public”: De Ligt 2001. For creation by the state, see Pliny’s fabri, Trajan’s bakers, or the ship owners of Miletus (above, n. 17 and 20).

318

Benedikt Eckhardt

jurists’ insistence that legitimate corpora were characterized by their contribution to the common good (utilitas publica) finds confirmation in this pattern, which does not include ethnic associations apart from the Jews, or associations named after deities of only private interest or pastime activities.26 It is this sphere of semipublic institutions with a defined purpose that is implied by the recognition of Jewish groups as “legitimate” associations. 4. The most likely purpose is the administration of Jewish privileges. The members of most legitimate collegia could expect freedom of certain municipal obligations in exchange for duties carried out by them. One of the functions of such organizations thus inevitably was, from a Roman point of view, controlling access to privileges. The number of members was usually limited; wives, children and others could apparently join, but were then distinguished from those who were eligible for privilegia, e.g., because they practiced the craft.27 In the case of veterans, managing privileges derived from a military career (and hedging in a potential threat to the social order) may have been the recognized purpose of the group. From the late Republican period to Augustus, Jews at several places in the Mediterranean world had also received significant privileges.28 The “usefulness” of Jewish associations may well have been seen in the fact that membership regulated access to these privileges. A difference lies in the fact that Jewish privileges, through information sharing and precedent, developed into a set of rights given to Jews anywhere,29 whereas the legitimate collegia received privileges on a purely local basis. Any Jew could hope to avoid appearing in court on the Sabbath, but not every faber received exemption from municipal duties. However, a faber who was admitted into the local collegium fabrum did, and I hold that a Jew who wanted to lay claim to privileges also had to join a Jewish association. The starting point was nevertheless different, and so was the outcome. The empire-wide regulations for a specific ethnic group went beyond the usual scope of legal regulations on collegia and required different concepts. A more plausible analogy for the Jews are the artists and athletes, whose associations were also used for administering privileges, but were created as ecumenical institutions by the Roman administration.30 Perhaps the translocal development of Jewish privileges was not  26 On utilitas publica, see especially Callistratus in Dig. 50.6.6.12. The relevance of Roman laws on collegia has been questioned, because evidence for actual dissolution of associations is sporadic, and many groups existed throughout the Roman period that certainly did not receive official permission as a collegium licitum (Arnaoutoglou 2002). Whatever one may think of the arguments advanced in this regard, they do not apply here. The question is not why so many associations without official recognition were not dissolved (for an answer, see Eckhardt 2018), but what characterized the relatively few that did enjoy such recognition. 27 Thus Callistratus in Dig. 50.6.6.12; for an application see AE 1983: 731 (Solva). This is what likely stands behind seemingly tautological formulations like [colle]gium fabror(um) et qui in eo | [sun]t (CIL VII 11 from Chichester, l. 6–7); scaenici | Asiaticia|ni et | qui in eo|dem cor|pore sunt (CIL XII 1929, Vienne). 28 See Pucci Ben Zeev 1998. They were more unusual than is argued there: Eckhardt 2017a: 22–28. 29 This comes close to the idea of a Magna Charta, despite Rajak 1984. 30 See now Fauconnier 2018 on these associations, and Eckhardt 2019 on the comparison with Jews.

Synagogues as Associations in the Roman Empire

 

foreseen when Caesar recognized Jewish associations as legitimate collegia; it also stands to reason that Jews were not regarded as equally relevant. 5. There is no free lunch. The privilegia of collegia licita went along with munera, obligations. The same mechanism that gave access to privileges also subjected people to state control. Jews were not originally obligated to fulfil services in return for their privilegia. However, it is likely that after 70, the Jewish tax came to be administered by the very associations that also administered Jewish privileges (which were apparently not withdrawn). Professional associations could be used for taxation purposes already in first-century CE Egypt.31 The implications are potentially severe. To legally practice circumcision after Hadrian, Jews would have had to be registered members of a Jewish association and pay the tax. From a Roman perspective, there was no “Judaism” outside associations. These associations were the obligatory organizational expression of belonging to a defeated people. It is unsurprising in this light that συναγωγή (and Latin synagoga) as a corporate designation remained restricted to Jewish groups: after 70 CE, to be confused with a Jewish organization was clearly undesirable.32 What all this means for “godfearers” and similar problems is one question to explore; another would be conversion. The development of translocal structures like the patriarchate and the transition to late antiquity (when associations in general came under much tighter control) also look different from here. That the concept of legitimate corpora was still applied to local Jewish groups in the early third century CE is shown by Caracalla’s rescript concerning the Jews of Antioch.33 It is nevertheless clear that the development of organizational features was not left solely to the individual groups, but that patterns were established that further distinguished the Jewish situation from local associations. Constantine already knew the official Jewish hierarchies, and the old argument that Rome contributed to their development is plausible in light of our findings.

Conclusion The emerging picture is different from the prevailing one that sees Jewish groups simply as “Greco-Roman Voluntary Associations”. They were associations of course, and at least until 70 CE, there were some others like them. However, the idea of a free competition between various “voluntary” or “private” associations  31

See Verboven 2016: 196; Langellotti 2016: 125–126. The evidence for συναγωγή as a common designator for associations does not stand up to scrutiny. The only certain non-Jewish examples are I.Perinthos 49a and 59 from the first century CE. The frequent designation ἀρχισυνάγωγος for a leader of an association in Macedonia is never coupled with συναγωγή, but with other corporate designators. I discuss the terminology in Eckhardt and Leonhard 2018: 67–74. 33 Cod. justin. 1.9.1 (213 CE) on the universitas Iudaeorum. The term universitas is used by jurists primarily for public institutions (like cities); Groten (2015: 47–71) even thinks that it could not be used for private associations, but this cannot be proven. 32

320

Benedikt Eckhardt

including Jewish groups cannot easily be realigned with the development sketched out here. Essential pieces are missing if we say that Jewish communities were “Greco-Roman Associations”, because the model presupposes that the Hellenistic situation (argument 1 above) already tells the whole story, when in fact it is merely a prelude. It certainly remains possible, by focusing on in-group activities, to see Jewish groups in the Roman empire as associations that did not essentially differ from associations in Hellenistic Greece. The observations supporting this are not wrong. The question is how important they really are when it is simultaneously possible to argue that Jewish associations in the Roman period were semi-public institutions with official functions, not open to everybody, and the obligatory context for legally practicing Judaism—something very different from Dionysiastai or Sarapiastai in Athens, Rhodes or Delos. In my view, a description of Jewish groups as associations that leaves out this second aspect contributes to misleading ideas about their role in society, and hence about very basic realities of what membership in such a group entailed. The comparison with other associations has much to offer, but erasing the differences ironically marginalizes its relevance. There are modes of (modern) observation that make all group life in antiquity look similar. But this does not mean that all group life took place under similar conditions, and one may wonder what is gained by pretending that it did.

Appendix: Case Studies To some degree, the reconstruction developed above is theoretical. It is based on a narrow understanding of the sources on the legal recognition of Jewish groups (argument 2), and then works from there. It therefore leads to the question how these assumptions play out “on the ground”, i.e., in studies of specific Jewish groups in specific places of the Roman world. Much of the evidence usually adduced in the debate on synagogues as associations is not helpful here, because the internal structure of public, semi-public and private institutions could look very similar, and the same is true for their activities. The mere observation that a group had magistrates or a dining hall does not prove any argument right or wrong. We thus cannot hope to find incontrovertible evidence for or against the model proposed here. What we can do is look at specific cities and see how Jewish groups do or do not fit in with the (inevitably fragmentary) evidence for local associational culture. The expectation would be that Jewish communities (unlike the Christians) had a solid legal standing, could possess land and collect money, and were perceived as durable organizations, the continued existence of which was more or less guaranteed. However, we should not expect to find much interaction with outsiders. As Jewish communities operated under different conditions than other associations, this difference should have left some traces in the

Synagogues as Associations in the Roman Empire



respective inscriptions. I offer here a broad overview of possible test cases, taken from different regions of the empire, and moving from East to West.34

Asia Minor As Asia Minor has already received much attention in debates on Jewish groups as “Greco-Roman Associations”,35 discussion in our context can be limited to a few observations on the most important phenomena. Leaving aside the late antique τόπος-inscriptions and two questionable cases of euergetism towards Jewish groups,36 the data that actually allow for comparison are largely limited to inscriptions dealing with grave care. The most instructive cases are perhaps Hierapolis in Phrygia and Nikomedeia in Bithynia. The first has been discussed many times because of a famous sarcophagus inscription: Publius Aelius Glykon gave money to two professional associations, the purple dyers and the carpet weavers, to finance commemorative celebrations on the feast of unleavened bread, the feast of the fiftieth day and the feast of the kalends.37 While various speculations have been offered, the situation seems reasonably clear in light of parallel inscriptions from Hierapolis: a man—who happened to be a Jew—had turned to established groups of a semi-public nature to ensure commemoration.38 In Hierapolis, many associations appear as beneficiaries of such endowments; they could thus assume a similar role to the imperial fiscus and the council of elders.39 All of these groups are professional associations; Jewish organizations are not among them. There was an organized Jewish community that could be named as guardian of a tomb, but these seem to be internal arrangements with  34 I skip Egypt as it is treated in detail by Kimberley Czajkowski in this volume and usually does not provide us with relevant information for the period after 117 CE. However, it should be noted that two new papyri of the early 4th century CE (Balamoshev 2017 and P.Oxy. LXXXIII 5364) strongly suggest that the reorganization of professional associations as obligatory tax units in the tetrarchy also applied to Jewish communities, which supports the reconstruction developed above. The Near East (Syria and Palestine) offers no relevant epigraphic data: the Theodotus inscription is no evidence for an association and in any case likely dates before 70 CE. The donor lists from Dura Europos cannot easily be compared to other material from that city, and the Roman context of Dura is a question in itself. 35 See Harland 2003; Stebnicka 2015: 54–60; Last 2016; Eckhardt and Leonhard 2018: 83–89. 36 For the τόπος-inscriptions, see IJO II 15–16 (Sardeis) and 37–39 (Miletus). On the (supposed) euergetism of Tation (IJO II 36 from Kyme or Phokaia) and Julia Severa (IJO II 168 from Akmoneia) see Eckhardt and Leonhard 2018: 78–80; neither case is clear evidence for a Jewish group attracting external benefactors, and there are no others that could be cited. 37 IJO II 196. 38 There is no reason to think that the purple dyers or carpet weavers were Jewish, that Glykon was a member of one or even both associations, or (as has been suggested at the Münster conference) that Glykon himself was not Jewish. The last point in particular is certainly possible (the inscription would then be irrelevant to our question), but the likelihood of someone celebrating the feast of unleavened bread and not identifying as Jewish does not seem very high. 39 Conveniently accessible in Ritti 2016.

322

Benedikt Eckhardt

no potential for outreach.40 The Jewish συναγωγή of Hierapolis clearly had a legitimate position within the city, and the fact that it was regarded as a possible guardian indicates that it was seen as a stable, durable organization. But if people like Glykon wanted to be commemorated in a civic context, they turned to other groups. The case of Nikomedeia is different because the inscriptions do not in themselves offer much information. The argument here is of a statistical nature. The number of inscriptions designating certain institutions as guardians of a tomb (by specifying the sum to be paid to them by violators of the tomb) is high enough to be certain about the pattern: such institutions could be the imperial fiscus, the city (or one of its subdivisions like a φυλή), a village—or a Jewish συναγωγή.41 The Jewish organization does not appear alongside private associations, but alongside other institutions that determined a person’s cultural and political identity. If we turn to the large centres of associational activity in Ionia, the picture does not change significantly. Despite the early attestation of Jewish communities in Ephesus through the Roman decrees preserved by Josephus, inscriptions are rare. Jews are designated as guardians of a tomb in two cases, but this in itself is irrelevant for the argument as long as the persons concerned were Jews, which is certain in one case and likely in the other.42 There is no evidence for interaction with outsiders or engagement with civic society—the typical reasons for Ephesian associations to put up an inscription.43 Smyrna offers a similar amount of data. Passing over a late antique building inscription set up by a πρεσβύτερος and “father of the tribe” (πατὴρ τοῦ στέμματος in analogy to pater synagogae in Western inscriptions),44 three inscriptions reveal aspects of Jewish organization in the city. Two of them are grave inscriptions set up by functionaries, an ἀρχισυνάγωγος and a “secretary of the people in Smyrna” (γραμματεὺς τοῦ ἐν Ζμύρνῃ λαοῦ); in the first case, “the ἔθνος of the Jews” appears alongside the Roman fiscus as recipient of fines for grave violation.45 While the hierarchies themselves are not unusual, such terminology is not found in other associations; the communicative context is clearly internal. The third inscription is a list of contributors to public building works that includes a contribution of 10,000  40

IJO II 205. That the obligatory copy of the decree is deposited not in the civic archive (as in all other cases from Asia Minor), but in a specific “archive of the Jews” clearly sets this arrangement apart from the ones involving other groups. 41 For the Jewish cases, see IJO II 154, 157, 158; for the argument, see Eckhardt and Leonhard 2018: 83–84. 42 IJO II 33 for a priest, IJO II 32 for an ἀρχιατρός (“chief-physician”). The title is explained by Ameling in IJO II: 155–156. Last (2016: 355) connects him to the association of physicians, but this confuses the fiscal title ἀρχιατρός with the leader of an association of ἰατροί and in any case would not add much to our discussion; cf. Eckhardt 2017b: 253. 43 An overview over associations in Ephesus can be found in Rohde 2012: 275–350. 44 IJO II 41. 45 IJO II 43, 44.

Synagogues as Associations in the Roman Empire



Denarii by the “former Jews” (οἱ ποτὲ Ἰουδαῖοι).46 This unique designation has caused some debate. If the reference was merely to a place of origin (“those formerly from Judea”),47 Greek epigraphic formulae would have offered numerous standard phrases to be used in this inscription, set up by civic magistrates. The point of the complicated designation must be that the donors formed a group formerly known as Ἰουδαῖοι but no longer so designated. It is tempting to connect this development with the Jewish tax (people no longer obligated to pay it?), but every solution must remain speculative. It is perhaps enough to note that the only example from Asia Minor (or indeed from the Roman empire as a whole) for a potentially Jewish association acting in an euergetic manner towards the city concerns a group that is decidedly designated as no longer Jewish. Jewish inscriptions have thus routinely been compared to a broad idea of “Greco-Roman Associations”, but a closer look reveals important differences even in those cases from Asia Minor that are cornerstones of the argument. It remains to be seen how the argument works in other regions that are less often at the centre of debates on “synagogues as associations”.

Macedonia In Macedonia, most attention has naturally been paid to the synagogue inscription from Stobi. In it, Claudius Tiberius Polycharmus, the “father of the συναγωγή at Stobi”, records that he has “built the houses for the sacred place, and the triclinium with the tetrastoon”, from his own resources.48 This can be regarded as a typical example of in-group euergetic practice, although Polycharmus did not (at least on this occasion) receive an honorific decree by the community as one might expect. The immediate context is purely local, but Polycharmus seems to refer to translocal structures on two occasions: He praises himself for having “held all the offices according to Jewish custom”, which seems to presuppose a general concept of how Ἰουδαϊσμός should be organized, and he states that anyone who modifies the rules set out here shall have to pay the patriarch (presumably in Palestine) 250,000 Denarii.49

 46

IJO II 40. As argued by Kraabel (1982: 455) and Harland (2003: 202). Cf. criticism by Cohen (1999: 78). Hallmannsecker 2017: 117–120 now proposes to change the text to οἱ τότε Ἰουδαῖοι, “the Jews who were present at the time”; this would of course change the interpretation significantly, but remains entirely hypothetical (Hallmannsecker assumes that the stonemason “committed the slip of reading the Ι of OI in his sketch together with the following Τ as a Π”, but we would still have to assume additional dittography, as I is present on the stone). 48 IJO I Mac1; cf. I.Stobi 20–23 for smaller inscriptions relating to Polycharmus. 49 Ll. 6–9: ὅς πολειτευσάμε|νος πᾶσαν πολειτεί|αν κατὰ τὸν Ἰουδαϊσ|μὸν. My translation is based on the explanation by Ameling (2016: 138–141), who then prefers a slightly different interpretation. 47

324

Benedikt Eckhardt

This translocal perspective certainly distinguishes the συναγωγή from other associations at Stobi. Excluding the unfounded restoration of a θίασος in a dedicatory inscription for Artemis (?),50 and leaving aside the evidence for Augustales in Stobi, we know of a βακχεῖον πρεσβύτερον setting up a short inscription for an ἀρχιμύστης, a κολλήγιον possibly of referees in gladiatorial games, and— likely—a group of συνήθεις led by an ἀρχισυνάγωγος.51 The latter two groups seem to have buried their leaders. The nature of the συνήθεις cannot be determined; they may have formed a private group. The βακχεῖον may have been involved in civic cult, but there is no way to know. What we can say is that the three associations fit a Romanized organizational context in the Eastern empire well, as should be expected from a city that had received municipal status in 69 CE. The Roman origin of κολλήγιον (= collegium) is obvious. Referees had their collegia in other places as well, including Rome itself; these seem to have been status groups with a professional aspect.52 The terms βακχεῖον and συνήθεις were introduced into Greek associational terminology by Roman rule and were rather widespread in second and third century CE Greece and Macedonia.53 The influence of Romanized organizational concepts is thus clearly visible in Stobi. If the Jewish συναγωγή is regarded as a Roman conception of legitimate Jewish organization, Polycharmus’ group fits this context. However, the peculiar phrasing of his inscription also supports the view developed above: Jewish groups belonged to the Roman associational order but operated under quite specific conditions. The epigraphic evidence from Stobi is altogether quite limited. In Macedonia, a more instructive case is Thessalonike, where more than 40 associations are on record.54 As a civitas libera, the city was not bound by Roman law, which probably made founding and sustaining an association easier here than elsewhere.55 Many of these associations appear in funeral contexts, so it is on first sight not surprising to find Marcus Aurelius Jacob naming “the συναγωγαί” as recipients of a fine in case someone used the sarcophagus bought by him.56 However, given the large number of association inscriptions from the city, it is noteworthy that none of them ever names an association as recipient of a fine. Jacob clearly presupposes that the συναγωγαί will be there for a long time, perhaps because they were part  50 I.Stobi 3: Ἀρτέμιδ[ι – – – – –] | τὸν βωμ[ὸν ἔθηκε] | Θεούχρη[στος μετ]|ὰ τῶν το[ῦ θιάσου ἐ]|πιμελητῶν – – – –] | Ζωΐλου κ[αὶ Διοσκου]|ρίδου κα[τὰ – – – – –]. Instead of a θίασος, one may think of a ἱερόν, and Artemis could be Artemidorus. 51 I.Stobi 7, 100, 119 with SEG LXII 439 (restorations according to Chaniotis: ἀρχισυ|[νάγωγος δ]ούμου and προνο|[ητὴς τῶν συ]νήθων). 52 See I.Ancyra 148, ll. 5–6 on a Pergamene referee: κολλήγιον ἔχον|τι ἐν Ῥώμῃ τῶν σουμμαρούδ[ων]. 53 I will deal with the evidence elsewhere. Briefly put, I think that βακχεῖον is derived from baccanal and that συνήθεις translates sodales. 54 Nigdelis 2010 plus recent additions; I give an overview in Eckhardt and Leonhard 2018: 151–159. 55 Nigdelis 2010: 22; see Pliny, Ep. 10.92–93. 56 IJO I Mac15: M(ᾶρκος) Aὐρ(ήλιος) Ἰακὼβ ὁ καὶ Eὐτύχιος | ζῶν τῇ συμβίῳ αὐτοῦ Ἄννᾳ | τῇ καὶ Ἀσυνκριτίῳ καὶ ἑαυτῷ μνί|ας χάριν· εἰ δέ τις ἕτερον καταθῇ, | δώσι ταῖς συναγωγαῖς λαρὰς | (Den.) μ(υριάδας) ζ ε.

Synagogues as Associations in the Roman Empire



of the institutional inventory of Thessalonike. The plural also seems to suggest that Jacob expected an overarching organization behind the (two or more) συναγωγαί, so that he did not have to specify (as would have been common) how the money would have to be divided between the organizations.57 Despite several attestations of ἀρχισυνάγωγοι as leaders of συνήθειαι, this is the only piece of evidence from Thessalonike for Jewish corporate organization, and for the term συναγωγή. It clearly falls outside the pattern established by (other) local association inscriptions.

The Danubian Provinces In the Danubian provinces, the only place that possibly allows for comparisons between a Jewish and other associations is Intercisa, a military camp with an accompanying village in Pannonia.58 The lack of civic status did not preclude the formation of associations: Veterans from Syrian Emesa and a collegius (sic) Fortunae dedicated altars to Diana Augusta and Fortuna respectively,59 a temple for Jupiter Optimus Maximus Hero was erected by “the worshippers of his association” led by a “father”,60 and another group of “worshippers” made a quite enigmatic dedication iudicio sacramenti—“upon the decision (to take) the oath of loyalty” (?).61 All of these inscriptions come from the time of Severus Alexander (222–235 CE) and stress the loyalty of the dedicators towards him. The members of the associations were likely soldiers and veterans from Syria and Thracia.62 It is plausible to connect the sudden appearance of these clubs (which contradicts the rule that soldiers should not have collegia in camps) with their special relationship to the Syrian emperor,63 although the precise nature of the connection remains debatable. It is in this context that we may want to place a notoriously problematic dedicatory inscription, also from the time of Severus Alexander.64 A man possibly  57 Normally one would have to specify; see for examples Harter-Uibopuu and Wiedergut (2014: 157– 158). Nigdelis (1994: 306) assumes that the sum would go to whichever group reported the deed first. 58 There does not seem to be a point in comparing the arcisina(gogus?) Ioses in the Roman colony of Oescus (late third or even fourth century CE) with the patrons of the local collegium fabrum, the only association known in the city (ILBulg 16, 18; AE 1987: 893; 2005: 1325). 59 RIU V 1056; 1062. Alföldy (1958: 181–182) dismisses the veterans because a corporate designation is missing (not necessary in my opinion) and argues that the cult of Fortuna was actually a local manifestation of the imperial cult. 60 RIU V 1084: cultores collegii eius. 61 RIU V 1115: iudicio | sacramenti | (vacat) | cultores; the empty line in-between apparently did not contain letters. Alföldy (1958: 185–188) argues for the usual understanding of sacramentum as a military oath, not some sort of initiation into mysteries. 62 Alföldy 1958: 196–197; cf. also Agócs (2013). 63 Alföldy 1958: 197–198. 64 IJO I Pan3 = RIU V 1051. The text given there runs (ll. 5–7) vot(um) | red(dit) l(ibens) Cosmius pr(aepositus?) | sta(tionis) Spondill(– – –) a synag(oga), but see the commentary ibid. for the by that time

326

Benedikt Eckhardt

called Cosmianus and possibly serving as a customs officer had fulfilled a vow by dedicating a plaque to Deus Aeternus, a common designation that in Pannonia (but nowhere else) was also applied to the Jewish god.65 The last line of the inscription mentions a synag(oga), to the right of the tabula ansata we find Iudeor(um). Due to many uncertainties in the reading, the nature of the synagoga’s involvement is unclear. That Cosmianus held an office there, too, is possible but not verifiable. It seems that the reign of Severus Alexander (who is credited in the Historia Augusta with a reform of professional collegia at Rome)66 triggered the foundation, or at least the public manifestation, of several associations in Intercisa. But unlike them, the synagoga did not appear as a corporate actor but was simply mentioned as a context for Cosmianus’ dedication. Nothing much can be derived from this for our question.

Italy As for Italy, the history of the Jewish community at Rome would require a detailed study of its own that cannot be provided here. However, it should be noted that even though this is the place that originally gave credence to the idea that Jews could be seen as collegia,67 none of the many Jewish inscriptions found there are of much interest to this investigation.68 There was clearly more than one Jewish organization in the city, but the same seems to be true for Thessalonike and perhaps other places where many Jews came together. Both for the sake of brevity and because of the nature of the evidence, Ostia is the more promising test case. Rome’s harbour not only features ample evidence for associations, an

 already extensive discussion, featuring a station called Spondilla, an a(rchi)synag(ogus) or the office of spondaules. A new reading has been offered by Szabó (2014: 201–202 = AE 2014: 1059): vot(um) | redeg(it) Smianus Pr|estas pondo IL l(ibens) a synag(oga), i.e., a man called Smianus Prestas collected money (49 pounds of silver) from the synagoga as a present to the emperor. This seems unlikely, as noted by Tibor (2016: 15 n. 5), who also rejects the otherwise unattested name Smianus. I do agree with Szabó that the ligature in l. 6 is more complex than previous editors have realized: MIA, not MI. An obvious solution that keeps most of the earlier reading in place would be Cosmianus, attested as a Jewish name in IJO I Thr1–2 (Philippopolis). This is more plausible than assuming a rather large gap between Szabó’s G and S given that the writer clearly ran out of space in the last two lines, although admittedly O is hardly visible. E could indeed be read instead of L as the fourth letter in l. 6, but that could perhaps be solved through the (unusual) abbreviation rede(dit). I see no traces of an E at the beginning of l. 7. As for the end, I still prefer Spondilla(e); that there is no word divider before a synagoga has been demonstrated by Fülep (1966: 95–97). Instead of an a(rchi)synag(ogus) or a “member of the synagogue” (IJO I), I would consider synag(ogae), “to the advantage of the synagoga”. 65 On Deus Aeternus, see Bartels and Kolb (2011). 66 SHA, Sev. Alex. 33.2. 67 Schürer 1879: 8–10. 68 They are conveniently accessible in JIWE II.

Synagogues as Associations in the Roman Empire

 

early Jewish presence already in the first half of the first century CE, and a building that has frequently been called the earliest synagogue in Europe,69 but also three inscriptions that offer some insights into Jewish organizational history. Not much can be learned from the epitaph of the archisynagogus Plotius Fortunatus, except perhaps that Jews were already organized in a corporate fashion at an early date.70 An inscription introduced by pro salute Aug(usti), but continuing in Greek records the building of (presumably) a meeting place and the dedication of the “ark for the holy law” by (in the final version of the text) Mindius Faustus.71 More information is provided by another inscription likely deriving from Ostia and possibly datable as early as the second century CE. The “gerusiarch” Gaius Julius Justus constructed a sepulchral monument for himself, his wife and his freedmen, on a plot of land that a group of Iud(a)ei had bought for this purpose and given to him, following a motion put forward by a pater, another gerusiarch and an official dia biu (“for life”).72 There are many things to note here. The lost beginning of line 1 must have contained a corporate designation; despite attempts to insert collegium or even universitas, I regard synagoga as the most plausible reconstruction.73 A participle is attached to Iudeorum; if [commor]antium is right,74 the Jews referred to themselves as foreigners in Ostia—an impression further strengthened by the use of Greek words like gerusia and (likely) synagoga, and of course by the Greek inscription of Mindius Faustus. The structure of the group seems to revolve around a board of magistrates with supreme authority. No gerusia or gerusiarchae are known in the West apart from another Jewish inscription from Puteoli.75 But as it has justly been observed that the general impression fits other evidence for “governing boards” in Roman collegia, an adaptation of Western forms of corporate organization can be  69 On the building, see Runesson (2001). Early Jewish presence is now attested through ILOP 108 (= AE 2009: 103). 70 JIWE I 14 (1st/2nd century CE). On associations in Ostia, see Rohde 2012: 79–274. 71 JIWE I 13; cf. the commentary ibid. on the reuse of the inscription by Mindius Faustus, and Runesson (2001: 85–88) on the relation (or not) of this inscription to the excavated synagogue building. 72 The text according to JIWE I 18: [synagoga (?)] Iudeorum | [in col(onia) Ost(iensi) commor]antium qui compara|[verunt ex conlat(?)]ione locum C(aio) Iulio Iusto | [gerusiarche ad m]unimentum struendum | [donavit, rogantib(?)]us Livio Dionysio patre et | [– – – – –]no gerusiarche et Antonio | [– – – dia] biu anno ipsorum, consent(iente) ge[rus(ia). C(aius) Iulius Iu]stus gerusiarches fecit sib[i] | [et coniugi] suae lib(ertis) lib(ertabusque) posterisque eorum. | [in fro]nte p(edes) XVIII, in agro p(edes) XVII. 73 White (1997: 43 n. 58) prefers collegium “given the tone of the rest of the text and the kind of honors bestowed”. The tone of the text is set, in my view, by the Greek word gerusiarches; as for honours, none are mentioned (see below). Universitas (the original proposal, based on the case of Antioch mentioned above) is justly rejected in JIWE I: 33–34. 74 There is no parallel for [commor]antium, but the common [negoti]antium (considered in JIWE I: 34) is excluded if, as seems clear from some of the photos available, the remains of R can be identified on the stone. Theoretically, there are other options, e.g., [perseve]rantium—see CIL VI 33875 for an enigmatic corpus perseverantium; one would perhaps need something like in fide or in concordia to make sense of this. 75 JIWE I 23.

328

Benedikt Eckhardt

assumed.76 Finally, one might consider taking the acquisition of land as an indication of legal status, but this argument is necessarily weak.77 There are good arguments for comparing the Jewish organization of Ostia to other associations. The presence of a pater may suggest that the fitting comparison would not focus on the many professional groups and their patrons, but on Mithras groups (where this was a grade of initiation) or “reed-carriers” (personnel of the Mater Magna cult).78 This would also leave some room for the terminological peculiarities: such groups tended to have their technical vocabulary which differed from the mainstream associational sphere. Perhaps this choice of comparison might also mitigate the odd inversion of hierarchies: usually, one would expect prominent members or magistrates to give something to the association, while in our case, it is the association that donates something to a prominent member.79 However, as the cannophori were an institution of the civic cult, we still would not know whether or not the Jewish group can be seen as a private association. Some more general observations are also pertinent here. While the evidence for the Ostian collegia usually shows their interaction with outsiders (patrons, benefactors, emperors or the Roman senate), our inscriptions document only internal procedures. We may legitimately infer from them that some Jews in Ostia were quite wealthy,80 but that in itself tells us nothing about the integrative capacities of Jewish organizations. Nor can we infer from the pro salute Augustiheading of the Greek inscription that Jews were linked to non-Jews “in important social and economic ways”,81 or from a similar architectural design of synagogue buildings and the meeting halls of associations that there was an “implied rivalry  76

White (1997: 46), who might push things too far when he states that the term gerusia “would have been a commonly recognized label for the governing board or trustees of the collegium”. Governing boards are a common phenomenon in Italian professional associations and could also be imported to the East; cf. Cracco Ruggini (1976: 471 and 491 n. 113) on (according to her, isolated) cases, one of which actually uses the term γερουσία (I.Side 109 from the late third century CE: γερουσία τοῦ μεγάλου συνεργίου). 77 It would require a lot of precision, namely a distinction between the Iudei who bought the place and the synagoga who therefore owned it (as a legitimate corpus). People could always buy something together and regard it as their common possession; the advantage of official recognition as a corpus was that ownership could be treated as legally independent of the current members of an organization. 78 Mithras: CIL XIV 62, 70, 286, 311–312. Cannophori: CIL XIV 37. White (1997: 43) proposes patre et | [col(legii) patro]no gerusiarche in l. 6 of the Julius Justus-inscription, but I do not know a parallel for someone being both pater (a position within the association) and patronus (a position usually held by outsiders) of an association at the same time (CIL III 8837 is erroneously included in the list of Clemente (1972: 144–156); Terentius Mercurius is pater et patronus of the person he commemorates, who happens to be a vexillarius collegii fabrum). It is much more plausible to regard …no as part of the name of another gerusiarch. 79 This in fact seems so odd that I doubt [donavit] in l. 5. Dedit would certainly suffice; it allows for some sort of reciprocal exchange. 80 Zevi (2014) now offers arguments for regarding the eminent Fabii of Ostia as Jewish. He also tentatively connects them to the synagogue building, but this remains a conjecture. 81 White 1997: 42. Rohde (2012: 50) notes that immigrant groups demonstrating loyalty to the emperor might have done so from a quite precarious position.

Synagogues as Associations in the Roman Empire

 

or even overt competition among associations”, including the Jews.82 At the very least, there is no evidence for this. If anything, the impression provided by the Jewish inscriptions is that as far as their corporate organization was concerned, Jews stayed among themselves. That they were organized in a way that resembles other associations is not surprising as the origins of their communal formation lie in the laws on legitimate collegia (see above), but this fact alone does not tell us what we want to know. Comparison with the actual inscriptions of other associations reveals different patterns of behaviour. Rome and Ostia are hardly representative of the West, or even Italy. A more instructive case—both in terms of the “average” situation and in terms of methodology—is provided by the Augustan colony of Brixia (modern Brescia) in Northern Italy. Two short Jewish inscriptions are known from the city. One is too fragmentary to stimulate discussion, but the other is more interesting: a short inscription dated by various commentators somewhere between the second and the fourth century CE reads Coeliae Paternae | matri synagogae | Brixianorum.83 This is certainly not a lot of information, but we must take what we get. In this case, we know that Jews formed an organized group that could set up a commemorative inscription, was explicitly defined as belonging to the city of Brixia and counted a woman among its prominent representatives. These facts can be compared to the extraordinary evidence for associations at Brixia. As a colony, Brixia operated under Roman law. We should expect a certain restriction of the associational sphere to “useful” groups, and despite the very large number of association inscriptions from the city, this is indeed what the evidence suggests. The local seviri Augustales were apparently recognized as a legitimate corpus by Antoninus Pius, a fact they stressed (for reasons that remain unclear) by emphasizing their right to have a common treasury (arca). They were referred to as socii or collegium and received sizeable donations.84 In another inscription from Brixia, the dendrophori honoured their patron “because our immunity (from obligations) has been confirmed due to his effort”.85 Clearly they had at some point been recognized as a legitimate corpus, and had now used their patron (who may have acted as an actor vel syndicus) to defend their privileges against an unknown threat. Dendrophori are mentioned in other inscriptions;86 apart from these, the evidence is largely dominated by the collegium centonariorum and the collegium fabrum, who often acted together, occasionally including  82

Richardson 2004: 213. JIWE I 5. The other inscription is JIWE I 4: – – – [ἀρ]χισυνά|γωγο[υ?] – – –. 84 InscrIt X.5 223, ll. 4–6: VIvir(i) Aug(ustales) soci(i) | quibus ex permiss(u) divi Pii | arcam habere permiss(um); cf. on the case Tran (2006), p. 358–362. For a donation of 1,000 Sesterces to the coll(egium) VIvir(orum) socior(um), see InscrIt X.5 985. 85 InscrIt X.5 135, ll. 11–12: quod eius industria immuni[t]|as collegi(i) nostri sit confirma[ta]; cf. Eck 1979: 265. 86 InscrIt X.5 16, 211, 266, 932. 83

330

Benedikt Eckhardt

the dendrophori as well.87 Alongside this typical constellation of the tria collegia, we find a collegium iuvenum Brixianorum (one of the civic age classes mentioned above, perhaps also involved in the cult of Iuventus),88 a collegium nautarum Brixianorum (ship owners),89 a collegium iumentariorum (a transport association with an official function in the cursus publicus),90 a collegium aeneatorum (musicians, or another transport association?),91 as well as professional groups of woolworkers, criers, and the “public sellers of medicine”.92 The enigmatic sodales Pacatiensium known from a late antique sarcophagus may have been a group of veterans;93 in addition, three inscriptions refer to a collegium Larum, which in one case is specified as worshipping the Lares of a local aristocrat.94 The latter case may have been a household association tied to a member of the local elite and hence operating under special circumstances.95 Otherwise, the associational sphere that emerges from these inscriptions fits the model of legitimate associations outlined above remarkably well. Most importantly, the dominance of the fabri and the centonarii and the treatment of Augustales as legitimate corpora fits the pattern. Brixia’s collegia (the term used almost invariably in the inscriptions) appear as semi-public institutions. We do not know if the professionals of lowly regarded crafts or trades received official recognition by emperors or the senate. It is interesting to note that even traders of questionable standing like the farmacopolae called themselves publici, but just how technical a term that was is debatable.96 It is also noteworthy that there was some degree of connectivity within this associational sphere. The frequent cooperation of fabri and centonarii has already been mentioned; one fascinating result is a “freedman of the collegia” called Fabricius Centonius.97 Augustales regularly appear as a mediator between (other) prominent associations and the local high society, as can be

 87

Centonarii: InscrIt X.5 110, 180, 208, 221, 239, 283, 289, 299. Fabri: InscrIt X.5 183, 226, 237, 280, 808. Both together (especially ins honorific inscriptions for members of the municipal elite): InscrIt X.5 120, 157, 189–190, 200, 202, 209, 216, 241, 266 (plus dendrophori), 274, 279, 299, 996, 1211; AE 1991: 822. 88 InscrIt X.5 145, 209 (a priest of the collegium iuvenum); for the cult of Iuventus, see AE 2010: 592. 89 InscrIt X.5 1065, 1070. 90 InscrIt X.5 17, 77; cf. Boscolo 2006/07. 91 AE 1991: 823. Boscolo (2006/07: 357–358) considers their possible role in transport; the term normally designates musicians in the military. 92 InscrIt X.5 294 (lanarii pectinarii sodales), 297 (lanarii coactores), 875 (lanarii carminatores sodales), 282 (collegium praeconum), 280 (collegium farmacopolarum publicorum). On the terms qualifying the lanarii, see Perry 2016: 503–504; on praecones, see Bond (2016: 27–29); on pharmacopolae, see Totelin 2016. 93 InscrIt X.5 1075 (from Phrygia Pacatiana? Pacatienses formed military units, and the use of collega for the deceased would fit military usage well). 94 InscrIt X.5 134 (the cultores of the Lares of a clarissimus iuvenis and Quindecemvir sacris faciundis), 134, 231. 95 Cf. for parallels Laubry 2012: 105 n. 13. 96 Tran (2012) offers a differentiated treatment of the evidence. 97 InscrIt X.5 216.

Synagogues as Associations in the Roman Empire



observed elsewhere.98 The fabri and centonarii honoured a priest of the iuvenes,99 and prominent people could act as patrons or even (honorary) magistrates of “all the collegia” in the city.100 “All” may or may not refer only to the tria collegia, but it is clear that even the less prominent professional associations were integrated into this network. An Augustalis took part in a joint donation of 400 Sesterces to the collegium iumentariorum; the collegium aeneatorum received a donation of 1,000 Sesterces by the same man’s wife for annual celebrations; the farmacopolae publici received one half of a property while the fabri received the other.101 There was clearly common ground. Does the synagoga fit in here? As an organized group able to put up an inscription, we have to compare it not to putative associations below the radar, but to the ones we know from other inscriptions. While no connection to the city’s associational network is apparent, two parallels immediately come to mind: the prominent role of a woman is nothing unusual in Brixia, and the addition Brixianorum is found in the names of two local associations. However, both observations also reveal differences. The women mentioned usually appear as outsiders who leave an association money (sometimes to commemorate their husbands) or, in one case, act as its patrona; sometimes they are honoured ob merita, but at the same time identified by an external reference—as someone’s wife, or a civic priestess.102 Matres of collegia are attested elsewhere, but not in Brixia. The  98

See on the phenomenon Cracco Ruggini 1973: 284–290. InscrIt X.5 209. 100 InscrIt X.5 238, 275. 101 InscrIt X.5 77, 280; AE 1991: 823. 1,000 Sesterces was an acceptable amount even for the more prominent dendrophori (who also received the sum from an Augustalis and his wife), as well as for the seviri socii themselves (InscrIt X.5 211, 985). 102 In most cases from Brixia, women receive honorific inscriptions ob merita or bene merenti; we should also place in this category inscriptions containing the Brixian formula titulo usa (which indicates that the person in question paid for the monument; see Gregori 2012: 363–367). The evidence can be grouped in the following categories. 1.) Honorific inscriptions without further explanation. This is the most common type. We do not know what Baebia Nigrina, Aemilia Aequa, Petronia Baebiana or Aemilia Synethia have done for the centonarii (InscrIt X.5 110, 180, 239, 932), or Sextia Asinia Polla for the iuvenes (InscrIt X.5 145), or Bittalia Festa and Clodia Secunda for the centonarii and the fabri (InscrIt X.5 190; AE 1991: 822 where ob merita vel sim. is missing, but it is clearly the same type of text; see also InscrIt X.5 241, an honorific inscription set up locus datus decreto decurionum for – – – Picatia). Some of these women are further qualified by reference to important family members (Baebia Nigrina was the daughter of the sister of a man of consular status) or civic offices (Aemilia Aequa and Clodia Secunda were priestesses of Trajan’s wife and stepmother); this does not suggest a special relationship to the association. In the case of Bedasia Iusta, honoured by the centonarii (InscrIt X.5 189), it is the husband who sets up the inscription titulo usus. 2.) Honorific inscriptions with some explanation. The dedication by the Augustales to the pietas of Hostilia Hostiliana is set up primae bene merenti (InscrIt X.5 223); perhaps she was the first donor after the Augustales had received the right to have an arca. The cultores collegii Larum honoured their patrona whose name has not survived (InscrIt X.5 225). 3.) Endowments to an association to ensure annual commemorative celebrations. Women could be the objects of such commemoration, in which case the husband gave money to the association (InscrIt X.5 280 for the fabri and the farmacopolae to commemorate Valeria Ursa; 985 for the Augustales to com99

332

Benedikt Eckhardt

position, which should not be confused with patronage in a legal sense, seems to have been more common in less prestigious, usually religious or ethnic groups.103 The addition of Brixianorum raises another question. In the case of the iuvenes and the nautae, the adjective seems to qualify the people, whereas in the Jewish case, it qualifies the institution. The synagoga Brixianorum thus finds a better parallel in the ordo Brixianorum than in the iuvenes Brixiani;104 one may think of a status group—the synagoga-people of Brixia, not “the Jewish people of Brixia who have a synagoga”. Finally, we have to consider the communicative dimension of the inscriptions mentioned here. The large majority of them demonstrate connections of associations and their members to higher-ranking outsiders (who receive honours for donations or patronage); in a few cases, the associations themselves appear as agents vis-à-vis lower-ranking members (who receive burials). They can thus be read as attempts to define the position of those responsible for their publication within the civic hierarchies.105 In the synagoga-inscription, things are different. It only communicates within the context of the synagoga, and if we really regard the inscription as an epitaph,106 the order is reversed: the supposedly (from the association’s point of view) high-ranking benefactor would have been buried by the association. The difference to the pattern we can reconstruct elsewhere should not go unnoticed. In any case, Coelia Paterna is the only woman in Brixia who is clearly referred to by reference to her position within the association. That this exact position is found in several other synagogae elsewhere also points to a rather exceptional case. There are evidently limits to what can be done with a five-word inscription. But yet again, we see that there is no straightforward case for the assumption that  memorate Clodia Achilla), but they could also give money to an association to commemorate their husband (InscrIt X.5 1065 for the nautae, 823 for the aeneatores). Neither case presupposes membership of either the deceased or the donor. It is noteworthy that the small sample of inscriptions belonging to this category shows more variety (5 associations) than the large number of inscriptions falling into category 1 (3 associations). Clearly, associations were approached differently according to status (the centonarii would perhaps not have bothered to deal with such requests). 4.) Inscriptions documenting concrete action on behalf of the association. The only example is InscrIt X.5 17, –mia Firmia Tertia dedicating an altar to the genius of the collegium iuvenum (of which she could not have been a member). Apart from category 3, nothing in this dataset presupposes or suggests a sepulchral context. 103 The list by Clemente (1972: 144–156) provides a helpful overview precisely because Clemente did not distinguish properly between patronus/patrona and pater/mater; on the relevance of this distinction, see Hemelrijk 2008. 104 This is the parallel mentioned in JIWE. That Brixianorum in the other examples qualifies the people and not the institution is clear from InscrIt X.5 1070, l. 2: [dedit coll]eg(io) naut(arum) B(rixianorum) praesent(ium). 105 See the insightful remarks by Perry (2016: 504–505). 106 Noy is rightly cautious in JIWE I: 8: “The wording is probably that of an epitaph, but could be from an honorific inscription.” However, if it was an honorific inscription, one would expect the body who has voted the honours to be mentioned (as in all cases cited above); we also miss the typical titulus usa or ob merita. The most likely interpretation seems to be “the grave of” Coelia Paterna.

Synagogues as Associations in the Roman Empire



the synagoga of Brixia operated under the same circumstances as the amply attested collegia. The evidence is equally and perhaps a bit more consistent with the view developed above: the synagoga was an undoubtedly legitimate, but socially marginalized institution that could not attract external benefactors.

The Far West Gaul, Germany and Britain provide no relevant material from antiquity. The situation for the Spanish provinces is largely similar, but we can at least point to the fact that even this far in the West, we find the same organizational features of Jewish groups as elsewhere in the empire, suggesting translocal uniformity beyond the level local associations could have achieved. A mosaic floor in Elche attests to the existence of ἄρχοντες and πρεσβύτεροι,107 and a relatively recent find from Merida is particularly remarkable: an inscription records that Annianus Peregrinus Honorificus had served as exarchon of two synagogae,108 thus not only confirming the possibility of overarching Jewish organization that we found in Thessalonike, but also extending the evidence for the function of exarchon, previously known only at Rome.109 Both texts likely date to the fourth century CE and thus fall outside our timespan, but they do show how specific concepts of Jewish organization spread empire-wide, with no apparent connection to the organizational framework provided by local collegia.110

Bibliography Agócs, N. 2013. “People in Intercisa from the Eastern Parts of the Roman Empire”. Specimina nova dissertationum ex institutis historiae antiquae et archaeologiae universitatis Quinqueecclesiensis 21/22: 9–27. Alföldy, G. 1958. “Collegium-Organisationen in Intercisa”. ActAnt 6: 177–198. Ameling, W. 2016. “Epigraphische Kleinigkeiten III”. ZPE 198: 138–144. Arnaoutoglou, I. 2002. “Roman Law and collegia in Asia Minor”. RIDA 49: 27–44. Ascough, R. S. 2015. “Paul, Synagogues, and Associations. Reframing the Question of Models for Pauline Christ Groups”. Journal of the Jesus Movement in its Jewish Setting 2: 27– 52. Ascough, R. S. 2017. “Methodological Reflections on Synagogues and Christ Groups as ‘Associations’: A Response to Erich Gruen”. Journal of the Jesus Movement in its Jewish Setting 4: 118–126.

 107

JIWE I 181. Published by García Iglesias (2010), who thinks that Annianus Peregrinus was honorificus exarchon; I adopt the text of Clauss (ll. 1–3): Annianus Peregrinus (H)ono|rificus duarum synagog(rum) | exarc(h)on. 109 On the Roman title see Williams 1994: 139–140. 110 On collegia in Spain, see Santero Santurino 1978. None are attested at Colonia Augusta Emerita (Merida) despite the relatively large number of inscriptions found. 108

334

Benedikt Eckhardt

Balduinus, F. 1557. Ad edicta veterum principum Rom(anorum) de Christianis. Basel: Oporinus. Balamoshev, C. 2017. “The Jews of Oxyrhynchos Address the strategos of the Nome: An Early Fourth-Century Document”. JJP 47: 27–43. Barclay, J. M. G. 2006. “Money and Meetings: Group Formation among Diaspora Jews and Early Christians”. In: A. Gutsfeld and D.-A. Koch (eds.), Vereine, Synagogen und Gemeinden im kaiserzeitlichen Kleinasien. STAC 25. Tübingen: Mohr Siebeck, 113–127. Bardtke, H. 1961. “Die Rechtsstellung der Qumrān-Gemeinde”. TLZ 86: 93–104. Bartels, J. and Kolb, A. 2011. “Ein angeblicher Meilenstein in Novae (Moesia Inferior) und der Kult des Deus Aeternus”: Klio 93: 411–428. Baslez, M.-F. 2019. “Les communautés juives de la Diaspora dans le droit commun des associations du monde gréco-romain”. In: Eckhardt, B. (ed.), Private Associations and Jewish Communities in the Hellenistic and Roman Cities. JSJSup 191. Leiden: Brill, 97– 114. Bendlin, A. 2011. “Associations, Funerals, Sociality, and Roman Law: The Collegium of Diana and Antinous in Lanuvium (CIL 14.2112) Reconsidered”. In: Öhler, M. (ed.), Aposteldekret und antikes Vereinswesen: Gemeinschaft und ihre Ordnung. WUNT 280. Tübingen: Mohr Siebeck, 207–296. Bond, S. 2016. Trade and Taboo: Disreputable Professions in the Roman Mediterranean. Ann Arbor, MI: University of Michigan Press. Boscolo, F. 2006/07. “Gli iumentarii e il sistema dei trasporti in area Veneta in età romana”. Atti e Memorie dell’Ateneo di Treviso N.S. 24: 345–361. Clemente, G. 1972. “Il patronato nei collegia dell’impero romano”. SCO 21: 142–229. Cohen, S. J. D. 1999. The Beginnings of Jewishness: Boundaries, Varieties, Uncertainties. Berkeley: University of California Press. Cracco Ruggini, L. 1973. “Stato e associazioni professionali nell’età imperiale romana”. In: Akten des VI. Internationalen Kongresses für Griechische und Lateinische Epigraphik München 1972. Munich: Beck, 271–311. Cracco Ruggini, L. 1976. “La vita associativa nelle città dell’Oriente greco: Tradizioni locali e influenze romane”. In: Pippidi, D. M. (ed.), Assimilation et résistance à la culture grécoromaine dans le monde ancien. Bucharest: Editura Academiei, 463–491. De Ligt, L. 2001. “D. 47,22,1,pr.–1 and the Formation of Semi-Public Collegia”. Latomus 60: 345–358. De Rossi, G. B. 1864. La Roma sotteranea Cristiana, Vol. 1. Rome: Cromo-Litografia Pontificia. Eck, W. 1979. Die staatliche Organisation Italiens in der hohen Kaiserzeit. Munich: Beck. Eckhardt, B. 2017a. “Rom und die Juden – ein Kategorienfehler? Zur römischen Sicht auf die Iudaei in später Republik und frühem Prinzipat”. In: Hasselhoff, G. K. and Strothmann, M. (eds.), “Religio Licita”? Rom und die Juden. SJ 84. Berlin: De Gruyter, 13–53. Eckhardt, B. 2017b. “Craft Guilds as Synagogues? Further Thoughts on ‘Private JudeanDeity Associations”. JSJ 48: 246–260. Eckhardt, B. 2018. “Religionis causa? Zur rechtlichen Lage der Vereine ‘fremder’ Götter in der römischen Kaiserzeit”. In: Blömer, M. and Eckhardt, B. (eds.), Transformationen paganer Religion in der römischen Kaiserzeit: Rahmenbedingungen und Konzepte. RVV 72. Berlin: De Gruyter, 113–152. Eckhardt, B. 2019. “Associations beyond the City: Jews, Actors and Empire in the Roman Period”. In: Eckhardt, B. (ed.), Private Associations and Jewish Communities in the Hellenistic and Roman Cities. JSJSup 191. Leiden: Brill, 115–156. Eckhardt, B. and Leonhard, C. 2018. Juden, Christen und Vereine im Römischen Reich. RVV 75. Berlin: De Gruyter.

Synagogues as Associations in the Roman Empire



Fauconnier, B. 2018. “Ecumenical Synods: The Associations of Athletes and Artists in the Roman Empire”. PhD thesis, University of Amsterdam. Fülep, F. 1966. “New Remarks on the Question of the Jewish Synagogue at Intercisa”. Acta Archaeologica Academiae Scientiarum Hungaricae 18: 93–98. Gabrielsen, V. and Thomsen, C. A. 2015. “Introduction: Private Groups, Public Functions?”. In: iidem (eds.), Private Associations and the Public Sphere. Copenhagen: The Royal Danish Academy of Sciences and Letters, 7–24. García Iglesias, L. 2010. “Nueva iscripción judía del Museo Nacional de Arte Romano de Mérida”. Anas 23: 11–26. Gregori, G. L. 2012. “Peculiarità dell’orizzonte epigrafico bresciano”. In: Donati, A. and Poma, G. (eds.), L’officina epigrafica romana: In ricordo di G. Susini. Faenza: Fratelli Lega, 361–371. Groten, A. 2015. Corpus und universitas: Römisches Körperschafts- und Gesellschaftsrecht. Zwischen griechischer Philosophie und römischer Politik. Ius Romanum 3. Tübingen: Mohr Siebeck. Grüll, T. 2016. “Jewish Presence in the Danubian Provinces of the Roman Empire”. In: Hrbácsek, M. (ed.), Židovský kultúrny fenomén v stredoeurópskom kontexte: Zsidó kultúra közép-európai kontextusban. Nitra: UKF, 9–18. Gruen, E. S. 2016. “Synagogues and Voluntary Associations as Institutional Models: A Response to R. Ascough and R. Korner”. Journal of the Jesus Movement in its Jewish Setting 3: 125–131. Haentzschelius, J. F. 1729. De hetaeriis veterum Christianorum. Leipzig: Titius. Hallmannsecker, M. 2017. “Heracles Hoplophylax, Iudaioi, and a Palm Grove: A Fresh Look at I.Smyrna 697”. Epigraphica Anatolica 50: 109–127. Harland, P. A. 2003. Associations, Synagogues, and Congregations: Claiming a Place in Ancient Mediterranean Society. Minneapolis, MN: Fortress. Harland, P. A. 2015. “Associations and the Economics of Group Life: A Preliminary Case Study of Asia Minor and the Aegean Islands”. SEÅ 80: 1–37. Hemelrijk, E. 2008. “Patronesses and ‘Mothers’ of Roman Collegia”. ClAnt 27: 115–162. Harter-Uibopuu, K. and Wiedergut, K. 2014. “‘Niemand anderer soll hier bestattet werden …’: Grabschutz im kaiserzeitlichen Milet”. In: Thür, G. (ed.), Grabrituale: Tod und Jenseits in Frühgeschichte und Altertum. Vienna: ÖAW, 147–171. Heinrici, G. 1876. “Die Christengemeinde Korinths und die religiösen Genossenschaften der Griechen”. ZWT 19: 465–526. IJO I = Noy, D., Panayotov, A., and Bloedhorn, H. 2004. Inscriptiones Judaicae Orientis, Vol. I: Eastern Europe. TSAJ 101.Tübingen: Mohr Siebeck. IJO II = Ameling, W. 2004. Inscriptiones Judaicae Orientis, Vol. II: Kleinasien. TSAJ 99. Tübingen: Mohr Siebeck JIWE = Noy, D. 1993. Jewish Inscriptions from Western Europe, Vol. I: Italy (excluding the City of Rome), Spain and Gaul; Vol. II: The City of Rome. Cambridge: Cambridge University Press. Kayser, F. 2013. “Les communautés ethniques du type politeuma dans l’Égypte hellénistique”. In: Delrieux, F. and Mariaud, O. (eds.), Communautés nouvelles dans l’Antiquité grecque: Mouvement, integrations et representations. Chambéry: Université de Savoie, 121–153. Kraabel, A. T. 1982. “The Roman Diaspora: Six Questionable Assumptions”. JJS 33: 445– 464. Kruse, T. 2015. Ethnic Koina and Politeumata in Ptolemaic Egypt. In: Gabrielsen, V. and Thomsen, C. A. (eds.), Private Associations and the Public Sphere. Copenhagen: The Royal Danish Academy of Sciences and Letters, 270–300.

336

Benedikt Eckhardt

Langellotti, M. 2016. “Professional Associations and the State in Roman Egypt: The Case of First-Century Tebtunis”. Chronique d’Égypte 91: 111–134. Last, R. 2016. “The Other Synagogues”. JSJ 46: 330–363. Last, R. and Harland, P. A. 2020. Group Survival in the Ancient Mediterranean: Rethinking Material Conditions in the Landscape of Jews and Christians. London: T&T Clark. Laubry, N. 2012. “Ob sepulturam: Associations et funérailles en Narbonnaise et dans les Trois Gaules sous le Haut-Empire”. In: Dondin-Payre, M. and Tran, N. (eds.), Collegia: Le phénomène associatif dans l’Occident romain. Paris: de Boccard, 103–133. Liu, J. 2009. Collegia Centonariorum: The Guilds of Textile Dealers in the Roman West. Columbia Studies in the Classical Tradition 34. Leiden: Brill. Nigdelis, P. M. 1994. “Synagoge(n) und Gemeinde der Juden in Thessaloniki: Fragen aufgrund einer neuen jüdischen Grabinschrift der Kaiserzeit”. ZPE 102: 297–306. Nigdelis, P. M. 2010. “Voluntary Associations in Roman Thessalonikē: In Search of Identity and Support in a Cosmopolitan Society”. In: Nasrallah, L., Bakirtzis, C., and Friesen, S. J. (eds.), From Roman to Early Christian Thessalonikē: Studies in Religion and Archaeology. Cambridge, MA: Harvard University Press, 13–47. Perry, J. S. 2016. “Sub-Elites”. In: Cooley, A. E. (ed.), A Companion to Roman Italy. Malden: Wiley-Blackwell, 498–512. Pucci Ben Zeev, M. 1998. Jewish Rights in the Roman World: The Greek and Roman Documents Quoted by Josephus Flavius. TSAJ 74. Tübingen: Mohr Siebeck. Rajak, T. 1984. “Was There a Roman Charter for the Jews?”. JRS 74: 107–123. Richardson, P. 2004. “An Architectural Case for Synagogues as Associations”. In: idem, Building Jewish in the Roman East. JSJSup 92. Leiden: Brill 2004, 207–221. Ritti, T. 2016. Per la storia sociale ed economica di Hierapolis di Frigia: Le fondazioni sociali e funerarie. Rome: Bardi. Rohde, D. 2012. Zwischen Individuum und Stadtgemeinde: Die Integration von collegia in Hafenstädten. Mainz: Verlag Antike. Runesson, A. 2001. “The Synagogue at Ancient Ostia: The Building and its History from the First to the Fifth Century”. In: Olsson, B., Mitternacht, D., and Brandt, O. (eds.), The Synagogue of Ancient Ostia and the Jews of Rome: Interdisciplinary Studies. Stockholm: Paul Åströms Förlag, 29–99. Santero Santurino, J. M. 1978. Asociaciones populares en Hispania Romana. Sevilla: Publicaciones de la Universidad de Sevilla. Schumann, G. A. 1723. De collegiis et corporibus opificum: Von denen Zünfften und Innungen derer Handwerker. Halle: C. Henckel. Schürer, E. 1879. Die Gemeindeverfassung der Juden in Rom in der Kaiserzeit. Leipzig: Hinrichs. Stebnicka, K. 2015. Identity of the Diaspora: Jews in Asia Minor in the Roman Imperial Period. Warsaw: Fundacja Taubenschlaga. Szabó, Á. 2014. “Jüdische Funde aus dem römischen Pannonien”. In: Gross, R. et al. (eds.), Im Licht der Menora: Jüdisches Leben in der römischen Provinz. Frankfurt am Main: Campus, 199–210. Totelin, L. M. V. 2016. “Pharmakopōlai: A Re-Evaluation of the Sources”. In Harris, W. V. (ed.), Popular Medicine in Graeco-Roman Antiquity: Explorations. Leiden: Brill, 65–85. Tran, N. 2006. Les membres des associations romaines: Le rang social des collegiati en Italie et en Gaule sous le haut-empire. Rome: École française de Rome. Tran, N. 2012. “Associations privées et espace public: Les emplois de publicus dans l’épigraphie des collèges de l’Occident romain”. In: Dondin-Payre, M. and Tran, N. (eds.), Collegia: Le phénomène associatif dans l’Occident romain. Paris: de Boccard, 63–80. Van Bynkershoek, C. 1719. Opuscula varii argumenti. Leiden: Van der Linden.

Synagogues as Associations in the Roman Empire

 

Verboven, K. 2016. “Guilds and the Organisation of Urban Populations During the Principate”. In: Verboven, K. and Laes, C. (eds.), Work, Labour, and Professions in the Roman World. Impact of Empire 23. Leiden: Brill, 173–202 White, L. M. 1997. “Synagogue and Society in Imperial Ostia: Archaeological and Epigraphic Evidence”. HTR 90: 23–58. Williams, M. H. 1994. “The Structure of Roman Jewry Re-Considered: Were the Synagogues of Ancient Rome Entirely Homogeneous?”. ZPE 104: 129–141. Zevi, F. 2014. “I Fabii ostiensi e gli Ebrei di Ostia”. Mélanges de l’École française de Rome: Antiquité 126/1, 83–94.

Synagogues in Inscriptions from Asia Minor: The Iulia Severa Inscription Reconsidered

Markus Öhler In 1909 Johann Oehler, a teacher at a school in Vienna’s ninth district and author of one of the first comprehensive overviews on Greek associations (Oehler 1905), published a list of Jewish inscriptions, thus preceding Jean-Baptiste Frey’s impressive (but highly debatable) collection by decades (Frey 1936–1952). In his tripartite contribution, “Epigraphische Beiträge zur Geschichte des Judentums”, which appeared in Monatsschrift für die Geschichte und Wissenschaft des Judenthums, Oehler listed not only 283 epigraphical records but also provided a short list of criteria (“sichere Kennzeichen”) for assessing whether an inscription is of Jewish/Judaean origin: (1) the terms Judaeus/Judaea, (2) the depiction of a menorah as well as the word “peace” in Hebrew, Greek or Latin, (3) the formula εἷς θεός, and (4) the consolation θάρσι, οὐδεὶς ἀθάνατος (Oehler 1909). He explicitly refrained from using names as indicators for Jewish provenance. Some of those criteria are still held in high esteem, while others, such as number (4) above, have been dismissed as inconclusive. It is also remarkable that many of the instances in Oehler’s list comprise none of these markers.1 In the subsequent hundred years, several scholars have dealt with the problem of establishing criteria for considering an inscription as Jewish. In his collection of Jewish epitaphs from 1991, Pieter W. van der Horst formulated an extensive list of indicators, such as the location, the use of Biblical names or Ἰουδαῖος, allusions or quotes of Biblical texts, references to an archisynagōgos and the display of Jewish symbols (menorah, etrog, shofar).2 At the same time, van der Horst pointed out that these criteria should be used with great caution, as “the matter is far from being simple” (van der Horst 1991: 18).3 He emphasizes that—for the sake of clarity—one has to be strict about applying these indicators: Only if two or three signs are present can one with some certainty label an inscription as Jewish/Judaean.4  1 E.g., the Iulia Severa inscription from Akmoneia (Oehler 1909: 298 no. 65) and many other instances from Asia Minor (nos. 53, 60–61, 63–64, 67, 70, 73, 76–80, 82, 88). 2 van der Horst 1991: 16–18; see also van der Horst 2015: 8–11. 3 van der Horst 1991: 18. 4 In the following I will use the label “Jewish” instead of “Judaean”, however keeping in mind that Ἰουδαῖος was predominately an ethnic term; see Öhler 2017.



Markus Öhler

In the same year, Ross S. Kraemer devoted a lengthy article to the methodological difficulties of distinguishing Jewish from pagan or Christian inscriptions (Kraemer 1991). She proceeds along the lines of previous scholars but is more sceptical on each of the usually mentioned indicators: Ἰουδαῖος, which is in any case rather seldom attested, is not always an indication of religious identity, but could also point to a place of origin. Συναγωγή or προσευχή are mentioned in pagan inscriptions, too, and the same goes for ἀρχισυνάγωγος and other terms for functionaries, which are commonly thought to be typically Jewish. Phrases and formulas are also not as sufficiently clear indicators as one might hope and Jewish symbols were used by Christians too. Names are notoriously difficult and specific locations of inscriptions are far from conclusive. In her conclusions, Kraemer points out that we should reckon with the problem that the borders between Jewish/Judaean, pagan and Christian were much more blurred than in our reconstructions. In addition, each inscription previously classified as Jewish should be examined again, together with a rigorous discussion of criteria (Kraemer 1991: 161–162). In 1996 Bij de Vaate and Jan Wilhelm van Henten presented their perspective on this problem (Bij de Vaate and van Henten 1996). Their conclusions were much the same as Kraemer’s: One has to work with “extreme caution”, especially if only terminological signs are used for classifying an inscription as Jewish/Judaean. On the contrary, they suggest one should refrain from distinct classifications for a significant number of inscriptions.5 Walter Ameling, in his seminal collection of Jewish inscriptions from Asia Minor that appeared in 2004,6 named several criteria for including inscriptions in his anthology. He considers the use of Ἰουδαῖος, references to specifics of Jewish life (synagogues, festivals, Scripture), provenance from a Jewish building, depiction of Jewish symbols, and the usage of Hebrew script as the most decisive signs for Jewish origin, whereas names and formulas are classified as less significant. Finally, Jonathan Price and Haggai Misgav introduced their discussion of Jewish/Judaean inscriptions with the formulation: “Despite 150 years of scholarship on Jewish epigraphy, no firm criteria have been developed—or are likely to be developed—to distinguish Jewish inscriptions from others” (Price and Misgav 2006: 461). The following review of inscriptions from Asia Minor mentioning synagogues will not contribute significantly to the general discussion of criteria for establishing a Jewish/Judaean origin of several inscriptions. From my point of view, everything has been said, and no entirely new arguments have been brought forward. However, it is significant for our purpose in this volume that the term συναγωγή  5 In their article they discuss attestations for curse terminology in inscriptions from Phrygia (mostly Akmoneia) which are widely believed to be Jewish, esp. by Trebilco (1991) in his work on Jews in Asia Minor. Their conclusion is sceptical: “In fact, more than half of the inscriptions from Acmonia assumed to be Jewish by Trebilco may very well be non–Jewish”, Bij de Vaate and van Henten 1996: 23. 6 Ameling 2004: 8–21. See also Noy, Panayotov, and Bloedhorn 2004: V; Noy and Bloedhorn 2004: V.

The Iulia Severa Inscription Reconsidered



is consistently mentioned as a signifier or—e.g., in the case of Johann Oehler— implicitly used as such. And the task to discuss every epigraphical instance anew and evaluate again the possibilities of Jewish/Judaean, pagan or Christian provenance, is still (and will be) a necessary exercise. Thus, a review of inscriptions from Asia Minor mentioning synagogues might still offer new insights. We will survey the inscriptions, taking a tour from the north of the Anatolian peninsula to the very south, starting with those that are usually classified as Jewish. In the subsequent section we will focus on presumably pagan inscriptions before finally reconsidering the famous inscription mentioning Iulia Severa, for which I will suggest that it might not be a Jewish inscription at all.

Presumably Jewish Synagogues in Inscriptions from Asia Minor Nicomedia in Bithynia IJO II 1547 1

5

10

τὴν θήκην ἔθηκα καὶ τὸν βωμὸν τῇ γλυκυτάτῃ τεκούσῃ Οὐλπίᾳ Καπιτύλλῃ· καὶ βούλομε ἕτερον μηδέν ἀνασκεβάσ[ε]· εἰ μή, ἕξει πρὸς τ[ὴν] κρίσιν καὶ δώσε[ι] τῇ συναγωγῇ (δηνάρια) ͵α καὶ τῷ ταμείῳ (δηνάρια) φʹ. χαίρετε. I have set up the urn and the altar for my sweetest mother, Ulpia Capitylla, (5) and I want no other to destroy it. Otherwise he will face judgement and will pay 1000 denarii to the synagōgē (10) and 500 denarii to the treasury. Farewell!

IJO II 1578 1

[... μηδένα] [ἕτερον τ]εθῆν[αι· ἐὰν] [δέ τι]ς τολμήσῃ, δ[ώ]-

 7 Here, and for the following instances, I refrain from listing previous editions of the cited inscriptions. This has already been done by Ameling (2004). Later re-editions will be noted. For IJO II 154 an online edition has been published by Phil Harland: http://philipharland.com/greco-roman-associations/graveof-ulpia-capitylla-with-fines-payable-to-a-synagogue-ca-250-ce/ (accessed 19 March 2020). 8 See also http://philipharland.com/greco-roman-associations/grave-frag-with-fines-payable-to-asynagogue-of-judeans-undated/ (accessed 19 March 2020).



Markus Öhler σει προστείμ[ου] τῇ συναγωγῇ τῶν Ἰουέων (δηνάρια) ͵α καὶ τῷ ταμίῳ (δηνάρια) ͵β. χαίρετε.

5

[. . . and no one else] to be buried in it. [Now if anyone] dares, he will pay a fine of 1000 denarii (5) to the synagōgē of the Jews and 2000 denarii to the treasury. Farewell!

Both inscriptions can be dated roughly to the middle of the third century CE and come from Nicomedia in Bithynia, where in all five Judaean inscriptions were found. Although the first inscription (IJO II 154) could be understood as referring to a non-Judaean synagogue, the second one (IJO II 157) is most probably Jewish. It covers two oft-cited criteria by mentioning a synagogue and Ἰουδαῖοι.9 In light of this attribution, the inscription for commemorating Ulpia Capitylla (IJO II 154) should also be attributed to the Jewish minority in Nicomedia, which must have been quite substantial. This epitaph includes also a threat that violators of the grave will be doomed (ἕξει πρὸς τ[ὴν] κρίσιν). The absolute wording suggests that God is to be perceived as the judge, although he is not explicitly mentioned here.10 The reference in IJO II 154 to an altar (βωμός) might be considered awkward for a Jewish inscription, but βωμός could also mean a non-cultic installation: a pedestal carrying a bust of Ulpia or the urn with Ulpia’s ashes (see Ameling 2004: 423). Both inscriptions feature a threat of a financial penalty, which should go to the synagogue and to the city-treasury.11 This split of fees is quite common in Nicomedia and other places, also for non-Jewish groups.12 It works as a stimulus for two institutions: On the one hand, members of the synagogue would have a special objective for checking the state of the burial place and, on the other hand, the city-treasurer had the means to enforce the stipulations.13 From the usage of συναγωγή in these inscriptions, it is clear that the term denotes the community of the Jews, not a building. The form Ἰουδέων (IJO II  9

Of course, it is possible that a pagan or a Christian thought that the fine should be payed to the synagogue, but it is not really probable. 10 In IJO II 155 (Nicomedia, after 212 CE) the deceased Aurelius Kyrion explicitly threatens a violator of his grave with judgement from God: ὃς δὲ ἂν παρὰ ταῦτά τι ποιήσει, ἕξῃ κρίσιν πρὸς τὸν θεόν; see van der Horst 1991: 125. 11 The fines inflicted on violators of the provisions in IJO II 154 and II 157 do not exceed normal amounts (cf. IJO II 43, Smyrna III CE). See also IJO II 158 (Nicomedia, III CE), in which the fee is 20,000 denars. The word συναγωγή in this inscription is by the way only reconstructed. 12 See in Nicomedia TAM IV,1 234, 236, 238, 239; see van der Horst 1991: 59; Rebillard 2003: 264; and Ameling 2004: 190. 13 A combination of fines payable to an association and the local treasury was not uncommon; see, e.g., TAM III,1 765 (Termessos; after 212 CE); I.HierapolisJudeich 218 (128–300 CE). For associations as recipients of a fine see, e.g., I.Kyzikos I 211, 291; I.Ephesos 2212. 2446; I.Smyrna 204; Nijf 1997: 56–57.

The Iulia Severa Inscription Reconsidered



157) is attested at other places in Asia Minor,14 but the phrase “synagogue of the Jews” has been found only one more time in Asia Minor (IJO II 36, Kyme or Phokaia; see below).

Kyme/Phokaia in Ionia IJO II 3615 1

5

Τάτιον Στράτωνος τοῦ Ἐνπέδωνος τὸν οἶκον καὶ τὸν περίβολον τοῦ ὑπαίθρου κατασκευάσασα ἐκ τῶ[ν vvvvvvvvv ἰδ]ίων ἐχαρίσατο τ[οῖς vvvvvv Ἰο]υδαίοις. ἡ συναγωγὴ ἐ[vvvvvv τείμη]σεν τῶν Ἰουδαίων Τάτιον Σ[vvvvvvvv τράτ]ωνος τοῦ Ἐνπέδωνος χρυσῷ στεφάνῳ καὶ προεδρίᾳ. Tation, daughter of Straton, son of Empedon, having constructed the house and the columned hall around the portico from her own resources (5) granted them to the Jews. The synagōgē of the Jews honoured Tation daughter of Straton, son of Empedon, with a gold crown and the privilege of the front seat.

This now lost inscription comes from either Phokaia or Kyme on the Ionian coast and is commonly dated to the 3rd century CE, although the basis for this dating is very tenuous. Neither the form of the letters nor the names point to a date any more specific than “imperial times” (Ameling 2004: 164). Since the inscription mentions the Jews as recipients of the gift and also their synagogue, we can be quite sure that it was erected by them. It is however not so certain if Tation, the donor for the community building—οἶκος and περίβολος τοῦ ὑπαίθρου—was a Jewess. Her name and the names of her father and grandfather are Greek, Tation being used quite often in Ionia.16 Thus it is possible that the phrases “she donated to the Judaeans” (ἐχαρίσατο τοῖς Ἰουδαίοις) and “the synagogue of the Jews honoured” (ἡ συναγωγὴ ἐτείμησεν τῶν Ἰουδαίων) reveal a certain distance between Tation and the Jewish community.17 But since Tation is offered a seat in

 14 Ἰουδέων: IJO II 5a (Ikaria; V/VI CE), 187 (III CE; Hierapolis) 231 (1st half III CE; Diokaisarea); Ἐιουδέων IJO II 37 (II/III CE?; Milet), 179 (after 212 CE; Apameia), 241 (IV CE; Korykos). 15 See also Harland 2014: GRA II 106; http://philipharland.com/greco-roman-associations/ honorsby-a-synagogue-of-judeans-for-a-female-benefactor/ (accessed 19 March 2020). 16 See the records in Corsten and Fraser (2010). 17 Cf. Kraemer 1992: 119; see also Brooten 2020: 263; Feldman 1996: 54; Avemarie 2009: 68 n. 135; and Rajak 2001b: 475 (“an outsider”). On euergetism for Jewish synagogues see Feldman 1996: 51–55; and Rajak (2001a).



Markus Öhler

the first row, which she would occupy during their meetings,18 it is more probable that she was herself a wealthy Jewess. She donated the house (οἶκος), which is not called “a synagogue”, to her own Jewish community, the συναγωγή. The inscription was very probably intended to inform outsiders about the generous gift of Tation and the subsequent honours of the συναγωγὴ τῶν Ἰουδαίων. That would explain the somewhat distanced language.

Chios in Ionia SEG LXII 60019 εἰρήνη ταῖς συναγωγαῖς εὐλογία τῷ Ἰσραήλ … Peace to the synagōgais Blessing to Israel …

The mosaic inscription, which is cited here only partly and has not yet been edited, as well as the remains of the synagogue on the island of Chios, are dated to the 3rd–5th century CE. It also mentions two donors and their vows. The references to Israel and the synagogues are sufficient evidence to consider the inscription as Jewish. For the use of the plural συναγωγαῖ see also an inscription from a certain Markos Aurelios Jacob (Thessalonica, 250–300 CE), which refers to fines payable to the synagogues (ταῖς συναγωγαῖς).20 In this inscription as well as in the one from Chios one has to assume that at least two assemblies of Judaeans existed in the city at the time. This is also attested for Side (see below).

Samos in Ionia IJO II 5 face A, front. 1 [οἱ ἄρχοντες κ]αὶ οἱ πρεσβύτεροι καὶ [ … ] [ … τῶ]ν Ἰουδαίων τῆς κατὰ [Σάμον ?] [ … συ]ναγωγῆς ἐτίμησαν Αρ[ … ] [ … πρεσβύτ]ερον, τὸν πάσης δόξη[ς … ] 5 [ … ]Α κτίσαντα ἡμῖν μνήμ[ … ] [ … ]ΕΝΟΝ τῷ τ[ῆ]ς λαμπρ[οτάτης ?] [ … ]ΑΙΔΙΑΔΗ[ … ] ΧΑΡ[ … ]

 18 See Trebilco 1991: 110; Rajak 2001a: 384 (“more likely”); Levine 2005: 509 n. 58; Lieu 2016: 105; and Harland 2014: 97. 19 See also Panayotov 2014: 69f. Pictures of the inscription have been published on naturedigital.blogspot.co.at/2010/09/old-synagogue-chios.html (accessed 19 March 2020). 20 SEG XLIV 556; see Nigdelis 1994: 305.

The Iulia Severa Inscription Reconsidered



[ … ἀ]νεθηκα[ν … ] [ … ] ΑΙΧΑ [ … ] face B, back. 1 …]ΚΤ[… [The leaders] and the elders and […] of the Jews of [Samos …] of the synagōgē honoured Ar[…], the elder, the … full of honour … (5) erected for us …

The inscription from the island Samos is dated to the 3rd century CE or later and honours an unknown presbyter from the local synagogue. Again, the references to Jews and the synagogue are sufficient evidence to categorize the text as Jewish, the mention of presbyters confirms this. The text is however quite fragmentary, e.g., the reference to leaders (οἱ ἄρχοντες) is reconstruced.21 The inscription was possibly fixed to the wall of their place of assembly. In the typical form of inscriptions of a city or an association it honours a donor or a meritorious functionary. But it is obvious that συναγωγή is to be understood as the Jewish community, who is honouring a member.

Sardis in Lydia I.Sardis VII,1 1722 […] καὶ ὅσοις ἀπενεμ[ήθη ὕδωρ· …] γυμνασίῳ γερουσια[κῷ ἐναντία· κρήνη] μυστηρίοις δυσὶ[ν ἐναντία …· κρήνη] Δομιτίας, (ἑκατοντά)χ(ους) εʹ· κρήνη Ε[…] κρήνη Ληναεῖτ(ις) πρὸς τοῖς […]· κρήνη μυστηρίῳ Ἄττει ἐνα[ντία ἀπόρρυ-] τος εἰς τὸ Διός· συναγωγ[ῆς κρήνη …] κρήνη πρὸς τῷ Ὠδείῳ, (ἑκατοντά)χ(ους) γʹ· [κρήνη] του πρὸς τῇ διστέγῳ τῆς [… οὗ καὶ] πύργοι, (τετρ)ά(μφορα) βʹ· ἀνδροφυλακίο[υ κρήνη ἣ ἐπὶ] τῇ καθόδῳ ἀγορᾶς πρὸς [… ὕδωρ] πέμπει· κρήνη ἐν τοῖς […] ρίου· vacat ἀπὸ ὑδρείο[υ κρήνη· κρήνη Λυσι(?)] μάχου, (ἑκατοντά)χ(ους) γʹ· κρήνη Π[… πρὸς τῷ] Μηνογενείῳ, (ἑκατοντά)χ(ους) γʹ· κρή[νη Ἀρσινόης …] ἀπὸ τῆς Ἀρσινόης κρ[ήνη ἀπόρρυτος] περὶ τὸ Μηνός· κρήνη ἐ[ν …] τοῦ πρὸς τῇ πύλῃ· κρ[ῆναι ἃς κατεσκεύασαν]

1

5

10

15

 21

See on archontes in synagogues Claußen 2002: 273–278; and Ameling 2004: 49 n. 43. Buckler and Robinson 1932: 37–40. See also IJO II 53; http://philipharland.com/greco-romanassociations/list-of-fountains-used-by-mystery-associations-a-synagogue-and-others/ (accessed 19 March 2020). The stone was already lost when Buckler and Robinson edited the inscription in 1932. The following text is slightly corrected according to the drawings of the inscription, esp. to the one of G. Hirschfeld (see below). 22

  20

1

5

10

15

20

Markus Öhler Ῥοῦφος καὶ Λέπιδος[…] Αὐρηλία Ἰουλία Μηνο[…] αρα, (τετρ)ά(μφορα) βʹ· Ἀσίννιος N[…] νος, (τετρ)ά(μφορον) αʹ· Φλ Σεκ[…] And to whom water was allocated … the fountain opposite the Gymnasion of the Gerousia, [the fountain opposite of the two mysteria, [the fountain of Domitia, 500 chous. The fountain of E[… the fountain of Lenaeitis at [… the fountain across from the mysterion of Attis, [which flows into the sanctuary of Zeus; [the fountain of?] the synagōgē; the fountain at the Odeieon, 300 chous ... [the fountain of … at the two vessels of [… which has also? towers, two four-amphore; [the fountain of] the guard-house [which is on? the descending way of the agora at [… water? sends; the fountain in the [… from the reservoir [… the fountain of Lysimachos, 300 chous; the fountain of P[… at the? Menogeneion, 300 chous; the fountain [of Arsinoe ? [which flows?] from that of Arsinoe [… at the sanctuary of Men; the fountain in [… which is near the gate; the fountains(?) [build by? Rufus and Lepidus [… Aurelia Iulia Meno[… two four-amphore; Asinnios M[… one four-amphore; Fl(avius) Sek[…

The dating of this inscription from Sardis to approximately 200 CE is quite certain, its character far less.23 Basically, it is a list of fountains including references to sanctuaries, public buildings and houses of individuals. At the end, the sponsors of some fountains might be mentioned. It is obviously not a Jewish inscription but one produced by the city of Sardis. A συναγωγή is listed without any further specifications, although it is possible that the now lost text to the right contained more information. According to the criteria for considering an inscription as Jewish, it is highly doubtful that the “synagogue” is a Jewish institution or building, since no other indications in the inscription for that identification are given. Nevertheless, it is frequently treated as a document referring to a Jewish synagogue building in Sardis.24 As it is not stated that the synagogue belonged to the Jews or to any other group, this might indicate that the writer and the presumed readers of the inscription must have known what “the synagogue” was. Since in this part of the  23 The dating to ca. 200 CE has been established by Herrmann 1993: 247, 253, 257, on the basis of the mentioned Asinnius (Nikomachos Frugianus) in line 21; see also Ameling 2004: 212. I cannot see why some scholars claim that the date of the inscription is “uncertain”; e.g., Binder 1999: 284 n. 96; and Magness 2005: 460 n. 110. 24 See, e.g., Buckler and Robinson 1932: 40; Frey 1936–1952: CIJ II 751; Seager and Kraabel 1983: 169; Trebilco 1991: 41; Binder 1999: 284; Lieu 2003: 204; Rajak 1998: 231; and Ameling 2004: 213.

The Iulia Severa Inscription Reconsidered

 

inscription the list seems to enumerate only buildings, it is also more probable that the συναγωγή is a physical structure, not a community. Other buildings mentioned in the inscription have similar descriptions: three μυστήρια (lines 3 and 6—possibly halls used by mystes [see Buckler and Robinson 1932: 40]), a sanctuary of Zeus (τὸ Διός line 7, but see below) and one of Men (τὸ Μηνός line 17), the hall for musical performances and poetry competitions (Ὠδείον line 8), the quarter of the guards (ἀνδροφυλάκιον line 10), all those are physical structures—something that has to be presupposed for συναγωγή too.25 To the scribe or a city councillor in Sardis around 200 CE the character of “the synagogue” was obvious, to the reader today this is not so clear: Was the synagogue a building owned by the Jewish minority or did it belong to a pagan association? A Jewish community was already established in Sardis more than 250 years before our inscription was made. Josephus (A.J. 14.259–261) cites a decree by the Sardians from 47 BCE according to which the Jewish citizens ( Ἰουδαῖοι πολῖται) were—among other things—permitted to assemble (συνάγωνται) and have a communal life (πολιτεύωνται).26 In addition, it is stipulated that they should be given a place, “in which they may gather with women and children” (τόπος …, εἰς ὃν συλλεγόμενοι μετὰ γυναικῶν καὶ τέκνων), to perform their ancestral prayers and sacrifices to God (τὰς πατρίους εὐχὰς καὶ θυσίας τῷ θεῷ).27 Also, a place (τόπος) for building and habitation—perhaps a certain part of the city—should be set apart for them (εἰς οἰκοδομίαν καὶ οἴκησιν αὐτῶν). If we assume that Josephus cited the original text of the decree this could be a perfect supplement to our inscription. It might refer to “the synagogue” without any details because the place of the synagogue in 200 CE had been granted to the Jews a long time ago and was known as “the synagogue” without any further specifications.28 A distinction between place and community can be seen in another text from Josephus’s account (A.J. 14.235): In an earlier decree by the Roman magistrate Lucius Antonius, dating to the year 50 or 49 BCE, it is stated that some time ago the Jews in Sardis had been permitted to have their own association (σύνοδον ἔχειν ἰδίαν) and their own place (τόπον ἴδιον) (Pucci Ben Zeev 1998: 176–181).  25 One should note, however, that κρήνη in line 7 (as in many other lines) is a reconstruction by the editors. Also the genitive συναγωγῆς is dubious. It is also possible to read the dative συναγωγῇ, which might have been followed by ἐναντία (see lines 2–3, 6). Equally possible is that τῶν Ἰουδαίων followed to the right, since no one knows how long the lines really were. 26 See Pucci Ben Zeev 1998: 217–225; and Catto 2007: 30–31. 27 For the use of τόπος in the context of synagogues see Pucci Ben Zeev 1998: 221–222; Claußen 2002: 144. The appearance of sacrifices might point to the authenticity of the cited decree; see Pucci Ben Zeev 1998: 223. 28 On the date of the excavated synagogue building in Sardis see, e.g., Botermann (1990); and Magness (2005). The debate whether the synagogue was constructed in the fourth or sixth century, is irrelevant to our discussion of I.Sardis VII,1 17, which is at least 150 years older than the archaeological remains (but see, e.g., Trebilco 1991: 41; and Lieu 2003: 204). It is also improbable that an earlier structure at the same place is meant, since the 4th/6th-century synagogue was an adaption of a Roman bath and gymnasium complex; see Magness 2005: 444.

 

Markus Öhler

For a Roman official in the first century BCE the Judaeans in Sardis assembled as an (ethnic) group, although it is not entirely clear whether the τόπος was a building or a civic meeting-place (see Pucci Ben Zeev 1998: 178). If our list of fountains (I.Sardis VII,1 17) indeed contains a reference to a Jewish synagogue, then it is highly probable that this was the structure owned by the Jewish minority, even if συναγωγή is not mentioned in Josephus.29 But as clear as this might seem, there are three caveats to this interpretation: 1) The stone is broken on the right side. No one knows how long the lines were and which words were inscribed after ΣΥΝΑΓΩΓ[Η]. 2) According to Hirschfeld’s drawing of the inscription, line 7 ends with a Gamma, not an Eta (as in Rayet’s sketch and the edition in Buckler and Robinson).30 The last preserved letters of line 7 are therefore ΣΥΝΑΓΩΓ. 3) The following κρήνη in line 7 is a reconstruction by Buckler and Robinson. It has the disadvantage that normally the list of fountains names the owner of the fountain after κρήνη, not the other way round. The other two instances of the sequence genitive + κρήνη (lines 10 and 13) are also only reconstructed. These three points offer at least two alternative interpretations of line 7. For the first alternative reading it should be noted that line 7 starts with three letters belonging to a word in the line above, followed by ΕΙΣ ΤΟ ΔΙΟΣ. Ameling already speculated that the reading might be εἰς τὸ Διὸς συναγώγιον, but rejects it as too desperate (“zu verzweifelt”).31 In my opinion this was premature, especially since I.ApamBith 35 indeed refers to a synagōgē of Zeus: ἐν τῆι τοῦ Διὸς συναγωγῇ (see below). The reading in I.Sardis VII,1 17 might thus be reconstructed as εἰς τὸ Διὸς συναγωγήν, which would imply that συναγωγή was the label for a building of a religious association devoted to Zeus. Such a group is actually attested in Sardis.32 Several inscriptions contain references to the worship of Zeus by groups who call themselves μύσται and θεραπευταί,33 although it is not clear whether those two groups belonged to one and the same or to different associations. The existence of an association of Zeus-initiates in Sardis, at least from I BCE to II CE, is beyond doubt. Admittedly, there is no other reference to a synagogue in these inscriptions, but, in my opinion, it is quite probable to  29 And at an even later time the Judaean minority of the city acquired a new building and transformed it into a large synagogue right in the middle of the city. See, e.g., Kraabel (1978/1992); and Kraabel (1983/1992). 30 The pictures are reproduced in Buckler and Robinson 1932: 38. 31 Ameling 2004: 213 n. 53. Liddell, Scott, and Jones 1996: 1692, refer for συναγώγιον to Philo, Som. 2:127, and Pollux 6:71. 32 See Hanfmann 1983: 131–132; and Herrmann 1996: 321–335. 33 See Herrmann 1996: 323 (I.Sardis VII,1 22; I BCE); Herrmann 1996: 321 (SEG XLVI 1529; I/II CE); Herrmann 1996: 328–329; (SEG XLVI 1530; I BCE); see also Herrmann 1996: 330 (SEG XXIX 1205; II CE). I.Sardis VII,1 8 mentions a sacred precinct of Zeus Polieus and Artemis (1 BCE; lines 133–134) and a document from Pergamon a priest for Zeus Polieus (τοῦ δὲ Διὸς τοῦ Πολιέως Ἀλκαίου; I.Perge 268,35– 34; 98–94 BCE).

The Iulia Severa Inscription Reconsidered

 

interpret the synagogue of the fountain-list as a reference to a group of Zeus worshippers and their place of assembly. A second alternative is opened up by a different understanding of συναγωγή. As many have observed, the noun is used also to describe the assembling of things like crops, various objects, money, or animals.34 In the LXX the term is used also for the gathering of water: Gen 1:9 formulates καὶ εἶπεν ὁ θεός συναχθήτω τὸ ὕδωρ τὸ ὑποκάτω τοῦ οὐρανοῦ εἰς συναγωγὴν μίαν, so that συναγωγή would mean a basin or water reservoir. In Lev 11:36 the author lists πληγαὶ ὑδάτων (springs of water), λάκκοι (cisterns) and a συναγωγὴ ὕδατος, presumably a collection basin or pond for water. In the context of our list of fountains in I.Sardis VII,1 17 this would make perfect sense: After a number of fountains it refers to a place where water is collected, maybe a concourse of water pipes or a small pond.35 Both alternatives indicate that it is not likely that the list of fountains (I.Sardis VII,1 17) would contain an actual reference to a Jewish synagogue. It should rather be treated as evidence for the use of συναγωγή as a term denoting a building, more probably one of an association of Zeus-initiates.

Philadelphia in Lydia IJO II 49

5

10

[τ]ῇ ἁγιοτ[άτῃ] [σ]υναγωγῇ τῶν Ἑβραίων Εὐστάθιος ὁ θεοσεβὴς ὑπὲρ μνίας τοῦ άδελφοῦ Ἑρμοφίλου τὸν μασκαύλην ἀνέθηκα ἅμα τῇ νύμφι μου Ἀθανασίᾳ To the most holy synagōgē of the Hebrews I, Eusthatios (5) the Godfearer, for the remembrance of the brother Hermophilos, (10) donated the basin together with my sister-in-law Athanasia.

 34 See Liddell, Scott, and Jones 1996: 1692. Epigraphical evidence for this use from Asia Minor: CIG 3056 (170/140 BCE; Teos); I.Erythrai 21 (277/5 BCE; Erythrai); IK.Rhod.Peraia 303 (II BCE?; Syrna/Bayir); I.Iasos 612 (127 BCE; Bargylia); OGIS 748 (280–275 BCE; Kyzikene). 35 To my knowledge the use of συναγωγή for such an installation has not been found in any other inscription. On terminology for fountains see Dorl-Klingenschmid 2001: 18–20.



Markus Öhler

This inscription from Lydian Philadelphia dates to the late 3rd or 4th century CE. It identifies the synagogue as one “of the Hebrews”, so that it is doubtless a Jewish inscription.36 In addition, the synagogue is called “the most holy one”, which in my opinion shows that not the building but the community is signified.37 The use of ἁγιοτάτος/ἁγιώτατος is an indication that the synagogue was perceived as something special, be it the place or the community.38 The benefactor Eustathios was quite probably a Jew who described himself as God fearing.39 The identification as a community of Hebrews is potentially a hint on the specific role of Hebrew as the holy language of Jews at that time. In addition, it might be important that this inscription was erected in the community building of the synagogue, so that “synagogue of the Hebrews” was perhaps more an insider designation than the official title of the Jewish community in Philadelphia.40

Hierapolis in Phrygia IJO II 19141 A

Ἡ σορὸς καὶ τὸ ὑπὸ αὐτὴν θέμα καὶ ὁ περὶ αὐτὴν τόπος Νεικοτείμου Λυκίδα τοῦ Ἀρτεμίσιου, ἐν ᾗ κεκήδευται Ἀπφιάς, ἡ γυνὴ αὐτοῦ. ταύτης ἀντίγραφον ἀπετέθη εἰς τὸ ἀρχεῖον. Ἰουδαηκή. Ἡ σορὸς καὶ ὁ περὶ αὐτὴν τόπος Αὐρ. Ἑορτασίου Ἰουλιανοῦ Τριπολείτου Ἰουδέου, νοῖν οἰκοντ[ος] ἐν Εἱεραπόλι, ἐν ᾗ κηδευθήσεται αὐτὸς καὶ ἡ γυνὴ αὐτοῦ Γλυκωνίς, κηδεύωνται δὲ καὶ τὰ τέκνα αὐτῶν, ἑτέρῳ δὲ οὐδενὶ ἐξέσται κηδευθῆναι, ἐπεὶ ὁ παρὰ ταῦτα ποιήσας δώσει τῇ ἁγιωτάτῃ συναγωγῇ ἀργυροῦ λείτρας δύο.

B

5

A

The sarcophagus and the base beneath it and the place around it (belong to) Neikotimos Lykidas, the son of Artemisios; in it Apphias has been buried, his wife. A copy of this has been deposited in the archive. Judaean.

 36

A similar designation is preserved from Corinth: IJO I Ach47 (late III CE or later). Jewish groups could also call themselves “the Hebrews”; see Noy 1993: 2, 33, 578, 579. 37 On the topic of holiness in connection with Judaean synagogues see Fine (1997); and Avemarie 2009: 57. 38 For other examples of the use for synagogues see below IJO II 191 and Ameling 2004: 125. It is also very prominent in Christian inscriptions, so that mutual influence should be taken into account. Other institutions like the city treasury could also be described as “holy” (ἱερώτατος), e.g., in Hierapolis: IJO II 193 (after 212 CE); 199 (II CE); 200 (after 212 CE); 202, around 200 CE; see also IJO II 204 (around 200 CE). 39 See Trebilco 1991: 162; and Ameling 2004: 205–206. 40 See on that also Stebnicka 2015: 159–163, who however strains the implications of the term. 41 See online http://philipharland.com/greco-roman-associations/149-grave-of-a-judean-familyinvolving-a-synagogue/ (accessed 19 March 2020).

The Iulia Severa Inscription Reconsidered B



The sarcophagus and the place around it (belong to) Aurelios Heortasios Ioulianos, citizen of Tripolis, Jews, now living in Hierapolis. In it he himself will be buried and his wife Glykonis, and also (5) their children shall be buried (here). No one else is allowed to be buried (here), whoever does otherwise will give two Litra of silver to the most holy synagōgē.

These two doubtlessly Jewish inscriptions on a sarcophagus from Phrygian Hierapolis are from different times, although they are quite similar.42 Whereas inscription A is from the end of the 2nd or beginning of the 3rd century CE, inscription B is from the second half of the 3rd century CE or even later. In this second text, the writer threatens people who use the grave against the owner’s will with a fee payable to “the most holy synagogue”.43 This description of the synagogue community is, as we have seen, not unusual.44 And, obviously, Συναγωγή does not mean the building, but the community of Jews in Hierapolis. In this respect, it is worth noting that in an earlier inscription (IJO II 205) from the 2nd half of the 2nd century CE the community of Jews in Hierapolis is called κατοικία τῶν ἐν Ἱεραπόλει κατοικούντων Ἰουδαίων. We need not deal with this here in too much detail,45 but it is certainly remarkable that in this inscription the writer does not mention the synagogue, but uses a different title for the community: κατοικία. This designation might stem from a specific living quarter of the Jews in Hierapolis. By the late 3rd century CE the κατοικία might have been called with the more common name “synagogue”. While living as a close community, like in Sardis (see above), the Jews adopted the designation as a synagogue at a later time, possibly due to a greater dispersion through the city or because of the growing importance of the title “synagogue” in the 3rd and 4th centuries CE.

Hyllarima in Caria IJO II 20 1

ὑπὲρ ὑγίας [.]ΝΟ[…]ΙΑΙ […] βασιλέως Ι[․․]․ Αὐρ. Εὐσανβάτιος πρεβύτερος καὶ Αυρ. Ἐπιτυνχούσα ὑπὲρ τῆς ἑατῶν σωτηρίας καὶ παίδων αὐτῶν καὶ ΚΙΦΙΙ[...]ΟΥ

 42 It is of course possible that “Judaean” in inscription A is barely a geographical label, but in view of inscription B this is not probable. Ἰουδαηκή might even label the sarcophagus as Judaean; see Ameling 2004: 407; and van der Horst 2009: 285. 43 In other Jewish grave inscriptions from Hierapolis fees should be payed to the gerousia (IJO II 189, II CE) or to the fiscus (IJO II 192, II CE; 198, II CE); see Harland 2006: 240. 44 See above; and Harland 2006: 227 n. 20. 45 But see the connections between this designation of the local synagogue with decrees cited in Josephus (A.J. 14.172, 259); see Miranda 1999: 150–151; Ameling 2004: 433; and Harland 2006: 226.

 5

Markus Öhler τῇ ἁγιωτάτῃ συναγωγῇ τὰ ΠΡΑ[... ἐκ τῶν ...]άτων. Α For the health … the emperor … Aurelius Eusanbatios, elder, and Aurelia Epitynchanousa (gave) for their own salvation and their children and … (5) to the most holy synagōgē the [from their own means].

The use of the names Aurelius and Aurelia show that this very fragmented inscription from Hyllarima in Caria is post-212 CE, perhaps from the 3rd century CE. Since a presbyter and a “most holy synagogue” are mentioned, it is quite probable that the inscription was donated by a Jewish couple. They gave something to their community, which is—again—called the most holy.46 A donation “for salvation” is intended as a gift to a deity and should secure health and wellbeing of the donors. This is very widespread in Hellenistic inscriptions and not typically Jewish.

Side in Pamphylia IJO II 219 1

5

[Ἰσ]άκις φροντιστὴς τῆς ἁγιωτάτ[ης] πρώτης συναγωγῆς ἔστην εὐτ[υχ]ῶς καὶ ἀνεπλήρωσα τὴν μαρμάρωσιν ἀπὸ [τοῦ] ἄμβωνος ἕως τοῦ σῖμμα καὶ ἔσμηξα [τὰ]ς δύο ἑπταμύξους καὶ τὰ δύο κιονοκέφαλα· ἰνδ(ικτιῶνος) ιεʹ, μη(νὸς) δʹ. I, Isakis, was happily appointed as curator of the most holy and first synagōgē and accomplished the installation of the marble-floor from the Ambo to the Sigma and let (5) the two seven-armed chandeliers and the two capitals be cleaned, 15th indiction, 4th month.

This is a rather late example from Side (IV/V CE) for the use of the phrase “the most holy synagogue”. Its Jewish character is also indicated by the name Isakis, which is a Greek form of Isaac and very probably not Christian. It is not clear whether “synagogue” should be understood as a reference to a building or to the community. The activities of the honouree were mostly renovation and cleaning of the interior of the synagogue building, so that it is possible to think of “the most holy and first synagogue” as a physical structure.47 The genitive τῆς ἁγιωτάτης πρώτης συναγωγῆς could accordingly be understood as a genitivus objectivus: Isakis was the curator of an edifice, which was the first synagogue building in Side or a renovation of it.  46 See, e.g., Ameling 2004: συναγωγή “kann hier nicht immer das Gebäude bezeichnen, sondern muß eine Institution meinen: Heilig ist die ‘Versammlung’ des auserwählten Volkes” (125). 47 See Fine 1997: 140; and Claußen 2002: 122.

The Iulia Severa Inscription Reconsidered



If we read it as a subjective genitive, “the most holy and first synagogue” would be the congregation of Jews in Side, who appointed Isakis as their curator for the renovation. The implication of the label “first synagogue” would then be to praise this one over other Jewish communities in Side.48 Both possibilities are plausible, a decision depends on the perspective of the reader.

Pagan Use of συναγωγή in Inscriptions from Asia Minor Triglia in Bithynia I.ApamBith 3549 οἱ θιασῖται καὶ θιασίτιδες [ἐ]στεφάνωσαν Στρατονίκην Μενεκρ[ά][τ]ου ἱερωτεύσασαν ἐν τῶι ηʹ καὶ οʹ καὶ ρʹ [ἔ]τει Μητρὶ Κυβέλῃ καὶ Ἀπόλλωνι στεφά[ν]ωι γραπτῶι ἐν στήλλῃ καὶ κηρυκτῶι σὺν ται[νί]αι καὶ ἄλλωι στεφάνῳ κηρυκτῶι σὺν τα[ινί]αι ἐν τῆι τοῦ Διὸς συναγωγῇ φιλαγαθήσασ[αν].

5

The male and female members of the thiasos crowned Stratonike, daughter of Menekrates, who was priestess of Mother Cybele and Apollo in the 178th year, with a crown with a band (5) engraved on a plaque that was announced and another crown with a band that was announced in the synagōgē of Zeus, since she acted in a benevolent manner.

The inscription, coming from Triglia in the Apameia Area (Bithynia), can be dated either to 119 or to 104 BCE, perhaps even later. It is engraved into a stele with two reliefs: One shows two deities—Cybele and Apollon—and a group of worshippers, one of them presumably Stratonike. The bottom relief depicts a banquet scene with dancers, musicians and servants which is quite similar to dedicatory reliefs from nearby Mysia (see Brehm 1996: 488–489). Two other inscriptions (I.ApamBith 33 and 34) belonged to the same association, although they do not refer to any of the deities mentioned in I.ApamBith 35 (see Corsten 1987: 50–51). The cult group worshipped apparently three deities, Cybele, Apollon and Zeus. It assembled both men and women, which is rather seldom.50 This unusual constellation might have something to do with specific  48

See Ameling 2004: 463. IJO II 220 is another inscription from Side (late IV/V CE), maybe from a different synagogue. 49 See also Harland 2014: GRA II 99; http://philipharland.com/greco-roman-associations/honors-bya-society-for-a-priestess-with-a-relief/ (accessed 19 March 2020). 50 For other examples see TAM V 1539 (= GRA II 117; Philadelphia, about 100 BCE), a householdbased cult group for Zeus and other deities, and IGUR I 160 (Torre Nova, 160–165 CE), a Dionysiac association connected to a prominent family.



Markus Öhler

local conditions that we are not able to grasp. It could be due to a Dionysiac orientation of this association,51 but since Dionysos is mentioned in none of the three inscriptions, this is rather unlikely. The honoured Stratonike however was a priestess of Cybele and Apollo, which is also shown in the accompanying relief.52 In the last line of the inscription it is stated that the honours for Stratonike should be announced ἐν τῆι τοῦ Διὸς συναγωγῇ.53 There are three different options for an interpretation of this clause: First, συναγωγή could mean the building owned by the community (Harland 2014: 53). The Theodotus inscription from Jerusalem uses the word in this way (CIJ II 1404 = CIIP I 9; before 70 CE: ᾠκοδόμησε τὴν συναγωγήν), as do late antique inscriptions from Syrian Tafas (IJO III Syr34, IV CE: τὴν συναγωγὴν οἰκοδόμησαν) and Greek Aegina (IJO I Ach58, 300–350 CE: τὴν σ[υναγωγὴν] οἰκοδόμησα). It is already attested in a list of donors for the renovation of the synagogue building in Berenike (CJZC 72; 55/56 CE: εἰς ἐπισκευὴν τῆς συναγωγῆς).The author of Luke-Acts employs the term in the same way in Luke 7:5 (τὴν συναγωγὴν αὐτὸς ᾠκοδόμησεν ἡμῖν) and Acts 18:7 (ἡ οἰκία ἦν συνομοροῦσα τῇ συναγωγῇ). The sequence ἐν τῆι τοῦ Διὸς συναγωγῇ would, however, imply that the building was metaphorically owned by Zeus, not by the community.54 This is a considerable difference from the use in Jewish contexts, but has a possible parallel in the fountain-list from Sardis, where a building is definitely meant (I.Sardis VII,1 17; see above). Second, συναγωγή could mean an actual gathering of the association. The fact that the second crown for Stratonike should be announced in the synagogue of Zeus (κηρυκτῶι … ἐν τῆι τοῦ Διὸς συναγωγῇ), not installed in a building, speaks in favour of this interpretation. As Kybele and Apollo are also mentioned and depicted on the accompanying relief, it was perhaps important to state that Zeus was the god worshipped during the specific meeting, in which Stratonike’s service to the association should be celebrated. The use of συναγωγῆ for actual meetings of people is widely attested and correlates with the root of the term (see below). Third, συναγωγή could mean the association itself. If we connect ἐν τῆι τοῦ Διὸς συναγωγῇ with φιλαγαθήσασ[αν], this would mean that Stratonike acted within the συναγωγή τοῦ Διὸς. Accordingly, this would imply that the whole  51

See Harland 2014: 52. Women were important for the plays of the Dionysiac myths. Corsten 1987: 55, interprets the group of worshippers of Cybele and Apollon as a sub-section of the main association. 53 For the use of κηρυκτός for the proclamation of honours see also I.ApamBith 34. The word is to my knowledge used only in these two inscriptions. 54 For the combination of συναγωγή with the name of a “deity“, see also Rev 2:9; 3:9 (συναγωγὴ τοῦ σατανᾶ): Although the author of the Book of Revelation has the intention to describe the character or behaviour of people attending the community (or assembling in the building?), he does so by polemically referring to their patron “deity”. 52

The Iulia Severa Inscription Reconsidered



group was an association of worshippers of Zeus.55 Two examples for a similar use of συναγωγή in a non-Jewish context are attested, both from Perinthos in Thrace: One is a “synagogue of small-wares dealers (or oar-dealers)” (I.Perinthos 59; I/II CE),56 the other a “synagogue of barbers” (I.Perinthos 49; I CE).57 All of these pagan examples specify the group with a description of their trade, apparently because synagogue alone was not clear enough. It is fairly comparable with the usage of the word in a Jewish Diaspora context, in which the synagogue is also specified by the label of the people gathered in this association. But again, in I.ApamBith 35, συναγωγή is constructed with the name of the deity worshipped, not with a group of people. A decision between these three different options is hard to make, although I would tend to interpret it as reference to a building, especially with regard to the Sardis inscription (I.Sardis VII,1 17).

Teos in Ionia GRA II 14158 This relatively long inscription, now lost, is located at Ionian Teos and is dated to the period between 146–133 BCE. It is cited here only partially, as most of the text is not relevant for our question.59 1

vac. ψήφισμα Ἀτταλιστῶν. vac. γνώμη τοῦ κοινοῦ τῶν Ἀτταλιστῶν· ἐπειδὴ ὁ εὸς τῆς συνόδου Κράτων Ζωτίχου ἔν τε τῶι ζῆν πολλὰς καὶ μεγάλας ἀποδείξεις ἐποιεῖτο τῆς πρὸς τοὺς Ἀτταλιστὰς εὐνοίας καὶ κατ’ ἰδίαν ὑπὲρ ἑκάστου καὶ κατὰ κοινὸν τῶν ὑφ’ ἑαυτοῦ συνηγμένων καὶ κειμένων, τὴν πλείστην ποιούμενος πρόνοιαν, σπουδῆς καὶ φιλοτιμίας οὐθὲν ἐλλείπων, καὶ πολλὰ μὲν [ἀγαθ]ὰ καὶ φιλάνθρωπα τῆι συνόδωι παρὰ τῶν βασιλέων

5

 55

See Corsten 1987: 55, who distinguishes a larger cultic association comprising men and women (the thiasos) from a smaller group with only men (the synagogue, which is represented in I.ApamBith 33 and 34). This would also explain the duplication of crowns bestowed on behalf of Stratonike; cf. already Foucart 1873: 6; see also Steinhauer 2014: 95. 56 ἀγαθῇ τύχῃ· Δρ[άκων ὁ καὶ?] | Χρῆστος Χρήστ[ου ἀνέ]- | θηκεν συναγωγῇ [ῥ]ωπο | πωλῶν τῶν περὶ Σωκρά- | την Μενίσκου. “To good fortune. . . . Drakon who is also called (?). . . Chrestos, son of Chrestos, set this up for the synagōgē of small-wares dealers (or oar-dealers) who are gathered around Sokrates, son of Meniskos.” 57 […] τῇ συναγω|[γ]ῇ τῶν κουρέω|[ν π]ερὶ ἀρχισυνάγ|[ωγ]ον […]. “[…] to the synagōgē of barbers, those around the archisynagōgos […].” See below. 58 See also CIG 3069; OGIS 326; http://philipharland.com/greco-roman-associations/honors-by-theattalists-for-kraton-152-bce/. The following text has been slightly changed (with OGIS) compared to the edition in Harland 2014: 325–326; see Le Guen 2007: 263. 59 The inscription continues after line 13 with further descriptions of the deeds of Kraton and the honours he should receive. Another inscription on the reverse side of the stone (CIG 3071) contains a list of items left to the Attalists by Kraton in his testament.

 10

Markus Öhler ἐποίησεν, ἀποδεχομένων αὐτῶν τήν τε ἐκείνου ἅπαντα τρόπον πρὸς ἑαυτοὺς εὔνοιαν καὶ τὴν ἡμετέραν αἵρεσιν καὶ συναγωγὴν ἀξίαν οὖσαν τῆς ἑαυτῶν ἐπωνυμίας, … Decree of the Attalists. Motion of the Koinon of the Attalists: Since the priest of the synodos, Kraton son of Zotichos, made during his lifetime many and great demonstrations of his goodwill towards (5) the Attalists, and did not fall short of zeal and pursuit of honour individually for each one as well as jointly for those who were gathered by him and chosen, doing this with greatest foresight, and created many good things and privileges for the synodos from the kings, so that they accepted his goodwill towards them in every respect and our hairesis and synagōgē, which is worthy of their name. …

The group behind this inscription was devoted to the worship of the Attalid dynasty, who ruled Pergamon from 188 to 133 BCE. The honoured Kraton was a very prominent flute-player and is also mentioned in honorific decrees at other places.60 He not only founded the association for the worship of the Attalid dynasty in Teos, he also secured funding and privileges by the kings. In this context the terms αἵρεσις and συναγωγή appear. They are used as alternative expressions for κοινόν (“community”) and σύνοδος (“gathering”), which shows that συναγωγή does not simply denote an actual meeting. It is the group as a meeting of people with common interests and the use of συναγωγή emphasizes that aspect of their communal activities. It is however not the name of the association, which is κοινὸν τῶν Ἀτταλιστῶν.

Kasossos in Caria I.Mylasa 94261 […?...] 1 […]ΩΙΙ[...]ΕΙΛΕ[…λή][ψε]ται ἐ[ξ] ἁπασῶν [τῶν β]οῶν, [καὶ ὅταν κα][τὰ] κοινὸν θύωσιν Κασωσσεῖς, κ[αθ’ ἑ]κ[άσ][τη]ν θυσίαν ἀφ’ ἑνὸς ἱερείου τῶν θυομέ5 [νω]ν· ὅταν δὲ ποιῶσιν τυράλφιτον κ[αὶ] θύω[σι]ν βοῦν, λήψεται τοῦ βοὼς κωλεὸν ἕλ[κο]ντα μὴ ἔλασσον μ(η)ῶν δέκα, λήψε[τα]ι δὲ ἐν τῆι θυσίαι ταύτηι δέρμα καὶ [σ]κέλος ἀφ’ οὗ ἂν κριοῦ βούληται, λήψε10 [τ]αι δὲ ἐν ταῖς συναγωγαῖς πάσαις [δ]ιμοιρίαν· ὑπάρξει δὲ αὐτῶι καὶ ἀτέ[λ]εια ἐγ Κασωσσῶι ὧν ἂν Κασωσ-

 60

See Le Guen 2007; and Harland 2014: 327–329. See for text and translation Carbon and Peels in CGRN 184 (http://cgrn.ulg.ac.be/file/184/; accessed 19 March 2020). 61

The Iulia Severa Inscription Reconsidered



[σε]ῖς κύριοι ὦσιν· ὁ δὲ πριάμενος τὴν ἱ[ερω]σύνην διορθώσεται τὸ ἀργύριον [τοῖς ἀν]θειρημένοις ταμίαις τοῖς πε[ρὶ Διομ]ήδην, ποιούμενος καταβολὰς [τὴ]ν μὲν πρώτην ἐν μηνὶ Δίωι, τὴν [δὲ δευ]τέραν ἐμ μηνὶ Ξανδικῶι, τὴν δὲ [τρίτη]ν ἐμ μηνὶ Ὑπερβαρεταίωι τῶι μετ[ὰ] [στεφανηφόρον Ἀπελ]λαῖον· οἱ δὲ ἀνθε[ιρη][μένοι] ταμίαι [...]Τ[...] […]

15

20

[... he will receive... ] from all the female oxen, [and also whenever] the Kasosseis sacrifice jointly, for each sacrifice from one of the animals among those sacrificed. (5) And whenever they make a “cheese-and-barley” offering and sacrifice an ox, he will receive a thigh weighing no less than 10 minai from the ox, and, during this sacrifice, he will also receive the skin and the leg from whichever ram he wishes, and he will receive a double portion (10) during all of the synagōgais. He will also be exempt at Kasossos from all (of the taxes and liturgies) over which the Kasosseis have authority. The purchaser of the priesthood will pay the money to (15) the appointed treasurers who (work) with Diomedes, making (three) instalments: the first in the month of Dios, the second in the month of Xandikos, the third in the month Hyperberetaios, the one [after the (20) stephanephoros] Apellaios. The appointed [treasurers ...]

Kasossos, a place 15 km south of Mylasa in Caria, is not known from literary sources, although a couple of inscriptions have been preserved. I.Mylasa 942 is dated to 200–100 BCE and is documentation of a contract of sale for the priesthood of a deity, probably Zeus (see I.Mylasa 941, 943). The inscription lists certain prerogatives of priests concerning shares of the sacrificed animals, exemptions from taxation and the modalities of payment.62 The use of συναγωγή in line 10 is typical for the original meaning of the word: In the context of social behaviour it means “gathering”, in this case gatherings of festive communities. Those were connected with banquets, during which priests should receive double portions of the sacrificial meal, in addition to the parts of the animals described in the previous lines. An interpretation of ἐν ταῖς συναγωγαῖς πάσαις as buildings would make sense in principle, but is very unlikely; the plural speaks against it. Equally improbable is the notion that the συναγωγαῖ were Zeus-associations, again because of the plural. Instead it is one of the many instances for the meaning “gathering/meeting”.63  62

See for a thorough description and commentary Carbon and Peels in CGRN 184 (see previous note). For a similar use see, e.g., Robert, Le Carie II no. 9 (Tabai, no date), the regulations of Epikteta (IG XII,3 330; Thera, 210–195 BCE), the document of an association for Zeus (P.Lond. VII 2193; Fayum, 69– 58 BCE) or a decree of an association of farmers (SEG VIII 529; Psenamoisis near Alexandria; 67/64 BCE). Another example from Alexandria (SB XXII 15460; 5 BCE) refers to the assembly of an association in the house of the σύνοδος. From Thessaliotis (Greece) comes a short mentioning of a synagogue of the Young, i.e., a meeting of young people from the city (IG IX,2 259; 117 BCE). See also JIGRE 20 (no date, Alexandria): This inscription mentions a προστάτης of a synagogue, but it could be Jewish and the wording is not secure. For the same usage in a Christian context see Jas 2:2: ἐὰν γὰρ εἰσέλθῃ εἰς συναγωγὴν ὑμῶν. 63

 

Markus Öhler

An Interim Conclusion Before we turn to the Iulia Severa inscription from Akmoneia, let us summarize our overview of the use of συναγωγή in Asia Minor: 1) The pagan attestations of the word are threefold: Apart from the common meaning, “gathering/meeting”, it is used at least one time for a building (I.Sardis VII,1 17; 200 CE), perhaps also much earlier in Triglia (I.ApamBith 35; late II BCE). For I.ApamBith 35 it is however possible that it denotes an association or a single meeting. 2) Almost all of the Jewish inscriptions come from the third to the fifth century CE and use συναγωγή always for the community as an association, never for a building or a gathering. The only exception would be I.Sardis VII,1 17, if a building of Jews is meant. It would then also constitute the earliest epigraphic attestation of the term as a label for the building of a Jewish community in Asia Minor.

The Iulia Severa Inscription (Akmoneia in Phrygia) IJO II 16864 τὸν κατασκευασθέ[ν]τα ο[ἶ]κον ὑπὸ Ἰουλίας Σεουήρας Π. Τυρρώνιος Κλάδος ὁ διὰ βίου ἀρχισυνάγωγος καὶ Λούκιος Λουκίου ἀρχισυνάγωγος καὶ Ποπίλιος Ζωτικὸς ἄρχων ἐπεσκεύασαν ἔκ τε τῶν ἰδίων καὶ τῶν συνκαταθεμένων καὶ ἔγραψαν τοὺς τοίχους καὶ τὴν ὀροφὴν καὶ ἐποίησαν τὴν τῶν θυρίδων ἀσφάλειαν καὶ τὸν [λυ]πὸν πάντα κόσμον. οὕστινας κα[ὶ] ἡ συναγωγὴ ἐτείμησεν ὅπλῳ ἐπιχρύσῳ διά τε τὴν ἐνάρετον αὐτῶν δ[ι]άθ[ε-] σιν καὶ τὴν πρὸς τὴν συναγωγὴν εὔνοιάν τε καὶ σ[που]δήν.

1

5

10

The oikos built by Iulia Severa restored P. Tyrronius Klados, archisynagōgos for life, and Lukios, Son of Lukios, archisynagōgos, and Popilios Zotikos, archōn, from their own and from collected resources. They decorated the walls and the ceiling, restored the security of the windows and the whole rest of the decoration. The synagōgē honoured them with a gilded shield because of their virtuous disposition and their goodwill and their diligence towards the synagōgē.

 64

See also Harland 2014: GRA II 113.

The Iulia Severa Inscription Reconsidered

 

Fig. 1. IJO II 168: Photo from MAMA VI 264 (public domain).

This famous inscription from Phrygian Akmoneia is regarded as the earliest epigraphic attestation of a synagogue of Judaeans in Asia Minor.65 It matches the criteria for establishing a Jewish origin in two points, as, apart from the synagogue, two archisynagōgoi are mentioned. When William Ramsay first published the inscription in 188866 he regarded it as “plausible” that Iulia Severa was a Jewess. In his Cities and Bishoprics, he compared her with Rufina, the Jewess ( Ἰουδαία) from Smyrna (IJO II 43; III CE?).67 Later interpreters, starting with Emil Schürer, did not adopt this very speculative opinion, but presupposed the Jewish identity of the mentioned synagogue and its functionaries (Schürer 1898: 16). So far, no scholar ever questioned that the

 65

See, e.g., Kraabel 1983/1992: 274–275; Trebilco 1991: 59; van der Horst 2009: 287; Levine 2005: 111– 112; White 1997: 307–310; Rajak 2001b: 463–475; Claußen 2002: 140–141; Gruen 2002: 108; Ameling 2004: 349–355; Catto 2007: 68–69; Runesson, Binder, and Olsson 2008: 134–135; Harland 2013: 199–200; Chiai 2012: 123–125; Edelmann-Singer 2013; and Stebnicka 2015: 275–276. 66 The first publication was in French, translated by Salomon Reinach; see Reinach 1891: 502–504; see also Oehler 1909: 298. 67 “It is obvious that Julia Severa was a Jewess” (Ramsay 1897: 650).

 

Markus Öhler

inscription should be regarded as Jewish at all.68 But as we will see, this is by far less secure than generally thought.69 Several problems are discussed in relation to this inscription. First, Iulia Severa lived in the middle of the 1st century CE.70 Some of her activities in Akmoneia are documented through inscriptions and coins:71 Iulia Severa was a priestess (ἀρχιερεία) of the local imperial cult, an office she performed with her husband L. Servenius Capito (see MAMA IV 139). She herself acted also as leader of the city’s games (ἀγωνοθέτις) and was a benefactor of the gerousia of Akmoneia (MAMA VI 263).72 Iulia Severa was related to a prominent C. Iulius Severus, who acted as high-priest of the provincial imperial cult in the 2nd century. Her son L. Servenius Cornutus became senator under Nero and legatus to the proconsul of Asia.73 If we consider her involvement in the imperial cult and in other cultic affairs connected with her official duties, it is extremely unlikely that she was a Jewess. But even to picture Iulia Severa as a Godfearer, in the sense of sympathizing with the practice and belief of Diaspora Judaism, is not as probable as it may seem to many commentators.74 Apart from our inscription, every other testimony of her activities speaks to the contrary, although it might be possible that Iulia Severa had other motives for benefaction: either an interest in the Jewish community as an important part of the local society, which would indicate the deep integration of the Jewish minority in the city, or a personal interest in adapting imperial behaviour in a local context.75 If the donor of the building is clearly a member of the Roman elite, even a priestess of the emperor-cult, one should initially presume that the whole inscription is not Jewish. Pagan donors to Jewish institutions are rarely attested. The only one before 70 CE is Markus Tittius, son of Sextus (Berenice, 24 CE). The local prefect was honoured by the politeuma of Judaeans for his advocacy on their  68

A modicum of doubt can be suspected from Phil Harland’s most recent treatment of this inscription (Harland 2014: “Although the self-designation ‘synagogue’ (συναγωγή) could be used by associations of various kinds (beyond Judaism), the leadership title ἀρχισυνάγωγος … is somewhat more characteristically Judean. … So we are likely dealing with a synagogue of Judeans here at Akmoneia” (152). 69 I owe Dr. Wolfgang Ernst (University of Vienna) for giving me the incentive to question this wellestablished view. 70 Coins point to her official functions in the years 55 to 65 CE (RPC I 3171, 3173–3176); see Edelmann-Singer 2013: 94. 71 MAMA VI 153, 263, 265. See, e.g., Halfmann 1979: 102; Rajak 2001b: 465–472; Ameling 2004: 351– 353; Harland 2013: 121–122, 199–200; Harland 2014: 151–154; and Edelmann-Singer 2013: 89, 94. 72 On women as agōnothetai see van Bremen 1996: 73–76. 73 Halfmann 1979: 105. For other members of this important family and their careers see Harland 2013: 122; and Ameling 2004: 352. 74 See, e.g., Trebilco 1991: 59; Rajak 2001b: 472–473; and Öhler 2003: 311–312. Differently, e.g., White 1997: “She was not, as once assumed, Jewish, nor was she even a Jewish ‘sympathizer’” (309). 75 For the latter see Edelmann-Singer 2013: 94–95, who points to Nero’s second wife Poppaea and her interests in Judaism. This was already an idea of Ramsay 1897: 651, and upheld by Groag 1918: 948. This presupposes, however, that the inscription is Jewish.

The Iulia Severa Inscription Reconsidered

 

behalf.76 Luke 7:5 narrates of the donation of a synagogue building by a Roman centurio, but this is very probably a Lukan redaction of the story and not a historical account. Capitolina, a woman from the local elite of Tralleis, donated in the course of a vow some parts of a synagogue building (IJO II 27, mid III CE), but she identified herself as a Godfearer. Iulia Severa would constitute the only example for such a deed, where we can be quite sure that she did not sympathize with the Jewish culture and religion. Therefore this issue led to sophisticated speculations about the relationship between a priestess of the emperor cult and the Jewish community.77 But this is only necessary if our understanding of the inscription is not determined by the religious and societal role of the original donor of the building, but by the words συναγωγή and ἀρχισυνάγωγος. The honoured men Publius Tyrronius Klados, Lukios, Son of Lukios, and Popilios Zotikos bear Roman or Greek names. The first one was definitely a Roman citizen, which of course neither proves nor denies a Jewish identity. Other considerations, however, make it rather probable that Publius Terronius Klados, the archisynagōgos for life, was not a Jew.78 On the one hand, he was a member of the Tyrronii, a family quite prominently involved in the affairs of the city. C. Tyrronius Rapo was in office together with Iulia Severa (MAMA VI 265), so the connection between her and Publius Tyrronius Klados might have rooted in this. It is of course also possible that Klados was a freedman of this family, so that he could have been a Jew after all (Ameling 2004: 353). On the other hand, as Tessa Rajak pointed out, a lifetime office of archisynagōgos is an indication of an honorific appointment, not of a real executive office.79 Klados might have been a nonJew anyway. The office of an archisynagōgos is well attested in Asia Minor.80 Six inscriptions mention an archisynagōgos or the feminine archisynagōgisa in a clearly Jewish context.81 All of these inscriptions are from the 3rd century CE or later.82 Six others are too fragmentary or are categorized as Jewish only because of the title  76 See Lüderitz 1983: no. 71; on the date see Runesson, Binder, and Olsson 2008: no. 132. The above mentioned Tation was herself a Jewess (IJO II 36, III CE). 77 E.g., Rajak 2001b: “… a philanthropic exchange arising out of a patronal relationship … some sort of friend of the Jews … some real attraction towards the God of Israel” (472–473); Trebilco 1991: 83, 179, 188: “evidence of ‘integration’ [of the Jews]”. 78 See Rajak 2001b: “another unlabelled god-fearer” (474). 79 Rajak 2001b: 474. The only Jewish instance from Asia Minor is IJO II 46 (Teos, III CE), for further references see Ameling 2004: 196. See Harland 2014: 151, for references in association documents. Runesson, Binder, and Olsson 2008: 135, consider the possibility that Klados was a retired synagogue leader, although with no further reference. 80 For an overview of the title and a collection of inscriptions see Horsley (1984); and Rajak and Noy (2001). 81 IJO II 1 (no place, Imp. time), 43 (Smyrna, III CE or later), 46 (Teos, III CE), 184 (Dokimeion, III CE), 217 (Beth Shearim/Pamphylia, III/IV CE), 255 (Nevsehir, no date). 82 I cannot see why the inscription mentioning the archisynagōgos Rufina from Smyrna (IJO II 43) is sometimes dated to the 2nd century CE, e.g., again by Stebnicka 2015: 147; see on the contrary Ameling 2004: “nach der Schrift nicht vor dem 3. Jh.” (188).

 

Markus Öhler

archisynagōgos, which is not sufficient to establish a Jewish origin.83 The literary attestations of the office are all of Christian origin, starting with the Gospel of Mark (5:22, 35, 36, 38 par Luke 8:49) and continuing in Luke-Acts (Luke 13:44; Acts 13:15; 18:8, 17).84 Josephus and Philo of Alexandria never mention it.85 Second, although the title is found more often in Jewish inscriptions, it is used by other groups as well. Kloppenborg and Ascough list nine instances for the title in a clear non-Jewish context, which are mostly from Macedonia and from the same time as the Iulia Severa inscription.86 In addition, some of the “Jewish” inscriptions from Asia Minor mentioning an archisynagōgos might in fact be also non-Jewish: IJO II 28 (Tralleis, end of I BCE) is so fragmentary that its Jewish character is only established on the basis of the word ἀρχισυνάγωγος (see Ameling 2004: 144) The same is true for IJO II 214 (Synnada, I/II CE), which also comes from Phrygia (see Ameling 2004: 448). Again, these two instances for the title are much earlier than those with a clear Jewish origin. Thus, there might have been non-Jews bearing the title in Asia Minor as well, and if our archisynagōgos for life P. Tyrronius Clados was not a Jew either, our inscription would be another instance for the non-Jewish use of the title. Third, the function of an archōn, here attributed to a Popilios Zotikos, is frequently mentioned in pagan inscriptions, not so much in a Jewish context.87 Since the function of an archōn was very prominent in city affairs, associations of all kinds adopted this title and different combinations with the prefix ἀρχ-.88 It is not at all a marker of Jewish origin. Fourth, the Iulia Severa inscription distinguishes two phases of the building. The first one from the 50s or 60s of the first century when Iulia Severa donated it, the second one sometime later when the honoured officials renovated it. The date of the inscription is not certain, but the end of the 1st or beginning of the 2nd century CE is very probable; mentioning Iulia Severa makes much more sense some decades after her donation than two or three centuries later, when the family was no longer prominent. In addition, a renovation of the building might have been necessary 20–30 years after the donation, so that 80–90 CE is quite  83 IJO II 25 (Myndos, IV–VI CE), 28 (Tralleis, after I BCE), 30 (Ephesos, V CE), 148 (Kyzikos, V/VI CE), 214 (Synnada, I/II CE), 256 (Nevsehir, no date). 84 See Rajak and Noy 2001: 398–404. They also point out that the Hebrew title rosh ha-kneset is a secondary development, not the root of the Greek title, Rajak and Noy 2001: 407–410. 85 See Claußen 2002: „Erstaunlicherweise ist dieser so prominente Titel eines ἀρχισυνάγωγος weder bei Philo noch bei Josephus belegt“ (260). Unfortunately, Claußen did not consider the implications of this. Rajak and Noy 2001: “The title is the one most widely represented in the ancient literature in association with the synagogue, and it is revealed there as the best known to outsiders” (398). 86 See, e.g., GRA I 66 (II CE), 72 (250 CE), 75 (90–91 CE), 86 (I/II CE); Kloppenborg and Ascough 2011: 311–312. Rajak and Noy 2001: 399, 428–429; and Horsley 1984: 219. 87 Apart from the Iulia Severa inscription in IJO II 160 (Sebastopolis, IV CE or later) and 220 (Side, after 390 CE). In IJO II 14 line 10 (Aphrodisias, IV/V CE) it is conjectured, whereas the Jewish character of IJO II 28 (Tralleis, late I BCE) is speculative. 88 For some instances see Harland 2014: 478.

The Iulia Severa Inscription Reconsidered

 

probable.89 But if the inscription is of Jewish origin, the question arises: how likely would it be that Iulia Severa or her family wanted to be connected that closely to the Jewish community, which had just been levied with a special tax because of the rebellion of the Judaean people in 66–70 CE?90 Fifth, if this dating to the late 1st or beginning of the 2nd century CE is correct, we should consider this in light of the other inscriptions from Asia Minor mentioning synagogues. As we have seen, those with a clear Jewish origin are from the 3rd century CE, whereas earlier instances are definitely or more likely nonJewish. Even if the use of synagōgē in the Iulia Severa inscription as a term denoting the community would be exceptional in light of the pagan inscriptions from Sardis and Amastris, in which it rather describes buildings, this should not be taken as indication for Jewish origin. On the one hand, the interpretation of the Amastris inscription is not clear and, on the other hand, the example from Teos has shown that synagōgē was sometimes used in pagan associations (see above). Sixth, the three men are honoured with a gilded shield (ὅπλον ἐπιχρύσον), which is otherwise not attested for Jewish communities (see Chiai 2012: 124– 125). Instead, this is quite usual in civic inscriptions or in documents of associations.91 Also, the language of the inscriptions is entirely in line with civic wording and even the form of the stone fits with that very well (Chiai 2012: 125). Seventh, the synagogue is here—again—the community honouring the donors and organizers of the renovations. The building is called οἶκος, which could mean a house, an assembly hall or something similar (White 1997: 308). Since Iulia Severa was an important woman, it is likely that her input for the synagogue must have been quite substantial. But, although the text could also be understood in a different way, meaning that Iulia Severa had been only the builder of a house which was subsequently bought by the community and then converted into their assembly building (e.g., Claußen 2002: 141), this is not likely. As this prominent woman is explicitly mentioned we should think of her as the donor of the house, not just as a previous owner.92 Eighth, Jews are well attested in Akmoneia, although the evidence might not be as strong as many think, as Bij de Vaate and van Henten have shown.93  89

See, e.g., Trebilco 1991: 59; Claußen 2002: 141; Ameling 2004: 350; and Harland 2014: 152–153. See Ameling 2009: “Honorary inscriptions for pagan benefactors are rare. … Was it somehow a stigma to be honoured by Jews, or was the number of interested pagan benefactors smaller than supposed?” (213). 91 See in Asia Minor, e.g., SEG LVII 1157 (Lydia, I CE); I.Sardis VII,1 8 (Sardis, 5–1 BCE); I.Kios 6 (Kios, no date); I.Perge 23 (Perge, after 20 CE), 67 (1–50 BCE). For an example for an association from the 1st century CE see IGRR IV 144 (Kyzikos, 41–54 CE). 92 See White 1997: 309; and Ameling 2004: 350. 93 See Bij de Vaate and van Henten 1996: 23–24. On two fragments of a synagogue building from the vicinity of Akmoneia see the commentary in Ameling 2004: 345, 351 n. 29. A connection between these fragments and the inscription (e.g., Catto 2007: 69) is not probable, since the fragments seem to be rather late. The earliest depictions of the seven-branched menorah in the Diaspora come from the middle of the 3rd century; see Hachlili 1998: 343. 90

 

Markus Öhler

Ameling includes eleven inscriptions in his corpus, mostly epitaphs from the 3rd century CE or later. Still, there is no doubt that by the 3rd century some Jews had established themselves in Akmoneia and this might have been the case even before that. But the quoted epitaphs do not refer to a Jewish community.94 Ninth, concerning the combination of two criteria for classifying this inscription as Jewish—συναγωγή and ἀρχισυνάγωγος—two observations are necessary. One, it can be expected that a synagogue is led by an archisynagōgos, the “leader of the synagogue”. Two, this combination is also attested for a pagan group. GRA I 86 (IGRR I 782; I.Perinthos 49) from Perinthos in Thrace (I CE) reads the following95: […] | [ὁ] διοικητὴς ∙ κα[ὶ] | [Μᾶ]ρκος ∙ Πομπήϊ[ος] | [Κ]ωμικὸς Κω[μικοῦ] | [υἱ]ὸς τὸν βω||[μ]ὸν τῇ συναγω|[γ]ῇ τῶν κουρέω|[ν π]ερὶ ἀρχισυνάγ|[ωγ]ον ∙ Γ(άιον) ∙ Ἰούλιον | [Ο]ὐάλεντα δῶ|[ρ]ον ἀποκατέστη|[σα]ν κα[ὶ] τὸν τόπο[ν] | [παρεσκεύ]ασα[ν]. […] the administrator and Marcus Pompeius Komikos, son of Komikos, restored the altar to the synagōgē of barbers, those around the archisynagōgos Gaius Iulius Valens, and provided the place.

No one would think of this inscription as one that was produced by Jews: It mentions an altar and the synagogue is an occupational association. How can we be so sure that the synagogue from Akmoneia was not an association of woodworkers, merchants or barbers?96 In sum, considering the unquestionable non-Jewish identity and religious focus of Iulia Severa, the names and functions of the honoured man, the character of the honours, the use of the terms συναγωγή and ἀρχισυνάγωγος in non-Jewish communities, the late attestations of συναγωγή in Jewish contexts in Asia Minor and the rather early date of the Iulia Severa inscription, it is more than possible that the Iulia Severa inscription is not Jewish. Some kind of association, which gathered regularly in a building donated by an elite woman of their city, used the term as a label for their community. A re-evaluation of the prominent place of the Iulia Severa inscription in reconstructions of the history of Judaism in Asia Minor is therefore appropriate.

 94

IJO II 172, 174, 175, 178 include fines for violation of graves, which should be payed solely to the fiscus, not to the Jewish community. See also IJO II 171 with a donation to the local neighbourhood. In addition, we should consider that surviving Jewish inscriptions in Asia Minor from the time before the 3rd century CE are very rare: This might point to an epigraphic habit, although a certain degree of destruction should also be considered; see Price and Misgav 2006: 469–470. 95 See also Kloppenborg and Ascough 2011: no. 86. This is only face A; face B is younger and reads: Διὶ Λοφείτῃ Ε[ὐ]δίων Φιλλῦδ[ος] ἱερεὺς νέοις ὐ[ρα]ρίοις(?) δῶρον. 96 For occupational associations in Akmoneia see GRA II 114 (fullers, 98–117 CE) and the commentary by Harland 2014: 157–158.

The Iulia Severa Inscription Reconsidered

 

Bibliography Ameling, W. (ed.) 2004. Inscriptiones Judaicae Orientis, Vol. II: Kleinasien. TSAJ 99. Tübingen: Mohr Siebeck. Ameling, W. 2009. “The Epigraphic Habit and the Jewish Diasporas of Asia Minor and Syria”. In: Cotton, H. (ed.), From Hellenism to Islam: Cultural and Linguistic Change in the Roman Near East. Cambridge: Cambridge University Press, 203–234. Avemarie, F. 2009. “Jüdische Diasporagemeinden in der Antike. Ihr Selbstverständnis im Spiegel der Inschriften”. In: Frey, J. and Standhartinger, A. (eds.), Neues Testament und frührabbinisches Judentum. WUNT 316. Tübingen: Mohr Siebeck, 43–82. Bij de Vaate, A. J. and van Henten, J. W. 1996. “Jewish or Non-Jewish? Some Remarks on the Identification of Jewish Inscriptions from Asia Minor”. BiOr 53: 16–28. Binder, D. D. 1999. Into the Temple Courts: The Place of the Synagogues in the Second Temple Period. SBLDS 169. Atlanta: Society of Biblical Literature. Botermann, H. 1990. “Die Synagoge von Sardes: Eine Synagoge aus dem 4. Jahrhundert?”. ZNW 81: 103–121. Brehm, O. 1996. Hellenistische und kaiserzeitliche Weihreliefs aus dem nördlichen Mysien: Ikonographie und Typologie einer Denkmälergruppe und ihre Bedeutung für die Religionsgeschichte einer antiken Landschaft. Münster. Brooten, B. J. 2020. Women Leaders in the Ancient Synagogue: Inscriptional Evidence and Background Issues. BJS 36. 2nd ed. Atlanta, GA: Scholars Press. Available online from JSTOR: www.jstor.org/stable/j.ctvzpv5mr.7 (accessed 2 September 2020). Buckler, W. H. and Robinson, D. M. 1932. Greek and Latin Inscriptions Part I. Sardis VII. Leiden: Brill. Catto, S. K. 2007. Reconstructing the First-Century Synagogue: A Critical Analysis of Current Research. LNTS 363. London/New York: T&T Clark. CGRN = Carbon, J.-M., Peels, S. and Pirenne-Delforge, V. Collection of Greek Ritual Norms (CGRN), Liège 2015– . Available at http://cgrn.ulg.ac.be (accessed 19 March 2020). Chiai, G. F. 2012. “Norm, Kommunikation und Identität: die jüdische Lebenswelt in den Inschriften des kaiserzeitlichen Phrygien”. In: Chiai, G. F., Gauly, B. M., Hartmann, A., Zimmer, G., and Zapff, B. M. (eds.), Athen, Rom, Jerusalem: Normentransfers in der antiken Welt. ESt 66. Regensburg: Pustet, 117–146. CIJ – see Frey 1936–1952. Claußen, C. 2002. Versammlung, Gemeinde, Synagoge: das hellenistisch-jüdische Umfeld der frühchristlichen Gemeinden. SUNT 27. Göttingen: Vandenhoeck & Ruprecht. Corsten, T. 1987. Die Inschriften von Apameia (Bithynien) und Pylai. Inschriften Griechischer Städte aus Kleinasien 32. Bonn: Habelt. Corsten, T. and Fraser, P. M. 2010. A Lexicon of Greek Personal Names. 5A: Costal Asia Minor: Pontos to Ionia. Oxford: Clarendon Press. Dorl-Klingenschmid, C. 2001. Prunkbrunnen in kleinasiatischen Städten: Funktion im Kontext. Studien zur antiken Stadt 7. Müchen: Pfeil. Edelmann-Singer, B. 2013. “Herrscherfrauen als Leitfiguren: Iulia Severa, Poppaea und die "Matronage" der jüdischen Religion”. In: Kunst, C. (ed.), Matronage: Handlungsstrategien und soziale Netzwerke von Herrscherfrauen im Altertum in diachroner Perspektive. Osnabrücker Forschungen zu Altertum und Antike-Rezeption 20. Rahden/Westfalen: Leidorf, 89–99. Feldman, L. H. 1996. “Diaspora Synagogues: New Light from Inscriptions and Papyri”. In: Fine, S. (ed.), Sacred Realm: The Emergence of the Synagogue in the Ancient World. New York: Oxford University Press, 48–66.

 

Markus Öhler

Fine, S. 1997. This Holy Place: On the Sanctity of the Synagogue during the Greco-Roman Period. Christianity and Judaism in Antiquity Series 11. Notre Dame, IN: The University of Notre Dame Press. Foucart, P. 1873. Des associations religieuses chez les Grecs: thiases, éranes, orgéons, avec le texte des incriptions relatives à ces associations. Paris: Klincksieck. Frey, J.-B. 1936–1952. Corpus Inscriptionum Judaicarum: recueil des iscriptions qui vont du 3e sie cle avant Je sus-Christ au 7e sie cle de notre e re. Rome: Pontificio Istituto di Archeologia Cristiana. GRA = Kloppenborg, J. S., Harland, P. A., and Ascough, R. S. 2011–. Greco-Roman Associations: Texts, Translations, and Commentary. BZNW. Berlin: De Gruyter. Groag, E. 1918. “Iulia Severa”. PRE X, 1: 946–948. Gruen, E. S. 2002. Diaspora: Jews amidst Greeks and Romans. Cambridge, MA: Harvard University Press. Hachlili, R. 1998. Ancient Jewish Art and Archaeology in the Diaspora. HdO I 35. Leiden: Brill. Halfmann, H. 1979. Die Senatoren aus dem östlichen Teil des Imperium Romanum bis zum Ende des 2. Jahrhunderts n. Chr. Göttingen: Vandenhoeck & Ruprecht. Hanfmann, G. M. A. 1983. “Religious Life”. In: Hanfmann, G. M. A. (ed.), Sardis from Prehistoric to Roman Times. Cambridge, MA: Harvard University Press, 128–136. Harland, P. A. 2006. “Acculturation and Identity in the Diaspora: A Jewish Family and ‘Pagan’ Guilds at Hierapolis”. JJS 57: 222–244. Harland, P. A. 2013. Associations, Synagogues, and Congregations: Claiming a Place in Ancient Mediterranean Society. Minneapolis, MN: Fortress. Harland, P. A. 2014. Greco-Roman Associations: Texts, Translations and Commentary. North Coast of the Black Sea, Asia Minor. BZWN 204. Berlin: De Gruyter. Herrmann, P. 1993. “Inschriften von Sardeis”. Chiron 23: 233–266. Herrmann, P. 1996. “Mystenvereine in Sardeis”. Chiron 26: 315–348. Horbury, W. and Noy, D. 1992. Jewish Inscriptions of Graeco-Roman Egypt. Cambridge: Cambridge University Press. Horsley, G. H. R. 1984. “An archisynagogos of Corinth?”. NewDocs 4: 213–220. IJO I – see Noy, Panayotov, and Bloedhorn 2004. IJO II – see Ameling 2004. IJO III – see Noy and Bloedhorn 2004. JIGRE – see Horbury and Noy 1992. JIWE II – see Noy 1993. Kloppenborg, J. S. and Ascough, R. S. 2011. Greco-Roman Associations: Texts, Translations, and Commentary. 1. Attica, Central Greece, Macedonia, Thrace. BZNW 181. Berlin: De Gruyter. Kraabel, A. T. 1978/1992. “Paganism and Judaism: The Sardis Evidence”. In: Overman, J. A. and MacLennan, R. S. (eds.), Diaspora Jews and Judaism: Essays in Honor of, and in Dialogue with, A. Thomas Kraabel. SFSHJ 41. Atlanta: Scholars Press, 237–255. Kraabel, A. T. 1983/1992. “Impact of the Discovery of the Sardis Synagogue”. In: Overman, J. A. and MacLennan, R. S. (eds.), Diaspora Jews and Judaism: Essays in Honor of, and in Dialogue with, A. Thomas Kraabel. SFSHJ 41. Atlanta: Scholars Press, 269–291. Kraemer, R. S. 1991. “Jewish Tuna and Christian Fish: Identifying Religious Affiliation in Epigraphic Sources”. HTR 84: 141–162. Kraemer, R. S. 1992. Her Share of the Blessings: Women’s Religions among Pagans, Jews, and Christians in the Greco-Roman World. New York: Oxford University Press.

The Iulia Severa Inscription Reconsidered

 

Le Guen, B. 2007. “Kraton, Son of Zotichos: Artists’ Associations and Monarchic Power in the Hellenistic Period”. In: Wilson, P. (ed.), The Greek Theatre and Festivals: Documentary Studies. Oxford Studies in Ancient Documents. Oxford: Oxford University Press, 246–278. Levine, L. I. 2005. The Ancient Synagogue: The First Thousand Years. 2nd ed. New Haven, CT: Yale University Press. Liddell, H. G., Scott, R., and Jones, H. S. 1996. A Greek-English Lexicon: With a Revised Supplement 1996. Oxford: Clarendon Press. Lieu, J. 2003. Image and Reality: The Jews in the World of the Christians in the Second Century. London: T&T Clark. Lieu, J. 2016. Neither Jew nor Greek? Constructing Early Christianity. 2nd ed. London: T&T Clark. Lüderitz, G. 1983. Corpus jüdischer Zeugnisse aus der Cyrenaika, mit einem Anhang von Joyce M. Reynolds. TAVO Beihefte, B 53. Wiesbaden: Reichert. Magness, J. 2005. “The Date of the Sardis Synagogue in Light of the Numismatic Evidence”. AJA 109: 443–475. Miranda, E. 1999. “La Comunità Giudaica di Hierapolis di Frigia”. Epigraphica Anatolica 31: 109–155. Nigdelis, P. M. 1994. “Synagoge(n) und Gemeinde der Juden in Thessaloniki: Fragen aufgrund einer neuen jüdischen Grabinschrift der Kaiserzeit”. ZPE 102: 297–306. Nijf, O. M. v. 1997. The Civic World of Professional Associations in the Roman East. Dutch Monographs on Ancient History and Archaeology 17. Amsterdam: J.C. Gieben. Noy, D. 1993. Jewish Inscriptions of Western Europe, Vol. II: The City of Rome. Cambridge: Cambridge University Press. Noy, D. and Bloedhorn, H. 2004. Inscriptiones Judaicae Orientis, Vol. III: Syria und Cyprus. TSAJ 102. Tübingen: Mohr Siebeck. Noy, D., Panayotov, A., and Bloedhorn, H. 2004. Inscriptiones Judaicae Orientis, Vol. I: Eastern Europe. TSAJ 101. Tübingen: Mohr Siebeck. Oehler, J. 1905. “Zum griechischen Vereinswesen”. Jahresbericht des K.K. MaximiliansGymnasiums in Wien: 3–30. Oehler, J. 1909. “Epigraphische Beiträge zur Geschichte des Judentums”. MGWJ 53: 292– 302, 443–452, 525–538. Öhler, M. 2003. Barnabas: Die historische Person und ihre Rezeption in der Apostelgeschichte. WUNT 156. Tübingen: Mohr Siebeck. Öhler, M. 2017. “Judäer oder Juden? Die Debatte ‘Ethnos vs. Religion’ im Blick auf das 2. Makkabäerbuch”. In: Avemarie, F. et al. (eds.), Die Makkabäer. WUNT 382. Tübingen: Mohr Siebeck, 157–185. Panayotov, A. 2014. “Jews and Jewish communities in the Balkan and the Aegean until the twelfth century”. In: Aitken, J. K. and Paget, J. C. (eds.), The Jewish-Greek Tradition in Antiquity and the Byzantine Empire. Cambridge: Cambridge University Press, 54–76. Price, J. J. and Misgav, H. 2006. “Jewish Inscriptions and Their Use”. In: Safrai, S., Safrai, Z., Schwartz, J., and Tomson, P. J. (eds.), The Literature of the Sages. CRINT II.3b. Assen: Van Gorcum, 461–483. Pucci Ben Zeev, M. 1998. Jewish Rights in the Roman World: the Greek and Roman Documents Quoted by Josephus Flavius. TSAJ 74. Tübingen: Mohr Siebeck. Rajak, T. 1998. “The Gifts of God at Sardis”. In: Goodman, M. (ed.), Jews in a Graeco-Roman World. Oxford: Oxford University Press, 229–239. Rajak, T. 2001a. “Benefactors in the Greco-Jewish Diaspora”. In: Rajak, T. (ed.), The Jewish Dialogue with Greece and Rome: Studies in Cultural and Social Interaction. AGJU 48. Leiden: Brill, 373–391.

 

Markus Öhler

Rajak, T. 2001b. “The Synagogue in the Greco-Roman City”. In: Rajak, T. (ed.), The Jewish Dialogue with Greece and Rome: Studies in Cultural and Social Interaction. AGJU 48. Leiden: Brill, 463–478. Rajak, T. and Noy, D. 2001. “Archisynagogoi: Office, Title and Social Status in the GrecoJewish Synagogue”. In: Rajak, T. (ed.), The Jewish Dialogue with Greece and Rome: Studies in Cultural and Social Interaction. AGJU 48. Leiden: Brill, 393–429. Ramsay, W. M. 1897. The Cities and Bishoprics of Phrygia. Oxford: Clarendon Press. Rebillard, É. 2003. “Groupes religieux et élection de sépulture dans l’Antiquité tardive”. In: Belayche, N. and Mimouni, S. C. (eds.), Les communautés religieuses dans le monde gréco-romain: Essais de définition. Bibliothèque de l’École des Hautes Études. Sciences religieuses 117. Turnhout: Brepols, 259–274. Reinach, S. 1891. Chroniques d’orient: Documents sur les fouilles découvertes dans l’orient hellénique. Paris: E. Leroux. Runesson, A., Binder, D. D., and Olsson, B. 2008. The Ancient Synagogue from its Origins to 200 C.E.: A Source Book. AJEC 72. Leiden: Brill. Schürer, E. 1898. Geschichte des Jüdischen Volkes im Zeitalter Jesu Christi III. Leipzig: Hinrich. Seager, A. R. and Kraabel, A. T. 1983. “The Synagogue and the Jewish Community”. In: Hanfmann, G. M. A. (ed.), Sardis from Prehistoric to Roman Times. Cambridge, MA: Harvard University Press, 168–190. Stebnicka, K. 2015. Identity of the Diaspora: Jews in Asia Minor in the Roman Imperial Period. The Journal of Juristic Papyrology. Supplements 26. Warsaw: University of Warsaw. Steinhauer, J. 2014. Religious Associations in the Post-Classical Polis. PAWB 50. Stuttgart: Franz Steiner. Trebilco, P. 1991. Jewish Communities in Asia Minor. MSSNTS 69. Cambridge: Cambridge University Press. Van Bremen, R. 1996. The Limits of Participation: Women and civic life in the Greek East in the Hellenistic and Roman periods. Dutch Monographs on Ancient History and Archaeology 15. Amsterdam: J.C. Gieben. Van der Horst, P. W. 1991. Ancient Jewish Epitaphs: An Introductory Survey of a Millenium of Jewish Funerary Epigraphy (300 BCE – 700 CE). CBET 2. Kampen: Kok Pharos. Van der Horst, P. W. 2009. “The Jews of Ancient Phrygia”. EJJS 2: 283–292. Van der Horst, P. W. 2015. Saxa Judaica Loquuntur: Lessons from Early Jewish Inscriptions. BIS 134. Leiden: Brill. White, L. M. 1997. The Social Origins of Christian Architecture, II: Texts and monuments for the Christian domus ecclesiae in its environment. Valley Forge, PA: Trinity Press International.

Dress Codes in the Synagogue of Dura Europos?

Katrin Kogman-Appel

Introduction One oft-noted phenomenon typical of the murals in the famous synagogue of Dura Europos (244–245 CE) is the juxtaposition of Graeco-Roman and Persian dress:1 Greek heroes and nymphs appear next to Persian kings, Roman and Persian soldiers on foot and on horseback share the same pictorial space, and priests in Persian garb assist the Israelite high priest. In the wake of frequent observations by anthropologists about how groups define themselves by means of dress,2 art-historical research often relates to costumes as an iconographic tool to delineate religious, ethnic, cultural, social, and/or gender identities.3 In the Dura synagogue the protagonists’ costumes seem to reflect a different mode, one that scholars are struggling to deal with. Most commonly they have been associated with certain compositional settings, such as court scenes with a range of Persian costumes along with pictorials that convey more intense religious messages, where Graeco-Roman garb seems to prevail. Owing to its location in the northeast of modern Syria on the western shore of the Euphrates, Dura was home to different communities of varied cultural and religious backgrounds. That Persian- and Greek-style garb also appears in works of art in Dura apart from those in the synagogue suggests cultural encounters, and some scholars have even contemplated the possibility that these images may have been related to everyday life. First excavated in 1932, the synagogue bears an inscription that dates the latest phase of the building to 244/245 CE, and, with the exception of the paintings immediately above the Torah shrine, the murals date from this stage.4 Both the  1

My thanks go to Hagit Sivan and Steven Fine for reading an earlier draft and their comments. For brief notes with a focus on antiquity, see Bonfante 1994 and Harlow and Nosch 2014; this approach also governs the recent contribution by Baird 2016, with further references to anthropologic literature; see also Pohl 2006 discussing various markers of ethnicity and concluding that dress is not a particularly strong one, and that it defines social rather than ethnic groups. I am grateful to Hagit Sivan for bringing this paper to my attention 3 See, e.g., Lovén 2014; specifically with regard to Dura Europos, see Klaver 2016. 4 For the published excavation report, see Kraeling 1956. 2

370

Katrin Kogman-Appel

initial construction of the synagogue and its restoration are associated with the period of Roman rule in the city (164–256 CE). Located near the eastern frontier of the Roman Empire Dura was destroyed by the Sasanians in 256 and subsequently abandoned. The interior of the synagogue was designed as a broad house, with a niche for the Torah shrine located in the western wall facing Jerusalem.5 Adjacent to the niche we see a set of apparently eschatological images, whereas the four walls beyond the niche area display a rich array of biblical themes in three registers arranged to follow a no longer detectable rationale other than chronological order.6 Upon their discovery, the paintings began to arouse the interest of a diverse group of scholars—archaeologists, historians, and art historians alike. At first approached as a remote garrison town at the periphery of the Roman Empire, in more recent scholarship Dura emerges as a trade centre at a hub of commercial roads, a multicultural environment, where worshippers of many different cults were culturally and ethnically entangled.7 Earlier art-historical scholarship looked at the synagogue murals from two perspectives: the role that these pictorials may have played in developing an iconographic tradition of the Hebrew Bible/the Old Testament from late antiquity to the late Middle Ages and the dependence of their imagery on midrashic literature and thus their relationship to a well-defined rabbinic worldview.8 Recent scholarship poses a different set of questions often criticizing earlier emphases on both the notion of iconographic chains and the dominant dependence of the imagery on the midrash.9 Recent approaches highlight the uniqueness of the imageries within their local setting or deal, for example, with the ritual and liturgical dimensions of the paintings.10  5 The murals were installed in the National Museum of Damascus; a replica (75 percent in size) is displayed in the Diaspora Museum at Tel Aviv University; a full set of photographs was published in Goodenough 1953–1968, vol. 11 (1964). 6 Scholars have often speculated about the nature of the underlying iconographic program, but given that about 40 percent of the murals did not survive, it has proven impossible to reconstruct any such plan; for a summary of the different attempts, see Gutmann 1992; for a critique of such endeavours, see Wharton 1994, who argues that the murals are a pastiche echoing in fact the essence of the midrashic method; recently the question of the underlying program with regard to the western wall was taken up by Levine 2012: 111–117; Xeravits 2017a. 7 Wharton 1995: 15–22, critiquing earlier scholarship for its Orientalist approach to the city; Jensen 1999: 178–79. 8 The literature on both questions is too vast and can be cited here only selectively: Sukenik 1947; Weitzmann 1964; Kessler and Weitzmann 1990; and Schubert 1992; both questions were recently revisited by Kessler 2012. 9 See a relatively recent summary by Fine 2012: 20–24; it is altogether doubtful that the patrons and artists of the synagogue would have had any text to work with and the midrashic iconography in Dura Europos may very well be the outcome of an oral midrashic culture that we cannot easily pinpoint in terms of fixed dated texts; see, e.g., Wharton 1995: 48, who stresses that the visualizations of biblical stories convey interpretations of their own; thus, the portrayal of Pharaoh to the left of the Torah shrine together with that of Ahasuerus to its right are suggestive of a “historical morality tale” of diasporic existence. 10 See, e.g., Levine 2012: 97–118; Laderman 1997; Fine 2005a; Fine 2005b.

Dress Codes in the Synagogue of Dura Europos?

371

The following paragraphs are an attempt to elucidate the rationale behind a possible “dress code” in the synagogue paintings and suggest that it functioned as an essential tool for visual story telling. I argue that it was employed to underscore the role the protagonists assumed in the flow of the visual narratives, while these roles merge with the protagonists’ ethnic or cultural identities only to a very limited degree. In the first part of the paper I revisit the various costumes in order to distinguish common denominators among the protagonists in different attires. This survey is followed by a discussion of the role that these costumes play in the visual language of the murals as they are approached against the background of Jewish life in Dura.

The Costumes Let me begin with no costume at all: the nude figure of Pharaoh’s daughter rescuing Moses from the river (Fig. 1). Moses’s birth is mentioned only briefly in the Bible and rarely depicted in art, certainly not in late antiquity and particularly not in Jewish art. In a way, the discovery of Moses in the river functions in lieu of a birth narrative in the sense that it visualizes the coming into the world of the future leader who will rescue the Israelites from bondage and through whom they will receive the law that establishes Israel’s identity as a nation. Erwin Goodenough observed that Pharaoh’s daughter holding the infant in her arm recalls images of Anahita, the Persian goddess of fertility and motherhood, whom the Greeks and Romans associated with Aphrodite.11 The choice of this motif defines the princess’s role as that of a mother figure in terms of a visual language that was part and parcel of late antique syncretism. However, the key to a full understanding of the princess’s role lies in the way Moses is pictured in the paintings as an adult. There we meet Moses having grown into a Graeco-Roman hero of sorts. He is wearing a striped tunic (chiton with clavi) and a draped mantle (pallium in Latin; himation in Greek) with some ornaments at the edges typical of that type of attire (Fig. 2).12 Moses’s identity as  11

Goodenough 1953–1968, vol. 9 (1964): 200–201; Moon 1992: 588–98; Anahita was explicitly associated with water; Goodenough’s interpretation was part of his overall conclusion that the murals reflect a mystic approach to the Bible embedded in a Hellenized Judaism, a conclusion that subsequent scholarship has largely rejected: for a summary of the critiques, see Avi-Yonah 1992. See Weitzmann and Kessler 1990: 30; Moon 1992: 598; and Xeravits 2017b for interpretations of the princess’s nudity as a sign of pagan identity and otherness; this makes sense, but I do not think that it is at the core of the image’s message. 12 Goodenough 1953–1968, vol. 10 (1964): 119, comparing Moses with Hercules; in Goodenough 1953–1968, vol. 9 (1964): 156–62, and Goodenough 1966, he argued that the bright coloured tunic with the pallium underscores ritual meaning and speculated that such garments were worn by 3rd-century Jews for ritual purposes; he also contended that the bright pallium foreshadows early Christian iconography, in which this wear would become typical of saints and angels; Moon 1992: 592–95, associated the mantle with Roman togati.

372

Katrin Kogman-Appel

a Jewish hero in Graeco-Roman garb is foreshadowed in the discovery scene not only by the appearance of the princess as the mother goddess, but also by that of her maidservants in the guise of Greek nymphs. They are dressed in the Greek costumes that are often found in depictions of nymphs and carry gifts as well as ointments and other substances. Greek nymphs are often shown in GraecoRoman art attending to newly born heroes or gods.13 Several more prophetic figures are shown in the murals in Greek attire, but such portrayals are not limited to prophets. Jacob, the patriarch, sleeping and dreaming at Beth-El is wearing the himation (Gen 28:10–19). David is imaged wearing a purple himation to highlight his royal identity while being anointed by Samuel (1 Sam 16); his brothers and Samuel himself are likewise clad in the Graeco-Roman costume. Samuel is thus identified as a prophet, as are Elijah and Ezekiel (even though not in all his appearances; Figs. 3, 4, and 5). The same applies to the two unidentified men flanking the Torah niche; even though there is no scholarly consensus about whom they represent, it is quite clear that they are prophets, authors of biblical books, or otherwise significant figures in Israelite history.14 Two more prophets flank David’s throne in one of the layers above the Torah shrine. All of these figures are the principal protagonists in the stories being told, and their roles as dignitaries, religious leaders, and forefathers are defined by their Graeco-Roman appearance.15 Other palliati are anonymous, and these are more difficult of interpretation. Among them are the representatives of the twelve tribes in the scene of the departure from Egypt (Fig. 2). A mural in the middle register of the western wall depicts the Ark of the Covenant, which had been captured by the Philistines (1 Sam 5–6; Fig. 6). As its presence caused a plague among the latter and the breaking of the idols of their god Dagon, it was sent away on its own, eventually reaching the Israelites through divine intervention. The Ark is accompanied by the two Philistines mentioned in the text, who are dressed in Persian garb, but three other men in tunics and pallia are overlooking the scene or observing the Ark’s journey. Whether representing Philistines or, more likely, Israelites at Beth Shemesh, their passivity and frontal appearance are striking. They stand somewhat apart from the action and seem not to assume any role in the story; they are static observers and witnesses of a divine miracle. More such anonymous figures appear in the story of Elijah and the priests of Baal, which is shown in two panels in the lower register of the southern wall (1 Kgs 18; Figs. 3 and 4). Confronting the Israelites near Mount Carmel, Elijah rebukes them, demanding that they decide between God and Baal and suggests  13 Du Mesnil du Boisson 1939: 125; Goodenough 1953–1968, vol. 9 (1964): 203–207, 221, who saw in the nymphs a sign of the flow of “divine fluid” typical of “Wunderkinder”. 14 The various opinions are summarized in Hachlili 1998: 112–113. 15 Goodenough 1953–1968, vol. 11 (1964): Pls. IV, VII, XIX, Figs. 75, 93, 323. There are a few more signs of Greek wear in scenes that have not fully survived, some of them completely unidentifiable, as in the upper register on the western wall near the southern corner: Kraeling 1956: 86–93.

Dress Codes in the Synagogue of Dura Europos?

373

putting God and Baal to a test. Elijah is wearing a tunic and is wrapped in a pallium. We also see three other men in the same dress. In his speech Elijah emphasizes that he is the only Israelite prophet present, whereas Baal is attended by 450 of his prophets. When he began praying to God, Elijah called to “all the people” to gather near to his altar to witness the miracle. In the adjacent image, the altar of Baal is flanked on either side by four men also wrapped in pallia. Their static and frontal appearance in apparent contrast with the biblical text that tells of the priests dancing, praying, and hurting themselves with swords, again, imbues them with the features of witnesses. Their appearance says nothing about their identities as either Israelites or Canaanites, but underscores the moment of witnessing divine intervention. In fact, in both scenes these figures point to the visible signs of this intervention: the fire on Elijah’s altar and the snake who came to bite Hiel, who was to light the fire at the appropriate moment so as to trick Elijah.16 Four similar figures appear in the Esther story (Fig. 7). Although completely motionless they are clearly visible and mark the centre of the composition, as the dynamic narrative, dominated by wearers of Persian garb, develops on either side of them. They cannot be identified with any of the figures in the story, but seem to be anonymous representatives of a collective group of observers, witnesses of the favourable turn of events.17 Many scholars have related to the puzzling fact that in the lengthy narrative of his vision in the valley of dry bones Ezekiel (37:1–14) appears first in Persian garb, whereas towards the end of this impressive narrative of resurrection and restoration to the homeland he is shown wearing a tunic and a himation (Fig. 5). Ezekiel’s first four appearances on the left show him as a prophet in communication with (the hand of) God, and it is his agency as a prophet that brings about the restoration of the bodies and their resurrection. The change of costume seems to take place in the actual moment the dead are restored to life. Wearing the Greek garb in his last two appearances, Ezekiel thus becomes a witness of divine action as do the resurrected, shown as well in tunics and mantles.18  16 Based on an aggadic motif surviving only in a much later source from the 13th century: Yal. Shim‘oni on 1 Kgs 18, § 214. 17 These anonymous figures in Greek garb were occasionally identified as “witnesses” in the literature, usually of historical events: Kraeling 1956: 156; Tawil 1979: 104; and Goldman 1994: 173. The common denominators in their appearance and their anonymity suggest, rather, that they were meant as witnesses of divine intervention. 18 Kraeling 1956: 193; and Du Mesnil du Boisson 1939: 94, suggested that the Persian garb refers to Ezekiel’s life during the Babylonian exile, whereas the Roman garb shows him in an eschatological situation. Emil G. Kraeling 1940 took this eschatological reading further and argued that the man wrapped in a Greek mantle represents “the David of the future”, as did Wischnitzer 1941 and Wischnitzer 1948: 45– 46, but given that this part of the narrative focuses on the resurrection, this is rather unlikely: see Xeravits 2018: 82; on the mural, also see Revel-Neher 2004, who suggests reading it as part of a greater eschatological scheme that unites the program of the paintings as a whole; Wischnitzer interpreted three of the figures in Persian garb as representatives of Judah and the ten palliati as representatives of the ten lost tribes.

374

Katrin Kogman-Appel

There are also several female counterparts, women who are wearing Greek attire: a long tunic and a mantle (palla) wrapped around the hips, covering parts of the upper body and the head. Unlike the men’s costume, which is almost always white, beige, or light pink, the women’s wear comes in strong colours. All of these figures are either Israelite or described as gentile but God-fearing women: Moses’ mother and sister in the narrative of his discovery and the midwives, who according to a midrash are identical with the former (Fig. 1),19 as well as the widow of Zarephath, whose infant son was resurrected through the intervention of Elijah (1 Kgs 17). As the male witnesses in the other scenes described above, the widow, too, is turned into a witness of a divine miracle.20 Several murals display figures, usually in a group, dressed in short Roman tunics, some in white, others in pink or red. These tunics are reserved for the common people: the bottom row of Israelites leaving Egypt (Fig. 2); the carriers of the Ark in the battle at Even Ezer (1 Sam 4:1–11; Fig. 8); some of Elijah’s attendants pouring water onto the altar on Mount Carmel (Fig. 3). Finally, a partly destroyed painting just above the Elijah scene shows a procession in which the Ark is being carried by four young men in Persian garb, who are accompanied by musicians in short Roman tunics (Fig. 9). Men in short Roman tunics are always Israelites. The Persian garb is more difficult to deal with. Most prominently, enthroned kings appear as Sasanian rulers. Seated on lavish thrones and accompanied by attendants, they are shown in trouser suits and tailored mantles. If we assume different ethnicities, theirs can be either a benign or an inimical function in the narration of Israelite history: Pharaoh giving the order to kill the Israelite newborns in the Exodus story (Fig. 1); Ahasuerus first following Haman’s suggestion to persecute the Jews and then changing his policy; and David in one of the layers of the central mural above the Torah shrine.21 Elsewhere similarly dressed kingly figures appear on horseback. One of them can be identified as Saul accompanied by his army seeking David in the wilderness of Ziph (1 Sam 26; Fig. 10). The other represents Mordecai, who while not a king, is being honoured by Ahasuerus and thus assumes quasi-royal status (Fig. 7). The most common costume is a simpler Persian trouser suit with a tailored tunic. In a painting that shows the initiation of the sacrificial cult, Aaron, the high priest is assisted by priests dressed in this Persian attire, and he himself is wearing it underneath his mantle (Fig. 11). As suggested by Elisabeth Revel-Neher, this portrayal of Aaron is of a type often found associated with priestly figures in the  19

See b. Soah 11b; for an English translation, see Epstein 1961. Goodenough 1953–1968, vol. 11 (1964): Pl. VIII. The tunic and palla have occasionally been described in the literature as typically Jewish dress: Moon 1992: 596; Croom 2010: chap. 7; but see Goldman 1994: 173. 21 Goodenough 1953–1968, vol. 11 (1964): Pl. IV, Figs. 75, 93, 323. This image was recently revisited by Xeravits 2018: 84. 20

Dress Codes in the Synagogue of Dura Europos?

375

east of the Roman Empire.22 In the procession scene described above, the men carrying the Ark are likewise wearing the trouser suit (Fig. 9). The same is true of the angels climbing up and down Jacob’s ladder. Pharaoh’s and Ahasuerus’s courtiers are shown that way (Figs. 1 and 7), as are the two Philistines who accompanied the Ark when it set out to Beth Shemesh, the fallen idols in front of Dagon’s temple (Fig. 6), and the representatives of the Israelite tribes near an enthroned David. Soldiers on horseback are wearing trousers and tunics (Figs. 8 and 10). Four of the Ezekiel figures in the vision of the dry bones are wearing the same costume (Fig. 5), as are two figures in the foreground of a scene that concludes the Ezekiel series whose identity is still subject to question.23 In short, whereas the Greek wear is mostly associated with Israelites, the Persian garb cannot be linked to any ethnic, religious, or political identity. Several of the narratives picture armed soldiers, some of them in actual battle. Most of them appear on foot in Roman armour (Figs. 2 and 8). As Dura was a garrison town, it would have been easy for the artists to find information about the various kinds of armour in use at the time, and fragments of military garb have been found there.24 Others are imaged on horseback, but it is not always clear if they are wearing Persian trousers or Roman breeches.25 As is true of the Persian trouser suit, the different kinds of armour do not reflect any political or other identity.

Dress, Identity, and Visual Language The array of costumes in the imagery of the Dura synagogue attests to a Jewish visual culture that operated in a zone of entangled and overlapping cultural identities. As other people in Dura, the artists and patrons were familiar with both Roman and Persian fashions and tastes, and they were well aware of both visual cultures. They also seem to have known of the potential of the costumes as carriers of meaning associated with identity, but they employed them in a very different way. Out of a large pool of entangled Iranian and Graeco-Roman cultural elements, they approached dress as an essential component of the visual language. It is this function of the costumes considered in relation to contemporary  22 Goodenough 1953–1968, vol. 11 (1964): Fig. 242. Revel-Neher 1992: 48; see also Fine 2014: 112; Schenk 2010 argues that the panel shows the continuous sacrificial worship performed in the Temple. 23 Goodenough 1953–1968, vol. 11 (1964): Pl. XXI; Kraeling 1956: 194–202. 24 James 2004: 41; for a detailed description of the different types of armour, see Hoss 2012: 100–101, who argues that the scale-mail armour appeared both in Sassanian and Roman depictions, while some of the cuirasses combine Hellenistic with Roman elements paralleled in images of military deities in the Temple of Bel. 25 These kinds of soldiers also appear elsewhere in Durene art and graffiti: Braid 2016: 34–35; soldiers in Roman-style cuirasses and helmets are also pictured in the ambiguous scene adjacent to Ezekiel’s vision of the dry bones.

376

Katrin Kogman-Appel

fashions and tastes in the region and dress-related identity issues that are at the core of this section. The various costumes shown on the walls of the building have been dealt with by several scholars. It has been observed that Persian garb appears most often in court scenes, whereas the Greek tunic with the himation demarcates religious contexts and is most often associated with Israelite prophets.26 It was also suggested that the attires with their obvious echo of various ethnicities are reminiscent of the use of different models of either Graeco-Roman or Persian background evidencing different artistic conventions.27 Dalia (Levit-)Tawil analysed the portrayals of Ahasuerus and Mordecai in the context of Iranian art and was able to show that the artists must have been very familiar with that iconography.28 Indeed, in Sasanian art not only the outfits of the kings are strikingly similar to those of Pharaoh and Ahasuerus, but the figures are also similarly seated on their thrones.29 Other scholars have pointed out that the different garb worn by Ezekiel in the valley of dry bones refers to his various functions in the narrative.30 Steven Fine suggests that scenes depicting episodes in Persia or Babylonia tend to show the protagonists in Persian garb, whereas those of events in the Land of Israel display figures in Greek wear, and that the images “project the garments worn by Jews at this time in the eastern Empire, including at Dura, into the biblical past”.31 None of these approaches considers the entire array of costumes. Neither do they deal with the different dress styles beyond observing the compositional types and artistic influence or relate to them in terms of the complex cultural fabric that brought them about.

 26 Kraeling 1956: 73; and Goodenough 1953–1968, vol. 9 (1964): 164–73, speculated that the tunic and mantle were reserved for religious dimensions and in Graeco-Roman art often for the Divine as well, and that the Jews borrowed this import in their preference for white gowns; Goldman 1992: 53–78, pointed out that in Buddhist art in Gandhara similar distinctions were made between draped and tailored costumes; Goldman 1999. 27 Goldman 1992: 73–74; similarly Hachlili 1998: 141, speaks of different visual topoi, such as reclining figures, enthroned figures, frontally shown figures, and the like; the identification of such conventions in terms of Roman or Persian origin is tricky; thus, Mordecai on horseback, e.g., was seen as both “shown in the stately mode of triumphal entry, the (Roman) adventus”, Moon 1992: 590, and a typical example of Sasanian imagery of kingship, Tawil 1979: 95–96. 28 Tawil 1979; and Levit-Tawil 1983. 29 Whereas the costumes are Sasanian, the frontal position is not; frontality was often described as a typical feature of Parthian art, Rostovtzeff 1935, and it was, among others, on this notion that (Levit-) Tawil 1979 based her analysis of the Durene murals in their relation to Parthian art. Rostovtzeff’s views were recently convincingly challenged by Dirven 2016, who argues that there have not been enough finds of Arsacid art to define it properly and that the frontal representations in the synagogue murals are an echo of Palmyrene art. However, early Sasanian reliefs occasionally show a king on horseback in an almost frontal position: see, e.g., in Naqsh-i-Rustam, https://oi.uchicago.edu/gallery/sasanian-rock-reliefsnaqsh-i-rustam-and-naqsh-i-rajab# 8C1_72dpi.png (accessed 23 March 2020). 30 Most recently, see Xeravits 2018: 82. 31 Fine 2013: 23–24; this was also suggested by Tawil 1979: 106, and Sabar 2000: 160.

Dress Codes in the Synagogue of Dura Europos?

377

As Fine also points out, ancient Jews in the Roman Empire were not distinguished from the surrounding societies in terms of dress.32 Thus, if the depicted garb played any role in the Durene Jews’ attempts at self-definition it was via the fashions of the surrounding cultures and the ways Jews related to them. Let us take a brief look at what these fashions might have been. Syria, a region with a range of Semitic populations, was exposed to varying degrees of Hellenistic and Roman influence. Apart from Dura with its abundant artworks, Palmyra, another famous site, further to the west, with finds from the late antique period offers further insight. Whereas Dura had been Hellenistic and was later occupied first by the Arsacids (113 BCE–164 CE) and then by the Romans (since 164), Palmyra remained independent for most of the period in question. Dura, as has often been noted, was home to a heterogeneous population (Palmyrenes among others),33 whereas the population of Palmyra was more homogeneous. Both cities left a rich visual and material culture, including a wealth of textile fragments, which, owing to these different historical and demographic backgrounds, are strikingly different. In Palmyra a great wealth of tombstones with portraits of the deceased was unearthed, whereas in Dura, the extant artworks reflect primarily mythological and ritual themes. Bernard Goldman relied on works of art from both sites in an attempt to reconstruct clothing norms in Syria. He noted that their potential to offer information on real-life clothing is very different owing to their dissimilar character and background. Whereas he believed that the Palmyrenes portrayed some measure of real life on their tombstones, showing the deceased in their best clothes, he argued that the dress of the protagonists of the mythical scenes in Dura are of historical significance and reflect long-gone norms. He also suggested that the Persian trouser suit would have been the costume commonly seen in the streets of any Asian city, whereas the Roman garb would have been considered exotic in eastern Syria.34 His study did not consider the actual textiles found at either site. Thus, in some contradiction to Goldman’s suggestions, it appears that in Dura numerous fragments of Roman tunics and mantles were found, but there were only rare traces of apparently Persian garb.35 We do not know which of the ethnic groups in Dura wore this Roman garb and if it was also used by Jews, but it can  32

Fine 2013: 21; Cohen 1999: 31; see also Pohl 2006: 112, noting that “the evidence that costume marked social distinctions is overwhelmingly stronger than its significance in displaying ethnic identities...”; similarly also Hoss 2012: 98–103; the same applies to later medieval Europe, where dress was not always a marker of ethnicity and minority groups; rather, such distinctions were imposed on them by the authorities so that they would be distinguishable. 33 Jensen 1999: 180–81; on the Palmyrenes of Dura, see Dirven 1999. 34 Goldman 1994; 1999. 35 Pfister and Bellinger 1945: 11, 14–21, state that no Persian garb was unearthed in Dura; decorations on one or two fragments, however, may point to a Persian origin; either way the scarcity of Persian textiles strongly contradicts Goldman’s suggestion, at least as far as Dura is concerned; the finds were recently revisited by Baird 2016 with special attention to yet unpublished material.

378

Katrin Kogman-Appel

be dated to the Roman period and specifically to the years just prior to the Sasanian conquest. In contrast, the textiles found in the Palmyrene tombs, even though they tell us very little about the shape of the costumes, reveal a wealth of different ornamental designs, many of which are clearly of Iranian origin.36 The tombstone portrayals offer further information. While the mythical protagonists of the Durene wall paintings, in particular those in the synagogue, are wearing either GraecoRoman or Persian garb and do not mix fashions, the men portrayed in Palmyra wear costumes of mixed styles.37 They are often shown wearing the trouser suit, but it is usually combined with one or another Graeco-Roman item, either a himation or a short shoulder mantle (chlamys) particularly suitable for horseback riding. Others apparently wore Persian-style tunics, but judging from the tomb reliefs these appear folded and belted Roman-style. Alongside these, there are also portraits that show the deceased in full Greek garb. This mixture of fashion is also true in regard to the attire chosen by women, who had a certain preference for Graeco-Roman garments, but favoured Persian-style headwear and jewels.38 In the Dura synagogue art only Esther’s outfit reflects a combination of Greek garb and Persian headwear (Fig. 7),39 and all the other women appear in purely Greek attire. All this still does not tell us how Durene Jews dressed. Although it seems plausible to assume that Palmyrenes indeed wore mixed-style costumes, the textile remains from Dura seem to indicate that even though different ethnic groups lived in the town, they all wore Roman dress, or at least that that dress seems to have been quite common. One wonders whether the Jews of Dura shared this preference, and the fact that key Israelite figures in the murals are dressed the same way seems to support such a suggestion.40 There are two possibilities: the first is that Roman garb was typical for Dura’s population including the Jews and that people did not commonly wear the Persian trouser suit. This would clearly mean that the appearance of Persian garb in the murals does not reflect everyday practice, but was rather an artistic convention. The other possibility is that the textile finds in Dura do not offer a full picture of what was worn there, but perhaps only indicate the norms of the Roman population. Other groups may in fact have dressed similarly to what was common in Palmyra and worn mixed-style  36

See, e.g., Pfister 1934: 13–38, 61–63. Some of these outfits are ambiguous: the angel on Jacob’s ladder (only one angel has survived in full) may wear some draped piece of cloth on his shoulders, but the painting is not preserved well enough to make out the details; behind Mordecai, who is clad in the Persian royal costume, there is some draped cloth, seemingly not attached to his body, flapping in the wind (Fig. 7). 38 Goldman 1994. 39 For a thorough discussion of Esther’s outfit, see Levit-Tawil 1999, who concludes that Esther bears features of Julia Domna (d. 217), Septimius Severus’ Syrian wife, and that the Jewish patrons and artists meant to mark the end of a Roman era that was beneficent towards the Jews; Septimius Severus and Julia’s reign was followed by turmoil. 40 Moon 1992: 592–595, but he overemphasized the Roman elements in the synagogue imagery. 37

Dress Codes in the Synagogue of Dura Europos?

379

dress, which would mean that the Greek and Persian costumes in the murals reflect artistic conventions and were not even remotely related to real-life norms. Either way we are left with the question as to whether these artistic conventions were randomly chosen because the people who designed the murals were familiar with both worlds,41 or whether they followed a dress code with clearly delineated iconographic meaning. It is likely that the artists who designed the synagogue murals not only approached the costumes with a dress code in mind, but imbued it with complex meaning, which in turn shaped the visual language of the paintings. This dress code functions in a broader context of a whole set of cultural features, pieces of a puzzle, so to speak, that can be assembled only with great difficulty towards an image of the Jews of Dura. Nothing is known of this community apart from its synagogue with its paintings and inscriptions in three languages and small fragments of parchment with a liturgical text unearthed in the vicinity of the building.42 We do not know where these people originated or how and why they came to Dura. It is not quite certain what they wore. Let me briefly survey the information earlier scholars have brought to light about the culture of Durene Jews, their degrees of acculturation to the Graeco-Roman world, and their possible relationship to the Sasanian Empire across the frontier, as well as to fellow Jews who lived there. Despite the multicultural profile of Dura’s population and the city’s earlier occupation history, the material culture excavated in the most recent layer is primarily Roman.43 It is difficult to assess the ways that the Jews of Dura related to the Roman authorities. Dura’s importance as a garrison town increased significantly in the second decade of the third century, and a military camp was set up in its northern sector. Earlier scholars assumed that the Jews of Dura were merchants who attended to the needs of the Roman army.44 However, more recently it has been argued that some elements in the iconography of the synagogue murals and the tile decorations might point to the possibility that some members of the Jewish community actually served in the army.45 It was pointed out that several murals show soldiers in quite a realistic and accurate way, suggesting a high degree of familiarity with military matters.46 A brief paper by Shalom Sabar on the Purim panel discusses the Esther story (Fig. 7) against the background of the probably somewhat delicate position of the Jews within the setting of the  41 Most of the eastern Syrian region was under Roman rule (except Palmyra), and any Roman influence can thus apparently be easily explained; however, given the area’s proximity to Iran, Persian influence was strong everywhere, even in regions that were successful in opposing Iranian political power: Goldman 1994: 173. 42 For a recent discussion, see Fine 2014. 43 Baird 2016: 32–33, who also stresses that representations of dress in imagery can be used as tool to reconstruct the actual practice only to a limited degree. 44 Gutmann 1992: 140. 45 Rosenfeld and Potchebutzky 2009; and Weisman 2012: 24–31. 46 Weisman 2012; and Rajak 2013: 105.

380

Katrin Kogman-Appel

Roman-Sasanian conflict. Sabar interprets the four men in Greek wear in the centre of the composition as Jews observing the situation, demonstrating the Durene Jews’ identification with Roman culture as they anticipated a Persian attack.47 As several Greek donor inscriptions and graffiti indicate, there must, indeed, have been Greek-speaking Jews in Dura.48 Do the Greek donor’s inscriptions, the possible involvement in army life, and economic prosperity really tell us that these Jews were all “Romanized”?49 In a way, the Jews in the Land of Israel and the coastal cities along the Mediterranean can be defined in terms of their high levels of acculturation to Roman culture. They had lived under Roman rule for a long period, but they spoke Greek. But this does not necessarily mean that all or most of the Jews of Dura fell into the same category and saw themselves as Jews acculturated to Roman culture. The indigenous population in the region was Semitic and spoke Aramaic. The dominant donor and community leader, one Samuel, left a detailed inscription in Jewish Aramaic. The inscriptions embedded in the biblical pictorials are likewise in Jewish Aramaic, with only three one-word inscriptions in Greek (two of which are names: Solomon and Aaron, even though underneath his mantle the latter is wearing Persian trousers). There is no inscription in Hebrew, and, as Fine observes, some of the Aramaic inscriptions parallel wording in the Aramaic Targumim. As Fine notes, this points to certain cultural preferences that are different from those of the Jews in the Land of Israel, where biblical scenes were commonly accompanied by Hebrew legends. Whereas variants of Aramaic were commonly spoken among Jews everywhere, this distinction still suggests a different cultural identity. Palestinian Jews, despite Aramaic being their spoken language, had a different kind of affinity to Hebrew than their coreligionists in Syria and Mesopotamia. Moreover, the limited use of Greek also speaks for itself: the worshippers in the synagogue did not look at the Bible from a Greek-speaking perspective.50 Apart from the Roman and the local Semitic components of Durene Jewish identity, there are elements that indicate an affinity with Iranian culture. Arsacid and Roman administration and culture were closely entangled in Dura Europos. Unlike Palmyra, which, as I noted earlier, remained independent during the late antique period, Dura had been under Arsacid rule for more than 250 years and works of art from that period were still to be seen in the city when it was under  47

Sabar 2000; Goodenough 1953–1968, vol. 9 (1964): 182–83, and Goldstein 1995: 115–16, interpreted them as angels. 48 All the inscriptions were originally published by Kraeling 1956: 255–320, and recently again with detailed analyses by Noy and Bloedhorn 2004: 133–213. 49 This term has caused much discussion in recent scholarship, see, e.g., Dirven 2016, with particular reference to Dura; she defines the style of the murals and other sites at Dura as an “outcome of a meeting of Roman and Graeco-Semitic art”: ibid., 87. 50 For a recent discussion of the inscriptions leading to a definition of the identity of the Jews in Dura as “hybrid”, see Fine 2014.

Dress Codes in the Synagogue of Dura Europos?

381

the Romans.51 Thus, Graeco-Roman and non-Roman identities cannot be neatly distinguished and the different ethnicities and their norms and life styles were much entwined. More important, however, is the question of how the Jews of Dura might have viewed the region across the Euphrates. There, in 224, twenty years before the synagogue murals were installed, the Sasanian dynasty had taken over. Scholars assume that the administration and the local population of Dura must have been aware of the possibility of an approaching Sasanian attack since 239.52 Jews had lived in the Persian Empire since the sixth century BCE in better and worse times, and it is hard to say what the Jewish expectations of the approaching conquest might have been. The Achaemenids were associated with the end of the Babylonian exile, the permission to rebuild the Temple, and, as the Esther mural so clearly reminds us, with both Haman’s attempt to persecute the Jews and its prevention.53 We do not know much about Arsacid approaches to the Jews, whereas somewhat more is known about Jewish life under the Sasanians. Early Babylonian rabbinic culture emerged in parallel to the transition between the two dynasties,54 and most of the information available reflects the perspective of Babylonian rabbis, who, according to Richard Kalmin, were a relatively secluded elite.55 Despite the rabbis’ apparent tendency to live and act in seclusion, numerous signs of acculturation have been observed in rabbinic thought, or, as Shai Secunda puts it, “the Bavli and Middle Persian literature produced meaning in the same discursive space and across the shared ‘text-scape’ of Sasanian Iran.”56 Earlier scholarship about the Babylonian Talmud in particular and Babylonian Judaism in general held the Arsacid regime to be more tolerant towards the Jews than the Sasanians.57 This view and the subsequent claim that the Babylonian rabbis were concerned about the rise of Sasanian power has been challenged by others. Given, that talmudic material is, for the most part, undated, it is difficult to contextualize it along a timeline. Kalmin has reread passages that earlier scholarship had interpreted as expressing concern about the emerging Sasanian regime and argues  51 Braid 2016: 34; on Parthian Dura, see Millar 1998; Gregoratti 2016 pointed out that the population was also mixed during the Arsacid period and that the various ethnic groups were Hellenized to different, often high, degrees, a fact that has to do with the origin of Dura as a Hellenistic entity; see Goldman 1992: 61, for a discussion of other areas controlled by the Parthians; Czajkowski 2016 emphasizes that during the Roman occupation its administrative and institutional culture was a mixed one with numerous local and Parthian elements alongside Roman ones. 52 Baird 2012: 312. 53 Gruen 2007 describes a nuanced, not totally benign, occasionally subversive biblical and postbiblical image of the Persians. 54 Elman 2012 challenged Neusner 1992, who argued that during the late 2nd and early 3rd centuries Babylonian Jewry cannot be considered to have held a rabbinic worldview. 55 Kalmin 2006: chaps. 4 and 5. 56 Secunda 2014: 145. 57 For example, Gafni 1990: 39–40.

382

Katrin Kogman-Appel

that they refer to later periods. Yaakov Elman pursued a different approach and analysed rabbinic sayings against the background of different geopolitical settings all within an Iranian framework. Secunda describes how rabbis and Zoroastrians othered one another, but carefully considers the mentioned difficulty of contextualizing these rabbinic comments. Finally, Robert Brody argues that in the Sasanian Empire the Zoroastrians did not pressure its minorities to convert and that anti-Jewish and anti-Christian persecutions in that Empire were rare prior to the fifth century, with the exception of an episode under Kartir, a late third-century Zoroastrian priest who excelled in religious zeal. Kartir was active some decades after the Jews of Dura had shaped their worldview.58 Thus, if a Sasanian attack had been spoken of during the 240s, the Jews of Dura might very well have had positive expectations regarding that eventuality. LevitTawil suggests that the portrayal of Mordecai is reminiscent of early Sasanian royal iconography and thus conveys the expectation that Shapur I would protect the Jews.59 In the name of R. Huna, finally, the Babylonian Talmud says the following, apparently reflecting a situation from the late third century: “‘Bring my sons from far and my daughters from the ends of the earth (Isa 43:6).’ Bring my sons from far: R. Huna said: these are the exiles in Babylon, who are at ease like sons. And my daughters from the ends of the earth: these are the exiles in other lands, who are not at ease, like daughters.”60 This statement indicates that in the eyes of some the Jews were better off under Sasanian rule than elsewhere, and one naturally wonders if this also reflects the view of Dura’s Jews. There is some evidence that the Jews of Dura had actual contact with people on the other side of the frontier. First of all, the frequent appearance of Persian garb in the synagogue murals certainly speaks of a close affinity to Iranian culture, and we have seen that their accurate rendering indicates a great deal of familiarity with and understanding of the visual language of kingship. Secondly, there is evidence, recently discussed by J. A. Baird, that the Sasanian conquest took place in two phases. There may have been a first campaign in 253 CE, which left the city under Sasanian rule for a few months until the spring of 254, and a final conquest in 256.61 Several Iranian inscriptions dated between 253 and 256 are very prominent in the Esther pictorial, attesting to visits by Iranians who saw the images and “liked” them (or: “approved” of them). Sabar, Fine, and Touraj Daryaee have made several attempts to contextualize these inscriptions, but it is not clear whether it was Persian Jewish visitors or Sasanian officials who inscribed them. The Esther pictorial was not only physically within reach (it appears in the lowest register), but it must have had a particular significance for  58 Brody 1990; Kalmin 2006: chaps. 4–6; Elman 2004; 2006; 2007; 2012; and Secunda 2014: chaps. 3 and 4. 59 Tawil 1979: 107; and Levit-Tamil 1983. 60 See b. Mena 110a; and Elman 2006: 165. 61 The 253 CE campaign was suggested by Rostovtzeff 1943, but later dismissed by James 1985, but for more recent evidence based on a review of the numismatic finds, see Baird 2012: 312–14.

Dress Codes in the Synagogue of Dura Europos?

383

Persian Jews, which led them to add their signatures.62 This might indicate that beyond their familiarity with Persian dress and their recollections about biblical Persians, the Jews of Dura had contacts with Jewish communities across the border. Moreover, these contacts might not necessarily have been limited to the few months after the first Sasanian attempt to conquer the city, but had existed for some time in the past. However, if the inscriptions were added by Sasanian officials, as Daryaee’s recent rereading of them implies, they may attest to positive interactions between the Jews of Dura and the Sasanians. In the eyes of these Iranians this was less a visualization of a biblical tale than a powerful image of Persian kingship apparently telling something about the Jewish expectations of a Sasanian conquest.63

The Costumes and Visual Narration The previous discussion about the wider context of the costumes shown in the murals reveals obvious signs of entanglement, making it evident that Persian/ Mesopotamian cultural elements were as much features of Durene Jewish identity as were the Graeco-Roman ones. These signs offer strong support for the conclusion that the apparently dichotomous dress code in the murals does not have much to do with actual fashion, identity, or political loyalties. Rather, it functions in the pictorials as a set of idiomatic features of the visual language employed in a sophisticated way. I observed earlier that Greek wear in the murals is commonly associated with Israelites or with figures who witnessed the greatness of God. Let me look again at the depiction of Moses’ infancy, at the Greek nymphs attending him, and at Pharaoh’s daughter associable with Anahita/ Aphrodite. Approaching these elements not as signs of “Romanization” of Israelite history, one realizes that by means of the Graeco-Roman visual idiom the patrons and artists were able to highlight certain features of Moses’ persona embedded in biblical exegesis. I return to these features below.  62

They were first published by Bernhard Geiger in Kraeling 1956: 283–317; and again in Noy and Bloedhorn 2004: 177–208; for discussions, see Sabar 2000: 162–63, suggesting that while the imagery of the Esther panel could have been interpreted by Romans as a homage to them, the Iranian inscriptions indicate that it was also suitable for conveying a message to Persian-speaking visitors about Ahasuerus’s support of the Jews: Fine 2011; Fine 2014: 113–19, and Baird 2012: 312–14, suggest that the inscriptions were added by Jewish visitors; however, all three of these scholars emphasize that this cannot be proven and that it is also possible that the texts were added by Sasanian officials; on this, see Daryaee 2010: 31– 32; as Baird, ibid., underscores, the fact that they were written carefully indicates that they are not unofficial graffiti. 63 Daryaee 2010 argues that during the early days of the Sasanian regime the rulers tended to identify with the Achaemenid past (a trend in cultural politics that was later abandoned), hence the interest in the Esther story and Ahasuerus’s portrayal; for the accuracy of the Persian elements in relation to Persian art, see Tawil 1979 and Levit-Tawil 1983.

384

Katrin Kogman-Appel

Thus, the worshippers in the synagogue did not associate Moses with Greek or Roman heroes or with a Graeco-Roman lifestyle, especially not if they walked about the streets of Dura in Persian trousers. Lucille Roussin deals with the Graeco-Roman garb in the imagery in relation to what we know from the Mishnah about Jewish dress norms in the Land of Israel and about material remains from the Bar Kokhba period. She observes a high degree of correspondence: there are Hebrew terms for the Roman items, which were decorated in a style that was typical of Roman tunics and mantles.64 In Roman Palestine and in areas near the Mediterranean coast, where Graeco-Roman culture was dominant, fourcornered garments, alitot in Hebrew, were undoubtedly worn by Jews. From a wealth of references in early rabbinic literature it appears that the alit was not yet a ritual object as it would be understood in the medieval and the modern periods, but rather an upper garment suitable for various mundane purposes. It seems to have been of considerable value, was used primarily by the upper classes, could be considered part of an inheritance, and could serve as a token for the redemption of a firstborn.65 The late antique alit, as a rectangular garment with four corners, had to have the halakhic fringes attached to it as a constant reminder of the full array of the 613 divine precepts,66 and several scholars have noted that such fringes can be discerned in the Dura murals in two of the portrayals of Moses.67 Although the fringes have no real ritual implications, they are an important factor in reading the himation in these images as a sign of identity, and even though they are hardly discernible, they undoubtedly strengthen the Israelite/ Jewish identity of the Greek wear. That wear being a norm among Jews in the Land of Israel indicates that in the imagery in the Dura synagogue it was associated with a “Palestinian Jewish identity”, which may have been projected onto the biblical scenes that imaged episodes that took place in historical Israel.68 Even if some or many of the Durene Jews dressed in a similar fashion, we have seen that this Palestinian Jewish identity was not fully congruent with that of the synagogue worshippers, which turns the Greek garb into a tool of visual language demarcating biblical history, ancestry, and divine intervention. The other component of the bilingual visual language of the murals was Persian. Kingship was visualized in Arsacid and Sasanian terms. Goldman followed the Persian dress back to its origins and showed that it was long associated with  64

Roussin 1994 and more recently Fine 2005a: 179–80; on the textile remains from the Bar Kokhba period, see also Revel-Neher 1992: 53–54; for a survey on the information about Jewish clothing in the Land of Israel, see Shlezinger-Katsman 2010. 65 For example, m. Pe’ah 4:3; t. B. Me. 5:26, ed. Lieberman; t. Bek. 6:13, ed. Zuckermandel; for more sources, see Kogman-Appel 2012: 69–70. 66 Num 15:38–39; Deut 22:12. 67 Pfister and Bellinger 1945: 10–12; Goodenough 1953–1968, vol. 9 (1964): 127; Fine 2013: 24; RevelNeher 1992: 53; and again Revel-Neher 2012: 506–508. 68 See above, n. 64.

Dress Codes in the Synagogue of Dura Europos?

385

various Others. In Achaemenid art it defined foreign identity and low status associated with peoples from the north. In Greek art, where Persian identity is problematized as Other, the trouser suit was adopted, again as a sign of foreign identity. In the Arcasid context, however, the costume began to be associated with the rejection of Hellenistic/Roman dominance and became a symbol of Iranian political identity. In Sasanian art in particular it is to be understood as part of an anti-Hellenistic and anti-Roman message.69 In the synagogue, however, the Persian royal wear was certainly not meant to deliver an anti-Roman message. Neither is it likely that it was chosen solely to please Persian visitors. It visualizes both Israelite and foreign kingship, both beneficent and hostile. It was employed as a recurring visual means to define certain protagonists’ role in the narrative and to demarcate situations where Israelites had to cope with foreign rule. The idea of “kingship” was then also associated with Saul and David. The men in simple trouser suits convey a different message: wherever they appear, change occurs. That is, they function as dynamic agents of narrative change. The artists used the dress code to delineate key figures in Israelite history, witnesses of divine intervention, and kings notwithstanding their political identity. All these are staged in the compositions following attempts to develop a narrative method, that is, to resolve the gap between time and space.70 A number of pictorials are aimed at conveying complex plots as continuous flows of events set in different kinds of pictorial space. Developing visual plots within the tension between the dynamicity of the narrative and the fixity of visual representation required certain techniques. The design of recognizable protagonists, for example, was crucial, as was the need for the protagonists who figure several times in the same narrative to look the same in every instance. The use of a bilingual set of visual elements made possible by the intercultural framework of Dura enabled the artists to enhance the appearance of their protagonists and to turn them into powerful tools of visual story telling. By means of the dress code they could offer further differentiation among various kinds of protagonists. This differentiation does not underscore identities in political or religious terms; rather it provides visible information about the role these individuals assume in the visual narrative, compensating, in a way, for the spatial limits of monumental composition. When a dynamic narrative designed as a set of consecutive scenes is placed in a static picture space, the viewer receives it as a sequence of events, and intuitively complements the visual information towards a more complete image of the story.71 The more visual information there is in the pictorial the easier it is for the viewer to complement the narrative.  69

Goldman 1992. The narrative qualities of the murals were discussed repeatedly by Kurt Weitzmann, but only as far as models are concerned and not in terms of visual strategy, see, e.g., Weitzmann 1990; on the historiography of narrative techniques in art in general, see also Lewis 2006. 71 Abbott 2002: 7. 70

386

Katrin Kogman-Appel

Let us take a brief look at the scenes relating to the discovery of Moses in terms of visual story telling. A third-century viewer looks at the sequence (Fig. 1); he has no Bible text to help him follow the narrative; he heard the story while in the synagogue and in (oral) study and has a notion of some exegetical elements to embellish the biblical story and enhance its contents.72 Relying on his memory, he is expected to associate what he sees with what he knows: (A)

(B)

(C)

(D)

(E) (F)

A man on a throne with attendants, all in Persian garb: by means of the dress code the man on the throne is understood to be a king and the gesticulating attendants are perceived as agents of the narrative. The midrash tells a story about Pharaoh consulting his courtiers as to how to deal with the ever increasing number of Israelites.73 Although the midrash may be enshrined in the viewer’s memory, the Persian garb as an indication of agency will help him associate the figures with the midrash, in this case bringing about further hardship to the Israelites. The man who was identified as king consulting his courtiers is shown in conversation with two women in Greek garb: the Bible (Exod 1:15–19) speaks of midwives who “feared God”; the dress code indeed defines the women as Israelite or God-fearing. The viewer realizes that the midwives and the two women to the extreme left are dressed the same way and can now associate this detail with another exegetical motif: Jochebed and Miriam appearing before Pharaoh disguised as midwives. Thus, as the dress code indicates, the women are Israelite. The infant Moses in an ark is placed among the reeds by a woman in Greek garb (Exod 2:3): here the narrative lacks consistency, as the woman who puts the ark among the reeds is expected to wear the same outfit as Jochebed in the other parts of the narrative. However, instead of being imaged in a red tunic and a beige palla, she is cast in a red mantle. Pharaoh’s daughter takes a bath in the river, as her attendants walk along the riverbank (Exod 2:5). Standing naked in the water, Pharaoh’s daughter is shown with the infant in her arms, the small ark appearing next to her (Exod 2:5). Although the pre-Masoretic text is not explicit about who actually rescued Moses (the princess or one of her maidens), the image implies that it was the princess who took the child and thus guides the viewer to grasp the story according to that exegesis. Visually, the nude princess with the baby can be associated with images of Aphrodite/Anahita and thus conveys a message of motherhood and caretaking.

 72 Later these midrashic motifs appeared in texts: b. Soah 11b–12b; b. Sanh. 101b; Tg. Ps.-J. Exod 1:15; Tan Shemot 62b; Exod. Rab. 1:20; see also Schubert 1974: 41–43. 73 Pharaoh’s counsellors are prominent in rabbinic literature: see Baskin 1983.

Dress Codes in the Synagogue of Dura Europos?

(G)

(H)

387

The princess’s servants bring gifts as well as ointments and various other things needed to attend to the infant. Their garb and the utensils they carry identify them as nymphs, thereby enhancing the identity of the infant as a special child. They are associated with the midrashic embellishments about Moses’ special features at birth, underscoring his identity as a prophet: a painless pregnancy and birth, typical of the mothers of prophets; the birth room filling with light; Moses’ exceptional beauty; and other miraculous events that attest to that identity.74 Moses is given to Jochebed and Miriam, who are identical with the midwives, to be nursed.

By means of the dress codes the artist was able to specify numerous elements in the story: kingship; narrative agency of the courtiers; and the exegetical motif of Jochebed and Miriam appearing in the guise of midwives; it is the princess who rescues the child, even though the biblical text is not explicit about this; her Aphrodite-like appearance demarcates the depiction as an equivalent to a birth scene; nymph-servants identify Moses as a prophet and thus associate him with the Divine. In other words, Pharaoh’s daughter was not perceived as a pagan goddess, neither did her appearance underscore her otherness. She was rendered in a visual language that speaks of motherhood and caretaking more than of Aphrodite. The Greek nymphs are part of a visual vocabulary that made Moses into a figure with certain features that could be translated into a “hero” when a Roman looked at the painting and a “prophet” when a Jew looked at it. The visual language of Greek heroes later delineates Moses as a prophet and a leader in terms of his description in the midrash. Probably the most sophisticated narrative device is the addition of the witnesses. In most cases they cannot or can only vaguely be identified with figures mentioned in the biblical text, but their existence in the visual story significantly amplifies the narratives’ messages. The static appearance of these tall men in chiton and himation associated with the high status of key Israelite figures— prophets and patriarchs—puts them at the highest rank in the hierarchy of protagonists, partially in contrast to the anonymity of some of the others. Wherever they appear they deliver a message of divine intervention, miracle, or salvation. Occasionally, it is the plot that creates this identity: the men flanking the altar erected for Baal could be understood as the pagan prophets. However, in that case we would probably expect them to appear in Persian garb similar to that of the statues of Dagon or the Israelite Temple priests, thus paralleling the identity of ritual agents. Instead their appearance is that of major Israelite figures, as if to emphasize that they are shown at the moment that God’s powers are manifest and recognized. This moment of recognition and witnessing is shared by all of these figures: the four men in the Purim story that split the narrative into two  74

See above, n. 72.

388

Katrin Kogman-Appel

parts, one negative, one positive; David’s brothers witnessing divine selection for kingship; the heads of the Israelite tribes crossing the Reed Sea, who serve not only as representatives of their tribes, but as witnesses of liberation and rescue; the anonymous men observing the cart with the Ark on its way to Beth Shemesh, thus forestalling calamities among the Philistines; and, of course, the Israelites in the lengthy Ezekiel narrative at the moment of resurrection. The wearers of the different costumes not only assume certain roles and functions in the stories, but also determine the pace of the narrative flow. The palliati convey a sense of static presence, even if their role in the story is expected to be dynamic. Moses is supposed to be moving ahead leading the people out of bondage, but his enormous stature and his frontal appearance seem to stop his forward movement. The “witnesses” are all passive and static. Their dignified immobile bearing invites the viewer to slow down and grasp the significance of the moment initiated by divine intervention. Visually the enthroned kings in Persian royal garb likewise are static, but their entourage usually conveys a sense of dynamic bustling, while their attendants bring about changes in the history of Israel. We have seen that it is the simpler Persian trouser suit that demarcates agency. Its wearers function as agents sent from heaven, as the angels on Jacob’s ladder or as Ezekiel as a divine agent in the vision of the dry bones. In other scenes they are instrumental in bringing hardship to Israel, as Pharaoh’s counsellors, who advise him to kill the male newborns of the Israelites. They also appear as the high priests’ assistants who keep the sacrificial cult functioning. Some of them are soldiers who determine the outcome of battles and thus the course of history. Wherever they appear these figures are the most active agents in the narrative plot.75 Their Roman counterparts— men in short tunics—also represent dynamic action. However, the low social status associated with the short tunic seems not to allow for narrative agency. They assume the role of extras, enlivening the scenery but do not necessarily yield changes of significant theological weight.

Conclusions Given that (a) we know so little about the Jews of Dura Europos, and (b) only 60 percent of the visual narratives in their synagogue have been preserved, we cannot, at this point, assess what the patrons and artists wanted to tell their viewers with the colourful and lively narratives they attached to the walls of the building.  75 In two scenes the heads of the tribes are shown in Persian suits: in the portrayal of the desert well each stands in a tent, and they also attend the enthroned figure of the messianic king in one of the destroyed layers of the mural above the Torah shrine, still faintly discernible at the time of their discovery, Goodenough 1953–1968, vol. 11 (1964): Pl. IV, Figs. 75, 93, 232. Both these scenes have proven particularly difficult of interpretation.

Dress Codes in the Synagogue of Dura Europos?

389

The synagogue being a space of teaching and learning had undoubtedly much to do with their intentions. The fact that the viewers entering its space found themselves overwhelmingly surrounded by biblical protagonists, several of whom gazed at them intensely, set in colourful dense and dynamic compositions, enhanced their didactic value. These narratives were certainly not perceived as neutral equivalents of text. What exactly was important to be visualized and mediated in the artists’ eyes can only be guessed at. This essay does not offer any attempt to make such guesses. Rather, it points to a visual tool employed by the artists and patrons in promoting whatever their agenda might have been. Much remains to be told about the narrative strategies utilized by those who designed the visual stories in the Dura synagogue. For example, scholars have never dealt with the constant tension between stories that develop in continuous flows conveying a sense of dynamic movement and others that concentrate all the elements of the narrative in one framing device and thus force the eye to follow the story in different directions and at a varying pace. Studying the way these strategies were perceived and received will determine how we can, finally, contextualize these narratives in what we know about the life of the Durene Jews. These Jews lived in a multicultural environment at the Roman-Sasanian frontier, a situation that has been explored in recent scholarship in various ways. For example, relying on Homi Bhaba’s concept of hybridity in contact zones, Michael Sommer describes Dura as a frontier region and thus “a space of ambiguity”. In this setting the synagogue murals are interpreted as a document of Jewish identity within a pagan environment coping with acculturation, hostility, and various situations of adjustment and accommodation.76 Similarly, Fine defines the Jews of Dura as hybrids in terms of both the imagery in the synagogue and their epigraphic habits, in particular the use of different languages in the inscriptions.77 Some scholars argue that various elements in the imagery can be read as signs of religious self-affirmation of a society that had to cope with Roman syncretism, relatively high levels of tension, and even religious pressure, whereas, at the same time, they probably prospered economically within the Roman system.78 We have seen that messages of divine intervention and miraculous salvation make up another thread that is shared by several of the imageries. Approached from the point of view of visual language, it becomes apparent that notwithstanding the Durene Jews’ political stances and their expectations of  76

Sommer 2016. Fine 2011. 78 Moon 1992: 601–603; Elsner 2001 speaks of “cultural resistance”, but admits at the same time that there is no documentary evidence of any such “resistance”, and that any conclusions drawn from studying the imagery can only point to likely possibilities: ibid., 272–273; Sommer 2006; and Rajak 2013. Flesher 1998 argues that the paintings deal with non-Jews but not in a confrontational way, but rather in an attempt to sketch an image of cooperation with Dura’s non-Jewish population; the Persian garb in the Esther story is interpreted as a sign of otherness. These views have been rejected by Dirven 2004, who argues that nothing in the historical circumstances and our knowledge about religious life in Dura points to competition or resistance. 77

390

Katrin Kogman-Appel

the Sasanians, or any other agendas, the multicultural ambience of Dura and its frontier reality created the stage for the development of a multilingual visual idiom. The artists who designed the murals “spoke” Roman as much as they “spoke” Persian in terms of the visual language they employed. The dress code of the synagogue imagery functions differently than such dress codes in other cultures, where costumes often tended to demarcate self-identity and otherness. Their position among the different cultures enabled the Jews of Dura to employ these various attires differently. In a way, the dress code can be interpreted as a specific Jewish tool of visual narration, as it was a by-product of Jewish life among different powers and the ability of the Jews of Dura to adapt to changing circumstances. The dress code in the imagery of the Dura synagogue speaks precisely of this ability to adapt, but not in political terms. It demonstrates the ability to be multilingual not only in a linguistic sense, but in a broader cultural sense as well. Viewed in this way the costumes, at first sight eclectically juxtaposed, emerge as tools specifically employed in a Jewish visual language that has to be read with a dictionary other than that of either Roman or Persian art.

Figures

Fig. 1.

Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria, 244–245, The Discovery of Moses (image in the public domain).

Dress Codes in the Synagogue of Dura Europos?

391

Fig. 2. Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria, 244–245, Israelites Departing from Egypt (image in the public domain).

Fig. 3. Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria,244–245, The Sacrifice of the Prophets of Baal (image in the public domain).

392

Katrin Kogman-Appel

Fig. 4.

Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria, 244–245, The Sacrifice of Elijah (image in the public domain).

Fig. 5.

Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria, 244–245, Ezekiel in the Valley of Dry Bones (image in the public domain).

Dress Codes in the Synagogue of Dura Europos?

393

Fig. 6.

Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria, 244–245, The Ark of Covenant in the Philistine Camp (image in the public domain).

Fig. 7.

Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria, 244–245, Esther Story (image in the public domain).

394

Katrin Kogman-Appel

Fig. 8.

Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria, 244–245, The Battle at Even Ha’ezer (image in the public domain).

Fig. 9.

Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria, 244–245, The Ark of Covenant Being Brought to Jerusalem (?) (image in the public domain).

Dress Codes in the Synagogue of Dura Europos?

395

Fig. 10. Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria, 244–245, David and Shaul in the Wilderness of Ziph (?) (image in the public domain).

Fig. 11. Damascus, National Museum, mural from the Synagogue of Dura-Europos, Syria, 244–245, The Initiation of Sacrifices (image in the public domain).

396

Katrin Kogman-Appel

Bibliography Primary Sources: Epstein, I. (ed.) 1961. The Babylonian Talmud. London: Soncino Press. Midrash Tanuma 1964. Midrash tanuma al amishah umshe torah. Jerusalem: LevinEpstein. Shinan, A. (ed.) 1984. Midrash Shemot Rabbah, parashot 1–14. Tel Aviv: Dvir. Clarke, E. G. (ed.) 1984. Targum Pseudo-Jonathan of the Pentateuch: Text and Concordance. Hoboken, NJ: Ktav. Lieberman, S. (ed.) 1988. The Tosefta: According to Codex Vienna, with Variants from Codex Erfurt, Ms. Schocken and Editio Princeps (Venice 1521). The Order of Nezikin. New York: Jewish Theological Seminary. Zuckermandel, M. S. (ed.) 1970. Tosephta: Based on the Erfurt and Vienna Codices. New ed. Jerusalem: Wahrmann. Hyman, D., Lerer, N., and Shiloni, Y. (eds.) 1973–1977. Yalqut Shim‘oni: Osef midreshe azal – torah, nevi’im, ketuvim. Jerusalem: Rav Kook Institute. Secondary Literature: Abbott, H. P. 2002. The Cambridge Introduction to Narrative. Cambridge: Cambridge University Press. Avi-Yonah, M. 1992. “Goodenough’s Evaluation of the Dura Synagogue: A Critique”. In: Gutmann, J. (ed.), The Dura-Europos Synagogue. A Re-evaluation (1932–1992). 2nd ed. Atlanta, GA: Scholars Press, 117–135. Baird, J. A. 2012. “Dura Deserta: The Death and Afterlife of Dura-Europos”. In: Christie, N. and Augenti, A. (eds.), Vrbes Extinctae: Archaeologies of Abandoned Classical Towns. Farnham, UK: Ashgate, 307–329. Baird, J. A. 2016. “Everyday Life in Roman Dura-Europos: The Evidence of Dress Practices”. In: Kaizer, T. (ed.), Religion, Society and Culture at Dura-Europos. YCS 38. Cambridge: Cambridge University Press, 30–56. Baskin, J. R. 1983. Pharaoh’s Counselors: Job, Jethro and Balaam in Rabbinic and Patristic Tradition. BJS 47. Chico, CA: Scholars Press. Bonfante, L. 1994. “Introduction”. In: Sebesta, J. L. and Bonfante, L. (eds.), The World of Roman Costume. Madison: The University of Wisconsin Press, 3–12. Brody, R. 1990. “Judaism in the Sasanian Empire: A Case Study in Religious Coexistence”. In: Shaked, S. and Netzer, A. (eds.), Irano-Judaica: Studies Relating to Jewish Contacts with Persian Culture throughout the Ages, Vol. 2. Jerusalem: The Ben Zvi Institute, 52– 62. Cohen, S. 1999. The Beginnings of Jewishness: Boundaries, Varieties, Uncertainties. Berkeley: University of California Press. Croom, A. 2010. Roman Clothing and Fashion. Stroud, UK: Amberley Publishing. Daryaee, T. 2010. “To Learn and to Remember from Others: Persians Visiting the DuraEuropos Synagogue”. Scripta Judaica Cracoviensia 8: 29–37. Dirven, L. 1999. The Palmyrenes of Dura-Europos: A Study of Religious Interaction in Roman Syria. RGRW 138. Leiden: Brill. Dirven, L. 2004. “Religious Competition and the Decoration of Sanctuaries. The Case of Dura-Europos”. Eastern Christian Art 1: 1–20. Dirven, L. 2016. “The Problem with Parthian Art at Dura”. In: Kaizer T. (ed.), Religion, Society and Culture at Dura-Europos. YCS 38. Cambridge: Cambridge University Press, 68–88.

Dress Codes in the Synagogue of Dura Europos?

397

Du Mesnil du Buisson, R. 1939. Les peintures de la Synagogue de Doura-Europos, 245–246 après J.-C. Rome: Pontificio Istituto Biblico. Elman, Y. 2004. “Acculturation to Elite Persian Norms and Modes of Thought in the Babylonian Jewish Community of Late Antiquity”. In: Elman, Y., Halivni, E. B., and Steinfeld, Z. A. (eds.), Jubilee Volume for David Weiss Halivni. Jerusalem: Orhot Press, 31–56. Elman, Y. 2006. “Middle Persian Culture and Babylonian Sages: Accommodation and Resistance in the Shaping of Rabbinic Legal Tradition”. In: Fonrobert, C. E. and Jaffee, M. S. (eds.), The Cambridge Companion to the Talmud and Rabbinic Literature. New York: Cambridge University Press, 165–197. Elman, Y. 2012. “Jewish Acculturation to Persian Norms at the End of the Parthian Period”. In: Wick P. and Zahnder, M. (eds.), The Parthian Empire and its Religions. Studies in the Dynamics of Religious Diversity / Das Partherreich und seine Religionen. Studien zu Dynamiken religiöser Pluralität. Gutenberg: Computus Druck, 151–161. Elsner, J. 2001. “Cultural Resistance and the Visual Image: The Case of Dura Europos”. CP 96.3: 269–304. Fine, S. 2005a. “Liturgy and the Art of the Dura Europos Synagogue”. In: Fine, S. (ed.), Art and Judaism in the Greco Roman World: Toward a New Jewish Archaeology, New York: Cambridge University Press, 172–183. Fine, S. 2005b. “Liturgy and the Art of the Dura Europos Synagogue”. In: Langer, R. and Fine, S. (eds.), Liturgy in the Life of the Synagogue: Studies in the History of Jewish Prayer. Winona Lake, ID: Eisenbrauns, 41–72. Fine, S. 2011. “Jewish Identity at the Limus: The Earliest Reception of the Dura Europos Synagogue Paintings”. In: Gruen, E. (ed.), Cultural Identity in the Ancient Mediterranean. Los Angeles: Getty Research Institute, 289–306. Fine, S. 2013. “How Do You Know a Jew When You See One? Reflections on Jewish Costume in the Roman World”. In: Greenspoon L. J. (ed.), Fashioning Jews: Clothing, Culture, and Commerce. West Lafayette: Purdue University Press, 19–27. Fine, S. 2014. “Jewish Identity at the Cusp of Empires: The Jews of Dura Europos between Rome and Persia”. In: idem (ed.), Art, History and the Historiography of Judaism in Roman Antiquity. BRLJ 34. Leiden: Brill. Flesher, P. V. M. 1998. “Conflict or Cooperation? Non-Jews in the Dura Synagogue Paintings”. In: Ovadia, A. (ed.), Hellenic and Jewish Arts: Interaction, Tradition and Renewal, Tel Aviv: Ramot Publ., 199–222. Gafni, Y. 1990. The Jews of Babylon during the Talmudic Period. Jerusalem: Shazar Institute (in Hebrew). Gates, M. 1984. “Dura-Europos: A Fortress of Syro-Mesopotamian Art”. BA 47.3: 166–181. Goldman, B. 1992. “The Dura Synagogue Costumes and Parthian Art”. In: Gutmann, J. (ed.), The Dura-Europos Synagogue: A Re-evaluation (1932–1992). 2nd ed. Atlanta, GA: Scholars Press, 53–78. Goldman, B. 1994. “Graeco-Roman Dress in Syro-Mesopotamia”. In: Sebesta, J. L. and Bonfante, L. (eds.), The World of Roman Costume. Madison: The University of Wisconsin Press, 163–181. Goldman, B. 1999. “The Iranian Element in the Dura Synagogue Murals”. In: Shaked, S. and Netzer, A. (eds.), Irano-Judaica: Studies Relating to Jewish Contacts with Persian Culture throughout the Ages, Vol. 4. Jerusalem: The Ben Zvi Institute, 298–310. Goldstein, J. A. 1995. “The Judaism of the Synagogues (Focusing on the Synagogue of DuraEuropos)”. In: Neusner, J. (ed.), Judaism in Late Antiquity, Part 2: Historical Syntheses. HdO I 17. Leiden: Brill, 109–161.

398

Katrin Kogman-Appel

Goodenough, E. R. 1966. “The Greek Garments on Jewish Heroes in the Dura Synagogue”. In: Altmann, A. (ed.), Biblical Motifs: Origins and Transformations. Cambridge, MA: Harvard University Press, 221–37. Goodenough, E. R. 1953–1968. Jewish Symbols in the Greco-Roman Period. Bollingen Series 37. 13 vols. New York: Pantheon Books. Gregoratti, L. 2016. “Dura Europos: A Greek Town of the Parthian Empire”. In: T. Kaizer (ed.), Religion, Society and Culture at Dura-Europos. YCS 38. Cambridge: Cambridge University Press, 16–29. Gruen, E. S. 2007. “Persia through the Jewish Looking-Glass”. In: Rajak, T. et al. (eds.), Jewish Perspectives on Hellenistic Rulers. Berkeley: University of California Press, 53–75. Gutmann, J. 1992. “Programmatic Painting in the Dura Synagogue”. In: idem (ed.), The Dura-Europos Synagogue: A Re-evaluation (1932–1992). 2nd ed. Atlanta, GA: Scholars Press, 137–154. Hachlili, R. 1998. Ancient Jewish Art and Archaeology in the Diaspora. HdO I 35. Leiden: Brill. Hoss, S. 2012. “Jewish Dress in the Roman Period according to the Evidence of the NonTalmudic Sources”. In: Schrenk, S., Vössing, K. and Tellenbach, M. (eds.), Kleidung und Identität in religiösen Kontexten der römischen Kaiserzeit. Regensburg: Schnell and Steiner, 65–70. James, S. 1985. “Dura-Europos and the Chronology of Syria in the 250s AD”. Chiron 5: 111– 24. James, S. 2004. The Arms, Armour, and Other Military Equipment. The Excavations at Dura Europos, Final Report 7, London: British Museum Press. Jensen, R. 1999. “The Dura Europos Synagogue, Early-Christian Art, and Religious Life in Dura Europos”. In: Fine, S. (ed.), Jews, Christians, and Polytheists in the Ancient Synagogue: Cultural Interaction during the Greco-Roman Period. London: Routledge, 174– 189. Kalmin, R. 2006. Jewish Babylonia between Persia and Roman Palestine. Oxford: Oxford University Press. Kessler, H. L. 2012. “Judaism and the Development of Byzantine Art”. In: Bonfil, R. et al. (eds.), Jews in Byzantium: Dialectics of Minority and Majority Cultures. Jerusalem Studies in Religion and Culture 14. Leiden: Brill, 455–500. Kessler, H. L. and Weitzmann, K. 1990. The Frescoes of the Dura Synagogue and Christian Art. Dumbarton Oaks Studies 28. Washington: Dumbarton Oaks Research Library and Collection. Klaver, S. 2016. “Dress and Identity in the Syrian-Mesopotamian Region: The Case of the Women of Dura Europos”. ARAM Periodical 28.1–2: 375–391. Kogman-Appel, K. 2012. A Mahzor from Worms: Art and Religion in a Medieval Jewish Community. Cambridge, MA: Harvard University Press. Kraeling, C. H. 1956. The Synagogue. The Excavations at Dura-Europos Conducted by Yale University and the French Academy of Inscriptions and Letters, Final Report 8, pt. 1. New Haven, CT: Yale University Press. Kraeling, E. G. 1940. “The Meaning of the Ezekiel Panel in the Synagogue at Dura”. BASOR 78: 12–18. Laderman, S. 1997. “A New Look at the Second Register of the West Wall in Dura Europos”. Cahiers archéologiques 45: 5–18. Larsson Lovén, L. 2015. “Roman Art: What Can it Tell us about Dress and Textiles? A Discussion on the Use of Visual Evidence as Sources for Textile Research”. In: Harlow M. and Nosch, M.-L. (eds.), Greek and Roman Textiles and Dress: An Interdisciplinary Anthology. Oxford: Oxbow Books, 260–278.

Dress Codes in the Synagogue of Dura Europos?

399

Levine, L. I. 2013. Visual Judaism im Late Antiquity: Historical Contexts of Jewish Art. New Heaven, CT: Yale University Press. Levit-Tawil, D. 1983. “The Enthroned King Ahasuerus at Dura in Light of the Iconography of Kingship in Iran”. BASOR 250: 57–78. Levit-Tawil, D. 1999. “Queen Esther at Dura: Her Imagery in Light of Third-Century C.E. Oriental Syncretism”. In: Shaked, S. and Netzer, A. (eds.), Irano-Judaica: Studies Relating to Jewish Contacts with Persian Culture throughout the Ages, Vol 4. Jerusalem, The Ben Zvi Institute, 274–297. Lewis, S. 2006. “Narrative”. In: Rudolph, C. (ed.), A Companion to Medieval Art: Romanesque and Gothic in Northern Europe. Malden: Blackwell Publishing, 86–105. Millar, F. 1998. “Dura-Europos under Parthian Rule”. In: Wiesehöfer, J. (ed.), Das Partherreich und seine Zeugnisse / The Arsacid Empire: Sources and Documentation. Historia Einzelschriften 122. Stuttgart: Steiner, 473–492. Moon, W. G. 1992. “Nudity and Narrative: Observations on the Frescoes from the Dura Synagogue”. JAAR 60.4: 587–658. Neusner, J. 1992. “Judaism at Dura-Europos”. In: Gutmann, J. (ed.), The Dura-Europos Synagogue: A Re-evaluation (1932–1992). 2nd ed. Atlanta, GA: Scholars Press, 155–192. Noy D. and Bloedhorn, H. 2004. Inscriptiones Judaicae Orientis, Vol. III: Syria and Cyprus. TSAJ 102. Tübingen: Mohr Siebeck. Patterson, L. E. 2017. “Minority Religions in the Sasanian Empire: Suppression, Integration and Relations with Rome”. In: Sauer E. (ed.), Sasanian Persia: Between Rome and the Steppes of Eurasia. Edinburgh: Edinburgh University Press, 181–198. Pfister R. and Bellinger, L. 1945. The Textiles. The Excavations at Dura Europos, Final Report 4, pt. 2. New Haven, CT: Yale University Press. Pfister, R. 1934. Textiles de Palmyre, Vol. 1. Paris: Les Éditions d’Art et d’Histoire. Pohl, W. 2006. “Telling the Difference: Signs of Ethnic Identity”. In: Noble T. F. X. (ed.), From Roman Provinces to Medieval Kingdoms. London: Routledge. Rajak, T. 2013. “The Synagogue Paintings of Dura-Europos: Triumphalism and Competition”. In: Pearce S. (ed.), The Image and its Prohibition in Jewish Antiquity. Oxford: Journal of Jewish Studies, 89–109. Revel-Neher, E. 1992. The Image of the Jew in Byzantine Art. Oxford: Pergamon Press. Revel-Neher, E. 2004. “Ezekiel Chapter 37: An Additional Key towards the Interpretation of the Pictorial Program in the Synagogue of Dura-Europos”. In: Eshel Y. et al. (eds.), “And Let Them Make Me a Sanctuary”: Synagogues from Antiquity to Present Days. Ariel: The Academic College of Judea and Samaria, 67–75 (in Hebrew). Revel-Neher, E. 2012. “‘By Means of Colors’: A Judeo-Christian Dialogue in Byzantine Iconography”. In: Bonfil, R. et al. (eds.), Jews in Byzantium: Dialectics of Minority and Majority Cultures. Jerusalem Studies in Religion and Culture 14. Leiden: Brill, 501–535. Rosenfeld B.-Z and Potchebutzky, R. 2009. “The Civilian-Military Community in the Two Phases of the Synagogue at Dura Europos: A New Approach”. Levant 41.2: 195–222. Rostovtzeff, M. I. 1935. “Dura and the Problem of Parthian Art”. YCS 5: 155–304. Rostovtzeff, M. I. 1938. Dura Europos and Its Art. Oxford: Clarendon Press. Rostovtzeff, M. I. 1943. “Res Gestae Divi Saporis and Dura”. Berytus 8: 17–60. Roussin, L. 1994. “Costume in Roman Palestine: Archaeological Remains and the Evidence from the Mishnah”. In: Sebesta, J. L. and Bonfante, L. (eds.), The World of Roman Costume. Madison: The University of Wisconsin Press, 182–190. Sabar, S. 2000. “The Purim Panel at Dura: A Socio-historical Interpretation”. In: Levine L. I. and Weiss, Z. (eds.), From Dura to Sepphoris: Studies in Jewish Art and Society in Late Antiquity. JAR Supplementary Series 40. Portsmouth: Journal of Roman Archeology, 154–163.

400

Katrin Kogman-Appel

Schenk, K. 2010. “Temple, Community, and Sacred Narrative in the Dura-Europos Synagogue”. AJSR 34.2: 195–229. Schubert, K. 1992. “Jewish Pictorial Traditions in Early Christian Arts”. In: Schreckenberg H. and Schubert, K. (eds.), Jewish Historiography and Iconography in Early and Medieval Christianity. Maastricht: Van Gorcum, 139–260. Schubert, U. 1974. Spätantikes Judentum und frühchristliche Kunst. Studia Judaica Austriaca 2. Wien: Verlag Herold. Secunda, S. 2014. The Iranian Talmud: Reading the Bavli in its Sasanian Context. Philadelphia: University of Pennsylvania Press. Shlezinger-Katsman, D. 2010. “Clothing”. In: Hezser, C. (ed.), The Oxford Handbook of Jewish Daily Life in Roman Palestine. New York: Oxford University Press, 362–381. Sommer, M. 2006. “Kulturelle Identität im Zeichen der Krise: Die Juden von Dura-Europos und das Römische Reich”. Chilufim 1: 12–31. Sommer, M. 2016. “Acculturation, Hybridity, Créolité: Mapping Cultural Diversity in Dura-Europos”. In: Kaizer, T. (ed.), Religion, Society and Culture at Dura-Europos. YCS 38. Cambridge: Cambridge University Press, 57–67. Sukenik, E. L. 1947. The Synagogue of Dura Europos and Its Murals. Jerusalem: The Bialik Institute (in Hebrew). Tawil, D. 1979. “The Purim Panel in Dura in the Light of Parthian and Sasanian Art”. JNES 38.2: 93–109. Weisman, S. 2012. “Militarism in the Wall Paintings of the Dura-Europos Synagogue: A New Perspective on Jewish Life on the Roman Frontier”. Shofar 30.3: 1–34. Weitzmann, K. 1972. “The Question of the Influence of Jewish Pictorial Sources on Old Testament Illustration”. In: Kessler, H. L. (ed.), Studies in Classical and Byzantine Manuscript Illumination. Chicago: University of Chicago Press, 76–95. Wharton, A. J. 1994. “Good and Bad Images from the Synagogue of Dura Europos: Contexts, Subtexts, Intertexts”. Art History 17.1: 1–25 [repr. in: Hoffmann, E. R. (ed.) 2007. Late Antique and Medieval Art of the Mediterranean World. Malden: Blackwell Publishing, 19–39]. Wharton, A. J. 1995. Refiguring the Post Classical City: Dura Europos, Jerash, Jerusalem and Ravenna. Cambridge: Cambridge University Press. Wischnitzer, R. 1941. “The Conception of Resurrection in the Ezekiel Panel of the Dura Synagogue”. JBL 60: 43–55. Wischnitzer, R. 1948. The Messianic Theme in the Paintings of the Dura Synagogue. Chicago: University of Chicago Press. Xeravits, G. G. 2017a. “The Message of the West Wall of the Dura Synagogue”. ZDMG 1671: 111–125. Xeravits, G. G. 2017b. “Goddesses in the Synagogue”. JSJ 48: 266–276. Xeravits, G. G. 2018. “The Reception of the Figure of David in Late Antique Synagogue Art”. In: Xeravits, G. G. and Goering, G. S. (eds.), Figures who Shape Scriptures, Scriptures that Shape Figures: Essays in Honour of B. G. Wright III. DCLS 40. Berlin: De Gruyter, 71–90.

Contributors to this Volume

Benjamin Arubas is a fellow of the Institute of Archaeology at the Hebrew University of Jerusalem. He is co-director of several excavations in Israel, inter alia, Bet Shean/Scythopolis, MasadaRoman Siege System, Tel Shalem Roman Military Camp, and the Arbel Galilean Synagogue. Kimberley Czajkowski is a Senior Lecturer in Ancient History at the University of Edinburgh. Between 2014 and 2016 she was a postdoctoral research fellow in the Cluster of Excellence Religion and Politics at Westfälische Wilhelms-Universität Münster. Lutz Doering is Professor of New Testament and Ancient Judaism at Westfälische Wilhelms-Universität Münster and heads the Institutum Judaicum Delitzschianum, having previously taught at Kings College London and Durham University. He is a project leader in the Cluster of Excellence Religion and Politics. Benedikt Eckhardt has been a Lecturer in Ancient History at the University of Edinburgh since 2018, having previously been employed by the University of Bremen. Katrin Kogman-Appel is Professor of Jewish Studies at Westfälische WilhelmsUniversität Münster. Previously, she taught at the Ben-Gurion University of the Negev (1996–2015). In 2015 she won an Alexander von Humboldt Professorship. She is a principal investigator in the Cluster of Excellence Religion and Politics. Andrew R. Krause is the Interim Academic Director and on the biblical studies faculty at ACTS Seminaries of Trinity Western University in Langley, British Columbia, Canada. He was a postdoctoral research fellow in the Cluster of Excellence Religion and Politics at Westfälische Wilhelms-Universität Münster. Ruth Langer is Professor of Jewish Studies in the Comparative Theology Area of the Theology Department at Boston College. She is also directing the university’s Center for Christian-Jewish Learning. Her research and writing focus on Jewish liturgy and Christian-Jewish relations. Uzi Leibner is Associate Professor of Classical Archaeology in the Institute of Archaeology at the Hebrew University of Jerusalem.

402

Contributors to the Volume

Clemens Leonhard is Professor of Liturgical Studies in the Faculty of Catholic Theology at Westfälische Wilhelms-Universität Münster. His research interests are ancient Jewish and Christian liturgies in their Greek and Roman contexts. He is a project leader in the Cluster of Excellence Religion and Politics. Jutta Leonhardt-Balzer is an honorary Senior Lecturer at the University of Aberdeen. Hermut Löhr is Professor of New Testament and Ancient Judaism at the Faculty of Protestant Theology, University of Bonn. He formerly taught in the universities of Jena and Münster. Judith H. Newman is Professor of Hebrew Bible and Early Judaism at Emmanuel College and in the Department for the Study of Religion at the University of Toronto and is affiliated with the Centre for Jewish Studies. Markus Öhler is Professor of New Testament Studies in the Faculty of Protestant Theology at the University of Vienna. Mechael Osband is a fellow at the Zinman Institute of Archaeology at the University of Haifa and a Lecturer of Archaeology at Tel-Hai, Ohalo and Kinneret Colleges. He is the director of the Majduliyya excavations. Jordan J. Ryan is Assistant Professor of New Testament at Wheaton College in Illinois. He is a staff member with the excavations at Tel Shimron, and has been a member of the excavations at Magdala. Previously, he was Assistant Professor of New Testament at the University of Dubuque Theological Seminary. Monika Trümper is Professor of Classical Archaeology at the Freie Universität Berlin and Spokesperson of the Berlin Graduate School of Ancient Studies on behalf of the Freie Universität. Zeev Weiss is the Eleazar L. Sukenik Professor of Classical Archaeology in the Institute of Archaeology at the Hebrew University of Jerusalem.

Index of Ancient Sources

A. The Hebrew Bible and Greek Translations of Its Books Genesis 1:1–2:4 1:1–2:3 1:9 LXX 2:2 13:15 14:17–20 15:6 17:8 22:17 28:10–19 Exodus 1:15–19 2:3 2:5 3:15 15 15:20–21 20:1–17 20:21 24:8 25:1–2 25:23–40 25:26–27 25:29 25:40 26:35 27:2 27:20 30:2–3 30:2 31:12 37:10–28 37:10 37:16 37:25 40:17

168 147 349 248 247 248 247, 248 247 248 372

386 386 386 260, 261 224, 225 221 147 270 248 168 133, 135 137 133 248 135 144 157 135 144 168 133 135 133 135 148

Leviticus 16 11:36 LXX 18:5 23:4 23:15 23:26–32 24:10–23 24:2

147 349 247 258 137 147 193 157

Numeri 4:7 15:37–41 15:38–39 27:16–17 29:7–11

133 147 384 216 147

Deuteronomium 1:1–6:3 3:26 5:6–21 6:4–9 6:9 9:19 11:13–21 11:21 14:22–29 17:14–20 17:17–18 21:23 22:8 LXX 22:12 24:4 25:3 26:8–10 26:12–15 27:15–28:69 27:26

147 266 147 147 277 248 147 273 147 147 277 247 131 384 144 242 286 147 147 247

404

Index of Ancient Sources

31:6 31:10–13 31:10–12 32 32:18a 32:35–36 33

248 146 277 224 271 248 247

Judges 15

62

1 Samuel 4:1–11 5–6 6:10–17 16 26

374 372 158 372 374

2 Samuel 7 15:32

247 270

1 Kings 6:29, 32, 35 7:48-50 8:28 17 18

136 135 265, 271 374 372, 373

2 Kings 4:17–37 5:1–14 23:2

190 190 147

1 Chronicles 24

147

8:18

147

Job 18:4a 18:4b 36:5 36:5a

271 271 266 272

Psalms 1 2 2:4 12:9 31:1–2 LXX 40 55:19 69:14 82:1 95 110:4 113–118 118:6

247 247 248 270 248 248 266, 272 272 271 248 248 257, 258 248

Proverbs 8:34–35 8:34

273 267

Isaiah 26:20 43:6 50:2 50:10a, b 55:6 56:7 LXX 66:20

248 382 272 272 270 217 270

273 248 295 295

148, 163 148 148 148 148

2 Chronicles 4:19–22 6:19 34:30

135 265 147

Jeremiah 12:14 31 44:1 46:14

Nehemiah 8 8:1–12 8:1–8 8:2 8:4–8

221 204 147, 193 288 35

Ezekiel 1 1:4–28 3 3:12–13 10:1–22



Index of Ancient Sources 11:16 37:1–14 37:26–28 40–48

139 373 169 164

Daniel 7:9

148, 158

Habakuk 2:3–4

248

2:4

247

Haggai 2:6, 21

248

Zechariah 4:1–14 4:2–3

158 134, 157 134



B. Additional Books of the Greek Scriptures 18:10

1 Esdras 9:3–55 9:38

147 204, 206

Judith 6:16 9:8

193 170

Tobit 1:8 7:12–13 13:10 (LXX 13:11)

193 193 170

1 Maccabees 1:21–24 2:34–41  3:48  4:49–51

135 193 193 135

2 Maccabees 2:4–8 2:13–15 15:9

170 146, 289 245

3 Maccabees 7:20

219

4 Maccabees 1:17–18 2:23

218 218

245

Wisdom of Solomon 8:7 218 9:8 170 Sirach 15:1 15:5 21:17 23:24 24 24:3 24:8–10 24:23 35:20–21 38:1–8 38:33 41:18 42:11 44:15

195 195 195 193 169 169 169 170 183 190 193, 195 193, 195 193, 195 193, 195

Baruch

247

Susanna 28 28 OG 41–43 60–62 62 OG

197 193 197 197 193

406

Index of Ancient Sources

C. The New Testament Matthew 4:13 4:23 5:17 6:2 6:5 7:12 9:18 9:35 10:17–18 11:23–24 11:20–24 11:23 12:2 12:9–14 12:9 12:10–12 12:14 13:54–58 13:54 22:40 23:1 23:6 23:34 Mark 1:16–20 1:15 1:21–28 1:21 1:23 1:27 1:38 1:39 2:1 2:13–18, 23 3:1–6 3:1 3:4 3:6 5:22 5:35, 36, 38 6:1–6

204 26, 189, 192, 193, 194 245 189 140, 189 245 189 192, 193, 194, 204 193 196 195 200 196 189, 195 192, 194, 196 189, 192 196 189 192, 194 245 196 196 193 197 195 26, 189, 196 192, 200, 204 192 190 189, 192 26, 192, 193, 194, 195, 204 204 197 189, 192, 197 102, 196, 200 189 196 189, 362 362 189, 196

6:1–2 6:1 6:2 6:5–6 6:6 12:28–34 13:9 13:15

204 204 192, 193 196 190 218 190, 193 131

Luke 4 4:14–15 4:15 4:16–43 4:16–30

245 242, 288 189, 192 192, 193, 194,204 26 139, 189, 192, 195, 242 35, 184, 289 167 143, 189, 192, 204 36, 189 193 195 190 189 196 192 196 189, 192 192, 204 189 196 192 189, 192 189 26 192, 200, 204, 354, 361 189 362 195 196, 200 193 196 189, 195, 197, 204

4:16–22 4:16–17 4:16 4:20 4:22–30 4:28–30 4:28 4:31–37 4:31–36 4:31–33 4:36 4:43–44 4:44 6:6–11 6:6 6:7 6:9 7:1–10 7:1–5 7:5 8:41 8:49 10:13–16 10:15 12:11–12 12:43 13:10–17



 

Index of Ancient Sources 13:10 13:14–17 13:17 13:18–21 13:44 16:16, 29, 31 21:12 24:27

192 189, 192 190, 195, 197 189 362 245 193 245

13:21, 27 15:21 16:13 18:7 18:8, 17 22:19 24:14 28:23

289 167, 184, 193, 289 223 354 362 193 245 245

John 1:45 4:46 6:25–71 6:25–59 6:31–58 6:30–33 6:41–42, 52 6:48–50, 58 6:59 6:60, 66 9:22 10:31–33 12:42 16:2 18:20

245 245 189 195, 196 192, 193, 197 221 189 195 189 192, 195, 200, 204 190, 195 190, 197 193 190, 196, 197 190, 197 189, 192, 193, 204

Romans 3:21 4

245 248

1 Corinthians 14:26

232

2 Corinthians 3:14 11:24

242 242

Acts 6:9 13 13:14–15 13:15

245 11 245 167, 184 139, 221, 242, 245, 289, 362

Hebrews 3–4 8–10 10 12:18–29 13:1–6 13:22

248 248 248 248 248 242

Revelation 2:9 3:9

354 354

D. Old Testament Pseudepigrapha Apocalypse of Moses 22:3 148 33:2–3 148 Letter of Aristeas 13–14

295

1 Enoch 14:8–25

158 148

71:5–12

148

Jubilees 1:19 4:26 8:19

148 148 148

Ps.-Philo, De Jona

247

408

Index of Ancient Sources

E. The Dead Sea Scrolls CD-A 11:17–21 11:21–12:1 1QS V 4–7 VI 3, 8 VIII 4–11 IX 3–6 XI 7–9

142 141, 142 141 220, 231 141 141 141

4Q273 (= 4QDh) 21

139

4Q405 20 ii–22 6–14

148

4Q421 13+2+8 2–3

139

4Q503

148

4Q504–506

140

4Q509

140

4QMMT

244, 245

11Q5 XXVI 9–15

148

139

4Q264a 1 4–5

139

4Q266 (= 4QDa) 5 ii 1–3

139

4Q267 (= 4QDb) 5 iii 3–5

142

4Q397 (= 4QMMTd) iv 14–21 10 244

4Q174 (= 4QFlorilegium) 247 1+2+21 i 6–7 141 4Q251 1–1 5

4Q271 (= 4QDf) 5 i 15

139

F. Philo of Alexandria De agricultura (Agr.) 44 50 79–83 79–82 80–81

216 224 221 224 224

De cherubim (Cher.) 98–106 218 De confusione linguarum (Conf.) 35 224 39 224 149 224

De vita contemplativa (Contempl.) 1 226 25 227, 245 28–29 227 27–29 226 28 227 29 226 30–33 280 30–31 139 30 226 31 227 32–33 221, 226 40–89 230 66–67 228 75–78 280



Index of Ancient Sources 80–88

225

De Decalogo (Decal.) 40 241 Quod deterius potiori insidari soleat (Det.) 114 224 De ebrietate (Ebr.) 79 94 111–118

224 224 224

In Flaccum (Flacc.) 4 41 45 47 48 49 53 73 120–123 121–122 121 122 137

306 306, 317 222, 306 222, 308 222, 308 222 223 222 306 223 224 183 221, 222, 306 306

Hypothetica (Hypoth.) 280, 281 6.9 279 7.11–14 184 7.11 218 7.12–20 279 7.12–14 220 7.12–13 139, 241 7.13–15 279 7.13 142, 193 7.14 221 Legum Allegoriae (Leg.) 2.102–103 224 3.105 224 3.162–168 221 Legatio ad Gaium (Legat.) 307 115 184, 280

132 134 137 138 139 145 148 156–157 156 157 165 191–192 210 256 311–313 311 346 371

306 223, 306 223, 306 223 306 219 223, 306 139 142, 167, 184, 193, 223, 225, 241 225 223, 306 223, 306 280 225 317 216, 306 223, 309 223

De migratione Abrahami (Migr.) 89–92 231 De vita Mosis (Mos.) 1.149 1.180 2.5 2.25–45 2.41 2.42 2.101–105 2.209–216 2.215–216 2.215 2.216 2.238 2.256–257 2.256

280 229 221, 224 229 231 221 223 135 280 184, 215, 218, 241 218 139, 142, 217, 218, 280 224 224 224

De mutatione nominum (Mut.) 115 224 253–263 221 De opificio mundi (Opif.) 128 218, 280

410

Index of Ancient Sources 1.75 1.214–215 2.38 2.126–127 2.127

De plantatione (Plant.) 46–61 224 126 224 De posteritati Caini (Post.) 67 216 104 224 106 224 121 224 167 224

2.268–271

De praemiis et poenis (Praem.) 66 241 Quod omnis probus liber sit (Prob.) 75–89 279 80–83 193 80 142, 226 81–82 139, 184 81 140, 141, 184, 217, 226, 306 82 226, 241, 280 83 226 84–88 226 88 279 De somniis (Somn.) 1.35 224 1.61–67 229

224 229 224 216 139, 142, 184, 219, 232, 241, 306 224

De Specialibus Legibus (Spec.) 280 1.113 229 1.66–67 229 1.193 225, 228 1.224 225 2.61–63 217 2.61–62 241 2.62 139, 142, 218, 219, 226, 241, 280 2.63 218 2.148 225, 229 2.199 224 3.131–132 229 3.171 228, 229 De virtutibus (Virt.) 48 224 72–75 224 135 224

G. Flavius Josephus Antiquitates 3.139–150 4.209–211 4.210 4.212 7.148–150 7.162 8.45–48 10.58–63 11.154–158 11.154–155 12.5 12.7–8 12.45–50

135 147 193 231 146 146 190 147 204, 206 147 134 295 295

12.250 12.318 13.62–73 13.65–66 13.66 14.72 14.172 14.213–216 14.235 14.256–258 14.259–261 14.259 14.260–261 14.260

135 135 182 300 182 133, 135 351 316 316, 347 223 347 351 228 141, 229



Index of Ancient Sources 14.227 16.42–43 16.43 16.164 17.200–201 18.15 19.281 20.173

141 184 139, 142, 193 142, 241 204 198 295 179

Bellum Judaicum 1.122 1.152 2.1–5 2.129 2.266 2.284–292 2.284–285 2.285–292 2.285–286 2.289–292 2.289 2.292 2.294–295 2.320–324 2.424 2.487 4.336 4.407–408 6.128 7.44–45 7.143–144 7.361 7.421–432 7.429

204 133, 135 204 141 178, 179 179 180 178 179 281 140 193 204 204 181 295 204 140, 178 178 140 140, 178 181 182 183

Contra Apionem 1.6–2.144 1.38–41 1.60 1.198–199 1.208–209 2.10–11 2.11 2.12–40 2.35 2.42 2.106 2.145–286 2.151–189 2.170–171 2.175–178 2.175 2.176–178 2.176–177 2.176 2.188 2.195–196 2.282

182 245 184 133 183 182 182 183 295 295 135 182 183 194 281 139, 142, 167, 183, 193, 215, 221 194 193 281 184 184 219

Vita 1–12 276–303 277–291 277–279 277 280

186 193, 196, 207 139 281 26 26



H. Rabbinic Literature and Targumim Mishnah Berakhot 1–5 1:1 2:4 4–5 4:3–4:4a 4:4b-6 5:5

259 268 260 257 260 260 190

6–9 7:3

259 260

Pe’ah 4:3

384

Bikkurim 1:4 3:6

261 260 144

412

Index of Ancient Sources

‘Erubin 3:5

204, 205

Yoma 7:1–3 7:1

Sanhedrin 10:1 11:2

190 204

147 35, 204

Makkot 3:1–15 3:12

193 193

Roš Haššanah 3:7 4:1–2 4:3 4:5–6 4:7 4:8

257 257 258 257 257 257

’Abot 1:1 3:10

274 261, 267

Ta‘anit 2 4:1–2 4:2–3

258 259 147

Tamid 2:1 4:3 5:1

259 161 147 147

Kelim 15:6

146

Megillah 3:1–3 3:1 3:2 3:3 3:4–6 4:1–2 4:1 4:3 4:4 4:5–9 4:5–6 4:5 4:6 4:10

255 196 255, 257 256 256 256 256 260 256, 257 256 256 261 256 256 256

Tosefta Berakhot 1:3 2:4 2:6–8 2:9 3:5 3:6 3:15 3:18

260 261 260 260 261 265 37 260

Pisa 10:8

257

Sukkah 2:10 4:6

257, 258 35, 305

Ta‘anit 2 2:4 3:3

258 259 259

Megillah 2:3 2:5 2:18

261 284 207, 256



Nedarim 5:5

193, 196, 255

Soah 3:12–13 3:12 7:1 7:7–8 7:7 7:8

257, 258 258 288 204, 288 147 35, 147





Index of Ancient Sources 3:10 3:20 3:21

258 193 37, 258

Soah 6:3

262

Baba mei‘a 5:26

384

Bekhorot 6:13

384

Talmud Yerushalmi Berakhot 4:4, 8b 5:1, 8d–9a 5:1, 9a 7:3, 11c 9:1, 13a 9:3, 12d

263, 265, 270 263 267 260 264 264

Bikkurim 3:3, 65c

258

Sukkah 3:14, 54a 5:1, 55a–b 5:1, 55b

258 220 35

Ta‘anit 4:2, 68a

146

Megillah 3:1, 73d 3:3–4, 74a 3:4, 74a 4:1, 74d

31, 32 205 32 284, 290

Talmud Bavli Berakhot 4:4, 8b 4b 5:1, 8d 5b–8b 6a

266 270 274 270 264, 271 265

7b 28b 31a 34a 34b 50a

272 257, 263 265 259 190 260

Sukkah 41b 43a 44a

258 258 258

Ta‘anit 5a–6a

266

Megillah 17b–18a 26a 27b 29a

257 32 255 139

Soah 11b–12b 22a

386 267

Sanhedrin 59a 90b 101b

285 240 386

Extracanonical Treatises ’Abot de-Rabbi Nathan 268, 274 Soferim 10:6 18:10

260 260

Midrashim Mekhilta de Rabbi Jischmael Wa-yassa 1, 155 259 Mekhilta de Rabbi Shimon bar Yoai 15:25, 103 259

414 Sifre Numeri § 135

Index of Ancient Sources 266

Sifre Deuteronomium § 29 31, 37 § 306 260 Exodus Rabbah 1:20

386

Tanuma Shemot 62b

386

Targumim Targum Ps.-Jonathan Exod 1:15 386

 

I. Papyri, Inscriptions, and Coins AE 1955: 175 1955: 177 1955: 184 1963: 17 1983: 731 1987: 496 1987: 893 1991: 822 1991: 823 2001: 854 2005: 1325 2009: 103 2010: 592 2011: 203 2014: 1059

317 317 317 317 317 317 325 330, 331 331 317 325 327 330 317 326

BGU IV 1137

298

CGRN 184

357

CIG 3056 3069 3071

349 355 355

CIIP I9

10, 130, 139, 142, 175, 185, 207, 231, 232, 241, 256, 354

CIJ II 751 II 1404

II 1432 II 1433 II 1440 II 1441 II 1442 II 1443 II 1447

346 10, 130, 139, 142, 175, 185, 193, 207, 231, 232, 241, 256, 354 305 305 302 298, 302 302 302 298, 301, 305

CIL II 1167 III 7060 V 7811 VI 85 VI 441 VI 1872 VI 4416 VI 29691 VI 33875 VII 11 IX 2213 X 1642–1643 X 1647 X 3699–3700 X 5198 XII 1929 XIV 10 XIV 37 XIV 62

317 317 317 317 317 317 317 317 327 318 317 317 317 317 317 318 317 328 328



Index of Ancient Sources XIV 70 XIV 168 XIV 169 XIV 256 XIV 286 XIV 311–312 XIV 4573

328 317 317 317 328 328 317

CIRB 70 71 985 1128

91 91 91 91

CJZC 72

354

CPJ I 129 I 134 I 138 II 432 III 473 III 1532a

303 140 303, 304, 316 300, 302 297 302

GRA I 66 I 72 I 75 I 86 II 113 II 114 II 117 II 141

362 362 362 362, 364 358 364 353 355

I.ApamBith 33 34 35 ID 1519 1520 1910 2111 2230 2294 2306

353, 355 353, 354, 355 348, 353, 355, 358 94 93, 94 103 103 101 103 105, 106

2328 2329 2330–2332 2330 2331–2333 2331 2332 2333 2348 2355 2369 2370 2378 2384 2390 2403 2410 2532 2593 2628

99, 106, 107 95, 97, 98, 99, 105, 106, 107 106, 107 100, 105, 106 119 101, 102, 104, 106 102, 104 103 103 103 103 103 96 103 103 103 103 105 103 101

I.Delta I 11 I 446 II 28

298 298 298, 299, 301

I.Ephesos 2212 2446

342 342

I.Erythrai 21

349

I.Fayum I9 II 204 III 205

298 298 298

IG IX,2 259 XI 1274 XI 1359 XII,3 330

357 103 103 357

IGRR I 782

364



416

Index of Ancient Sources

I 1077 I 1095 IV 144

298, 305 298 363

I.HierapolisJudeich 218

341

I.Iasos 612

349

IJO I I Ach47 I Ach58 I Ach60–63 I Ach60 I Ach61 I Ach62 I Ach63 I Ach64 I Ach65 I Ach66–67 I Ach66 I Ach67 I Ach70, 71 I Mac1 I Mac15 I Pan3 I Thr1–2 II 1 II 5 II 5a II 14 II 15–16 II 20 II 25 II 26 II 27 II 28 II 30 II 32 II 33 II 36 II 37–39 II 37 II 40 II 41 II 43

326 350 354 105 101 102 100 99 103 95, 96 316 88, 90, 94 91 105 140, 323 324 325 326 361 344 343 362 321 351 362 316 361 362 362 322 322 321, 343, 361 321 343 323 322 322, 342, 361

II 44 II 46 II 49 II 53 II 148 II 154 II 155 II 157 II 158 II 160 II 168 II 171 II 172 II 174 II 175 II 178 II 179 II 187 II 189 II 191 II 192 II 193 II 196 II 198 II 199 II 200 II 204 II 205 II 214 II 217 II 219 II 220 II 231 II 241 II 255 II 256 III Syr34

322 361 349 345 362 322, 341, 342 342 322, 341, 342, 343 322, 342 362 321, 358, 359 364 364 364 364 364 343 343 351 350 351 350 321 351 350 350 350 322, 351 362 361 352 353, 362 343 343 361 362 354

I.Kios 6

363

IK.Rhod.Peraia 303

349

I.Kyzikos I 211, 291

341

 

Index of Ancient Sources ILBulg 16, 18

325

ILOP 108 129

327 317

I.Mylasa 941 942 943

357 356 357

InscrIt IX.5 1075 X.5 16 X.5 17 X.5 77 X.5 110 X.5 120 X.5 134 X.5 135 X.5 145 X.5 157 X.5 180 X.5 183 X.5 189–190 X.5 189 X.5 190 X.5 200 X.5 202 X.5 208 X.5 209 X.5 211 X.5 216 X.5 221 X.5 226 X.5 231 X.5 223 X.5 225 X.5 237 X.5 238 X.5 239 X.5 241 X.5 266 X.5 274 X.5 275 X.5 279 X.5 280

330 329 330, 332 330, 331 330, 331 330 330 329 330, 331 330 330, 331 330 330 331 331 330 330 330, 331 330, 331 329, 331 330 330, 331 330 330 317, 329, 331 331 330 331 330, 331 330, 331 329, 330 330 331 330 330, 331

X.5 282 X.5 283 X.5 289 X.5 294 X.5 297 X.5 299 X.5 808 X.5 823 X.5 875 X.5 932 X.5 985 X.5 996 X.5 1065 X.5 1070 X.5 1211

330 330, 331 330, 331 330 330 330, 331 330 332 330 329, 331 329, 331 330 330, 332 330, 332 330

I.Perge 23 67 268

363 363 348

I.Perinthos 49 59

355, 364 355

I.Sardis VII,1 8 VII,1 17 VII,1 22

363 345, 347, 348, 349, 354, 355, 358 348

I.Smyrna 204

342

I.Stobi 3 7 100 119

324 324 324 324

JIGRE 9 13 17 18 20 22 24

302 140, 305 140, 305 140 298, 305 298, 301, 305, 357 140, 302 140, 298, 302

418

Index of Ancient Sources

25 26 27 28 117 125 126

140, 302 298, 299, 301 140, 302 140 140, 302 140, 303 303

JIWE I4 I5 I 13 I 14 I 18 I 23 I 181

332 329 329 327 327 327 327 333



OGIS 326 573 737 742 748

355 299 298 305 349

MAMA VI 153 VI 263 VI 265

361 361 360, 161

P.Lond. III 1177 VII 2193

300 298, 357

P.Oxy. IX 1205 XVIII 2192 LXXXIII 5364

297 287 321

P.Polit.Iud.

308

RIU V 1051 V 1056 V 1062 V 1084 V 1115

325 325 325 325 325

RPC I 3171 I 3173–3176

360 360

SB I 4211 XXII 15460

298 298, 357

SEG VIII 529 XXIX 1205 XLIV 556 XLVI 1529 XLVI 1530 LVII 1157 LXII 439 LXII 600 LXIII 974

357 348 344 348 348 363 324 344 317

TAM II 175 III,1 765 IV,1 234 IV,1 236 IV,1 238 IV,1 239 V 1539

317 341 342 342 342 342 353



J. Ancient Christian Literature Augustin, De civitate Dei 6.11 282 Eusebius, Praeparatio Evangelica 8.7.12–13 241 8.7.13–15 279

Justin Martyr, First Apology 67.3 286





Index of Ancient Sources

K. Greek and Latin Authors Asconius Pedianus, Pro Cornelio 75 317 Cicero De natura deorum 2.159

224

Pro Flacco 69

225

Codex Justinianus 1.9.1

319

Herodotus, Historiae 2.111 183 Ovid, Ars Amatoria 1.76 223 Philostratus, Vita Apollonii 3.39 190 4.20 190 Pliny the Younger Epistulae 10.33–34

317

10.92–93

324

Panegyricus 54.3

317

Sophocles, Antigone 256 181 775 181 Suetonius, Tiberius 32.2 184 Tacitus, Historiae 5.4.3

282

Ulpianus frg. Vat. 233

317

Vitruvius, De Architectura 1.7.1–2 179 1.7.1 180 3.1.3 180 3.2.1 180 7.1.2 180



Index of Modern Authors

Adler, Yonatan 11, 19, 143, 149 Agócs, Nándor 325, 333 Aimé-Giron, Noël 295, 310 Alexander, Philip S. 193, 254, 274, 305, 310 Alföldy, Géza 325, 333 Allison, Dale C. 190, 194, 208 Ameling, Walter 322, 323, 333, 335, 340, 341, 342, 343, 345, 346, 348, 350, 351, 352, 353, 359, 360, 361, 362, 363, 364, 365, 366 Amir, Roni 45, 65, 71, 78 Amiran, David H. K. 64, 65 Amit, David 11, 12, 19, 36, 37, 39, 129, 131, 139, 145, 149 Anderson, Gary A. 142, 149 Arieh, Eliahu J. 64, 65 Arnaoutoglou, Ilias N. 306, 310, 318, 333 Arubas, Benjamin 16, 47, 56, 57, 61, 65, 74, 199, 220 Ascough, Richard S. 192, 315, 333, 362, 364, 366 Atrash, Walid 56, 58, 66, 68 Attridge, Harold W. 21, 68, 211 Audiat, Jean 117, 121 Auwers, Jean-Marie 239, 250 Avemarie, Friedrich 250, 343, 350, 365, 367 Aviam, Mordechai 13, 15, 19, 27, 32, 33, 34, 35, 36, 38, 39, 130, 131, 132, 133, 134, 135, 136, 137, 138, 143, 144, 145, 149, 150, 156, 157, 158, 159, 160, 161, 162, 163, 164, 171, 172, 199, 202, 205, 206, 208 Avigad, Nahman 40, 57, 66 Avi-Yonah, Michael 57, 66, 236, 396 Avshalom-Gorni, Dina 10, 15, 19, 26, 33, 39, 49, 66, 76, 78, 127, 128, 130, 131, 143, 146, 150, 156, 164, 172, 198, 201, 202, 208

Bacher, Wilhelm 250 Baird, Jennifer A. 369, 377, 379, 381, 382, 383, 396 Balamoshev, Constantinos 321, 334 Balduinus, Franciscus B. 314, 334 Balty, J. 9 Barclay, John M. G. 182, 183, 187, 295, 310, 314, 334 Bardtke, Hans 314, 334 Barrett, Caitlín Eilís 119, 121 Bartels, Jens 326, 334 Baskin, Judith R. 386, 396 Baslez, Marie-Françoise 316, 334 Bauckham, Richard 15, 19, 33, 39, 128, 131, 132, 133, 134, 135, 136, 137, 138, 139, 143, 144, 149, 150, 156, 157, 158, 160, 161, 162, 163, 164, 172, 202, 208, 236 Becker, Hans-Jürgen 268, 274 Belayche, Nicole 104, 105, 106, 121, 368 Bellinger, Louisa 377, 384, 399 Ben David, Chaim 71, 72, 73, 78 Bendlin, Andreas 317, 334 Ben Ephraim, Y. 73, 78 Berlin, Andrea M. 205, 208 Bernabò, Massimo 62, 69 Bernier, Jonathan 190, 192, 193, 197, 198, 200, 204, 208 Bij de Vaate, Alice J. 340, 363, 365 Bijovsky, Gabriela Ingrid 50, 67 Bilde, Per 181, 187 Billerbeck, Paul 220, 233 Binder, Donald D. 10, 11, 15, 19, 22, 33, 34, 35, 39, 85, 86, 88, 89, 90, 91, 96, 99, 100, 101, 102, 104, 108, 111, 121, 122, 128, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 143, 144, 150, 152, 156, 157, 158, 159, 160, 161, 162, 164, 172, 181, 183, 187, 192, 193, 195, 196, 198, 199, 200, 201, 202, 204, 206, 208, 209,

Index of Modern Authors 211, 215, 232, 236, 299, 301, 310, 346, 359, 361, 365, 368 Blenkinsopp, Joseph 168, 172 Bloedhorn, Hanswulf 44, 66, 121, 335, 340, 366, 367, 380, 383, 399 Böhm, Martina 89, 90, 121 Bond, Helen 190, 209, 330, 334 Bonfante, Larissa 369, 396, 397, 399 Borgen, Peder 184, 187, 217, 218, 221, 233 Boring, M. Eugene 204, 209 Boscolo, Filippo 330, 334 Botermann, Helga 347, 365 Bouet, Alain 119, 121 Boustan, Ra‘anan 9, 19, 62, 66 Bowman, Alan K. 296, 310 Braslavsky, Joseph 46, 66 Breccia, Evaristo 298, 310 Brehm, Oliver 353, 365 Brighton, Mark Andrew 181, 187 Britt, Karen 9, 19, 62, 66 Brody, Robert 382, 396 Brooke, George J. 172, 247, 250 Brooten, Bernadette J. 191, 209, 343, 365 Bruneau, Philippe 82, 83, 85, 88, 89, 90, 91, 92, 93, 94, 95, 96, 98, 101, 102, 103, 105, 106, 107, 108, 109, 111, 113, 115, 116, 117, 119, 120, 121 Brünnow, Rolf-Ernst 55, 56, 57, 58, 66 Buckler, William H. 345, 346, 347, 348, 365 Butler, Howard C. 58, 66 Butz, Patricia A. 93, 121 Campbell, Jonathan G. 244, 250 Capponi, Livia 296, 310 Carbon, Jan-Mathieu 356, 357, 365 Carruthers, Mary 166 Catto, Stephen K. 10, 19, 86, 88, 90, 91, 96, 104, 121, 143, 150, 191, 209, 347, 359, 363, 365 Cerfaux, Lucien 222, 223, 233 Chachel-Laureys, Rachel 21, 210 Chancey, Mark A. 143, 152, 192, 210 Chaniotis, Angelos 9, 19, 324 Charlesworth, James H. 142, 149, 150, 171, 211 Chatzidakis, Panagiotis 119, 121



Chazon, Esther G. 140, 148, 150, 236, 275, 276 Chiai, Gian Franco 359, 363, 365 Clarke, Ernest George 396 Claußen, Carsten 10, 11, 13, 19, 240, 250, 345, 347, 352, 359, 362, 363, 365 Clemente, Guido 328, 332, 334 Cohen, Barak S. 258, 262, 274 Cohen, Naomi G. 218, 221, 231, 233, 247, 250 Cohen, Shaye J. D. 184, 187, 261, 275, 323, 334, 377, 396 Cohn, Leopold 224, 233 Collar, Anna 88, 90, 104, 105, 107, 121 Collart, Paul 57, 66 Collingwood, Robin George 17, 190, 209, 211 Colson, Francis Henry 279, 280, 291 Conder, Claude Reignier 46, 66 Corbo, Virgilio C. 10, 19, 26, 39, 44, 66, 130, 198, 209 Corsten, Thomas 343, 353, 354, 355, 365 Cotton, Hanna 11, 19, 130, 131, 150, 365 Cowey, James M. S. 308, 310 Cowley, Arthur E. 295, 310 Cracco Ruggini, Lellia 328, 331, 334 Cresswell, Tim 180, 187 Crook, Zeba 195, 209 Croom, Alexandra 374, 396 Crowfoot, John W. 56, 66 Czajkowski, Kimberley 18, 219, 226, 296, 307, 308, 310, 321, 381 Damati, Emmanuel 53, 67, 69 Dar, Shim‘on 57, 66, 69 Daryaee, Touraj 382, 383, 396 de Bakker, Mathieu 177, 187 de Jonge, Henk Jan 239, 250 De Luca, Stefano 29, 39, 57, 66, 128, 131, 134, 143, 144, 150, 164, 172, 201 De Ricci, Seymour 298, 310 De Rossi, Giovanni Battista 314, 334 Deines, Roland 196, 209 Delorme, Jean J. 113, 115, 120, 121 Deonna, Waldemar 96, 101, 102, 103, 119, 120, 121 Der Nersessian, Sirarpie 62, 66 Dion, Paul-Eugene 300, 301, 303, 310

422

Index of Modern Authors

Dirven, Lucinda 376, 377, 380, 389, 396 Dochhorn, Jan 148, 150 Doering, Lutz 16, 139, 141, 150, 167, 172, 202, 222, 226, 227, 229, 233 Domaszewski, Alfred von 55, 56, 57, 58, 66 Dorl-Klingenschmid, Claudia 349, 365 Dothan, Moshe 50, 51, 66 Dray, Yehoshua 78 Du Mesnil du Buisson, Robert 372, 373, 397 Ducat, Jean 82, 93, 106, 109, 116, 120, 121 Dunbabin, Katherine M. D. 9 Eck, Werner 329, 334 Eckhardt, Benedikt 18, 232, 297, 298, 299, 306, 307, 310, 311, 315, 318, 319, 321, 322, 324, 334 Edelmann-Singer, Babett 359, 360, 365 Edwards, Douglas R. 27, 40, 151 Edwards, James R. 209 Ehrenkrook, Jason von 179, 180, 188 Ehrlich, Uri 37, 40 Elbogen, Ismar 220, 233, 260, 275 Elman, Yaakov 381, 382, 397 Elsner, Jaś 389, 397 Engel, Helmut 197, 209 Epstein, Isidore 72, 78, 374, 396 Erickson-Gini, Tali 31, 40 Erlich, A. 9 Eve, Eric 190, 209 Fai, Stephen 178, 179, 187 Falk, Daniel K. 140, 148, 150, 231, 234 Fauconnier, Bram 318, 335 Feldman, Louis H. 147, 224, 225, 233, 343, 365 Fine, Steven 15, 19, 35, 40, 130, 131, 132, 133, 134, 135, 137, 138, 139, 140, 141, 142, 143, 150, 152, 153, 162, 163, 172, 188, 202, 209, 237, 243, 250, 276, 288, 291, 292, 350, 352, 365, 366, 369, 370, 375, 376, 377, 379, 380, 382, 383, 384, 389, 397, 398 Finkelberg, Margalit 239, 250 Fiore, Benjamin 219, 234 Fischer, Moshe 44, 67

Fleischer, Ezra 253, 256, 275, 276 Flesher, Paul V. M. 10, 19, 389, 397 Foerster, Gideon 44, 46, 55, 67, 198, 209 Foucart, Paul François 355, 366 Foucault, Michel 177, 187 Fraisse, Philippe 87, 117, 119, 121 Fraser, Peter M. 343, 365 Frey, Jean-Baptiste 235, 339, 346, 365, 366 Friedman, Shamma 261, 262, 275 Fuglseth, K. 217, 233 Fülep, Ferenc 326, 335 Gabrielsen, Vincent 311, 313, 335 Gafni, Isaiah M. 254, 275, 381, 397 Ganor, Amir 12, 22, 27, 30, 31, 41, 130, 153 García Iglesias, Luis 333, 335 Garza Diaz Barriga, Andrea 127, 153, 201, 205, 212 Gates, Marie-Henriette 397 Gauthier, Philippe 93, 121 Geiger, Bernhard 383 Gendelman, Peter 13, 19, 133, 134, 151 Gersht, Rivka 133, 134, 151 Geva, Hillel 29, 40, 151 Goldberg, Arnold 250 Goldman, Bernhard 373, 374, 376, 377, 378, 379, 381, 384, 385, 397 Goldstein, Jonathan A. 380, 397 Gonen, Ilana 78 Goodacre, Mark 190, 209 Goodenough, Erwin R. 370, 371, 372, 374, 375, 376, 380, 384, 388, 396, 398 Goodman, Martin 184, 187, 274, 310, 367 Gordon, Richard 9, 222, 234 Grabbe, Lester L. 295, 311 Gregoratti, Leonardo 381, 398 Gregori, Gian Luca 331, 335 Grey, Matthew J. 62, 66, 67 Griffiths, John Gwyn 295, 302, 311 Groag, Edmund 360, 366 Groten, Andreas 317, 319, 335 Grüll, Tibor 326, 336 Gruen, Erich S. 295, 311, 315, 335, 359, 366, 381, 397, 398 Guijarro, Santiago 190, 209

Index of Modern Authors Gutmann, Joseph 216, 234, 370, 371, 379, 396, 397, 398, 399 Gutman(n), Shmarya 10, 49, 72, 78 Habas, Li-hi 132, 151 Haber, Susan 11, 19 Hachlili, Rachel 10, 11, 15, 20, 37, 40, 66, 130, 131, 132, 133, 134, 135, 138, 143, 144, 145, 151, 161, 162, 163, 172, 191, 192, 202, 209, 241, 250, 276, 363, 366, 372, 376, 398 Haentzschelius, Johann Gottfried 314, 335 Hakohen, A. 257, 275 Halbwachs, Maurice 166, 172 Halfmann, Helmut 360, 366 Hallmannsecker, Martin 323, 335 Hanfmann, George M. A. 348, 366, 368 Haran, Menahem 36, 38, 40, 134 Har-Even, Benjamin 13, 20, 27, 28, 29, 39, 40, 130, 151, 199, 209 Harland, Philip A. 89, 90, 94, 314, 315, 321, 323, 335, 336, 341, 343, 344, 351, 353, 354, 355, 356, 358, 359, 360, 361, 362, 363, 364, 366 Harlow, Mary 369, 398 Harter-Uibopuu, Kaja 325, 335 Hatch, Edwin 216, 234 Hauptman, Judith 261, 263, 275 Hayward, C. T. Robert 146, 151, 172 Hegermann, Harald 216, 234 Heinrici, C. F. Georg 314, 335 Heirman, Jo 177, 187 Hemelrijk, Emily 332, 335 Hempel, Charlotte 142, 151 Hengel, Martin 196, 209, 216, 217, 234, 236, 240, 250 Herrmann, Peter 346, 348, 366 Hersey, George L. 181, 187 Hezser, Catherine 40, 69, 152, 205, 259, 275, 278, 279, 291, 400 Hill, Stephen 55, 67 Hirschfeld, Yizhar 12, 20, 29, 40, 345, 348 Honigman, Sylvie 296, 300, 308, 311 Hooker, Morna 190, 209 Horbury, William 217, 219, 220, 221, 222, 223, 234, 299, 366



Horsfall, Nicholas 283, 291 Horsley, Gregory H. R. 361, 362, 366 Horsley, Richard A. 201, 209 Hoss, Stefanie 375, 377, 398 Hunt, Arthur Surridge 297, 311 Hüttenmeister, Frowald Gil 181, 187 Hyman, Arthur B. Dov 396 Ilan, Zvi 46, 53, 67, 69 Israel, Yigal 31, 40 Jacobs, Louis 262, 275 James, Simon 398 Jensen, Robin 370, 377, 398 Johnson, William A. 66, 278, 282, 283, 284, 285, 286, 287, 288, 291 Jones, Henry Stuart 235, 348, 349, 367 Jördens, Andrea 296, 311 Juster, Jean 216, 223, 234 Kaiser, Otto 224, 234 Kalman, Ya‘akov 21, 210 Kalmin, Richard 381, 382, 398 Kartveit, Magnar 89, 90, 91, 93, 95, 121 Kasher, Aryeh 21, 180, 187 Kayser, François 316, 335 Kazen, Thomas 193, 209 Kee, Howard C. 10, 20, 129, 130, 139, 151, 196, 201, 210, 216, 234 Keith, Chris 190, 209, 210, 212 Kenyon, Kathleen M. 56, 66 Kershner, Isabel 15, 20, 131, 136, 138, 139, 151, 162, 172 Kessler, Herbert L. 370, 371, 398, 400 Kister, Menachem 268, 275 Kitchener, Horatio H. 46, 66 Klaver, Sanne 369, 398 Kloppenborg Verbin, John S. 11, 20, 25, 40, 130, 151, 179, 180, 187, 191, 197, 199, 210, 315, 362, 364, 366 Knibb, Michael A. 197, 210 Kogman-Appel, Katrin 18, 138, 384, 398 Kohl, Heinrich 13, 20, 43, 44, 55, 57, 58, 59, 64, 67, 71, 78 Kolb, Anne 326, 334 Korner, Ralph 192, 195, 210 Kraabel, Alf T. 45, 53, 68, 108, 216, 234, 323, 335, 346, 348, 359, 366, 368

424

Index of Modern Authors

Kraeling, Carl H. 35, 40, 56, 67, 369, 372, 373, 375, 376, 380, 383, 398 Kraeling, Emil G. 295, 311, 373, 398 Kraemer, David 275 Kraemer, Ross S. 226, 234, 340, 343, 366 Kratz, Reinhard G. 244, 250 Krause, Andrew R. 14, 17, 20, 141, 151, 166, 172, 176, 178, 182, 184, 187, 192, 200, 210, 215, 227, 228, 229, 230, 234, 301, 311 Krauss, Samuel 216, 311 Kreeb, Martin 93, 96, 121 Krencker, Daniel 55, 58, 67 Kruse, Thomas 300, 308, 311, 316, 335 Laderman, Shulamit 370, 398 Langellotti, Micaela 319, 336 Langer, Ruth 17, 18, 231, 253, 257, 263, 275, 290, 291, 292, 397 Larsson Lovén, Lena 369, 398 Last, Richard 311, 314, 315, 321, 322, 336 Laubry, Nicolas 330, 336 Le Donne, Anthony 190, 209, 210, 212 Le Guen, Brigitte 355, 356, 367 Lefebvre, Henri 176, 177, 187 LeFebvre, Michael 193, 210 Leibner, Uzi 9, 14, 16, 20, 21, 37, 40, 43, 46, 50, 54, 60, 67, 68, 69, 131, 132, 151, 152 Lena, Anna 29, 39, 134, 144, 150, 164, 201 Leone, Francesco 132, 151 Leonhard, Clemens 17, 227, 292, 297, 311, 321, 334 Leonhardt-Balzer, Jutta 17, 215, 217, 221, 224, 225, 229, 230, 233, 235, 247, 295, 296 Lerer, Isaac Nathan 396 Levenson, Jon D. 168, 172 Levine, David 254, 275 Levine, Lee I. 9, 10, 11, 12, 13, 15, 18, 20, 21, 25, 26, 31, 32, 35, 37, 40, 53, 63, 67, 68, 86, 108, 122, 129, 130, 139, 140, 147, 148, 151, 180, 183, 184, 185, 186, 187, 191, 192, 193, 197, 200, 207, 209, 212, 215, 216, 235, 253, 254, 255, 256, 257, 262, 263, 264, 269, 275, 288, 296,

298, 303, 305, 311, 344, 359, 367, 370, 399 Levinskaya, Irina A. 217, 235, 300, 311 (Levit-)Tawil, Dalia 373, 376, 378, 382, 383, 398, 399, 400 Lewis, Ioan Myrddin 190, 210 Lewis, Suzanne 385, 399 Liddell, Henry George 216, 217, 235, 348, 349, 367 Lieberman, Saul 258, 259, 261, 275, 276, 284, 292, 384, 396 Lieu, Judith M. 344, 346, 347, 367 Lifshitz, Baruch 222, 232, 235, 298, 311 Ligt, Luuk de 317, 334 Lim, Timothy H. 250 Liu, Jinyu 317, 336 Loffreda, Stanislao 44, 67, 198, 209 Löhr, Hermut 17, 221, 248, 250 Lüderitz, Gert 361, 367 Lyttelton, Margaret 58, 67 Ma, John 93, 94, 122 Ma‘oz, Zvi Uri 11, 21, 44, 55, 57, 64, 68, 71, 78, 199, 210 Magen, Yiaq 10, 20, 50, 67, 130, 142, 151, 198, 210 Magness, Jodi 9, 12, 14, 20, 21, 22, 45, 62, 66, 67, 68, 142, 151, 153, 191, 199, 205, 210, 346, 347, 367 Maier, Johann 220, 231, 232, 235 Maisler, Benjamin 57, 68 Mandel, Paul 261, 276 Mann, Jacob 215, 218, 219, 235 Manns, Frédéric 220, 221, 231, 235 Maresch, Klaus 308, 310 Martin, Dale B. 21, 68, 211 Martin, Matthew J. 217, 225, 233, 235 Martyn, J. Louis 198 Mason, Steve 147, 179, 180, 187 Matassa, Lidia D. 11, 14, 16, 21, 87, 88, 89, 90, 96, 98, 99, 100, 101, 102, 103, 104, 105, 106, 108, 109, 110, 111, 112, 114, 117, 118, 120, 122, 129, 151 Mazor, Gabriel 56, 66, 68 Mazur, Belle D. 85, 104, 106, 108, 111, 114, 122 McCollough, C. Thomas 13, 21, 64, 68, 130, 152, 199, 210

Index of Modern Authors McDonald, Lee Martin 239, 250 McEwan, Indra Kagis 179, 180, 187 McKay, Heather A. 14, 21, 217, 218, 220, 222, 223, 227, 232, 235 McLaren, James S. 181, 188 McLean, B. Hudson 93, 108, 122 Meir, Dafna 71, 78 Meir, Eran 71, 78 Mélèze-Modrzejewski, Joseph 210, 220, 222, 235, 295, 298, 311 Mendelson, Alan 229, 236 Meshorer, Ya‘aqov 132, 133, 135, 137, 152 Meyer, Ben F. 192, 203, 210 Meyers, Carol L. 45, 50, 55, 57, 68 Meyers, Eric M. 13, 14, 21, 44, 45, 50, 53, 55, 57, 68, 143, 152, 192, 210 Millar, Fergus 381, 399 Miller, Shulamit 9, 21, 60, 68, 253, 276 Minns, Denis 286, 292 Miquel Pericás, Esther 190, 210 Miranda Elena 351, 367 Mitchell, Stephen 104, 105, 122 Monson, Andrew 296, 311 Moon, Warren G. 371, 374, 376, 378, 389, 399 Moore, Carey A. 197, 210, 219, 220 Moretti, Jean-Charles 87, 117, 119, 121, 122 Mosser, Carl 193, 210 Najar, Arfan 10, 15, 19, 26, 33, 39, 49, 66, 127, 128, 130, 143, 146, 150, 156, 164, 172, 198, 201, 202, 208 Netzer, Amnon 396, 397, 399 Netzer, Ehud 10, 11, 12, 21, 25, 26, 27, 31, 40, 128, 129, 130, 152, 198, 210 Neusner, Jacob 20, 21, 22, 67, 68, 210, 257, 263, 276, 381, 397, 399 Newman, Judith H. 16, 132, 138, 164, 169, 170, 172, 202, 229, 254, 276 Neyrey, Jerome H. 222, 236 Nielsen, Inge 86, 90, 122, 123 Nigdelis, Pantelis M. 324, 325, 336, 344, 367 Noam, Vered 139, 152 Nock, Arthur Darby 222, 236 Norelli, Enrico 239, 250



Nosch, Marie-Louise B. 369, 398 Notley, R. Steven 132, 134, 135, 152 Noy, David 121, 219, 222, 299, 332, 335, 340, 350, 361, 362, 366, 367, 368, 380, 383, 399 Oakeshott, Michael 190, 211 Oehler, Johann 339, 341, 359, 367 Oesterley, William Oscar Emil 216, 236 Öhler, Markus 18, 297, 334, 339, 360, 367 Oliphant, Laurence 71, 78 Olsson, Birger 10, 11, 15, 22, 86, 88, 89, 90, 91, 96, 99, 100, 101, 102, 104, 122, 130, 152, 192, 196, 198, 199, 209, 211, 212, 215, 232, 236, 299, 336, 359, 361, 368 On(n), Alexander 10, 22, 50, 68, 130, 145, 152, 198, 211 Osband, Mechael 16, 75, 199, 220 Oster, Richard E. 216, 234, 236 Panayotov, Alexander 121, 335, 340, 344, 366, 367 Parker, Holt N. 284 Parvis, Paul 286, 292 Patrich, Joseph 39, 41, 66, 132, 149, 152, 153 Patterson, Lee E. 399 Peels, Saskia 356, 357, 365 Peleg-Barkat, Orit 34, 39, 40, 41, 76, 78, 136, 137, 149, 152, 153 Perrot, Charles 221, 227, 236, 242, 250 Perry, Jonathan S. 330, 332, 336 Petersen, David L. 134, 152 Pfister, René 377, 378, 384, 399 Pirenne-Delforge, Vinciane 365 Pitre, Brant J. 203 Pitt-Rivers, Julian 195, 211 Plassart, André 81, 82, 96, 97, 98, 99, 100, 101, 102, 103, 105, 106, 107, 108, 112, 116, 118, 119, 120, 122 Pohl, Walter 369, 377, 399 Pomeranz, Jonathan A. 263, 276 Porath, Yosef 37, 40, 56, 68 Porten, Bezalel 295, 311 Potchebutzky, Rivka 379, 399 Price, Jonathan J. 130, 340, 364, 367

426

Index of Modern Authors

Pucci Ben Zeev, Miriam 223, 228, 236, 318, 336, 347, 348, 367 Qimron, Elisha 139, 152 Rahmani, Levi Yiaq 132, 133, 136, 137, 138, 145, 152, 161, 172 Rajak, Tessa 222, 236, 318, 336, 343, 344, 346, 359, 360, 361, 362, 367, 368, 379, 389, 398, 399 Ramsay, William W. 359, 360, 368 Rapuano, Yehuda 12, 22, 198, 211 Rathbone, Dominic 296, 310, 311 Ratner, Baer 32, 40 Raviv, Dvir 132, 152 Rebillard, Éric 342, 368 Redpath, Henry Adeney 216, 234 Reich, Ronny 129, 132, 134, 152, 153 Reif, Stefan C. 183, 184, 188, 254, 256, 276 Reinach, Salomon 359, 368 Revel-Neher, Elisabeth 373, 374, 375, 384, 399 Richardson, Peter 108, 130, 131, 152, 186, 188, 300, 314, 329, 336 Richter, Gisela M. A. 38, 40 Riesner, Rainer 216, 231, 232, 236 Ritmeyer, Leen 204, 211 Ritter, Bradley 295, 311 Ritti, Tullia 321, 336 Robert, Louis 357 Robinson, David Moore 345, 346, 347, 348, 365 Rohde, Dorothea 322, 327, 328, 336 Rosenfeld, Ben-Zion 379, 399 Rosenson, I. 257, 275 Rostovtzeff, Michael I. 376, 382, 399 Roth-Gerson, Lea 25, 41 Rough, Robert H. 44, 68 Roussel, Pierre 116, 120, 122 Roussin, Lucille 384, 399 Rozenberg, Silvia 50, 68 Rubenstein, Jeffrey L. 262, 276 Runesson, Anders 10, 11, 12, 14, 15, 22, 86, 88, 89, 90, 91, 96, 99, 100, 101, 102, 104, 122, 130, 146, 147, 148, 152, 162, 172, 184, 186, 188, 191, 192, 193, 194, 196, 197, 198, 199, 200, 206, 211, 215,

232, 233, 236, 299, 301, 311, 327, 336, 359, 361, 368 Rutgers, Leonard V. 66, 301, 311 Ryan, Jordan J. 10, 12, 13, 17, 22, 128, 146, 190, 191, 192, 195, 199, 200, 203, 204, 205, 211 Sabar, Shalom 376, 379, 380, 382, 383, 399 Safrai, Shmuel 191, 216, 220, 221, 231, 236, 253, 276, 367 Salzmann, Jorg Christian 220, 221, 231, 236 Sanders, Ed Parish 181, 188, 191, 211, 215, 218, 220, 229, 231, 232, 236 Sanders, James A. 239, 250 Sandmel, Samuel 219, 220, 231, 236 Sänger, Patrick 300, 311 Santero Santurino, José María 333, 336 Sanz-Rincón, Rosaura 127, 153, 201, 205, 212 Sappir, Iaq 37, 41 Sarason, Richard S. 220, 236, 263, 275 Schenk, Kära L. 375, 400 Schiffman, Lawrence H. 33, 41, 132, 134, 135, 139, 153 Schofield, Alison 142, 153, 177, 188 Schrage, Wolfgang 216, 236 Schröder, Bernd 219, 236 Schubert, Kurt 370, 400 Schubert, Ursula 386, 400 Schumacher, Gottlieb 72, 78 Schumann, Gotthilff A. 314, 336 Schürer, Emil 216, 220, 230, 236, 314, 326, 336, 359, 368 Schwartz, Daniel R. 13, 21, 22, 68, 225, 235, 236 Schwartz, Seth 31, 41, 45, 68, 298, 311 Scott, Robert 216, 217, 235, 348, 349, 367 Seager, Andrew R. 346, 368 Secunda, Shai 381, 382, 400 Shanks, Hershel 26, 41, 85, 122, 138, 153 Shiloni, Yiaq 396 Shinan, Avigdor 396 Shlezinger-Katsman, Dafna 384, 400 Siegert, Folker 247, 251

Index of Modern Authors Sirkis, Orna 10, 20, 50, 67, 130, 142, 151, 198, 210 Skarsten, R. 217, 233 Soja, Edward W. 176, 177, 188 Solomon, Avi 141, 142, 153 Sommer, Michael 389, 400 Spigel, Chad 66, 86, 87, 122, 145, 153, 202, 212 Stacey, David 199, 212 Stebnicka, Krystyna 321, 336, 350, 359, 361, 368 Stein, Markus 104, 122 Steinhauer, Julietta 355, 368 Stemberger, Günter 246, 251, 278, 279, 280, 281, 292 Stern, Menahem 282, 292 Steudel, Annette 141, 142, 153 Stewart-Sykes, Alistair 251 Steyn, Gert J. 248, 251 Stiebel, Guy 129, 152 Stock, Brian 167, 173 Strange, James F. 14, 22, 26, 41, 45, 50, 53, 55, 57, 68, 156, 200, 205, 206, 212, 276 Strange, James R. 14, 20, 21, 22, 39, 64, 68, 69, 150, 152, 199, 210, 212 Stroumsa, Guy G. 234, 239, 250 Struve, Vasily Vasilievich 91, 121 Sukenik, Eleazar Lipa 56, 66, 108, 370, 400, 402 Sussman, Varda 138, 153 Syon, Danny 12, 22, 41, 49, 69, 79, 152, 153, 198 Szabó, Ádám 326, 336 Tabory, Joseph 150, 153, 259, 276 Talgam, Rina 9, 57, 63, 65, 69, 131, 133, 136, 138, 139, 162, 173 Taylor, Joan E. 226, 236 Tcherikover, Victor 295, 306, 311 Thackeray, Henry St. John 247, 251, 281, 292 Theissen, Gerd 203, 212 Theobald, Christoph 239, 251 Thomsen, Christian A. 311, 313, 335 Thyen, Hartwig 224, 237, 246, 247, 251 Tondriau, Jules 222, 223, 233 Torgü, Hagit 13, 19



Totelin, Laurence M. V. 330, 336 Tov, Emanuel 36, 41, 146, 153 Tran, Nicolas 329, 330, 336 Trebilco, Paul R. 235, 340, 344, 346, 347, 350, 359, 360, 361, 363, 368 Trümper, Monika 14, 16, 81, 84, 86, 87, 88, 93, 95, 97, 98, 102, 103, 104, 106, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 120, 122, 123 Tsafrir, Yoram 44, 69 Tuan, Yi-Fu 17, 165, 171, 173 Turcotte, T. 64, 65 Turnheim, Yehudit 57, 69 Tweed, Thomas A. 173 Twelftree, Graham H. 192, 212 Tzionit, Yoav 10, 20, 50, 67, 130, 142, 151, 198, 210 Ulrich, Eugene C. 244, 251 Urman, Dan 71, 78 van Bremen, Riet 360, 368 van Bynkershoek, Cornelius 317, 336 van der Horst, Pieter W. 14, 22, 129, 140, 142, 153, 217, 237, 339, 342, 351, 359, 368 van Henten, Jan Willem 340, 363, 365 van Nijf, Onno 123, 342, 367 Verboven, Koenraad 319, 337 Vermes, Geza 191, 212, 236 Veyne, Paul 222, 237 Vicari, Jacques 57, 66 Watts, James W. 193, 212 Watzinger, Carl 13, 20, 43, 44, 55, 57, 58, 59, 64, 67, 69 Weber, Thomas M. 56, 69 Weigand, Edmund 57, 69 Weisman, Stefanie 379, 400 Weiss, Zeev 9, 13, 16, 21, 22, 30, 31, 34, 37, 39, 40, 41, 63, 69, 130, 131, 132, 145, 199, 220, 235, 257, 262, 275, 276, 399 Weitzmann, Kurt 62, 69, 370, 371, 385, 398, 400 Weksler-Bdolah, Shlomit 10, 22, 50, 68, 130, 145, 152, 198, 211 Wendland, Paul 224, 233

428

Index of Modern Authors

Wharton, Annabel Jane 370, 400 White, L. Michael 14, 22, 85, 86, 96, 108, 109, 123, 327, 328, 337, 359, 360, 363, 368 Wiedergut, Karin 325, 335 Williams, Margaret H. 333, 337 Winston, David 218, 237 Winter, Bruce W. 222, 237 Winter, Dagmar 203, 212 Wischmeyer, Oda 142, 153 Wischnitzer-Bernstein, Rachel 373, 400 Wise, Michael O. 142, 278, 281, 283, 288, 289, 292 Witmer, Amanda 190, 212 Wulzinger, Karl 58, 69 Xeravits, Géza G. 370, 371, 373, 374, 376, 400 Yadin, Yigael 198, 212, 243, 251 Yarden, Leon 132, 153



Yavor, Zvi 10, 12, 22, 26, 29, 31, 41, 49, 69, 71, 76, 79, 129, 130, 145, 146, 152, 153 Yeivin, Zeev 45, 53, 57, 59, 69 Zangenberg, Jürgen 9, 21, 22, 38, 41, 68, 143, 145, 153, 156, 173, 185, 188, 211 Zapata Meza, Marcela 127, 129, 152, 153, 201, 205, 212 Zarmakoupi, Mantha 93, 119, 123 Zeitlin, Solomon 183, 188 Zerubavel, Eviatar 167, 173 Zevi, Fausto 328, 337 Zissu, Boaz 12, 22, 27, 30, 31, 39, 41, 130, 149, 153 Zschietzschmann, Willy 55, 58, 67 Zuckerman, Constantine 307, 312 Zuckermandel, Moses Samuel 384, 396 Zunz, Leopold 276



Index of Subjects

Akmoneia 321, 340, 358–364 Alexandria 35, 104, 215, 219–220, 226, 232, 246–247, 281, 287, 295–300, 302, 304–309, 357, Aposynagōgos 197–198, 208 Arbel (synagogue) 55, 59 Archisynagōgos 35, 144, 185–186, 195, 197, 204, 298, 305, 327, 339, 355, 358– 359, 361–364 Ark (tevah) 32, 37–38, 53, 138, 158, 257– 262, 327  see also Torah shrine Arsacid(s) 376, 380–381, 384 Asia Minor 156, 321–323, 339–364 Associations 12, 18, 86, 89, 93–94, 114, 117, 184, 197, 216, 232, 278, 297–299, 306–309, 313–333, 339, 341–343, 348, 357, 360, 362–364

Chariot throne, see Merkavah Chios 344 Chiton 371, 387 Chorazin, see Korazim Columns 10, 12–13, 26, 29–31, 33, 38, 49–51, 54–55, 57, 59–60, 62, 73, 75, 128–129, 135, 137–138, 156, 158, 199– 200, 206 Community 15, 17, 18, 25, 32, 36, 43, 46, 85, 88, 93, 95, 105, 107, 108, 118, 141– 142, 148, 165–167, 171, 175–181, 197– 198, 208, 216, 222, 228, 230, 239, 243, 254–261, 264, 266, 270, 272, 285, 288, 295, 297, 303, 307, 314, 321, 323, 326, 342–345, 347, 350–354, 356, 358, 360– 364, 379–380  local c. 46, 118, 146, 165, 181 Costume 371–379, 383, 385, 388, 390

Babel, Tower of 62–63 Baraita/baraitot 240, 258, 262–270 Bar‘am 55, 58–59 Beit ha-kneset 241 Beit ha-midrash 241 Bema 28, 31–32, 35, 37, 53, 63, 143, 191 Benches 10, 12–13, 25–31, 33, 49–62, 74– 77, 82, 84, 87, 98–99, 109, 114–117, 129–130, 146, 156, 200–201, 220, 243, 278 Benefactor(s) 93–95, 140, 189, 222–224, 231, 313, 328, 332–333, 350, 360, 363 Boundaries 73, 175, 176, 179, 189, 315, 350 Byzantine period 9, 14, 16, 31, 33, 45, 51, 57, 62, 64–65, 71–73, 82, 139, 145, 162, 199

Delos 14, 16, 81–120, 320 Diaspora 9, 12, 14–15, 25, 37, 86, 139– 141, 197, 216–217 , 223, 225–233, 241– 243, 289, 304, 355, 360 Donor (donorship) 32, 88, 96, 100, 104– 106, 108, 117, 222, 315, 321–323, 343– 345, 352, 354, 360–361, 363, 380 Dura Europos 18, 33, 35, 138, 321, 370– 395

Canon, canonization 239–250 Capernaum 11, 14, 25–26, 44–45, 55, 57, 59, 64, 130, 138, 158, 193–200, 204– 205

Eagle imagery 56–57, 59 Egypt 139–140, 169–170, 215, 222–223, 287, 289, 295–310, 316, 319, 321, 372, 374, 391 Elder(s) 185, 220, 279, 321, 345, 352 Elites, Roman 277, 285 Essenes 12, 17, 139–142, 184, 216, 217, 225–227, 231, 279, 306 Ethnicity 177, 182, 283, 299, 315–318, 332, 339, 348, 369, 371, 375, 377–378, 381 Exodus (story) 222, 374

430

Index of Subjects

Fleischer, Ezra 253, 256 Galen 283, 285, 289 Galilee 11, 13, 25–26, 43–50, 129, 137, 140, 149, 192–194, 199–201, 205–207, 223, 242  see also Synagogues, Galilean Gamla 9–12, 25–26, 29, 31, 49–50, 71, 76–77, 129–130, 136, 145–146, 198, 201 Gellius 282, 284–285, 289–290 Godfearer(s) 319, 349, 360–361 Golan 11, 13, 16, 71–77, 129, 199 Gush alav 44–45, 50, 55, 57, 59, 65 Haftarah 221, 242, 247 Halakhah 31, 37, 142, 273, 287 Hellenistic period 9, 15, 18, 47, 54, 82, 93, 98, 104–105, 107–108, 118, 130, 161, 163, 170, 193, 198–200, 205, 219, 313, 316–317, 320, 352, 375, 377, 381  see also Ptolemaic (era, dynasty) Herodium 9–10, 25, 130, 198 Heterotopia 177, 185 Hierapolis 321–322, 343, 350–351 High Priest, Jewish 35, 133–134, 147, 167, 229, 241, 285, 288, 369, 374, 388 Himation 371, 373, 376, 378, 384, 387 Honour (and shame) 89, 92, 94, 195–197, 224, 242, 256, 261 Honorary practice 93–95 orvat ‘Ammudim 55, 57, 59, 63 orvat Etri 12, 16, 27, 30–33, 130 orvat Kur 9, 38, 135, 145 uqoq 9, 62–63 Instruction 17, 139–140, 168, 170–171, 183–184, 218–221, 231, 241, 245 Islamic period 73 Italy 82, 326–333 Iulia Severa 18, 339–364 Jericho 129–130, 198–199 Josephus, Flavius 14, 16, 26, 133–135, 139–142, 146–147, 159, 161–162, 166– 167, 175–187, 190, 193–198, 205–207, 215–216, 219, 221, 223, 227–230, 232,

241, 245, 277–283, 290, 295, 298, 300– 301, 316, 322, 347–348, 362 Khirbet Diab 13, 16, 27–33, 130, 199 Khirbet e-uwani 13, 27, 130, 199 Khirbet Qana 13, 27, 64, 130, 199–200 Khirbet Qumran 12, 142 Khirbet Shema‘ 44–45, 53 Khirbet Wadi amam 9, 14, 16, 43–65, 131 Korazim/Chorazin 12, 45, 53, 57, 59, 145, 200, 206 Late antiquity 31, 37–38, 200, 202, 253– 254, 319, 321–322, 330, 354, 370–371, 377, 380, 384 Law and Prophets 242, 244–245, 249 Law, Roman 194, 314–316, 324, 329 Legal status 18, 307, 309, 314, 328 Levine, Lee I. 13, 18, 31, 86, 108, 140, 186, 207, 262, 269, 288 Levite(s) 147, 259 Library 146, 169, 227, 287–289 Literacy 17, 278, 285, 288, 290 Liturgy 16, 27, 33–38, 184, 220, 231, 253– 263, 290 Ma‘amad/ma‘amadot 144, 147–148, 259 Macedonia 140, 319, 323–325, 362 Magdala 10–11, 16, 25–27, 29, 33–38, 49– 50, 76–77, 127–149, 155–171, 198– 206, 229, 233, 291 Majduliyya 13, 16, 71–77, 199 Marcus Aurelius 282, 287, 289, 324 Masada 9–11, 25, 129–130, 141–142, 169–170, 181, 198, 243, 288 Meals, dining 12, 33, 145, 228, 230, 232, 257, 260, 274, 313 Meeting(s) 10–12, 84, 93–94, 114–115, 117–118, 142, 146, 167, 179, 196, 200, 202, 216–219, 221, 225–227, 230–232, 243, 266, 278, 281, 283, 288, 298, 300, 303, 313, 327, 328, 344, 348, 354, 356– 358, 380 Meiron 14, 44–45, 55, 58–59 Menorah 14, 33–34, 36, 131–135, 138, 143, 155–159, 163–164, 185, 202, 339, 363

Index of Subjects Merkavah 138, 143, 148–149, 158, 163– 164, 167–168, 170 Meroth 53, 63 Miqweh/miqwa’ot (stepped pool[s]) 10– 11, 29–30, 96, 111, 129, 142–143, 186, 201, 243 Modi‘in (Umm el-‘Umdan) 10, 25, 50, 130, 198 Monumental architecture 43, 47, 57, 60, 62, 71, 77, 86, 199, 385 Mosaic(s) 9, 26, 33, 54, 60–63, 76–77, 127–128, 131, 136, 169, 179, 185, 333, 344 Nabratein 14, 44–45, 65 Narrative 62, 164, 168, 175, 177–178, 196–198, 204, 220, 224, 259, 269, 285, 301, 371, 373, 374, 385–389 Nazareth 11–12, 25, 35, 192, 195, 204, 206, 242 Nicomedia 341–342 Ostia 12, 14, 317, 326–329 Otium 282 Pallium/palla 371, 373, 374, 386 Petichah 17, 221, 239–250 Pharisees 194, 196–198, 241, 247 Philadelphia 298, 349–350, 353 Philo of Alexandria 14, 17, 135, 139–142, 161–162, 167, 183–184, 193, 215–233, 241, 246–247, 279–287, 290, 295–310, 316–317, 348, 362 Philosopher(s) 218, 282, 286 Politics 7, 383 Poseidoniasts of Berytos 93, 112 Prayer 14–15, 32, 34, 36–38, 86, 89–90, 96, 140, 143, 164, 183–184, 217, 220– 232, 253–272, 280, 305  orientation of 36–37, 183, 260 Priest(s), Jewish 132–133, 139, 146–148, 157, 163, 185–186, 220, 257–259, 279, 322, 369, 374, 387 Priest(s), Non-Jewish 313, 330–331, 348, 353–354, 356–357, 360–361, 372–373, 382 Proseuchē 89, 91, 104, 108, 117, 139–140, 182, 215–233, 300–305



Psalms 220, 224, 227, 231, 244, 247, 257– 58 Pseudo-baraitot 262, 269 Ptolemaic (era, dynasty) 104, 134, 140, 222–223, 295–297, 299–304, 307–309, 316 Qiryat Sefer (Khirbet Badd ‘Isa) 10–11, 25, 50, 129–130, 142, 198 Qumran (caves, scrolls) 36, 140, 170–171, 220, 231, 242–243, 246  Q. yaad (community) 140–141, 226, 254, 260, 314  see also Khirbet Qumran Rabbis, the 15, 17, 31, 139, 162, 207, 254– 274, 278–279, 283–290, 381–382 Rhetoric 175–187, 266, 281, 284 Ritual(s) 11, 15–17, 25–28, 37, 94–95, 111, 147, 166, 176, 184–185, 239–249, 257–259, 269, 277–291, 370–371, 377, 384, 387 Roman period 9, 13–16, 18, 25, 31, 33, 37–38, 43, 45–51, 53–57, 60, 64, 71–77, 82, 130–131, 138–140, 145, 149, 161, 163, 170, 193, 196, 199–202, 205, 207, 254, 264, 267–269, 304, 306–309, 314, 318, 320, 370, 377–378 Rome, Roman 17–18, 26–27, 38, 46, 50, 81–82, 97, 107, 134, 139, 141, 146, 157, 166, 178–182, 185, 194, 219, 223, 225230, 243, 277–279, 281–290, 314, 316– 324, 326–329, 333, 347–348, 360–361, 369–371, 374–381, 383–385, 389–390 Sabbath 32–33, 35, 139, 142, 148, 168, 179, 182, 184, 189, 193, 202–204, 216– 227, 230–332, 241–242, 279–283, 290, 318 Sacred 25, 51, 89, 104, 139–143, 155, 162, 165, 171, 175, 177–181, 184, 222, 225, 228, 239, 241, 245, 246, 259–260, 279, 305, 323, 348 Sacrifice(s) 35, 94, 107, 140–142, 147– 148, 179, 181, 228–229, 232–233, 264, 286, 301, 306, 307, 347, 357, 397 Sadducees 194 Safrai, Shmuel 253

432

Index of Subjects

Samaritan(s) 33, 83, 85–88, 90, 92, 95, 98, 105–108, 117–119, 135, 316 Samos 344–345 Samson 62 Sardis 228–230, 316, 345–351, 354–355, 358, 363 Sasanian(s) 374–389 Scrolls 11, 32–38, 129–130, 141–146, 149, 164, 170, 243, 255, 286, 288–289 Seat(ing), special 35, 82, 84, 87, 112, 114, 116–117, 144–145, 164, 343–344, 374, 376 Shikhin 14, 64, 199–200, 205 Side (town) 328, 344, 352–353, 362 Side room(s) 26, 36, 38, 49, 52–53, 129, 131, 142, 145–146, 243 Silent reading 279, 283 Spain 333 Stele/Stelae 82, 88, 90–95, 98, 105, 107, 119–120, 353 Stobi 140, 323–324 Study room 49, 131 Synagogue(s) passim  association s. 11, 196–197  functions of 130, 167, 192, 200, 206, 253, 256, 320  Galilean s. 13, 16, 26, 43–64, 131, 155, 171, 195, 205  public s. 11–12, 14–15, 18, 47–52, 73–77, 86, 93–94, 129, 131, 133, 155–156, 165, 167, 178, 192–208, 263, 317, 319–322, 346  readings in 17, 35, 185, 221, 249, 278, 281, 287, 289  Second Temple(-period) s. 9–14, 16, 27, 29, 32, 37, 49, 51, 129–130, 139–140, 143, 145, 148–149, 215, 242, 253, 283

 service in 141, 220–221, 230–231, 242, 245, 256, 264, 290  terminology of 108, 142, 218– 219, 241, 296, 297–301, 306 Tanakh, structure of 240, 246, 249 Tation 321, 343–344, 361 Tax, Jewish 319, 323 Tel Rekhesh 13, 27, 29, 33, 130, 199–200 Temple, Jerusalem 7, 9, 15–17, 25, 34–38, 127–149, 155–171, 200, 202, 204, 206– 207, 222, 225, 228–229, 231, 233, 253, 257–260, 264, 273, 288–289, 307, 375, 381, 387 Temple, heavenly 148–149, 248 Theodotus Inscription 10, 11, 17, 25, 130–131, 139, 142, 175, 185–186, 207, 231–232, 241, 256, 289, 321, 354 Theos Hypsistos 86, 89, 99–108, 118 Therapeutae 17, 139, 216, 221, 225–227, 228, 230–232, 245 Thirdspace 176–177, 185 Tiberias 26, 37, 46, 50–51, 56, 139, 196, 205, 207, 273 Torah-reading 16, 35–38, 144–148, 206, 220–226, 232, 242, 253, 256, 258–260, 267, 277–291 Torah shrine 11, 13, 14, 191, 288, 369– 370, 372, 374, 388  see also Ark (tevah) Transgression 178–180, 193, 268 Triglia 353, 358 Women, in synagogues 101, 191, 221, 225–231, 330–331, 347, 353–355 Zeus 104–108, 346–349, 353–357,  Z. Hypsistos 104–108