1In this paper, I propose some general remarks about non-rabbinic Judaism, rabbinization, and what Simon C. Mimouni calls ‘synagogal Judaism’.1 My historical scope encompasses the periods of Late Antiquity and the Early Middle Ages in both Palestine and the Diaspora.

2The notion of rabbinization is at the heart of the present book, but it remains difficult to speak of without a grasp of non-rabbinic Judaism. Rabbinization is the process by which non-rabbinic Jews became rabbinic, but speaking of rabbinization may be problematic, because the very basis of this process, namely, the nature of non-rabbinic Jews or non-rabbinic Judaism, is far from clear: How can we identify these Jews? What is the nature of their Judaism?

3The main features of Christian non-rabbinic Judaism are well known.2 Here, however, I am discussing a Judaism that was both non-rabbinic and non-Christian—some third kind of Judaism.

4Non-rabbinic Judaism is a woolly, difficult notion. As Shaye J. D. Cohen admits in the conclusion of his seminal article on the epigraphical rabbis: “What was the nature of this non-rabbinic Judaism in the Diaspora and the synagogue, I do not know.”3 According to Cohen, the epigraphical evidence demonstrates the reality of non-rabbinic Judaism. However, from the same evidence, the nature of this Judaism remains uncertain. If the main part of Jewish society, both in Palestine and in the Diaspora, belongs to non-rabbinic Judaism, an enigma arises: What was the process by which all non-rabbinic Jews became rabbinic?

5In the last part of my paper, I will argue that the notion of ‘synagogal Judaism’, presented for the first time in a book by Mimouni (published in 2012),4 may help facilitate a better understanding of non-rabbinic Judaism and the process of rabbinization.

1.0. Non-Rabbinic Judaism: The Old Model and the New Model

6In the following pages, I shall discuss two historiographical models, which I propose calling ‘the old model’ and ‘the new model’, even if the terms ‘old’ and ‘new’ may be misleading. Indeed, the old model (mainstream Judaism passing from priests to rabbis after 70 CE) remains attractive to several scholars, particularly in Israel, while the new model results from over forty years of research.5

7One of the greatest figures of the old model is the Israeli historian Gedaliah Alon. In this model, post-70 Judaism is clearly dominated by the rabbis and is on the whole identical to rabbinic Judaism.6 Erwin Goodenough was the forerunner of the new model. Its most recent and radical formulations are currently found in the work of Seth Schwartz and Hayim Lapin.

8The new model includes two main arguments: the authority argument and the plurality argument.7 The authority argument claims that the rabbis are not the leading group of Jewish society and thus they do not control the Judaism of their time. They are a peripheral or even marginal minority. What scholars who advocate this approach intend by ‘authority’ is not always unambiguous. Authority could be defined as religious, economic, political, or legal. Official or semi-official authority is not the same thing as informal authority, such as influence or prestige. Authority with power to sanction differs from voluntarily accepted authority. The diversity argument emphasizes the persistent plurality of post-70 Judaism: the rabbis are only one of its components.

9Most ‘new’ scholars have tried to combine the authority argument with the diversity argument, and this is no simple task.8 According to the authority argument, the rabbis are not dominant and, consequently, another group necessarily leads Jewish society. The diversity argument is not so concerned with the authority issue. Jewish society may exist without a single leading group: the main Jewish authority would be local and vary from place to place.

10What evidence supports the new model?9 Some scholars emphasize the contrast between rabbinic and non-rabbinic sources. According to rabbinic literature, the rabbis would be the centre of Jewish society. In non-rabbinic sources (inscriptions, archaeological data, Christian literature, Roman legal codes), even when they are Jewish, the rabbis are marginal or simply absent. Consequently, the old model would have made the significant mistake of taking rabbinic literature literally and of failing to understand the ideological nature of this literature, which does not reflect historical reality objectively.10

11I think that the contrast between the sources is not so sharp. One also finds in rabbinic literature itself substantial evidence that supports the new model, as can be seen in the following examples:

1. The small number of rabbis mentioned. Even if we consider the rabbis an elite group among rabbinic Jews, this point remains puzzling.

2. The title rabbi is used only after the year 70. This fact suggests that the rabbis considered themselves a new group, and such a group must always struggle to achieve a prominent place in society.

3. The rabbis are concentrated mainly in Palestine and Babylonia.

4. As argued by Catherine Hezser, rabbinic authority does not appear in Talmudic literature as something official or formal. It does not work without the agreement of the other Jews.11 Schwartz states clearly: “The modern debate over whether the rabbis or someone else led the Jews after the destruction is rendered moot by the failure of rabbinic literature itself to claim a leadership role for its protagonists.”12

5. The existence of tensions between rabbinic norms and other norms or between rabbis and a number of groups, like priests, ʿamme ha-aretz, or several categories of heretics (minim).13

6. The performance of pagan rites and the persistent attraction of idolatry in many Jewish communities.14

12The new model is also supported by the fact that a number of texts have ambiguous relationships with rabbinic Judaism. This group includes some apocalyptic and pseudepigraphical texts, the Targumim, the Hekhalot literature, and piyyutim. All these sources are ambiguous because they show different degrees of both rabbinic and non-rabbinic (or perhaps, in some cases, anti-rabbinic) features. According to several scholars, they are better understood against a priestly background than a rabbinic one.15

13A last piece of evidence may strengthen the new model, that of Jewish magic, which forms an important, but often overlooked, aspect of Jewish culture in Late Antiquity, as pointed out by Gideon Bohak in his seminal work on the issue.16 In particular, Bohak emphasizes that what rabbinic literature teaches us on magical practices does not match what we know from Jewish magical sources themselves.17

14Thus, in the old model, Judaism was identified with rabbinic Judaism. Has the new model succeeded in drawing a new picture of ancient Judaism, corresponding more closely to historical reality? Regarding this question, historiography often remains elusive and the answers, when they do exist, diverge to a significant degree.

15What specific name could be given to non-rabbinic Judaism? From a terminological viewpoint, should one speak of ‘non-rabbinic Judaism’, ‘non-rabbinic Jews’, ‘peripheral rabbis’, etc.? Is non-rabbinic Judaism some sort of unity, or is it irreducibly plural?

16For Annette Yoshiko Reed and Michael Satlow, non-rabbinic Judaism, like Judaism itself, has no unity. Each variety of Judaism should be studied within its regional and local setting.18 According to the radical and nominalist view of Satlow, there is no Judaism, but only Jews and local communities.19

17When we consider the distinction between the authority argument and the diversity argument within the new model, it is clear that proponents of the former are more inclined than proponents of the latter to admit the possibility of a single, non-rabbinic Judaism. Indeed, authority and unity often go hand-in-hand. An authoritative Judaism, whether non-rabbinic or rabbinic, could not exist without some minimal unity.

18If we admit that non-rabbinic Judaism, in spite of its diversity, was unified in some way, is it possible to describe its main features? How was it organized? What were its institutions or structures? Did it only comprise the Jewish masses, or did it also include specific elites? If it did, who were these elites? What were its rituals, its theological conceptions, its means of expression? What was its relationship with rabbinic Judaism?

19Regarding this last question, it is possible to emphasize points of tension and conflict between both types of Judaism. Conversely, they were also separated by porous frontiers, permitting a close, if not complementary, relationship between them. Relevant to this issue are the various phrases which Daniel Boyarin uses to describe ‘binitarian Judaism’ and its logos theology.20 The phrase ‘non-rabbinic Judaism’ or the emphasis on the difference between the ‘synagogue’ and the ‘house of study’ suggests a strong contrast with rabbinic Judaism, while the phrase ‘para-rabbinic Judaism’ indicates a greater proximity between both Judaisms.

20Does non-rabbinic Judaism have the same informal network structure that Hezser accords to the rabbinic movement?21 Should we say that one of the main differences between non-rabbinic Palestinian and diasporic Judaism lies in the fact that the former coexisted with rabbinic Judaism in the same limited space, which was not the case for the latter? If we admit diversity in each group, it becomes possible to conceive of non-rabbinic Jews who would be closer to rabbinic Jews than to other non-rabbinic Jews.

21One may finally ask to what extent rabbinic and non-rabbinic Judaism differ in their relationship to Hellenization and Romanization. It is not so evident that non-rabbinic Judaism would be more Hellenized and Romanized. Indeed, if we follow Saul Lieberman and the great number of scholars who agree with his perspective, rabbinic literature testifies to a high degree of Hellenization. Moreover, a recent book by Lapin argues that Palestinian rabbis could be considered ‘Romans’.22

2.0. The New Model: Five Examples

2.1. Annette Yoshiko Reed and Michael Satlow: Diversity Only

22An approach that emphasizes Jewish diversity is reflected in the work of at least two scholars: Annette Yoshiko Reed and Michael Satlow. In a ground-breaking study, Reed discusses non-rabbinic varieties of post-70 Judaism.23 She begins by presenting rabbinic evidence, particularly texts dealing with ʿamme ha-aretz, Sadducees, charismatic priests, and minim. She then explores three other bodies of texts: Hekhalot literature, the magical sources, and what I prefer to call the ‘synagogal corpus’ (Targumim, piyyutim).24 She lastly considers Christian sources, whose portrait of the Jews is not necessarily consistent with rabbinic evidence. It does not follow that the Christian texts are badly informed and wrong. Rather, they may indeed be relevant and give some evidence of non-rabbinic Jews, their beliefs, and practices.25 On a methodological level, Reed’s discussion remains very empirical. It describes varieties of non-rabbinic Judaism that are reflected within various groups of texts. The varieties are never incorporated into wider categories or groups. It is not so much the scholar’s care and caution that explains this approach, as it is her desire to highlight the local scale. Indeed, Reed understands the diversity of post-70 Judaism as a mainly regional diversity.26

23For his part, Satlow observes that speaking of ‘Judaism’ or ‘Jewish culture’ implies that there is a cultural system, Judaism, which is different from non-Jewish cultures and which may be ‘influenced’ by them. If a scholar chooses to discard the categories of ‘influence’ and ‘Hellenization’, he should also avoid those of ‘Judaism’ and ‘Hellenism’. Only Jews and their local communities exist, not Judaism.27 Jewish local communities are deeply integrated within their surrounding environment. Thus, they must be studied within that framework and not against the background of more general entities (Hellenistic Judaism, rabbinic Judaism, etc.) and related literary sources (Philo’s writings, Talmudic literature, etc.).28

24Reed and Satlow share the same basic view: evidence shows that post-70 Judaism was diverse, and it is not possible to reduce this diversity to more general groups. Priority should be given to regional diversity, meaning that there are as many Judaisms as places. Such a nominalist approach may be questioned, because groups larger than local communities frequently play an important role in history. Moreover, nominalism knows only particular facts and neglects that they may be considered as elements articulated within a structure. It is difficult to completely avoid the notion of structure in the humanities.

2.2. Stuart S. Miller: A Complex Common Judaism

25According to E. P. Sanders, the notion of ‘common Judaism’ is the most appropriate to describe Second Temple Judaism, centered on Temple and priesthood.29 Stuart S. Miller reworked this notion in order to apply it to the rabbinic period. Miller speaks of a ‘complex common Judaism’, which combines Sanders’ common Judaism and the ‘complex Judaism’ of Martin Hengel and Roland Deines. This Judaism is ‘common’, since all its components share the same common source, biblical tradition in the broad sense of the term. It is ‘complex’ because it has generated the monumental synagogues in all their diversity: some are decorated with a zodiac, some include a list of the priestly courses, others contain mosaics or texts which show rabbinic features. It takes into account both ethnic and religious dimensions of Jewish identity. Finally, it sheds light on the way pagan material culture was appropriated within a Jewish context.30 In contrast to the views of Reed and Satlow, complex common Judaism emphasizes the unity of ancient Judaism: the differences between the rabbis and other Jews should not be overstated. However, this approach tends to underestimate tensions and conflicts stemming from diversity, as pointed out by Mimouni: “There are many conflicts between the two surviving movements [those of the rabbis and the Christians] and the third category of Judeans [so-called synagogal Judaism]. They will lead at a date difficult to determine with accuracy to the victory of the descendants of the Pharisees/rabbis and the Nazoreans/Christians.”31

2.3. Daniel Boyarin: Jewish Binitarianism

26In his seminal book Border Lines, Daniel Boyarin argues that the belief in a second God was widespread among Jews both before and after 70 CE. It took the form of the logos theology in the writings of Philo, the Gospel of John, and the Targumim. As he himself admits, Boyarin is far from being the only scholar to claim the existence of Jewish binitarianism.32 Many scholars came to this idea by different paths: the problem of the Jewish roots of Christianity and Gnosticism, the study of rabbinic traditions about ‘two powers in heaven’, or interest in the figure of Metatron within medieval Jewish mysticism.33

27Border Lines is, however, one of the books in which Jewish binitarianism has the most important place.34 According to Boyarin, this conception should be considered an old Jewish tradition that finds support in the biblical text. Evidence for binitarianism may be found among both Greek- and Hebrew-/ Aramaic-speaking Jews. The energy devoted by the rabbis to fighting binitarianism suggests that it must have been highly attractive for a great number of Jews. The very presence of binitarian concepts within rabbinic traditions attests to their popularity and to the fact that the rabbis were unable to eradicate them completely.

28Boyarin’s main thesis is that the rabbis decided to consider Jewish binitarianism non-Jewish in order to strengthen the boundaries between Judaism and Christianity. In fact, binitarian Judaism seems to be the background against which rabbinic Judaism and Christianity emerged and in relation to which each is defined.35 Paradoxically, the rabbinic/Christian dichotomy remains at the centre of the book, while non-rabbinic or para-rabbinic Judaism related to binitarianism and logos theology is not explored enough and needs further investigation. Within the narrative centred on this pair, Christianity appears as a proto- or parent religion with rabbinic Judaism as its offspring.

2.4. Emmanuel Friedheim and Seth Schwartz: Judeo-Paganism

29Historians and scholars in the field of rabbinic studies have long argued that Jews of the Second Temple and rabbinic periods were no longer attracted to idolatry. Their opinion was based on explicit statements of the rabbis and other evidence, such as Judith 8.18. Regarding the Roman period, they also shared the conviction that idolatry was declining among the pagans themselves. The figure of the sceptical pagan is well known from aggadic literature.36 Only a few scholars, including Ludwig Blau, had different positions.37 Even Goodenough claimed that, in spite of its use of pagan symbols, Palestinian Judaism could not be considered a form of Jewish idolatry or polytheism. In his view, it remained fundamentally faithful to the Law of Moses.38 On this issue, one of Goodenough’s disciples, Morton Smith, did not agree with his master. His study of magical texts revealed the existence of a paganized Judaism that was in no way marginal.39

30It is only recently that the traditional narrative has really been challenged, and this by two scholars: Emmanuel Friedheim and Seth Schwartz. Both have emphasized the importance of Judeo-paganism in Palestine during the rabbinic period. According to Friedheim, during this period paganism was still a living religion in Palestine and remained attractive to Jews. Several laws in tractate Avodah Zarah presuppose this background. Talmudic literature also refers to explicit cases of Jewish idolatry. Finally, the aggadic corpus contains traditions which support Friedheim’s claims.40 On Schwartz’s telling, Palestinian Judaism collapsed after the Bar Kokhba revolt under the oppression of Roman imperialism. As a result, a great number of Jews were incorporated into a Greco-Roman framework, consisting of civic cults and pagan culture. By the second and third centuries, the cities of Tiberias, Sepphoris, and Lydda are mainly Jewish, but their material remains (coins, inscriptions, statues, mosaics) are pagan.41

31Both Friedheim and Schwartz are a long way from agreeing on all the points discussed. Schwartz focuses on archaeological evidence and chooses to dismiss rabbinic sources, which lack objectivity. By contrast, Friedheim gives more balanced consideration to both kinds of evidence and discusses rabbinic sources in more detail. His conclusions are also less radical than those of Schwartz. For him, speaking of a collapse or a virtually complete paganization of Palestinian Judaism between 135 and 350 CE is overblown, and the archaeological data used to support the contrary can be read differently. Friedheim holds that even the rabbis were partly Hellenized, and they were able to influence various sectors of Jewish society. Nevertheless, it remains striking that two scholars, using different methods and working independently from each other, reached a similar conclusion: the Mishnaic and Talmudic periods were marked by the growing importance of Judeo-paganism.

2.5. Rachel Elior: Priestly Judaism versus Rabbinic Judaism

32The basic claim of Rachel Elior is simple: when the priests were separated from the earthly Temple and Merkavah, they conceived alternative forms of the Temple and Merkavah in heaven. This process of substitution occurred three times in ancient Jewish history: after the destruction of the First Temple with Ezekiel’s vision, during the Second Temple period with the Qumran community, and after 70 with the Hekhalot mystics.42 According to Elior, the three periods form a historical continuity and can be considered three stages in the development of the same conception, that of the mystical chariot vision or Merkavah.43

33Elior’s book The Three Temples deals mainly with the Qumran community and corpus. The community is dominated by priests, and its corpus reflects priestly lore, whose main features are described by Elior.44 First, the priests of Qumran believe in the unity of heaven and earth, which has implications for their conception of space, time, and liturgy.45 Second, they tell us three myths about calendrical issues that involve, respectively, Enoch, the Watchers, and the sacred times of Sabbath and Shavuot.46 For the priests of Qumran, the only calendar in accordance with both divine revelation and the laws of nature is the solar calendar.47 Enoch, the first priest, brought it from heaven to earth.48 In contrast to Enoch, the evil Watchers taught the lunar calendar to the generation of the Flood.49 The third myth sees history as a succession of sabbatical cycles, patterned after the seven days of creation. Shavuot, which is also connected to the number seven, is the feast marking the renewal of the covenant. In fact, Qumran priests argue that the Sinaitic covenant is only the last in a long chain of covenants, all associated with the date of 15 Sivan.50 Within the priestly lore, angels play an important role.51 They share a great number of features and attributes with the priests. Angels and priests possess the same knowledge and observe the same rituals.52 Angels are described like priests and vice versa.53 Jubilees, the books of Enoch, and the Testament of Levi describe the origins of the relationship between angels and priests, while the Songs of the Sabbath Sacrifice and the Blessings Scroll reveal their liturgical affinity.54

34In chapter nine of The Three Temples, Elior holds that the early rabbis knew Qumran literature and excluded it from the rabbinic canon.55 They also marginalized its central concern, namely the heavenly Merkavah.56 Last but not least, they conceived of their Judaism as opposed to the Judaism of the secessionist priests. The same could be said of the later rabbis. Indeed, the rabbis do not recognize a priesthood predating Aaron and have a negative view of Levi.57 Moreover, they completely omit priests in the chain of transmission of the Torah.58 On a theological level, they establish clear boundaries between heaven and earth and are mainly interested in earthly issues.59 They reject the three myths that form the basic core of the priestly lore. According to the rabbis, the sin of the Watchers never occurred, and Enoch is now considered a sinner.60 As to the third myth, Tannaitic literature does not use the term ‘Shavuot’, and the Qumran feast of the renewal of the covenant has no place in the ritual world of the rabbis.61 The rabbinic lunar calendar, dependent on human initiative, contrasts with the eternal order of the priestly solar calendar.62 Rabbinic angels are never connected with the calendar or priests.63 Finally, the rabbis differ from the Qumran priests by advocating a Torah that is “no longer in heaven” (Deut. 30.12) and open to human interpretation, revealing a more democratic and individualistic conception of Israel.64 The rabbis believe in an Oral Torah, whereas the priests only give authority to revealed writings.65

35Elior admits that the picture is not so simple and that rabbinic attitude toward priestly traditions could be better described as a mixture of sanctification, conditional acceptance, and rejection.66 Nevertheless, she claims that the real heirs of the secessionist priests are not the rabbis, but the Jews of the synagogues and the mystics of the Hekhalot. In the synagogue Qedushah, angels are liturgical partners with Israel, as was the case in the Songs of the Sabbath Sacrifice. Synagogue iconography refers to the Temple and also probably to the affinity between heaven and earth. Synagogue inscriptions include lists of the priestly guards.67 Regarding the Hekhalot traditions, they share with Qumran literature a “common infrastructure” (the heavenly Temple and Merkavah and the centrality of the angels, who are considered the counterparts of the mystics) as well as many other details.68 Elior notes that the Hekhalot texts are only a partial continuation of Qumran priestly tradition. Within them, the secessionist and polemical dimension disappears and rabbinic authority seems to be recognized, even if it is only on an earthly level.69

36The broad and ambitious synthesis offered by Elior raises many questions as well as many problems. As pointed out by Jonathan Klawans, how could the so-called secessionist conceptions be so widespread among the sources of the Second Temple period?70 Is the continuity between Qumran and the Hekhalot literatures so obvious, particularly when we consider the texts within the framework of a mystical priesthood? Even Philip Alexander, who emphasizes a number of significant parallels between the two corpora, notes: “The Hekhalot texts are not as precise and detailed as Sabbath Songs in correlating the heavenly Temple and its liturgy with the earthly cult.”71 Elior’s discussion of the calendars and their ideological implications is far from being unanimously accepted.72 As with every synthesis, the work of Elior is not free from simplification. Like other scholars, she speaks of a Qedushah at Qumran, while there is no citation of Isa. 6.3 or Ezek. 3.12 within the Dead Sea Scrolls.73 She often neglects to mention that several important principles or doctrines ( “All Israel have a part in the world to come” [m. Sanh. 10: 1], the Oral Torah, the valorization of debate and different points of view) were not obvious for the Tannaim and only appeared at a later stage in rabbinic history.74

37Some problems have more direct bearing on our investigation. Elior focuses on the contrast between the Qumran priests and the early rabbis, while a systematic comparison between rabbinic and the Hekhalot literatures would have been more interesting for us. According to Alexander, it is difficult to conceive of a ‘priestly Judaism’, really autonomous and distinct from rabbinic Judaism.75 Sacha Stern even argues, probably too readily, that it is impossible.76 Finally, it is important to note that Elior remains faithful to the traditional view of mainstream Judaism passing from the priests to the rabbis after 70. That being said, it is clear that Elior is the scholar who offers the most articulate and systematic reflection on the differences between priestly and rabbinic forms of Judaism. Other scholars agree with Elior on the continued importance of priesthood and priestly concerns during the Roman and Byzantine period, and Mimouni is very close to Elior’s argument when he speaks of a priestly synagogal Judaism (judaïsme sacerdotal et synagogal) that would be mystical and the direct source of the Hekhalot literature.77

3.0. Rabbinization

38The term ‘rabbinization’ means, first and foremost, the way that rabbinic texts appropriate elements from various sources, Jewish or non-Jewish, literary or non-literary. Scholars often use it when they deal with rabbinic interpretation of the Bible and the Jewish past. The figure of Moses, when rabbinized, becomes Moshe Rabbenu. One of the authors who most frequently uses the term ‘rabbinization’ with this meaning is Jacob Neusner.78

39More recently, the term has acquired another meaning, that of a process by which non-rabbinic Jews become rabbinic. Two scholars have particularly explored this new understanding of the concept of rabbinization: Seth Schwartz and Hayim Lapin.79 Both have found evidence of rabbinization in the following items:

1. The invention of piyyut.

2. The growing presence of the Hebrew language, attested directly in inscriptions and indirectly in Justinian’s Novella 146.

3. The apparition of iconophobic and iconoclastic tendencies among Palestinian Jews.

4. The use of the term deuterosis with the meaning ‘rabbinic tradition’ by Jerome, Epiphanius, and Novella 146.

5. In some inscriptions (Dabbura: Rabbi Eliezer ha-Qappar; Venosa: duo rebbites),80 the title rabbi indisputably refers to real rabbis. The inscription of Rehov includes texts having close parallels in rabbinic literature.

6. The rabbinic figures and materials that are found in the Hekhalot and apocalyptic writings.

7. A number of halakhic traditions.

8. The references to the rabbis (rabbāniyyūna and aḥbār) in the Qurʾan (e.g., Q 3.146; 5.44, 63; 9.31, 34).

40As it is the case with the concept of non-rabbinic Judaism, the concept of rabbinization raises many questions.

3.1. Chronology

41When did the process of rabbinization begin? The chronological setting of the present book (400–1000 CE) indicates that it did not begin before 400 CE. Schwartz contends that the first signs of rabbinization may be recognized in the sixth century. However, the growing involvement of the rabbis in wider Jewish communal life, the expanding scope of their halakhic decisions, and the rabbinization of marriage contracts began largely before the sixth century.81 The phrase ‘rabbinic movement’ even implies that, given its very existence, rabbinic Judaism could do no other than spread in a non-rabbinic Jewish milieu (maybe as a continuation of so-called Pharisean proselytism82).

42When was rabbinization achieved? It is not easy to answer this question. The difficulty lies mainly in the ambiguity of Karaism. How should we interpret its emergence? It could be considered as proving the existence of an already established rabbinic authority that aroused opposition.83 It may also indicate that the rabbis were not yet powerful enough to impose their authority. More generally, the Islamic context seems to have been more favourable to rabbinization.

3.2. Geography

43Even if a substantial part of the evidence for rabbinization comes from Palestine, it is obvious that the Babylonian setting played a central role in the process. Thus, the foundation of Baghdad, the new capital of the Abbasid empire (762 CE), may have contributed to the strengthening or even institutionalization of the judicial power of Babylonian rabbis.84 It should be noted that some Babylonian magic bowls cite passages from the Mishnah, which is not the case for Palestinian amulets.85 Moreover, unlike Palestinian sources, the Babylonian Talmud shows a clear tendency to rabbinize the figure of Jesus.86

44As Christian Robin has recently argued, South Arabian Judaism or, more precisely, the Judaism of the Himyarite kingdom, obviously belongs to a priestly type and reveals no rabbinic features.87 By contrast, when describing North Arabian Jews, Islamic sources show no priestly features, starting with references to the rabbis in the Qurʾan.88

45Rabbinization within the Medieval Christian world—the Latin West and the Byzantine East—remains to be explored.89

3.3. Conception

46The rabbinization process may be considered from two different sides: the rabbinic and the non-rabbinic. Schwartz writes of the authors of the Hekhalot texts: “This means either that late antique rabbis were trying to annex magical practice, or that Late Antique magicians were claiming rabbinic origins for their teachings, presumably because such a claim would have enhanced their prestige.”90 Accordingly, rabbinization may consist in the adoption of rabbinic elements by non-rabbinic Jews or, conversely, of the adoption of non-rabbinic elements by the rabbis. In both cases, the elements are frequently modified in order to be integrated into the culture or the literature of the rabbinic or non-rabbinic group.

47As pointed out by Ra‘anan Boustan, rabbinic Judaism does not merely replace the varieties of non-rabbinic Judaism, but shapes them, while also being shaped by them: “From the sixth century on, rabbinic forms, themes, and modes of authority increasingly inflect even those genres or corpora that seem to have existed at the boundaries of rabbinic literary culture. It would seem that rabbinic culture was itself transformed in the process.”91 The notion of rabbinization, however, raises a serious methodological difficulty: it is not always possible to draw a clear distinction between what is rabbinic and what is non-rabbinic.92

3.4. Modalities

48The rabbinization process is not necessarily continuous, whether in time or in space. It has probably known periods of regression. The evolution of the Jewish patriarch would be a good example of ‘de-rabbinization’. Indeed, several authors have pointed out that the patriarch began its historical trajectory within the rabbinic movement before becoming the patron of non-rabbinic Judaism.93 According to Oded Irshai, the post-Amoraic period in Palestine is marked by the decline of rabbinic authority and the rise of the priests.94

3.5. Means of Spreading

49Rabbinization is not by definition the imposition of rabbinic norms and conceptions on other Jews, who are reluctant to accept them. It could be also conceived as the spreading of rabbinic way of life, which has become attractive for wider Jewish circles. Thus, for Ron Naiweld, Babylonian rabbis composed certain post-Talmudic tractates in order to spread their conception of the Torah among non-rabbinic Jews.95 Furthermore, the study of rabbinic travels or the notion of ‘religious network’ may contribute to shedding light on the issue of rabbinization.96

3.6. Content

50Rabbinization involves cultic sites, rites, power structures, and literary materials. Regarding texts, the ambiguous corpora already referred to (Targum, Hekhalot, apocalypses, and piyyut) are well-adapted for a reading in terms of (imperfect?) rabbinization. Two scholars have already applied the concept of rabbinization to the Hekhalot corpus as well as to the later apocalypses: Ra‘anan Boustan and Martha Himmelfarb.97

3.7. Context

51Finally, it is possible that rabbinization has been merely favoured by the context, and the role of rabbinic agency is less important than is usually thought. As Lapin notes, the change in the legal status of Jews in the Christian empire and the appearance of more exclusively Jewish communities made it easier for the rabbis to become communal leaders.98 The growing institutionalization of rabbinic academies in post-Amoraic Babylonia and rabbinization are probably connected, even if the nature of this connection needs further investigation.99

4.0. Rabbinization and the Hekhalot Literature: The Article of Ra‘anan Boustan

52A discussion of Ra‘anan Boustan’s article is relevant to our discussion, since it deals with both rabbinization and non-rabbinic Judaism within a framework mainly limited to the relationship between the Hekhalot and rabbinic literatures.100 At a methodological level, the sociological approach remains marginal in the article, whose keywords are clearly ‘literature’ and ‘culture’. The Hekhalot and rabbinic literatures are considered “sites of Jewish literary culture.”101 Boustan emphasizes a literary fact: the existence of shared materials and literary overlaps between the Hekhalot and rabbinic literatures. He also tries to understand it and asks what the meaning of the overlaps is. How were they possible? According to Boustan, the category of rabbinization and some sociological approaches, such as those of Michael Swartz and Moulie Vidas, may contribute to shedding light on both questions.

53Boustan is well aware that the Hekhalot texts are an example of what I have called an ‘ambiguous corpus’. In some ways, they differ clearly from rabbinic literature and it is even possible to say that the two corpora reflect opposing forms of piety.102 Their comparison reveals, however, a significant number of shared traditions. The rabbis borrowed from the Hekhalot literature, but the reverse also occurs. Thus, a model which aims to explain the relationships between the Hekhalot and rabbinic literatures should take into account this hybrid situation, composed of both similarities and differences.103 One may notice that the descriptive part of Boustan’s article emphasizes the similarities primarily, if not exclusively.

54Boustan begins his discussion by describing the two dominant views of the relationship between the Hekhalot texts and rabbinic literature: one of them is ‘dialectical’ and the other ‘binary’. The former is exemplified by Gershom Scholem and the latter by Goodenough and Elior. According to the dialectical view, there is only one Judaism, a ‘common Judaism’, namely rabbinic Judaism, and the Hekhalot traditions are its esoteric dimension.104 By contrast, the binary view distinguishes between two forms of Judaism: rabbinic Judaism and non-rabbinic Judaism. In this framework, the rabbis are depicted as only concerned with worldly issues and the Law. Accordingly, the Hekhalot literature stems from a single non-rabbinic Judaism, which is mystical. Halperin’s work also belongs to the binary school of thought.105 Boyarin’s dichotomy between the rabbis and binitarian Judaism appears only once before the conclusion of the article.106

55When describing the dialectical model, Boustan speaks of an “inner dialectic between the mystical and the halakhic normative dimensions.” Unfortunately, the precise meaning of the term ‘dialectic’ in this context is not further explained.107 In Boustan’s view, the dialectic pattern knows only one Judaism, which is rabbinic, but it seems to us that Scholem’s opinion is more nuanced. For instance, Scholem affirms the existence of both heterodox and orthodox (rabbinic) Jewish Gnostics and relates Hekhalot literature to the latter.108 When he understands the Hekhalot corpus as the expression of a ‘Jewish Gnosticism’, he is very close to the binary view of Goodenough and Elior.109 Goodenough himself was interested in Scholem’s work, which he saw as complementary to his own work.110 As to the binary view, if we follow Boustan’s description, it distinguishes between “two wholly discrete forms of Judaism” and mystical Judaism is “wholly autonomous” or “hermetically sealed from rabbinic Judaism.”111

56This presentation raises several questions. First, two forms of Judaism could not be wholly distinct or completely separated from each other, because, if that were the case, it would be impossible to call them both ‘Judaism’. Second, even if we admit the possibility of such a dichotomy, it does not fit within the approaches of Goodenough and Elior. Both make extensive use of rabbinic literature and find evidence for mystical Judaism within it. It follows that neither Goodenough nor Elior see rabbinic and non-rabbinic forms of Judaism as wholly separate. ‘Dichotomy’ does not necessarily imply separation, and the frontiers between the two forms of Judaism may have been porous. Third, according to the binary view (says Boustan), there is a conflict between rabbinic and mystical Judaism.112 How can mystical Judaism be both “wholly autonomous” from rabbinic Judaism and the result of a development in opposition to it? If there is a conflict between mystical and rabbinic Judaism, then they are strongly related to each other. It is interesting to note that in Elior’s terms, the conflict between priestly Judaism and the Pharisees-rabbis was particularly strong in the Second Temple period, but a shift occurred after 70 CE: the priests recognized the authority of the rabbis, at least on an earthly level, and tried to avoid conflicts with them.113 Finally, when Boustan states that the binary view considers rabbinic Judaism ‘mainstream Judaism’, he is right about Elior’s approach, but not about other versions of the binary view, which rather argue for a mainstream non-rabbinic Judaism.114 Goodenough, for example, holds that Hellenized Judaism is the most widespread form of Judaism in the rabbinic period.

57In the second part of his argument, Boustan claims that the dialectical and binary views reflect common assumptions and are more similar than usually thought. Therefore, they may be included in the so-called ‘perennialist tradition’. In fact, they share three attributes, which make the complex relationship between the Hekhalot and rabbinic literatures difficult to explain. First, the dialectical and binary views are both unilateral. Indeed, the dialectical view emphasizes the similarities between the Hekhalot and rabbinic literatures and thus fails to explain their differences. The binary view has the opposite approach.115 Second, both patterns are dichotomist in that they distinguish between a mystical and a non-mystical component: the dialectical view sets both components ( “the mystical and the halakhic-normative dimensions”) within the same rabbinic milieu, while the binary framework relates each component to a specific Judaism (‘mystical Judaism’ versus ‘rabbinic Judaism’). Dichotomist approaches oversimplify the complexity and diversity of Jewish culture in Late Antiquity.116 Obviously, there is no simple distinction or difference between halakhah and mysticism or rabbis and mystics. Third, both views are static. They share the same conception of mysticism as an ancient or even timeless religious experience.117 According to them, mysticism has an a priori definition, an unchanging essence. Thus, there are strong and stable boundaries between mysticism and the rest of Jewish culture.118 Mysticism only changes under the influence of the rabbis, who appear as the single active force within the Jewish cultural system. Boustan notes, however, that in the perennialist view, even the rabbis tend to be conceived of in a static and essentialist manner: the important cultural transformations that affected rabbinic culture in Late Antiquity are not taken into account.119

58At first sight, the notion of a ‘perennialist tradition’ including the dialectical and binary views seems to be relevant. As pointed out above, Goodenough himself regarded Scholem’s approach as complementary to his own work. Moreover, both Scholem and Elior distinguish between the esoteric and exoteric dimensions of Judaism.

59I noted that the main feature of the perennialist tradition is dichotomy. Thus, this tradition is principally understood on the basis of one of its components: the binary view. One passage of the article directly identifies the binary view with a “kind of Jewish philosophia perennis.”120 A few pages later, Boustan says that scholars connected with the perennialist tradition share the same “dichotomous view of rabbinic and mystical (emphasis is mine) forms of early Judaism.”121 However, it is only the binary view that distinguishes between rabbinic Judaism and non-rabbinic mystical Judaism. The dialectical view claims rather that there is only one (rabbinic) Judaism.

60Finally, I may observe that the idea of a mystical tradition maintaining itself through the centuries fits better with Elior’s view than with Scholem’s. Indeed, Elior often gives the impression that the same priestly worldview may be found within the book of Ezekiel, Qumran literature, and the Hekhalot texts, and that the destruction of the Second Temple did not fundamentally affect this worldview.

61In sum, since the perennialist tradition is primarily conceived on the model of the binary view, its conception is biased and unbalanced. The reason for the imbalance is the following: Boustan mainly criticizes the binary view, while showing a clear preference for the dialectical view. I shall discuss this important point further on.

62According to Boustan’s argument, the ‘perennialist tradition’ is dichotomist as a whole. It follows that he distinguishes between a dichotomist binary view and a dichotomist dialectical view. The former is a tautology: a binary view is necessarily dichotomist. The latter remains to be clarified: how can a view be both dichotomist and dialectic?

63At first glance, it is paradoxical, because, by definition, a dialectical view is not dichotomist and could even be said to be anti-dichotomist, as is showed by the Hegelian criticism of Kant. In Hegel’s view, Kant is a thinker of ‘understanding’. The moment of understanding is a moment in which the concepts are stable and form fixed dichotomies. By contrast, ‘reason’ sees opposed concepts within a dialectical framework. Accordingly, contradiction is a process leading to a third concept, in which two conflicting ideas are reconciled and raised to a higher level.122

64In order to understand how an approach could be both dialectical and dichotomist, we must return to the most prominent figure of the dialectical approach to Judaism: Gershom Scholem. It is true that Scholem often uses the term ‘dialectic’, but he does not always give the same meaning to it.123 At least two different conceptions of dialectic appear in his writings. According to the first conception, the opposing sides are Law and mysticism, and the third element, which is the synthesis, is Judaism and its historical evolution.124 The opposing sides of the second conception are myth and religion, and their synthesis is mysticism.125

65The first conception places little emphasis on synthesis and emphasizes rather the opposing sides: the constant tension between the principles of law and mysticism is the very life of Judaism. By contrast, the second conception highlights the synthesis provided by mysticism, which includes in the same whole the two opposing principles of myth and religion. Boustan obviously has the first conception in mind when he describes Scholem’s dialectical and dichotomist explanation of the relationship between Hekhalot and rabbinic literature. Indeed, for the first conception, the dialectic is one with its dichotomist component, that is, a tension between two opposing principles. In light of the first conception, we understand better the following statement of Boustan: “I show that, ironically, Scholem’s understanding of the inner dialectic between the mystical and the halakhic-normative dimensions within a single but multifaceted Judaism has unwittingly encouraged a binary view of the Jewish tradition, in which the mystical and the rabbinic represent two diametrically opposed forms of Judaism.”126 This statement is best illustrated by the case of Goodenough (a supporter of the binary view), who, like Scholem (a supporter of the dialectical view), speaks of a tension between law and mysticism.127 While Scholem places the tension within rabbinic Judaism, Goodenough connects Law with the rabbis and mysticism with another form of Judaism. Thus, Boustan is right when he says that a dichotomist dialectic (Scholem) has led to a pure dichotomy (Goodenough).

66If Boustan adopts Scholem’s first conception of dialectic, he neglects the second, according to which mysticism is not one of the opposing sides, but the very synthesis of the dialectical process. Boustan’s presentation of Scholem’s dialectic is therefore somewhat simplistic. It also raises a second difficulty, perhaps more problematic: Scholem does not use the term ‘dialectic’ when he deals with the specific subject of Hekhalot literature. His definition of the Hekhalot worldview seems to be more Gnostic than dialectical.

67As already pointed out, Boustan claims that the dialectical and binary views are both dichotomic. However, dichotomy is not the same in both cases. The dialectical view divides one (rabbinic) Judaism, whereas the binary view contrasts two different forms of Judaism. By putting forward the notion of a perennialist tradition, Boustan tends to play down the importance of this difference. In his view, what is significant is dichotomy and not the domain within which it operates. Nevertheless, it may be asked whether he is right about this. The fact that the binary and the dialectical views disagree on the very existence of non-rabbinic forms of Judaism is not an insignificant detail! Finally, it is not clear that the scholars connected with the perennialist tradition ignore the “historicizing approach to Hekhalot literature.”128 In fact, they also try to clarify the historical context in which the authors of the Hekhalot corpus wrote. The difference between Boustan and these perennialist scholars lies in the selection of different historical contexts: Scholem and Elior prefer an ancient (70–400 CE) Palestinian setting; Boustan (and other scholars) a later (400–800 CE) Palestinian or Babylonian setting.129

68Boustan not only criticizes the perennialist tradition, but also offers an alternative model about the relationships between Hekhalot and rabbinic literature. His model includes four components:

1. Descriptive: Textual data, which, while showing some differences, primarily illustrate similarities between Hekhalot and rabbinic literature.

2. Literary: The notion of rabbinization.

3. Sociological: The hypotheses of Swartz and Vidas.

4. Historical: Morton Smith’s approach to post-70 Judaism.

69Boustan’s model thus involves three levels: ‘literary’, ‘sociological’, and ‘historical’. On the literary level, Boustan emphasizes the notion of rabbinization, which he defines as follows: “From the sixth century on, rabbinic forms, themes and modes of authority increasingly inflect even those genres or corpora that seem to have existed at the boundaries of rabbinic literary culture. It would seem that rabbinic culture was itself transformed in the process.”130 Thus, rabbinization is essentially considered a literary and cultural process. This process develops in two directions: The rabbis exert their influence on ‘non-rabbinic’ corpora and vice versa. In a previous passage, Boustan gives another definition of rabbinization: “[…] the spread of rabbinic hegemony was gradual and remained incomplete throughout Late Antiquity; but […] its success also entailed willy-nilly both its diversification and its appropriation within other branches of Jewish literary culture— among synagogue poets and preachers, among magicians, and among mystics.”131 As in the first definition, Boustan emphasizes the notion of authority. He also understands the transformation of rabbinic culture as its appropriation within various corpora or discourses and therefore as its diversification. Thus, rabbinization is both a constraint and a source of “cultural creativity”.132

70Rabbinization being a reciprocal process, it results in a growing proximity between Jewish rabbinic and non-rabbinic texts. Boustan speaks of ‘convergence’, ‘amalgamation’, ‘harmonization’, and ‘dialogue’133. Finally let us note that, in Boustan’s view, rabbinization is also a specific period of Jewish history, which he dates from the fifth or sixth to the eighth centuries, following Schwartz’s claim that rabbinization really began in the sixth century.134

71The title of Boustan’s article might indicate that his model is based exclusively on the notion of rabbinization, but this is not the case. He also adds the sociological approaches of Swartz and Vidas.135 Both scholars try to identify the social milieu from which the Sar ha-Torah materials of Hekhalot literature emerged. They agree on at least one point: the Jews responsible for these texts are not the rabbis, but form a group close to the rabbis. They differ, however, on the identity of the group: synagogue functionaries (Swartz) or reciters in rabbinic academies (Vidas).

72Last but not least, Boustan finds his ‘general orientation’ in the conception of post-70 Judaism advocated by Smith.136 Unlike Goodenough, Smith argues that the plurality of post-70 Judaism cannot be reduced to a simple dichotomy between rabbinic Judaism and a single non-rabbinic Judaism. For Smith, there are no strong differences between various forms of Judaism, but rather differences of degree: Jewish groups and circles form a continuum, and the frontiers between them are highly porous.

73In sum, Boustan provides an alternative model designed to solve the problems posed by the perennialist approach. First, his model is not static, since it offers an ‘historicizing’ approach to the Hekhalot corpus in terms of rabbinization.137 Second, it is not dichotomist, since it builds on Smith’s criticism of the dichotomist view of Goodenough. Third, it is not unilateral, since it explains both similarities and differences between Hekhalot and rabbinic literature, emphasizing the convergence of the corpora (literary rabbinization) and the proximity of the groups that lie behind the corpora.

74Boustan’s approach remains vague on the articulation of its four components. He describes his alternative approach mainly in the introduction and the second part of the article. It is striking that none of these sections deal with all four components or with the question of how they are connected. Furthermore, Boustan’s discussion is more descriptive than explicative: the arguments of Smith, Vidas, and Swartz are presented briefly, and the category of rabbinization, despite being present in the title, appears only a few times in the article.

75However, it is possible to offer a hypothetical construction of Boustan’s argument: the notion of rabbinization (2) and the hypotheses of Swartz and Vidas (3) are two different and complementary ways of interpreting the textual data (1), literarily and sociologically, and Boustan refers to Smith’s conception of post-70 Judaism (4) as a more general historical framework within which the literary and sociological interpretations find their place. Basically, 4 is the basis of 2 and 3, which explain 1.

76One might well ask, nonetheless, whether rabbinization and the sociological insights of Swartz and Vidas are really complementary. Neither Swartz nor Vidas speak of rabbinization (the same could be said of Smith). To answer this question, we should first fully understand how each approach accounts for textual data and particularly the similarities between Hekhalot and rabbinic literature.

77The literary approach (in terms of rabbinization) emphasizes both the growing authority of rabbinic texts and the increasing interest of the rabbis in other Jewish corpora. The sociological approach, exemplified by Swartz and Vidas, claims that the Hekhalot traditions (or at least the Sar ha-Torah traditions) arose within circles close to those of the rabbis: the synagogue functionaries and the reciters. Sociological proximity naturally results in literary similarities and overlaps. Both approaches clearly differ on one point: the former speaks of convergence and is dynamic, the latter speaks of proximity and is static. Against this background, is it possible to call them ‘complementary’?

78Common sense would suggest that proximity is the result of convergence: it is because some groups are rabbinized that they become close to the rabbis. In this case, the concept of rabbinization is not simply literary, but also sociological, and may include the approaches of Swartz and Vidas, considered more dynamically. This seems to be confirmed by a statement of Boustan. While he stresses again and again that rabbinization is a literary and cultural category, at one point he mentions the role of rabbinization in “Jewish culture and society” (emphasis is mine).138 Even if the argument seems compelling, it is not fully satisfactory. Swartz’s synagogue functionaries could indeed have been rabbinized, but is it possible to think the same of Vidas’s reciters, who belong to the rabbinic academy?139 Moreover, it is striking to see that the passage of the article connecting textual data with the approaches of Swartz and Vidas does not refer to rabbinization.140

79At first sight, Boustan differs from the perennialist tradition, in that he uses the notion of rabbinization. Indeed, neither Scholem nor Elior discuss rabbinization, but the notion may be found in their works. As is well known, Scholem sees Hekhalot lore as an orthodox Jewish Gnosticism, that is, a Gnosis which has been revised in order to conform to a rabbinic framework. In other words, Hekhalot traditions are rabbinized Gnosis. As to Elior, she claims that the authors of the Hekhalot texts accepted rabbinic authority on an earthly level and avoid sectarian and polemical attitudes, in contrast with the Qumran priests. There is no reason not to consider this shift a form of rabbinization.

80Boustan himself alludes to the fact that perennialist scholars may use the notion of rabbinization as well when he argues for “a nuanced understanding of the process of rabbinization.”141 It follows that there may be, or may already have been, other interpretations of rabbinization that are not nuanced. What leads Boustan to assert that his conception of rabbinization is nuanced? He emphasizes both rabbinic agency and the transformation of rabbinic culture in the process. By contrast, the perennialist views would reduce rabbinization to the action of the rabbis on a passive non-rabbinic Judaism.142

81Occurring only a few times within his article, Boustan’s notion of rabbinization remains unclear on a number of issues. Is it a descriptive or an explanatory notion? He notes “patterns of similarity and difference,” “mutual literary appropriation,” and “permeable boundaries.”143 Are these phenomena identical to rabbinization or do they explain it? The question may even be raised whether rabbinization is an explanatory tool or, conversely, something that needs to be explained. In the following passage, ‘rabbinization’ is clearly explanatory: “This essay considers the role that rabbinization of Jewish culture and society at the end of antiquity (c. 500–900 CE) played in the formation of the distinctive registers of discourse found in Hekhalot literature.”144 Another passage suggests the opposite:

Instead, I wish to argue that a more nuanced mapping of the imperfectly intersecting terrains of Hekhalot and rabbinic literatures will open up new avenues for understanding both the extension of rabbinic hegemony and the enduring heterogeneity of Jewish culture during the transitional period at the end of Late Antiquity that saw the empires of the ancient European, Mediterranean, and Near Eastern world evolve into what Garth Fowden has called the ‘commonwealths’ of the early Middle Ages.145

82This long sentence is probably one of the most important in the article. It uses spatial terminology and metaphors ( “mapping,” “terrains,” “avenues”) and thus reflects the influence of the so-called ‘spatial turn’ in English humanities scholarship (and beyond). It is striking that in this sentence it is the study of the Hekhalot texts and their relationship with rabbinic literature ( “a more nuanced mapping of the imperfectly intersecting terrains of Hekhalot and rabbinic literatures”) which helps to explain rabbinization ( “will open up new avenues for understanding both the extension of rabbinic hegemony and the enduring heterogeneity of Jewish culture”). Indeed, “the extension of rabbinic hegemony” is the very definition of rabbinization, which also entails, for Boustan, a diversification of Jewish culture. In sum, the first sentence ( “This essay considers…”) suggests that rabbinization is a tool to better understand the relationships between Hekhalot and rabbinic literature, but the second reference states the contrary.

83Regarding rabbinization as “the gradual amalgamation of rabbinic and Hekhalot traditions and their attendant modes of authority,” Boustan makes the following remark: “Both rabbinic and Hekhalot literatures thus bear witness to the relatively early integration [emphasis mine] of what may appear, on phenomenological grounds, to be mutually exclusive modes of religious piety and authority.”146 In speaking of “relatively early integration,” Boustan supports the view that the Hekhalot texts were directly redacted in a rabbinized form by authors both different from and close to the rabbis. Consequently, Hekhalot literature appears in its very conception as an ambiguous corpus, including both rabbinic and non-rabbinic features. However, it seems to me that the following remarks suggest another approach to the relationships between rabbinization and the redaction of the Hekhalot texts: “I find myself persuaded […] that the very specific configuration of ideas, themes, imagery, and practices that defines ‘Merkavah mysticism’ […] is absent from rabbinic sources from the third and fourth centuries […] Something changed quite palpably from the late fifth to eighth centuries.”147 This passage emphasizes the indifference of the rabbis towards Hekhalot traditions up to the fifth century. It is tempting to suppose that a similar indifference characterized the ‘Merkavah mystics’ of the same period. It is also tempting to argue for the existence of a first version of Hekhalot literature devoid of rabbinic features. In a recent lecture, Philip Alexander distinguished between a first stage of Hekhalot literature (‘the old-fashioned Hekhalot’), which was not preserved, and its final version. The first stage would reflect the conceptions of non-rabbinic mystics. By contrast, the final version is rabbinized. Alexander asks further whether the rabbis are directly responsible for the rabbinization of the text or whether it is a strategy of the mystics themselves.148

84The main purpose of Boustan’s article is to challenge Elior’s view of the Hekhalot traditions, which connects them with a non-rabbinic priestly form of Judaism. For Boustan, this view exemplifies the perennialist tradition, which is unable to account satisfactorily for the complexity of the relationships between Hekhalot and rabbinic literature. Indeed, Elior often gives the impression of describing a priestly Judaism whose worldview has remained substantially unchanged from Ezekiel to the Hekhalot texts. However, Boustan’s criticism of Elior’s view is not fully convincing. As already stated, Elior’s book, The Three Temples, focuses on Qumran literature and addresses the Hekhalot corpus only in its last chapter. Elior contrasts the views of the Qumran priests and the rabbis on many topics, and the result is impressive. Boustan criticizes Elior’s dichotomic approach, but only in general terms. It is necessary to discuss the picture in detail and to deal with the dichotomies one by one.

85Even if Elior is wrong, it remains possible to argue for a connection between Hekhalot literature and the priests. As Boustan himself acknowledges, Alexander also connects both Qumran and Hekhalot literature to Jewish priests in a more nuanced way than Elior.149 The differences between Qumran and Hekhalot literature are easy to understand: if priests are behind the Hekhalot texts, they share a lot of materials and concerns with the priests of Qumran, but they are involved in a very different historical context. Boustan sees Swartz’s and Vidas’s sociological views in a positive light, but he does not explain why connecting the Hekhalot literature with a priestly milieu would be less sociological or insightful. These priests could be as rabbinized as Swartz’s synagogue functionaries or Vidas’s reciters.

86Boustan is well aware that the Hekhalot texts are not the only corpus scholars have linked to Jewish priests. This is true as well for the Targumim and the piyyutim.150 Therefore, it is possible to argue that all these corpora emerged within the same priestly milieu. Boustan, however, claims that the plurality of non-rabbinic forms of Judaism should not be reduced to a single alternative Judaism. Accordingly, he adds that rabbinization affects different branches of “Jewish literary culture”, including synagogue poets, preachers, magicians, and mystics.151

87He does not say, however, how we should understand the priestly features shared by the Hekhalot texts, piyyutim, and Targumim, and their common proximity to the world of the synagogue. Swartz, whose work Boustan refers to concerning the original setting of Hekhalot literature, argues for the existence of a priestly piety connected with the synagogue and which coexists with rabbinic ideology.152 Moreover, Boustan does not mention one important point shared by Swartz and Vidas: the group behind the Sar ha-Torah texts is closely related to the synagogue.153

88As already seen, the issue of the relationships between Hekhalot and rabbinic literature may be explored within three different frameworks: one dialectical, one dichotomist, and the alternative view offered by Boustan. These frameworks have broader implications regarding the nature of Late Antique Judaism. The dialectical framework knows only one Judaism, that of the rabbis. The dichotomic framework distinguishes between rabbinic Judaism and a single non-rabbinic Judaism. Boustan’s framework, which is based on the reflections of Smith, seems to conceive of a single Judaism (a ‘continuum’), including many (rabbinic and non-rabbinic) groups separated by porous frontiers. It is close to what Miller has recently called ‘complex common Judaism’.

89The three frameworks (in four views) may be resumed in the following table:

90Boustan’s approach (view 4) is close to the dialectical view (view 3). Both share the idea of a single, but multifaceted Judaism, and Boustan only claims that the dialectical understanding of the relationship between the different facets of Judaism is too simplistic. Thus, he holds that Hekhalot and rabbinic literature could be described as two facets (among others) of a single Judaism, but they are not “merely complementary facets” (and the only ones), as Scholem and others thought.154 Therefore, it is not surprising that Boustan criticizes primarily the dichotomic view and conceives of the perennialist tradition, which includes both dichotomist and dialectical features, essentially on the basis of the dichotomic view.

91A fourth view appears in the table (view 2), which considers two Judaisms related to each other. View 1 represents a strong dichotomy and view 2 a soft dichotomy. Contrary to Boustan, it seems to me that Goodenough’s and Elior’s approaches are not so different from view 2. The hypothesis of a synagogal Judaism, which will be discussed in the last part of my paper, is probably the best example of view 2.

92Boustan may think that views 1 and 3 cannot encompass the notion of rabbinization. According to view 1, the two Judaisms are separated and could not influence each other, while view 3 assumes there is only one rabbinic Judaism and thus nothing to rabbinize. For Boustan, only view 4 can give proper weight to the process of rabbinization, since it recognizes both Jewish diversity and the possibility of reciprocal influences within a single, social continuum. In fact, however, rabbinization may also be present within view 2 as a form of relationship between the two Judaisms (and within views 1 and 3 as well, as pointed out below).

93Finally, since he recognizes the existence of one single Judaism which has the form of a continuum, Boustan tends to downplay differences and tensions between various Jewish groups.155

5.0. Synagogal Judaism

94In a book published in 2012, Simon Claude Mimouni suggested a new model to describe Palestinian Judaism after 70.156 It includes three Judaisms: rabbinic, Christian, and synagogal. Mimouni characterizes synagogal Judaism in two ways: negatively, as neither rabbinic nor Christian, and positively, to include the majority of Palestinian Jews. Synagogal Judaism finds its material basis and its primary area of expression in synagogues, which at that time were neither directed nor controlled by the rabbis. Unlike rabbinic Judaism, synagogal Judaism was well-integrated into the Greco-Roman world. Several inscriptions testify that Jews perform the function of agoranomos or bouletes. For Mimouni, these Jews are clearly synagogal Jews.

95Priests hold a dominant position within synagogal Judaism. In recent publications, Mimouni has increasingly stressed the priestly component of synagogal Judaism, speaking of priestly-synagogal Judaism (judaïsme sacerdotal et synagogal).157 Synagogal Jews may be Greek- or Aramaic-speaking. For Mimouni, some apocalyptic writings may be related to synagogal Judaism, which is both mystical and messianic. It is in many respects the institutional or official Judaism of that time, as it is the cult recognized by Romans. Dominant for a long period, it was finally overridden by the rabbinic and Christian movements. In the course of this development, however, both had assimilated elements from their ancient rival.158

96My aim is not to deal with all the questions raised by Mimouni’s hypothesis. I shall limit my discussion to some points directly connected with the major issues of my paper, the nature of non-rabbinic Judaism and rabbinization.

97Mimouni is concerned with non-rabbinic Judaism, like others who adhere to the new model. However, he departs from most by calling it ‘synagogal Judaism’. This label will probably raise some objections before being accepted (if it is) by the scholarly world. Did the rabbis not take part in synagogue life? Is the synagogue not a central component of common Judaism? If we admit the existence of many Judaisms, the synagogue would be a common point for all of them and not a criterion we could use to distinguish between them.

98It is, however, difficult to ignore Goodenough’s main thesis, that the ancient synagogue was not controlled by the rabbis, but by another form of Judaism. Since the 1970s, several studies have confirmed Goodenough’s thesis. These emphasize the rather marginal and sometimes problematic place of the synagogue in rabbinic literature. They also highlight contradictions between the archaeological and rabbinic evidence regarding the synagogue: these contradictions mainly touch upon architecture, figurative art, and conceptions of the sacred.159 A number of authors explicitly see the synagogue as a place of conflict between different trends or circles within Judaism.160

99The phrase ‘synagogal Judaism’ is not found in the writings of Goodenough, Neusner, Cohen, Lee I. Levine, or Schwartz. However, it is probably the most appropriate way to describe non-rabbinic Judaism, which, according to these very scholars, would be related to the synagogue setting. In a recent article, Fergus Millar has, I think, correctly applied the expression ‘synagogal Judaism’ to the approaches of Levine and Schwartz: “Should we therefore draw the conclusion, as Schwartz, following Levine, does, if with qualifications, that ‘rabbinic’ Judaism and ‘synagogal’ Judaism not only represent distinct spheres of religious practice, but were actually distinct, the one from the other.”161 The expression is all the more justified, in that the synagogue is not merely one of the elements supporting the new historiographical model; it is actually the central and unifying element of this model. Thus, Steven Fine has rightly understood the central role of the synagogue in what he (critically) called “the search for Nonrabbinic Judaism.”162

100The ambiguous corpora are directly (Targum, piyyut) or indirectly (Hekhalot literature) connected with the synagogue. Several magical practices are also related to the synagogue setting.163 The Jewish patriarch—an example of de-rabbinization—is recognized in the Theodosian Code as the leader of the synagogue network.164

101The main languages of Palestinian non-rabbinic Judaism, if it did include the majority of Palestinian Jews, could only be Aramaic and Greek, that is, the languages that are mostly attested in synagogue inscriptions. I have argued elsewhere that one can interpret the Qedushah as one of the prayers of synagogal Judaism. The oldest version of this prayer is preserved in Greek.165

102The synagogue is also connected with the priests, whom an increasing number of scholars consider as the elite, or one of the elite groups of non-rabbinic Judaism.166 At the end of his article about chains of tradition in ʿAvodah piyutim, Michael Swartz notes: “Chains of tradition in the ‘Avodah piyyutim add to the evidence that, along with rabbinic ideology, a form of cultic piety coexisted, in which the priesthood was valued, and perhaps even represented, in the ancient synagogue.”167

103Binitarianism, which, in Boyarin’s view, is one of the main features of non-rabbinic Judaism, is also related to the synagogue. Evidence of binitarianism is found in the Targum, while the Talmud interprets some problematic prayers as binitarian.168 Goodenough explains the synagogue art against the background of a mystical Philo. While his explanation has generally been rejected, Mimouni maintains that synagogal Judaism is mystical. According to Jodi Magness, the mystical interpretation of the ancient synagogue becomes relevant if it is based on the Hekhalot corpus.169

104Mimouni emphasizes the continuity of synagogal Judaism in Palestine. Schwartz sees discontinuity, holding that the history of post-70 Palestine is marked by rupture. For Schwartz, the period is first characterized by the adoption of a pagan lifestyle and practices among Jews. After 350 CE, radical changes occur in Palestine, and Judaism is increasingly organized around the synagogue, the local community, and the benefactors of these two institutions.170 This stark contrast, based mainly on the archaeological data, raises several problems that have been discussed elsewhere, particularly by Miller.171

105Like Mimouni, Levine gives an account of the ancient synagogue that builds on both archaeological and literary evidence and stresses continuity more than discontinuity. Consequently, a single synagogal Judaism would have taken different forms according to the local context, first pagan and then Christian. The diversity of the archaeological synagogues is a striking fact: each of them should be understood, if possible, within its historical and geographical setting, as Levine has argued in his last book, Visual Judaism. However, in the same book, Levine also holds that the synagogues and their art reflect a ‘common Judaism’, albeit different from rabbinic Judaism. Accordingly, there is no contradiction between the artistic diversity of synagogues and the existence of a single synagogal Judaism, the diversity being a part of the latter and responding to different and changing settings.

106As to rabbinization, how does it affect the synagogue? Obviously, the synagogue is one of the key places of this process. Most of the traces of rabbinization pointed out by Lapin and Schwartz are connected with the synagogue. This is also the case of the ambiguous corpora.

107Ezra Fleischer has argued that the rabbis composed the Amidah prayer after the destruction of the Second Temple. Building on this thesis, Ruth Langer claims that it took many centuries for the Amidah to spread from rabbinic circles to a wider Jewish world. The growing success of the Amidah from the fourth century onwards is related to the gradual sanctification and ‘templization’ of the synagogue. Thus, the diffusion of the Amidah in Late Antiquity is a good example of rabbinization within a synagogue setting.172 It is possible that the history of the Qedushah may be another example of such rabbinization, but this time we would be dealing with a non-rabbinic prayer that was finally accepted by the rabbis. Rachel A. Anisfeld argued that the rabbis used homiletical Midrashim to present their Judaism in a more accessible and attractive form and to spread it within the wider Judaism of the synagogues. They particularly used emotional rhetoric and emphasized the indulgence of God towards Israel.173

108Thus, I would argue that rabbinization seems to have consisted essentially in the rabbinization of the synagogues (and in the ‘synagogalization’ of the rabbis), even if the modalities of the process require further explanation.