A LANDSCAPE TRANSFORMED
Cities
THE LANDSCAPE of high imperial Palestine was dominated by the pagan city. It was there that wealth was concentrated, and there that patterns of expenditure generated by the Greco-Roman ideology of euergetism resulted in the production of monumental structures and public writing.1 These were rare outside the city, but where they existed they were unambiguously derived from urban models—bathhouses and basilicas, marked with inscriptions in Greek. In Syria, large villages had long been important, and in the high empire, many aspired to be cities of the standard Greco-Roman type.2 Smaller villages on the whole lacked monumental construction and public writing, except occasionally for monumental grave complexes.
By the fifth and sixth centuries, the landscape had been transformed. The cities remained important; indeed, some of them grew. Aelia Capitolina, a backwater in the high empire, became, as Christian Jerusalem, a metropolis with a population estimated at 50,000–80,000. Negligible desert settlements, such as Sobata, Mampsis, and Nessana, grew and some of even became cities—small (Elusa, the largest of the Negev settlements and the only one among them that was unambiguously a city, is thought to have had a population of about 10,000) and chaotically laid out, but cities all the same.
Jerusalem and the Negev cities were unquestionably anomalous. The tremendous growth of the former was obviously due to its importance in Christian theology; it was second only to Constantinople as an ecclesiastical center. The growth of the Negev cities is more problematic: recent surveys show that the northern Negev as a whole was surprisingly densely inhabited in the fifth and sixth centuries, and probably later. The well-known sites were all surrounded by villages and farmsteads, and winepresses are extremely common. The area as a whole is comparable to the limestone massif of northern Syria, another agriculturally marginal region that flourished under the later Roman Empire at a time when the density of population in adjacent, rainier areas was at its maximum.3 The presence at Nessana, which was perhaps no more than a village, of a caravansarai with ninety-six beds suggests something about the role of a possibly expanding commerce in the growth of the region.4 It may at least be suggested that cultural changes (e.g., the sedentarization of desert tribes perhaps connected to their christianization) played a role, too.5 Evidence for the growth of other Palestinian cities is more ambiguous: the walls of Caesarea and apparently Tiberias were extended, some formerly uninhabited districts of Scythopolis and Sepphoris became residential, and so on. But the consensual view that on the whole the urban population of Palestine increased in the late empire, an aspect of a general increase in population, seems entirely plausible.6
Less controversial is the transformation in character and physical appearance of the cities: where they had once been pagan in appearance, they were now Christian, though in many places older structures continued to be maintained and used. The Palestinian cities thus participated in changes that occurred throughout the eastern empire. Pagan temples were destroyed or turned into churches, theaters and amphitheaters fell into ruin or were filled in with market stalls or private houses, agorai and other public spaces became cramped bazaars and residential districts.7 Monumental construction certainly did not cease and was not restricted to churches. The Tiberians built a new bathhouse in the fourth century, and the Sepphorites built lavishly decorated public buildings and a classical-style colonnaded street in the fifth.8 But in general there was massively more public expenditure on religious buildings, and, conversely, much less expenditure on other types of public buildings than there had been previously.9
Though some of the Palestinian cities were notorious, at least among church fathers, for their resistance to Christianity—Gaza, Raphia, and Petra because of their inveterate paganism, and Tiberias and Sepphoris because of their Judaism—they all eventually became Christian or, rather, Christians came to predominate and the bishops to play leading roles in municipal affairs.10 Tiberias and Sepphoris are not known to have had bishops before 449, but the partial reconstruction of the latter in the fifth and sixth centuries was in part a monument to the glory of the Church—so an inscription informs us.11 Nevertheless, the coastal and Decapolitan cities retained some of their traditional diversity. Their populations included, in addition to Christians, Jews and probably dwindling numbers of Samaritans and pagans until the Muslim conquest.
Tiberias and Sepphoris, too, retained Jewish populations. Tiberias was of course the home of the patriarchs until the 420s, which obviously inhibited the spread of Christianity there, notwithstanding Epiphanius’s fantasy about the patriarchs’ crypto-Christianity. It apparently remained the center of the rabbinic movement in Palestine, and of the explosive literary production (the Palestinian Talmud, the Midrash collections, the piyyut, etc.) associated with it until and even for several centuries after the Muslim conquest.12 Whether the Jews remained a majority in this period is unknown. It is surely noteworthy, though hardly probative, that the only synagogue remains so far discovered are in the northern and southern outskirts of the city (though the Talmud mentions a kenishta deBoule, presumably located in the center of the city).13 Similarly, the insistence of the excavators that Sepphoris retained a Jewish majority under the late empire is baseless (though not necessarily false).14
Only during this period can we be certain that the Jewish inhabitants of the cities built synagogues. Though it would be perverse to doubt that there were synagogues in Tiberias, Sepphoris, and Caesarea earlier, the earliest archaeological remains of an urban synagogue—perhaps of any synagogue—in post-Destruction Palestine are from Hamat Tiberias, probably constructed in the early or middle fourth century.15 Rather surprisingly, the urban synagogues (unlike the prayer house of first-century Tiberias described by Josephus, Life 277) are invariably small structures, frequently situated in remote or inconspicuous locations. The synagogue complex of Hamat Tiberias measures only 13 by 15 meters, the newly discovered synagogue of Sepphoris is 20 by 8 meters, and the main hall of one of the largest, that of Hamat Gader, measures 13 by 13.9 meters—approximately the same size as most rural synagogues.16 No counterpart has yet been discovered in Palestine to the massive, centrally located synagogue of Sardis.17
The urban synagogues did differ from the rural ones, though: there seem often to have been more than one per settlement (two are known from Tiberias, Sepphoris, and Scythopolis), and they occasionally used Greek in inscriptions.18 Furthermore, urban synagogues were often much more lavishly decorated than rural ones: they tended to make more use of marble, and several, most strikingly those of Hamat Tiberias, Sepphoris, and Gaza, are decorated with magnificent figurative mosaic pavements. Nevertheless, they seem on the whole contemporaneous with the rural synagogues, and apparently they did not serve as models for the latter, unlike urban churches, which were imitated by architects in the Christian countryside. Though the essential structure and many of the motifs of the pavement of the synagogue of Hamat Tiberias were taken up in later synagogues, there is no evidence of direct imitation. The ideological underpinnings of the urban synagogues, as revealed in the inscriptions, likewise bear comparison with those in the countryside (see below). There is, in sum, no physical evidence for a city-based hierarchy of synagogues in late antique Palestine, a point that will be discussed in more detail below.
Villages
The transformation of the landscape of rural Palestine in late antiquity was if anything even more striking than that of the cities. The countryside was now packed with nucleated settlements, ranging in population from several hundred to several thousand inhabitants. If the surface surveys may be trusted (and though they are questionable in every detail, they may cumulatively paint a plausible picture of the true situation, especially since they correspond so well with the well-attested situation in northern Syria), many more such villages existed in late antiquity than at any earlier period.19
One novel characteristic of these villages was, if we follow a recent suggestion of Benjamin Isaac, that while as late as the early mid-fourth century they were commonly inhabited by mixed populations of pagans, Jews, and some Christians, there was a tendency later toward religious separation.20 Most villages came to be either Jewish or Christian, or perhaps pagan. As early as the 370s Epiphanius could regard the largest settlements of Galilee, Tiberias, Sepphoris, and Capernaum as exclusively Jewish, though this is likely to be a polemical exaggeration.
By about 500, the pattern of separation was clearly defined, and it is confirmed by another fundamentally important novelty of the late antique rural Palestinian landscape: all but the very smallest villages had at their center, often set in a paved square, a monumental, purpose-built, stone religious building.21 In fact, villages usually had either churches or synagogues, not both. There were exceptions. Some were like the extremely large village of Capernaum, located in the heart of Jewish Galilee, but also a Christian pilgrimage site. Others were Jewish villages that became Christian, perhaps like er-Rama, in western Galilee.22 A very few villages had two churches or two synagogues (Kefar Baram, for example), probably the result of the synoecism of two originally separate settlements.
The precise numbers of churches (there were many hundreds)23 and synagogues (commonly said to be around 100 or 120)24 indicated by the remains are impossible to establish because in many cases the remains are merely suggestive. For example, many of the Upper Galilean sites surveyed by Z. Ilan yielded only fragments of monumental structures. Some or all of these could indeed have been from synagogues, as Ilan assumed, especially since earlier observers had documented what they regarded as the remains of synagogues at many of the sites. But many could just as well have been from churches, or the large stones may have been brought from elsewhere for use in later structures.25 Notwithstanding the many doubtful cases, though, patterns of geographical distribution emerge clearly: the churches were heavily concentrated in Judaea and in the district of Akko-Ptolemais, northwest of Upper Galilee, while the synagogues were concentrated in eastern Galilee and western Golan. In western Galilee, the Carmel region, the Beth Shean Valley, and southern Judaea, there were looser concentrations of both churches and synagogues.26
Dates
There were certainly many more such structures than have been discovered. Given the small size of the settlements in which some of the buildings were found, it seems reasonable to follow the consensus of archaeologists and historians in concluding that by the later fifth or early sixth centuries, almost every village in Palestine, except perhaps the very smallest, had either a church or a synagogue.27
Unlike the Jews, the Christians tended to record the dates of construction and renovation of their religious buildings in dedicatory inscriptions. (The Jews’ general failure to date synagogues and graves is a puzzling fact that will not detain us here.) So we know that the great age of rural church construction in Palestine was the middle and later fifth century, with considerable activity continuing through the sixth. The buildings were remarkably uniform—overwhelmingly apsidal, and later triapsidal, basilicas, though a small number were octagonal. In both cases, the models were the great Constantinian urban and pilgrimage churches.28
By contrast, the dating of the synagogues is, as I have already suggested, controversial and their architecture was, uncontroversially, highly varied.29 What is often forgotten is that the controversy about dating attaches to only one category of synagogues. Otherwise, there is general agreement on several important issues. First, the basic version of the old tripartite typology (Galilean/“broadhouse”/apsidal basilica) elaborated by Avi-Yonah from the work of Kohl and Watzinger and Sukenik can no longer be maintained, since many of the synagogues discovered since the 1960s have failed to conform to any of the old categories. Second, even those who retain an early dating for the “Galilean” type do not deny a late antique date for the other types, and presumably would not hesitate to depend on stratigraphy and other ostensibly objective criteria to date synagogues that conform to no type. In sum, even followers of Avi-Yonah would have to admit, if they applied their own methods rigorously, that the synagogue did not reach its maximal diffusion until the fifth or sixth centuries. In practice, though, Israeli scholars, who are now almost alone in believing that the Galilean synagogues were built in the later Severan period (and even among them consensus is breaking down; see below), tend to regard them as already an essential component of the Jewish village.30 The matter is thus worth a brief discussion.
One of the most serious problems with the old chronology was a piece of information long known but easily dismissed because of its uniqueness. A dedicatory inscription carved on the meticulously classicizing lintel of the supposedly early synagogue at Nabratein unambiguously dates the construction—not restoration—of the synagogue to the 494th year after the Destruction, that is, c. 562.31 This fact, as Naveh long ago observed, should have caused archaeologists more unease than it apparently did.32 Recently, in another blow to the old typology all the more serious for having been delivered by an Israeli archaeologist, concerning synagogues regarded as “early” (midlate third century) by their American excavators, it has been argued that the synagogues of Gischala and Horvat Shema were actually built in the fourth century.33
There is no question, though, that the major blow to the early dating of the Galilean type synagogue was the discovery of thousands of low-denomination bronze coins of the fourth and early fifth centuries beneath the floor of the synagogue of Capernaum—a flagship example of the type. In fact, such deposits (which must be distinguished from treasury hoards) are commonly found in the excavation of ancient synagogues; whatever their purpose may have been, they were certainly intentional.34 In most such cases, archaeologists have not hesitated to draw the obvious conclusion: the synagogue was constructed after the date of the latest coin in the deposit. But deeply ingrained conservatism and loyalty to Avi-Yonah as chief representative of the new tradition of Israeli archaeology (combined perhaps with a mild suspicion of the Franciscan excavators—a suspicion in fact justified by the questionable quality of much of their work) were not the only reasons, important as they were, for the Israeli archaeologists’ denial of the obvious in the case of Capernaum.
In a major recent article, Z. U. Maoz offered three kinds of argument in support of the old chronology.35 The first two, from stratigraphy and history, need not detain us. His stratigraphic arguments for dating synagogues like those of Meron and Gischala to the early third century appear at first glance, at least to the nonarchaeologist, to have a certain rough plausibility; but in fact they are only rationalizations, amounting to no more than the claim that the stratigraphy can be reconciled with a dating to the early third century, not that it makes such a dating likely. Maoz’s argument about Capernaum—that the synagogue was built elsewhere in the early third century and then moved, stone by stone, to its present site in the fifth—is special pleading36 (though I have recently heard it publicly praised, if not precisely endorsed, by Mordechai Aviam). The argument from history may be unhesitatingly dismissed, depending as it does on the supposition that Rabbi Judah I was in effect king of the Jews—a supposition derived from a misreading of both rabbinic literature and the works of Origen. The final argument, from style, is more compelling. The resemblance of the Galilean synagogues to buildings constructed in the second and third centuries, especially the southern Syrian shrines surveyed by Butler around the turn of the century, is undeniable, and Maoz fortifies this main point with a good deal of circumstantial detail, the truly novel aspect of his article.37 But even this is unconvincing in the end. Why should the resemblance of the synagogues to the Syrian shrines prove that they were built at the same time? The shrines, after all, probably remained in use into the fifth century, and some of them are standing to this day. Some Galilean villagers, when they decided to build synagogues, may have wished them to look appropriately “sacred.” Before the fifth century, the only available models were the shrines built (and still maintained) by the pagan villagers in the neighboring districts; monumental churches, later an important model for synagogue builders, were still rare. Once in use for synagogues, there is no reason the Galilean style should not have remained in use through the sixth century.38 An important component of the old chronology, it is worth remembering, is a view no longer tenable since the emergence of late antiquity as a discrete field of study (and therefore lives on as a prejudice), that is, that high-quality “classical” architecture and design were impossible after the middle of the third century and that therefore high quality implies an early date, and crudity a late date. In reality, apparent differences in quality tell us more about social and economic differences, or differences in esthetic sensibilities or religious predispositions, or about the prejudices of scholars, than about chronology.
A Booming Economy?
Since the classic study of G. Tchalenko, it has been commonplace to explain the explosion in construction in Syria and Palestine in late antiquity by supposing that there was an economic boom between the fourth and sixth centuries; more recently it has been argued that the era of prosperity continued even beyond the middle of the sixth century, to the very eve of the Muslim conquest.39 Tchalenko argued that the boom was related to the rise of olive monoculture in some areas, but this view now needs to be qualified.40 It has more recently been suggested that the boom was fueled by developments outside the traditional eastern Mediterranean economy, especially in Arabia, the evidence for which is that pre-Islamic Arabic poems sometimes mention products imported from Syria.41 The influence of the rise of Constantinople on the Syrian economy should also be considered.
In addition to factors that affected Syria generally, Palestine was the beneficiary of Christian theology, which led to extensive investment by the emperors and other grandees in ecclesiastical and monastic foundations. But Avi-Yonah was right to be equivocal about the effects of Christianity on the Palestinian economy. Much Christian investment had only short-term consequences; for example, church construction created only brief spurts of demand for labor. And much investment was absolutely unproductive: all the silver plate deposited in the great churches came ultimately out of the taxpayers’ pockets and benefited only the invaders who plundered it.42
The remains of churches and synagogues are among the most important evidence for a boom in northern and central Palestine. In the aggregate, the construction seems to imply the availability of unexpected quantities of surplus in what we might have supposed was a standard rural subsistence economy. And both churches and synagogues were often built and decorated—sometimes surprisingly lavishly—by artisans from urban workshops who probably expected to be paid in cash.43 For example, though marble fixtures and mosaic pavements were more common in urban than in rural synagogues, they were not unknown in the latter. Marianos and Aninas, who signed their names to the mosaic of the Bet Alfa synagogue in Greek, also made part of the floor of the Samaritan synagogue of Scythopolis: probably they were based there.44 Likewise, Rusticus, the stonemason who built the synagogue of Dabbura, in the western Golan, signed his name in Greek and is likely to have come (despite his name!) from Caesarea Philippi or Akko (Naveh, On Mosaic, no. 7). This would imply the easy availability of not only surplus but also gold coins. A surprising level of monetization, which conforms with Tchalenko’s hypothesis that the economy of rural late antique Syria was heavily marketoriented, is also implied by the coin “treasuries” (which may in fact be private funds) discovered in at least thirteen synagogues.45
But we should not underestimate the extent of communal exertion involved in the construction, and the likelihood that it was often done piecemeal over many years. Here, the mosaic of the Bet Alfa synagogue is once again instructive, for the dedicatory inscription on it states, if it has been correctly read, that the villagers had to collect and sell one hundred se’in of wheat in order to pay the mosaicists, and this in a village only six kilometers—an hour’s brisk walk—from Scythopolis, the capital of Palaestina Secunda.46 As to the synagogue treasuries, though some are very rich, most of them contain only bronze coins, which had no stable value. In some cases the treasury hoards were the result of centuries of collection.47
In sum, it is not unlikely that late antiquity was a period of unprecedented prosperity in Palestine, as in Syria. But this alone cannot explain the burst of synagogue and church construction. Contrary to what some scholars have assumed, there is no reason to think that most Jewish villagers before the fifth century regarded a synagogue as an essential component of their settlement. And the Bet Alfa mosaic at least warns us that even in late antiquity, the construction and decoration of a monumental building was no trivial matter for a village. The source of surplus needed for the construction of the synagogues may well have been a strong economy, but the choice of how to use it has no necessary connection with the economy per se.
1 See G. Woolf, “The Roman Urbanization of the East,” in S. Alcock, ed., The Early Roman Empire in the East (Oxford: Oxbow, 1997), pp. 1–14.
2 See Millar, Roman Near East, pp. 17–23.
3 See G. Tate, Les campagnes de la Syrie du Nord I, Institut français d’archéologie du procheorient, Bibliothèque archéologique et historique 133 (Paris: Paul Geuthner, 1992).
4 See C. J. Kraemer Jr., Excavations at Nessana, vol. 3, Non-Literary Papyri (Princeton: Princeton University Press, 1958), no. 31; on trade, see the discussion on pp. 26–28.
5 On the Negev settlements, see the important discussion of C. Foss, “The Near Eastern Countryside in Late Antiquity,” JRA suppl. 14 (1995): 225–31.
6 See the surveys of Palestinian cities in Y. Tsafrir, Eretz Israel, pp. 317–32; Y. Dan, The City in Eretz-Israel during the Late Roman and Byzantine Periods (Jerusalem: Yad Ben Zvi, 1984), pp. 51–68. In general, Dan assumes more growth and provides slightly higher population figures, but his disagreements with Tsafrir are minor. For Sepphoris see, pending publication, the annual surveys in HA in the 1990s, and note also the comments of Netzer and Weiss, Promise and Redemption, pp. 9–10. For Tiberias, see the report of Y. Hirschfeld in HA 104 (1995): 32–38, and the discussion of Hirschfeld and G. Foerster in NEAEHL, s.v.
7 However, see Foss, “Near Eastern Countryside,” 226, who revives the view that urban planning broke down only in the seventh and eighth centuries.
8 On the Tiberian bathhouse, see Y. Hirschfeld, “Tiberias,” in NEAEHL 1466–67; on Sepphoris, Z. Weiss, “Sepphoris,” in NEAEHL 1327.
9 See Tsafrir and Dan; on the transformation of the late antique city, see M. Sartre, Bostra: Des origines à l’Islam (Paris: Paul Geuthner, 1985), pp. 119–39; H. Kennedy, “From Polis to Madina: Urban Change in Late Antique and Early Islamic Syria,” Past and Present 106 (1985): 141–83; M. Whittow, “Ruling the Late Roman and Early Byzantine City”; A. Zeyadeh, “Urban Transformation in the Decapolis,” Aram 4 (1992): 101–15.
10 See Dan, City, pp. 14–17, on paganism; 90–102, on the role of the bishops; on the Jewish cities, Epiphanius, Panarion 30; Theodoret of Cyrrhus, Church History 4.22 = GCS 44.260.
11 On the bishops, see Avi-Yonah, Jews of Palestine, p. 168; contra Z. Rubin, “Joseph the Comes,” Epiphanius, Panarion 30. 4.5 (= GCS 25.339) is very careful not to say that Tiberias had a bishop, either, it would seem, at the time of the writing in the 370s, or of the dramatic date of his story, in the reign of Constantine. On the reconstruction of Sepphoris, see E. Netzer and Z. Weiss, “Sepphoris, 1991–2,” IEJ 43 (1993): 190–96; the Duke University excavators claimed that the faunal remains from the city changed from mainly sheep and goats in the high empire to mainly pigs in the sixth century: E. Meyers, C. Meyers, and K. Hoglund, “Sepphoris,” IEJ 45 (1995): 68–71.
12 See R. Brody, The Geonim of Babylonia and the Shaping of Medieval Jewish Culture (New Haven: Yale University Press, 1998), pp. 100–109.
13 Kenishta deBoule: Y. Sheq. 7:5, 50c (the story assumes the central location of the synagogue), Y. Taan. 1:2, 64a; synagogues of Tiberias: Hirschfeld, “Tiberias,” 1468–70.
14 See Weiss and Netzer, Promise and Redemption, p. 9.
15 See Dothan, Hammath Tiberias, pp. 66–67.
16 For a full list of the dimensions of the synagogue remains, see Hachlili, Ancient Jewish Art, p. 148.
17 See A. T. Kraabel, “The Diaspora Synagogue: Archaeological and Epigraphic Evidence since Sukenik,” ANRW 2.19.1, pp. 483–88.
18 Contrast Naveh’s mainly rural corpus of Aramaic inscriptions and Lifshitz’s mainly urban corpus of Greek inscriptions.
19 See M. Kochavi et al., Judaea, Samaria, and the Golan: Archaeological Survey, 1967–68 (Jerusalem: Keter, 1972), and the important discussion in H. Lapin (forthcoming) containing a complete survey, analysis, and criticism of surface surveys of late antique material in Israel and the territories; see also Y. Hirschfeld, “Changes in Settlement Patterns of the Jewish Rural Populace before and after the Rebellions against Rome,” Cathedra 80 (1996): 3–18—noting the scattered evidence for farmsteads in late antique Judaea and Samaria, and the absence of such evidence for Galilee (implying, Hirschfeld believes, that the Jewishness of the Galileans, which meant that they were organized in communities, required them to live in nucleated settlements). This is an interesting phenomenon, which requires some sort of explanation. But Hirschfeld overlooks the obvious point that even in Judaea most people lived in villages, not on farmsteads. Note also J. Pastor’s point in response to Hirschfeld that the evidence for late antiquity is continuous with that of the Second Temple period: farmsteads were never common in Galilee and never absent from Judaea and Samaria (“Why Were There No Jewish Farmsteads in Galilee?” Cathedra 84 [1997]: 175).
20 B. Isaac, “Jews, Christians, and Others in Palestine: The Evidence from Eusebius,” in Jews in a Graeco-Roman World, pp. 65–74; on the separation of Jews and Christians, see M. Aviam, “Christian Settlement in Western Galilee in the Byzantine Period” (master’s thesis, Hebrew University, 1994).
21 It is less clear whether Galilean villagers, like those in northern Syria in the later fifth century, also built substantial, often heavily decorated stone houses—one of the striking characteristics of the region surveyed by Tchalenko. The evidence of the Meiron excavations suggests that on the whole they did not, but Meiron may have been abandoned before 500, when Syria reached the peak of its prosperity (E. Meyers et al., Excavations at Ancient Meiron, Upper Galilee, Israel, 1971–72, 1974–75, 1977 (Cambridge: ASOR, 1981), pp. 50–65; 158–61, on the abandonment of the site). In general, Galilee gives the impression of having been less prosperous than the northern Syrian limestone massif, but this impression may be false: the limestone massif was a marginal area, inhabited mainly under Roman rule and abandoned once and for all in the early Middle Ages. Galilee has been continuously inhabited, and ancient stone structures are in general unlikely to have survived later inhabitants’ need for building materials.
22 On the christianization of some Jewish villages in southern Judaea, see J. Schwartz, Jewish Settlement, pp. 107–9.
23 Ovadiah lists 265 sites in NEAEHL, but many of these were pilgrimage sites and monasteries that contained more than one church.
24 See Levine, “Synagogues,” in NEAEHL.
25 See Z. Ilan, “Survey of Ancient Synagogues in Galilee,” EI 19 (1987): 170–207.
26 See most conveniently the maps in Y. Tsafrir, L. DiSegni, and J. Green, Tabula Imperii Romani: Iudaea-Palaestina (Jerusalem: Israel Academy of Sciences and Humanities, 1994). Similar maps appear in NEAEHL.
27 E.g., Y. Tsafrir, in Ancient Churches Revealed (Jerusalem: Israel Exploration Society, 1993), p. 4; D. Urman, The Golan (Oxford: BAR, 1985), p. 93.
28 Tsafrir, Ancient Churches Revealed, pp. 2–16. There is no corpus of (the hundreds of) Palestinian church inscriptions, though they will presumably be included in the data-base/corpus now being prepared at Tel Aviv University. Meanwhile, most convenient, though seriously out of date, is M. Avi-Yonah, “Mosaic Pavements in Palestine,” in Art in Ancient Palestine, pp. 283–382, supplemented by A. Ovadiah and R. Ovadiah, Hellenistic, Roman, and Early Byzantine Mosaic Pavements in Israel (Rome: L’Erma di Brettschneider, 1987). For a discussion of the dated inscriptions, see L. Di Segni, “Epigraphic Documentation on Building in the Provinces Palaestina and Arabia, Fourth-Seventh Centuries,” in J. H. Humphrey, ed., The Roman and Byzantine Near East, (JRA suppl. 31 (1999)): 149–78.
29 A few synagogues are dated: that at Nabratein to 562 (see below); Gaza, 508 (Lifshitz, Donateurs 73a = CIJ 2.967); an inscription from Ascalon dates a gift to the synagogue there to 604 (Lifshitz, Donateurs 70 = CIJ 2.964); the mosaic pavement at Bet Alfa was made in the reign of Justin (Naveh, Al Psefas 43); the reasons given by most commentators for preferring Justin I (reigned 518–527) over Justin II (567–578) are inadequate. For a convenient set of plans of Palestinian synagogues, see Hachlili, Ancient Jewish Art, pp. 144–47.
30 E.g., Z. Safrai, Jewish Community. Strikingly, even L. Levine, who rejects the old typology, nevertheless regards it is indubitable that the synagogue functioned as the central institution in Jewish communities (i.e., presumably, towns) “everywhere” in the second and third centuries: The Ancient Synagogue (New Haven: Yale University Press, 2000), pp. 171–72. Even among scholars who retain the old typology, “early” has become subtly later in the past few decades; one no longer reads about second-century synagogues. It should also be noted that consensus is beginning to dissolve even among Israeli scholars: see below for E. Netzer’s recent redating of several Galilean-type synagogues to the fourth century. R. Hachlili, a prominent follower of Avi-Yonah in most respects, has now discarded his chronology of the synagogues in favor of an approach based on regionalism; see her contribution to S. Fine, Sacred Realm: The Emergence of the Synagogue in the Ancient World (New York: Oxford University Press, 1996), pp. 98–111. But the majority of Avi-Yonah’s students in Israel remain faithful, for reasons that seem to me to have more to do with the sociology of the Jewish studies establishment there than with anything else.
31 It seems to me likely that the inhabitants of Nabratein would have used an erroneous chronology like that of the Seder Olam Rabbah, rather than the more accurate tradition preserved by the Christian chronographers, to date the destruction of the Second Temple. For discussion of Nabratein, see Levine, Ancient Synagogue, pp. 298–99.
32 On Mosaic, p. 31, for the inscription; p. 4, for discussion of dating. Naveh rather precociously regarded the old typology as discredited already in this book.
33 See E. Netzer, “The Synagogues in Gischala and Khirbet Shema: A New Look”, EI 25 (1996): 450–54. I have recently heard it argued by the American archaeologist Jody Magness that these synagogues were built in the sixth century.
34 See, e.g., Z. Ilan, “The Synagogue and Bet Midrash of Meroth,” in R. Hachlili, ed., Ancient Synagogues in Israel, Third through Seventh Centuries CE, BAR International Series 499 (Oxford: Tempus Reparatum, 1989), pp. 21–41, esp. p. 27
35 “When Were the Galilean Synagogues First Built?” EI 25 (1996): 416–26.
36 This argument has now been published in English: “The Synagogue at Capernaum: A Radical Solution,” in Roman and Byzantine Near East 2: 137–48.
37 Much of which, though, consists of results of studies of architectural features that posit excessively specific chronologies of the sort one would have imagined long since discredited.
38 As even Y. Tsafrir admitted, in an earlier attempt to salvage the old chronology; see “The Synagogue at Meroth, the Synagogue at Capernaum, and the Dating of the Galilean Synagogues: A Reconsideration,” EI 20 (1989): 337–44. But he regarded the definitely late examples, e.g., at Meroth, to be characterized by crudity of workmanship, a view refuted by the synagogue of Nabratein.
39 See G. Tchalenko, Villages antiques de la Syrie du nord, 3 vols. (Paris: Paul Geuthner, 1953–1958); for reservations about some of Tchalenko’s conclusions, see G. Tate, “La Syrie à l’époque byzantine: Essai de synthèse,” in J.-M. Dentzer and W. Orthmann, eds., Archéologie et Historie de la Syrie ii (Saarbrücken: Saarbrücker Druckerei und Verlag, 1989, pp. 96–116. For arguments that the boom lasted into the seventh century, see M. Whittow, “Ruling the Late Roman and Early Byzantine City.”
40 Tate, Les campagnes, demonstrated that the limestone massif had been inhabited since the first century, that its economy depended heavily on grain culture and livestock, and that the villages in the region were more traditionally rural—more agriculturally oriented, less planned, less economically diverse—than Tchalenko had supposed. Nevertheless, he too attributed the burst of construction in the late fifth century to the production of olive oil, marketed not internationally but mainly to the surrounding cities.
41 See Sartre, Bostra, pp. 119–39.
42 See M. Avi-Yonah, “The Economics of Byzantine Palestine,” IEJ 8 (1958): 39–51.
43 Gifts commemorated in inscriptions are almost always in cash or at least are evaluated in cash.
44 See L. Roth-Gerson, Haketovot Hayevaniot Mibattei Keneset Be’eretz Yisrael (Jerusalem: Yad Ben Zvi, 1987), p. 33.
45 See A. Kindler, “Donations and Taxes in the Society of the Jewish Villages of Eretz Israel during the Third to Sixth Centuries CE,” in Ancient Synagogues, pp. 55–59.
46 See Naveh, On Mosaic, p. 43, for this interpretation of the inscription.
47 The largest hoard, at Meroth, contained 485 coins, of which 245 were gold, of various denominations, with a value estimated at 17,874 folles, or 2,235 man/days of work. The bulk of the coins are dated from the reign of Anastasius (491–518), to that of Phocas (d. 609), six are from the fourth century, and additional coins were added to the treasury until 1193. See Ilan, “Synagogue and Bet Midrash of Meroth,” 30–31.