1The Jewish nuclei of medieval Europe defined themselves— religiously, culturally and linguistically—as parts of the broader entity of a Jewish people historically anchored in the Middle East. Indeed, most of them can be traced back to the Middle Eastern Jewish populations of antiquity, although nowhere by direct evidence of actual migration, but rather by tenuous lines of ritual and literary traditions that must have been carried abroad by migrants and were often reworked into myth.1 Against this mainstream approach, a persistent strain in scholarship postulates non-Jewish origins for both Sephardi and Ashkenazi Jews, claiming that communities consist mostly of converts from other faiths, most notably the Khazars.2 The debate has not been restricted to academia, but resonates strongly with contemporary political contentions. Thus, the origins of the Ashkenazim have been tied to the fight over the legitimacy or illegitimacy of the Jewish claim to the Land of Israel. In the heat of the political argument, the issue of Ashkenazi origins has moved from the fringes into the centre of public debate, for instance, in the writings of Shlomo Sand. A similar assertion of non-Jewish origins has been made for central and northern France, where converts of Late Antiquity and the Early Middle Ages were thought to have been numerous enough to produce the substantial Jewish population emerging there in the eleventh century.3

2Running parallel to this mainstream/fringe dichotomy of views on the origin of European Jews, there is a similar one concerning the antiquity of their presence. Most scholars, the present one included, see the Jewries of northern Europe as recent—that is, ninth- or tenth-century—arrivals, with no continuity backwards to a sparse and hazy presence in Late Antiquity. In contrast, the Jews of the Mediterranean south are believed to have a much longer history, to the point that Italian Jewry has been called “millenary,” one that “has lived in one of the Diaspora countries for a millennium or more.”4 To the Italian Jews one must add their coreligionists in Byzantium. There is considerable overlap between the two in most regions of southern Italy.

3Contrary to this majority view, since the nineteenth century (and, lately, with renewed vigour), there has been a tendency to claim significant continuity of the Jewish presence not only in the Mediterranean region, but also north of the Alps, in northern Gaul and even in Germany. Here, too, present day concerns can be discerned behind scholarly opinion, for instance, the desire to present the city of Cologne as “the cradle of Ashkenazic Jewry” in a yet-to-be-built Jewish Museum that will cater to an expected torrent of Jewish tourists.5 Elsewhere, in Normandy, a sudden bloom of cultural creativity in the twelfth/thirteenth centuries could not be explained except by “a lengthy prehistory of Jewish settlement and legal rights […] apparently beginning during the period of Roman colonization of Gaul,” one thousand years earlier.6 In both cases, the evidence proffered for these assertions ranges from flimsy to non-existent.

4A short sketch of the general political and economic background of Europe might help flesh out the essential timeline. All regions of Mediterranean Europe (Byzantium, Italy, southern France, and Iberia) experienced a headlong economic and demographic crisis between the sixth and eighth centuries, in which the prime victim was the urban population, among which were, of course, the Jews. Each region variously witnessed slow (and sometimes more rapid) demographic and economic recovery from the ninth century or, in places, from the tenth century onwards.7 Politically, this means the mid-Carolingian period in France, northern Italy, and Catalonia; the Byzantine reconquista in the Mediterranean; and the formation of a Muslim state in Iberia. In terms of settlement structure, this was the time of a hesitant re-urbanization in the south—consider Venice and Amalfi— and of an altogether new semi-urban and urban formation in the north—to wit, the numerous commercial wik settlements on the northwestern seaboard and in England. It is our thesis that Jews, as with other urban and commercial elements, had little incentive to settle or expand in the crisis-ridden European regions of the first medieval centuries, but good reasons to do so from the ninth century onwards. This is indeed the picture our detailed examination of the evidence has illuminated: a double movement, consisting of re-population in the south and of immigration to the north.8 Except for Iberia, we have not found evidence for the migratory movement from the Islamic world into Europe that has been alleged in a recent book.9

5In more detail, the Jewish settlement history in the different regions of Late Antique and Early Medieval Europe can be characterized thus:10

6In the course of Roman antiquity, Jews came to make up a significant component of some town populations of the Eastern Roman Empire (later Byzantium). Of the one hundred sites of archaeological evidence for Jewish life in the Balkans, Greece, and Asia Minor, more than half are on/near the Mediterranean or Black Sea shores and on islands in the Mediterranean. Of the inland sites, the vast majority are in Asia Minor, mostly on the ancient trans-Anatolian highway leading from Smyrna (Izmir) and Ephesus on the Aegean coast via Iconium (Konya) to Mesopotamia. The archaeological evidence, though considerable, in no way supports the widely-held opinion of a vast Jewish population in the Roman Empire, a notion that has lately been subjected to well-deserved criticism.11

7Seen from the perspectives of extent and numbers of communities, medieval Byzantine-Jewish settlement never reached the Late Antique high point. For the earlier period (sixth to ninth centuries), it is difficult to decide whether only a lack of sources or an actual demographic low is reflected. I tend to accept the second view. For the later part (tenth to twelfth centuries), the decline in numbers, roughly half that of Late Antiquity, is substantial. Still, given the premise of an earlier dramatic drop in the Jewish population, such a ratio implies a remarkable recovery. The evidence also provides a further distinctive and apparently persistent feature—the migratory geography of Jewish Byzantium. It may thus be safely stated that two basic demographic phenomena mark Byzantine Jewry throughout our entire period. First, there was a continuity of Jewish presence in the Eastern Roman Empire from Late Antiquity into the High Middle Ages, though ebbing and surging at a pace apparently attuned to that of the population at large. Second, there was geographical dissemination and a migratory flow throughout the Byzantine space.

8Italy presents a complex picture. Home to a sizeable Jewish population in antiquity, especially in Rome, here, too, the beginning of the Middle Ages saw a general retreat of human settlement and of population numbers. In only a small number of locations—Rome and some towns in the south—is there a continued Jewish presence into the Middle Ages. A similar disparity between the south and the rest of the country holds for the more numerous places where Jews first settled during the Middle Ages. Many of these lay in regions ruled, to varying degrees, by Byzantium until the eleventh century. Culturally and demographically, southern Italian Jewry was much influenced, if not directly derived, from its Byzantine equivalent, and this involved some degree of migration. However, compared to Byzantium, the low rate of continuity indicates a considerable difference in the stability of the Jewish presence.12 Altogether, the number of communities everywhere in Italy is small, much smaller than in Byzantium. The ecclesiastical reformer Peter Damian (1007–1072), who spent all his life in Italy, remarked in the prologue to his treatise Against the Jews (1040– 1041) that writing such a tract is barely worth the effort, as “the Jews are now almost deleted from the face of the earth.”13 This might have been somewhat exaggerated, as witnessed by Benjamin of Tudela’s Italian itinerary a century later. Benjamin’s late-twelfth-century travels were situated in a new era of general demographic growth.

9On the Iberian Peninsula, the sparse Jewish population of Roman Late Antiquity seems to have barely survived into subsequent Visigothic times. In contrast, the data available by the late tenth and early eleventh century reflects a different order of magnitude, in terms of both the number of inhabited places and population figures. This appears to parallel the general demographic curve—upwards—in al-Andalus and is clearly linked to a more favourable Arab regime and to immigration from North Africa. These together produced a Sephardic Jewry showing no visible similarity to, and continuity with, the few indistinctive Jews of the Roman and Visigothic periods. In the Christian north, Barcelona and Gerona were the earliest places of residence (in the ninth century), due to the impetus given these parts by Carolingian colonization in Catalonia. In Leon-Castile, Jews came to be present in the capital during the tenth century and in other places in the course of the following one. In the kingdoms of Aragon and Navarra, they appear not earlier than the eleventh to early twelfth century. They came from the south of the peninsula, where a now highly intolerant Muslim regime caused significant numbers of Jews to flee to the Christian north. According to one opinion, there were also migrants from France. Population growth apparently reached its apogee towards the end of the thirteenth century and the beginning of the fourteenth, making Iberian Jewry the largest of all Europe.

10In Gaul, a transient presence can be noticed in the fourth century along the Roman borders, slightly later also in a few towns, primarily in the south. Only in Arles, Narbonne, and possibly Marseilles did Jewish habitation continue uninterrupted into the ninth–eleventh centuries. As in other parts of Europe, the post-Carolingian era saw considerable growth overall. The new communities in northern Gaul became the western branch of Ashkenazic Jewry, with congregations in the Île-de-France, Maine-Anjou, Burgundy, Champagne, Lorraine, and Normandy. As elsewhere, the greatest number and widest distribution was attained during the third quarter of the thirteenth century, surely due in some measure to immigration, which is, however, very hard to discern and could have come only from the Mediterranean south. There is little room for the hypothesis raised some years ago that the French part of Ashkenazic Jewry derived from immigrants from Germany.14 Given the very small numbers of souls in these tiny communities, population expansion is better explained by internal growth.

11In Germany, Jewish life was, for a long time, a small-scale affair wholly dependent on immigration. In the ninth century, migrants to Germany cannot have made up more than a few dozen families and, in the tenth century, maybe a few hundred. In the course of the eleventh century there was marked growth in numbers, nourished by ongoing immigration from France and, to a lesser degree, from Italy, in addition to internal demographic growth. It appears that proselytes, though present, contributed only a handful of persons to the early Jewish population.

12In Eastern Europe, the sizeable Roman-era Jewish settlement along the shores of the Danube and the Black Sea did not continue into the Early Medieval period. The earliest evidence for renewed presence—in the tenth/eleventh century—speaks of a transient one, of traders coming mostly from the west—Germany—and less frequently from the east—the lands of the Turks. These merchants crossed Bohemia, Hungary, and Poland on their way to and from Russia, but some of them were also active in the former countries. A stable resident community, possibly of Khazar origin, apparently settled in Kiev in the tenth century. In Bohemia, Hungary, and Poland, Jews inhabited individual communities in the eleventh century. Outside of these principal places, further settlement did not occur before the twelfth and thirteenth centuries. Thus, Eastern Europe was populated by Jews considerably later and at considerably lower density than all other regions of Europe. Of a possible pre-Ashkenazic stratum— Byzantine, Turkish-Khazar, or Slavonic—little can be discerned in the sources.

13To sum up: population numbers and distribution are a critical factor for realistic assessment of the weight and role of Jews in the economies and polities of medieval Europe. There can be no doubt that in Late Antiquity some groups in southern Europe—in Italy and Byzantium—were quite substantial, even though the millions proffered in scholarship do not stand up to scrutiny. In other parts, such as Spain and southern Gaul, much smaller numbers were present, while northern Gaul had few and the Roman parts of Germany had no Jews, except for some itinerant merchants or craftsmen. The administrative centre of Trier might have been an exception, and the same has been alleged for Cologne. Along the Danube border in eastern Austria and Hungary more Jews were present, in some places amounting to synagogue communities, and the same holds for the Black Sea shores and the Crimea. In Eastern Europe proper, as in Germany beyond the Roman border, no evidence of an antique Jewish presence, however slight, has been found. Not surprisingly, such existence was confined to the urban landscape of the Roman world, in its Western and even more in its Eastern parts.

14In the first centuries of the Middle Ages, Jewish life continued on a diminished scale in the Byzantine Empire. Elsewhere, the evidence dwindles to almost nothing, except for a few places in Italy and southern France. In Visigothic Spain, the total absence of archaeological finds and other evidence produced by Jews themselves is difficult to square with the spate of repressive legislation enacted by the Visigothic monarchy and church after the conversion to Catholicism. In a similar way, in northern and central Gaul, our investigation has raised doubts whether ecclesiastical literature can provide confirmation of actual Jews rather than the virtual ones serving polemical or rhetorical purposes. Everywhere, indicators point to a severely restricted Jewish population, although to different degrees in different regions.

15By the ninth/tenth centuries, new growth, slow at first and then accelerating, becomes visible everywhere. In Spain and possibly also in Sicily, it is clearly tied to the more favourable Arab regime, to immigration from North Africa, and to new links forged with the Middle Eastern centres of Jewish learning. In Italy and southern France, the factors contributing to growth are still obscure, but trends are similar. In central and northern Gaul and western Germany, the Jewish presence was a new phenomenon, wholly dependent on immigration from the south. From there it drew demographic and cultural resources, to be transmitted and transformed, with a time lag, to the north. Save for the enigmatic Khazar entity, whose Jewish character is much in doubt and which completely disappeared from the stage, the Jewish population in Eastern Europe began its growth at the very end of the period under consideration. By this time, the eleventh century, both northern and southern Jewries had come of age: part of the European landscape; strong enough to claim intellectual independence from the centres of religious authority in the Middle East; equipped with ready legal procedures to navigate a range of economic pursuits that were very different from the antique ones. In this and many other senses, the medieval Jewries of Europe represent a rupture, a new phenomenon quite dissimilar from the Greek-speaking Mediterranean Diaspora of Late Antiquity.

16In this view, the unmistakable demographic decline of the Late Antique Mediterranean Diaspora provided a clean slate for the reconstruction of a new medieval European Jewish population, one that was to exhibit a very different cultural and linguistic profile.15

17What are the implications for the topic of our conference? As I see it, the main problem is the way demography and culture interact. One example: in a talk at the Jerusalem World Congress of Jewish Studies in August 2013 entitled ‘The Origins of the Halakhic Culture of Ashkenaz: A Proposal’, Hayim Soloveitchik put forward an intriguing hypothesis.16 It flies in the face of hitherto accepted opinion, which sees southern France, Italy, and ultimately the Land of Israel as the places of origin of Ashkenazic culture: “Given their command of Babylonian Aramaic, their ignorance of and indifference to the Yerushalmi, and their exclusive preoccupation with the Bavli, the founding fathers of Ashkenazic halakhah hailed from Babylonia rather than from Palestine.” Here Soloveitchik did what most of us also tend to do unhesitatingly: to equate culture with origins. The same problem animates this conference. Rabbinization implies a process, the notion that antique, Mediterranean, Hellenistic Judaism in some way changed, developed, or morphed into rabbinic Judaism. From my point of view, if we accept that large segments of European Jewry had no antique antecedents, at least part of this assumed process of change is not really necessary. If so, we might want to examine methodically the assumption that the development of a culture necessarily needs a demographic carrier, a ‘mule’ (so to speak) on which to travel. To put it in an offhand way, with medieval Judaism being such an elite culture par excellence, did these few family groups really need more than a handful of family traditions? A second assumption to be queried says that one needs time for such changes to come into their own. Medieval European Jewish history has a number of examples where a rich local culture came into fruition within a very short time span, for instance the bloom of the German ShUM communities that took less than a century. In short, it is possible that Western European communities were never ‘rabbinized’ in the sense that these new Jewries were of the rabbinical persuasion from their very outset. This would still leave the question: where, or rather how, did they acquire this cultural profile?