In Russian
Contents Huns
Contents Tele
Contents Alans
Ogur and Oguz
Tengri, Khuday, Deos and God
Crescent And Star
Türkic Islam

Buddism, Nestoriansm, Islam

Russian Version needs a translation
Alan Dateline
Avar Dateline
Besenyo Dateline
Bulgar Dateline
Huns Dateline
Karluk Dateline
Khazar Dateline
Kimak Dateline
Kipchak Dateline
Kyrgyz Dateline
Sabir Dateline
Seyanto Dateline
Peter B. Golden

The World of the Khazars, Leiden, Boston, 2007, ISSN 0169-8524, ISBN 978 90 04 16042 2
© Copyright 2007 by Koninklijke Brill NV, Leiden, the Netherlands.


Adji M. Türks and Judaism
Golden P. Khazar Conversion
Elhaik E. Ashkenazim and Türks


W Huns
5th c. AD
6th c. AD
6th c. AD
W. Goktürk Kaganate
7th c. AD
Subar anabasis Kurbat Bulgaria
AD 630-660
ca AD 650-850
Kangar anabasis
AD 750
ca AD 800
per Gumilev
AD 850
AD 900
Kagan Domain
10th c. AD
Itil Bulgaria
ca AD 800
AD 900
AD 950.gif (92K)
AD 1050.gif (92K)
Introduction 123
The anatomy of conversions 125
Khazar Tengriism 131
Marwân forced Qagan to convert to Islam 137
Conversion to Judaism 139
Dating and Impact of the Conversion 151
Internalization 157
Some Conclusions 158

 Posting Foreword

Nobody bothers to delve into the conversion of the Khazars to Islam or Christianity, but the Khazars' conversion to Judaism is a perennial subject of heightened interest. Fortunately, the facts on that conversion require an abundance of the background information, and thus the works unwittingly enrich the Middle Age history in general, and Turkology in particular. This effect is not specific to particular researchers, its mirror image is the focus on Ottoman history that unwittingly provides a wealth of information on the Middle Age history and Turkology in general. The best contributors to the Türkic history were the scholars exploring self-centered research, wherever is the center.

The Khazars were a small tribe, at the most Khazars numbered 5% of the Kaganate population, covered under an umbrella supraethnic, or a politonym appellative “Khazars”. Somehow researchers tend not to define the term “Khazars”, allowing readers and other researchers to conclude what they please. The loose terminology makes the very subject or event of conversion indistinct. In the most general terms, “conversion” refers to a process of transition from a prior state to a later state, where the prior state is an ubiquitous sprawl of Tengriism among all Türkic people, from the Scythians, Cimmerians and Sarmats on the Atlantic seaboard to the Huns at the Ordos bend, and the later state is a spread of Judaism on the Khazar island of the same Türkic people in a greatly diminished geographical space. The initial stage is not contaminated by alternate influences, since the Huns started rolling their centers westward before the advent of the later religions. An exception to this state was a presence of the miniscule Sarmato-Greek Bosporan kingdom, which may have touched on the local Onogur Bulgars in its sphere of influence. Contamination started in the 4th c. with the advent of the Armeno-Agvan Christianity, then in the 7th c. with the advent of Islam. Gy. Moravcsik asserts that the Greek Chrisyianity made inroads into the


Page numbers are shown at the beginning of the page in blue. Posting notes and explanations, added to the text of the author and not noted specially, are shown in (blue) in parentheses and in blue boxes, or highlighted by blue headers. Diacritics may need verification against the original.

Peter B. Golden



The Khazar conversion to Judaism, an unusual, but not unique occurrence, must be viewed within several larger contexts. The broadest of these was the conversion of the nomadic, steppe peoples to one or another of the universal religious systems (Judaism, Christianity, Manichaeism, Islam or Buddhism). This was a process that was sometimes, but not exclusively, associated with the further growth and articulation of state structures. Although a divine mandate to conquer and rule was often part of the Eurasian nomads’ imperial ideology, they did not create any of the great world religions. For these they turned to the sedentary cultures with which they came in contact. Indeed, this is a further example, it has been argued, of their dependence, in the spheres of both material and spiritual culture, on sedentary society. Their choice of religion often grew out of very immediate, “mundane (rather) than spiritual considerations,” combined with “sound political sense.” Conversion of the elites, especially the ruler, almost always preceded large-scale conversions.1

In the lingo of the “great world religions” or “universal religions”, religion is an institution, a hierarchical edifice with leaders and flocks. The concept was created by would be and de-facto leaders, solely for the convenience and utility of the leadership. Other than to persuade the flocks, it does not serve the flocks, and satisfies only the needs induced among the flock by the same leadership. The primitiveness of the hierarchical ideology is amply illustrated by the uniforms worn by the members of the hierarchy, and the rituals that are depicted as the substance of the religion. Remove the superficial uniforms and rituals, and the remnants would be the substance called “religion”. Distinctions between superficiality and substance bothered “flocks” from the first days of every religion, great and small, and were one of the main causes for conflicts and fragmentation of every organized religion.

To call a religion “great” because it wielded great sticks and clobbered more than the others is not objective in any measure. Our consciousness has changed drastically, yesterdays missionaries today are called terrorists and treated as such. The sticks are no more viewed as potent conversion tools, they are universally disdained. From today's viewpoint, a religion that was embraced by great many peoples in Eurasia, that in its heyday millennia covered Eurasia with kurgans from the Qin imperial grave by Pacific to the Newgrange megalithic kurgan in the Ireland by Atlantic, that gave an etiological view to the Eurasian populations, and that kept guiding their daily life, and that is recited in ine form or another in the holy books of the today's “great world religions” can't be casually dismissed. The religion did not changed; it had gained new packagings, new labels, renewed lingos, new generations of salesmen, in places it replaced the phantom reincarnation with a phantom eternal life in a phantom eternal paradise, but the main idea of the “other world” and the eternal life remains the substance of the “great world religions” and the only reason they have their flocks. Not the packaging, not the dogmata hold the flocks who easily change dogmata with their passports, but the Tengrian beliefs in afterlife and practical methods to achieve it inspire imagination and induce hope. No organized religion at any era managed to suppress a common desire to retain the substance without their organized institutions.

It is quite natural that the politics of religion is sold as the religion of politics. Confusing institution with religion, and politics with institution was an eternal effective tool of salesmanship. The real difference was that the sedentary population was a captive audience, malleable by conviction with a stick, while the mobile nomads were not stick-convincible, the marketers had to use gentle persuasion.

A somewhat narrower context is that of the struggle by the great agrarian empires of the medieval Mediterranean world to win over to their respective faiths the “heathen barbarian” peoples to their north. More concretely, this meant the struggle of Western and Eastern Christendom and the Muslim Caliphate to convert the Germanic, Slavic and Altaic steppe peoples.

1 See A.M. Khazanov, “The Spread of World Religions in Medieval Nomadic Societies of the Eurasian Steppes” in M. Gervers, W. Schlepp (eds.), Nomadic Diplomacy, Destruction and Religion from the Pacific to the Adriatic (Toronto Studies in Central and Inner Asia, 1, Toronto, 1994), pp. 11–15 and his “World Religions in the Eurasian Steppes: Some Regularities of Dissemination” in G. Bethlenfalvy et al. (eds.), Altaic Religious Beliefs and Practices. Proceedings of the 33rd Meeting of the Permanent Altaistic Conference, Budapest, June 24–29, 1990 (Budapest, 1992), pp. 197–201. On the course of various religions among one Eurasian, Turkic, nomadic, tribal confederation, see P.B. Golden, “Religion Among the Qıpčaqs of Medieval Eurasia” Central Asiatic Journal, 42/2 (1998), pp. 180–237.

The fourth century conversion of the Goths to a moderate form of Arian Christianity through the activities of Ulfilas,2 was the first of several attempts by Constantinople to reach beyond the Danube using bilingual and bicultural agents. Evaluations of Byzantine success in the Pontic steppes vary. Moravcsik proclaimed Byzantine conversion activity in eighth century Khazaria “remarkably successful.” Although Christian communities could be found in the steppe and immediately adjoining areas of the North Caucasus, successes were sporadic and do not appear to have resulted in mass conversions.3 In this light, Noonan has recently suggested that Byzantium, in fact, did not fare particularly well here.4 One interesting example, of some relevance to the Khazars, is that of the so-called “North Caucasian Huns,” a subject people of the Khazars. In 682, according to Movsês Dasxuranci, the Albanian ruler Varaz Trdat sent the Albanian (Ałuanian) bishop Israyêl to the “Huns” to bring them into the Christian fold. The ruler, Ałp Iłutuêr (Alp Elteber),5 a son-in-law of the Khazar Qağan, “and his army” were converted. The policy was pronounced a success and the Huns were duly proclaimed “allies.”6 We know nothing of the subsequent fate of Christianity among the North Caucasian Huns. Khazar-led attacks into Albania, however, were still a commonplace in the early eighth century.

Since the “Khazars” were a splinter of the Western Türkic Kaganate headed by an progeny of the Ashina dynastic line, initially “Khazars” were “Türks” or “Göktürks”, till a coup brought another line to the power. Alp Elteber, or Great Viceroyal of the Land, was a relative of the newly self-proclaimed Kagan appointed as a governor, or rather a representative of the Kagan. In the 660's, when the splinter occurred, the existent power structure of the Caucasian Huns has not changed, although its leader was moved to a second place after Türkic Kaganate took over in the 550s, and the Caucasian Huns became a member of the Göktürk confederation. With the change in the centers of power, in 582 they became members of the newly hatched Western Türkic Kaganate, and in 630's they became members of the newly hatched Türkic Kaganate  that became known as Khazar Kaganate. Administratively, the Caucasian Huns did not experience any changes in 630's, they did not participate in the political squabbles, and did not join the newly Bulgar Kaganate formed by Kurbat, remaining with the restructured Türkic Kaganate. In light of these events, to call the Caucasian Huns “a subject people of the Khazars” is inaccurate. Neither Khazars were “Khazars”, nor the Caucasian Huns were “subject people”.

The Armenian consonant depicted as ł has a phonetic value of a some kind of g, hence  the Albania (Ałuania) is accurately phoneticized as Agvania. The Türkic title Ałp Iłutuêr has a clear l, and is usually accurately rendered Alp Ilutver or Alp Ilitver, in all cases rendering in Armenian the Türkic title Alp Elteber.

Movsês Dasxuranci is usually called Movses Khorenatsi, Movses Kalankatuatsi, or Movses Kagankatuatsi, the last based on -g- reading of the letter ł to reflect a Türkic title in the name of his village, the element -kat- in the name of the village is the Türkic dialectal of kang “village”. All readings reflect a hapax notation supposedly made by the ms. author. The reason for contrasting transliterations of a single record are quite clear, manipulations started still during the Imperial times, when history was supposed to reflect winners and losers, but the interchangeability of -v- and -u- (Ałuania vs Ałvania) is unclear, since the source is Armenian and not a Greek.

The Alans, an important element of Byzantine diplomacy in Western Eurasia, provide a further illustration of the difficulties. According to al-Masûdî, their rulers, previously pagans, had converted to Christianity “during the reign of the 'Abbâsids.”

Since Tengriism is a religion of individuals, and not institutions, the subject of the wholesale conversion is dubious. The concept of a religion-state preceded the modern concept of a nation-state, they are mirror-symmetrical, whereas one cared of your thoughts and did not care of your origins, the other cares of your origins and does not care of your thoughts. In either case, the objective is purely political, to unite “us” vs. “them”, and instill hatred against aliens. In the absence of the notion of the religion-state, religious conversion is an individual matter. It is as individual as the freedom of association and parliamentary system of power, innate to the mobile nomadic societies. The ease of association and disassociation befuddle scientists conditioned to stationary environment, pulling a rug from under all they know. A religious switch had nothing to do with conversion, on the individual level it was viewed in strictly political terms, and the political affiliation was decided on practical grounds. In most cases, a political switch did not impose new obligations, but was to provide direct and immediate benefits. In practical terms, the difference between religions is to supply or not supply your ancestors, children, and kins with travel necessities, enable or not enable them to come to their maker. It is clear that any child would be suspicious of any stranger advocating for not providing for the diseased, and instead relying on verbalized or silent appeals to Tengri. This suspicion had lasted for centuries, it took 10-20 generation to wean a predominant portion of the population from taking crucial care of their beloved diseased. Supplying the diseased was a duty independent of a political affiliation with a particular polity advocating a specific religious doctrine. In the following discourse, these points are amply illustrated by the cited historical records. Nobody abjures their faith, not now and not before, at best it is wrapped in the stranger's clothes. For the “great world religions”, even after a millennium of strict and sometimes cruel enforcement, the problem of regular attendance still remains acute and where possible, is forcefully compelled.

None of the Türkic peoples completely abandoned the term Tengri for the Creator; switching to a different religion just made Allah, Yahweh, Savaof צבאות‎, or Christ their Tengri: there goes Allah Tengri, Yahweh Tengri, etc.

2 This was initially, but not exclusively, the work of Ulfilas, a man of non-Gothic, or mixed Gotho-East Roman origins, see R. Fletcher, The Barbarian Conversion From Paganism to Christianity (New York, 1997), pp. 72–77; P. Heather, The Goths (Oxford, 1996), pp. 60–61, 85, 90–91, 131, 312–217.
3 Missionary activity began as early as the mid-fourth century, stemming from Armenian, Armeno-Caucasian Albanian, Georgian and Byzantine sources, see Gy. Moravcsik, “Byzantine Christianity and the Magyars in the Period of Their Migration”  The American Slavic and East European Review, 5 (1946), pp. 29–45; D. Ludwig, Struktur und Gesellschaft des Chazaren-Reiches im Licht der schriftlichen Quellen (Münster, 1982), I, pp. 318–325; M.G. Magomedov, Obrazovanie xazarskogo kaganata (Moskva, 1983), pp. 158–172.
4 Th. Noonan, “Byzantium and the Khazars: a Special Relationship?” in J. Shepard, S. Franklin (eds.), Byzantine Diplomacy (Aldershot, 1992), pp. 120–121.
5 This is a common title of a subordinate ruler in Turkic polities, see G. Clauson, An Etymological Dictionary of Pre-Thirteenth Century Turkish (Oxford, 1972), p. 134; P.B. Golden, Khazar Studies (Budapest, 1980), I, pp. 147–150.
6 Movsês Dasxuranċi, The History of the Caucasian Albanians, trans. C.J.F. Dowsett (London, 1961), pp. 153–171.

After 320/931–2, however, they drove out the bishops and priests sent to them by Byzantium and abjured that faith.7 Christianity had greater success in the Slavic lands, imposed, initially, by force of arms in Turko-Slavic Balkan Bulgaria and through a combination of diplomacy, charismatic missionary leadership (Cyrill and Methodius and their followers) and military pressure elsewhere.8 Islam was also not inactive in Western Eurasia. The 'Abbâsid Caliph al-Mamûn (813–833) composed a work to answer the questions of the ruler of the Burğar regarding the Islamic faith. Although there is some debate over the identity of these Burğar (Danubian-Balkan or Pontic-Bosporan),9 Islam subsequently found a warm reception among the kindred Volga Bulgars whose leader converted in the early tenth century.10 In the Eurasian steppe world and in Eastern Europe, the Khazar conversion (to Judaism) was one of the first, coming sometime after the embrace by the Uygur Qagan (762) of Manichaeanism, but before the conversion of the Balkan Turkic Bulgars (864) and the Slavs of Central and Eastern Europe.

The anatomy of conversions

As with mass religious conversions elsewhere, leadership was almost always provided by the rulers themselves and often had to overcome domestic resistance. In the first Türk Empire (Eastern Qaganate: 552–630, 682–742, Western Qaganate: 552–659, ca. 699–766), early rulers such as Mugan (553–572) and his successor Taspar (or Tatpar, 572–578) were interested in Buddhism as were their kinsmen in the Western Türk Qaganate, the progenitor of the Khazar state.11

By the time of Ibn Fadlan embassy, some Bulgars professed Islam for about 150 years (6 generations). Most remained Tengrians, and after religious squabbles lasted for another two centuries. In the outside world, the caliphate's recognition of Itil Bulgaria as an Islamic state in 922 AD was taken as a state conversion, quite a misleading milestone. That conceptual mistake of equating rulers with the state and the masses applies to all other examples.

Tengriism syncretized with Buddhism at an early date, as attested by the Buddhist-type toreutical inscriptions belonging to the western Hunnic milieu. The Huns brought to the Europe Tengriism already syncretized with Buddhism. In the east, Manichaeanism syncretized with Tengriism that was already syncretized with Buddhism. Since Manichaeanism syncretized with Buddhism from its beginning, syncretization with Buddhism-syncretized Tengriism was easy. Since Manichaeanism also syncretized with Christianity, the monotheistic version called Nestorianism, the syncretization of Tengriism with Nestorianism also was easy. These blendings produced numerous versions of the Tengriism in respect to the ritualistic side, symbology, and terminology. A high nomadic mobility tended to level off the differences, while the religious centers tended to stratify the population.

The Balto-Slavic people did not have a Tengrian tradition of provisioning deceased for travel, reflecting an absence of the Tengrian concept of reincarnation and eternal life. Without the burden of the Tengrian etiology, their acceptance of Christianity did not involve a torturous abandonment of the ancestors.

The Tengriist acceptance of the Judaism likely went along with syncretization, where the Tengrian traditions were wrapped into new ritualistic form that developed in the Middle East. The abundance of Turkisms in the Genesis chapter of the Old Testament, and the Tengrian concept of non-anthropomorphic Creator, and impossibility to picture the Supreme God attest to the Tengrian influence on the conceptual development of the Judaism. Since Judaism did not have a concept of human immortality by incarnation, resurrection, or Paradise, Tengrians did not have to abide by any Judaic prescripts for burial traditions.

7 Al-Masûdî, Murûj ad'-D'ahab wa Maâdin al-Jawhar, ed. C. Pellat (Beirut, 1966–1979), I, pp. 228–229.
8 See A.P. Vlasto, The Entry of the Slavs into Christendom (Cambridge, 1970); F. Dvornik, Byzantine Missions Among the Slavs (New Brunswick, 1970), C. Bowlus, Franks, Moravians and Magyars. The Struggle for the Middle Danube 788–907 (Philadelphia, 1995).
9 An-Nadîm, Kitâb al-Fihrist, ed. G. Flügel, J. Roediger (Leipzig, 1871), I, p. 111; B. Dodge (trans.), The Fihrist of al-Nadîm (New York, 1970), I, p. 254. For the Bosporan identification, see O. Pritsak, “The Role of the Bosporus Kingdom and Late Hellenism as the Basis for the Medieval Cultures of the Territories North of the Black Sea” in A. Ascher et al. (eds.), The Mutual Effects of the Islamic and Judeo-Christian Worlds: The East European Pattern (New York, 1979), pp. 6–7.
10 See the account of Ibn Fadlân, the secretary of the caliphal mission sent to Volga Bulğaria in 921–922: A.Z.V. Togan, Ibn Fadlân’s Reiseberich (Abhandlungen für die Kunde des Morgenlandes, 24/3, Leipzig, 1939).
11 See S.G. Kljaštornyj, V.A. Livšic, “The Sogdian Inscription of Bugut Revised” Acta Orientalia Academiae Scientiarum Hungaricae, 26/1 (1972), pp. 78–79; Liu Mau-tsai, Die chinesischen Nachrichten zur Geschichte der Ost-Türken (T’u-küe) (Göttinger Asiatische Forshungen, 10, Wiesbaden, 1958), I, pp. 36–38, 43; J.-P. Roux, La religion des Turcs et des Mongols (Paris, 1984), pp. 25–27. As for the Western Türks, Buddhism was already a presence in the Kušan and Hephthalite states, the principal hegemons in the region prior to the Türk take over, see B.A. Litvinskij, Vostočnyj Turkestan v drevnosti i rannem srednevekov’e. Étnos, jazyki, religii (Moskva, 1992), pp. 427–431.

These were, apparently, largely personal conversions which did not have a profound impact on society. Subsequently, when Bilge Qagan (716–734) proposed building Buddhist and Taoist temples, Tonyuquq, his Chinese-educated advisor, argued forcefully against it, noting that it was the mobility of the Türks, unencumbered by immovable property, which gave them their military advantage.12

Similar arguments would be advanced in later Turkic-nomadic societies by those who were fearful of the lure of urban life.13 In the Činggisid era, settlement in the cities was viewed as a crime against Činggis Qan’s Yasa.14 Conversion to a universal religion in the steppe invariably involved close contact with urban elements, a program that was not always attractive to the nomadic rank and file. The Sûfî wandering out into the steppe was far more effective in bringing Islam to the Turkic nomads than the learned ‘ulamâ’ of the cities.15 The third context in which the conversion should be viewed is the local one. What were the events and who were the personalities, if they can be determined, that brought about the conversion (to Judaism)? And finally, we must assess the legacy of the conversion (to Judaism). Was this an ephemeral event or did it set in motion long-term changes?

The dimension of urban influence appear to be a presumption contrary to the historical attestations. True, the nomadic Tengriism grew on a different soil than the sedentary religious models, but the interdependence is obvious, and the irrelevancy of the economical system in the etiology is profound. Chinese was a highly urbanized society, but its Qin emperor gave the world a most visible monument of the Tengriist thought. Abraham was a nomadic cattle breeder, but he became a Jewish patriarch. Tonyukuk, a military and civil leader, appreciated the military advantage of the mobility over sitting ducks populations. The concept of the self-propelled urban civilization may be popular among the urban populations, but it was the mobility of the nomads that spread technology and ideas from one encloistered population to another. Abaris was a Scythian sage credited with a number of theological works, and Pythagoras, among other things, was listening to the Hyperborean Abaris. He was a recent extract from South Siberia (the origin of the European Scythians) who lived at the end of 7th - beginning of 6th century BC, or 8 generations after cross-Eurasian migration to the proximity of the Greece; we know of him from the works of the Greek poet Pindar (522-442 BC), Plato's (429–347 BC) scholium to the “State”, philosopher Heraclides of Pontus (388-310 BC), sophist Himerius (315-386 AD). Being Hyperborean, Scythian, and a nomad did not prevent Abaris from teaching Greeks on theology and Pythagoras on geometry. These examples show that we are all mortals, and obsessed with continuation of our personal life, en urbi et orbi.

In his study of cross-cultural contacts in Eurasia, Jerry Bentley notes that “there is no single dynamic” that can be used to explain the “process of large-scale conversion in pre-modern times.” Similarly, there could be more than one reason for conversion, going beyond immediate “spiritual or cultural advantages.” Bentley discerns three broad patterns that come into play: “conversion through voluntary association; conversion induced by political, social, or economic pressure; and conversion by assimilation.”

12 Liu, Die chineschischen Nachrichten, I, pp. 172–173, 224.
13 Mahmûd al-Kâšγarî, Dîwân Luγât al-Turk. Compendium of the Turkic Dialects, ed. trans. R. Dankoff in collaboration with J. Kelley (Cambridge, Mass., 1982–1985), II, p. 103; see also Clauson, ED, pp. 453–454, referring to the sedentary Tats (Iranians), recorded the saying: qılıč tatıqsa iš yunčır, er tatıqsa et tınčır “If a sword rusts, (a man’s) work weakens, if a man/warrior adopts Iranian habits, his flesh becomes putrid,” On nomad-sedentary relations, see A.M. Khazanov, Nomads and the Outside World (Cambridge, 1984, 2nd ed. Ann Arbor, 1994).
14 V.V. Bartol’d (W. Barthold), Dvenadcat’ lekcij po istorii tureckix narodov Srednej Azii in his Sočinenija (Moskva, 1963–1977), V, p. 173.
15 M.F. Köprülü, Türk Edebiyatında İlk Mutasavvıflar (2nd ed., Ankara, 1966), pp. 13–15; Bartol’d, Dvenadcat’ lekcij in his Sočinenija, V, pp. 68–69.

As conversion often brought with it considerable, sometimes even radical changes in a variety of human activities, legal, social, culinary, sartorial, linguistic, it is difficult for the historian to reconstruct the motivation for such a change.16

Conversion accounts, whose task is to instruct and strengthen the faith of the newly converted, do not necessarily tell the full story. There is a strong emphasis on the miraculous. Of the various patterns noted, Bentley finds “voluntary association” to be “perhaps the most elusive.” He notes, however, a number of incentives involving “political, economic or commercial alliance with well-organized foreigners.” When viewing the actual examples of voluntary association, the role of long-distance merchants, often dispersed in trading diasporas across a continent, looms particularly large. Moreover, the local elites, in this way, could establish bonds with more powerful states from which political and military as well as commercial advantages could be gained. Association with a recognized imperial power could also confer domestic political benefits, bolstering otherwise weak regimes.17 The pace and depth of conversion must also be taken into consideration.

In much of the literature on conversions of Inner Asian peoples, attempts are made, as Devin DeWeese has noted, to “minimize the impact of the ‘new’ religion or to deny its significance beyond small circles within an Inner Asian state or people.”18 This has certainly been true of some of the scholarship regarding the Khazars. Richard Eaton, in his study of the Islamization of Bengal, provides an important model for the assimilation of a new faith in an agrarian community. His conclusions are equally valid for the pastoral-nomadic peoples. Eaton identified three stages in this process that began in Bengal in the thirteenth century and concluded in the eighteenth century. In the first stage, elements of the new faith were included in the already existing belief system. In Bengal this involved the activities of dynamic Sûfîs who pioneered the opening of new lands for cultivation and with this the inclusion of Muslim figures in the local pantheon. To proper Muslims this is simply širk (polytheism), the gravest sin in Islam. But, in this way, Islamic ideas, however distorted, became familiar and infiltrated the local belief system.

In the religious sales brochures, the role of the missionaries, the leaders, and the fake miracles and events obscure reality. It is easy to control and dominate a sedentary society, and still it took syncretization and five centuries to effect a conversion in an agrarian community, vulnerable to social pressure. For nomadic societies, violence and compulsion are useless. First, they elect their leaders in a parliamentary mode, and second, they have a tradition of voting with their feet. Applying example of  sedentary society to  nomadic society is like applying sheep control methods to the herds of cats. Cats behave differently from sheep, and most scholars know that. What takes a few months with sheep takes years with cats, and the results last only as long as the incentive remains valid. Examples of dethroning the nomadic leaders ascend to the earliest systematic records, e.g. Anacharsis in Herodotus 4.76, they are consistent for millennia, and so are the records of dispersion and migration away from unsuitable alliance. A faulty premise would certainly cause inaccurate conclusions. A sequence of inclusion, identification and displacement in essence invalidates the paradigm of conversion and turns it into a paradigm of evolution.

16 J. Bentley, Old World Encounters. Cross-Cultural Contacts and Exchanges in Pre- Modern Times (Oxford, 1993), pp. 8–9.
17 Bentley, Old World Encounters, pp. 9–10.
18 D. DeWeese, Islamization and Native Religion in the Golden Horde (University
Park, Penn., 1994), pp. 301–302.

In the second stage, Muslim notions, ideas and values came to be identified with elements of the earlier faith as the symbiosis of the no longer really competing belief systems deepens. Such a situation can last for a long time. Eventually, in the Bengali case, the Mughal government took an active role, alongside entrepreneurial-minded local Muslims, promoting more orthodox versions of Islam. This led to the third stage in which the old religion was displaced by a fully Islamic one. Further reform took place, but this was now in an Islamic setting.19 This pattern of inclusion, identification and displacement can also be observed in the steppe world. Within the Khazar orbit, the initial stages can be seen in the conversion tale of the North Caucasian Huns noted previously. The chopping down of the pagan holy forest by the priests and the conversion of the trees into Christian symbols and a site of Christian worship provides a graphic illustration of what DeWeese terms “assimilative displacement.”20 The burial practices of Islamicizing Volga Bulğaria 21 in the tenth and eleventh centuries still preserved elements of earlier Pre-Islamic custom and belief.22 Bulğar tomb inscriptions mixed Arabic and Turkic.23 By the eleventh century, Volga Bulğar scholars were known in the Islamic heartlands.24 We should bear these models in mind when we examine the course of Khazar Judaization.

Since the closest, contemporary conversion of a Turkic steppe state to a universal, world-religion is that of the Uyğurs, for comparative purposes we might briefly examine this event. In 762, in the Chinese capital, Luoyang, which his forces had “liberated” at the request of the Tang from rebels, Bögü, the Uyğur Qağan, converted to Manichaeanism, having been instructed in that faith by Soġdian Manichaean priests who had been residing there.

19 R. Eaton, The Rise of Islam and the Bengal Frontier 1204–1760 (Berkeley, 1993), pp. 268–303.
20 Dasxuranci/Dowsett, pp. 163–164; De Weese, Islamization and Native Religion, pp. 292–294.
21 F.Š. Xuzin, Volžskaja Bulgarija v domongol’skoe vremja (X-načalo XIII vekov) (Kazan’, 1997), pp. 110–111 suggests that Volga Bulğaria needed a new ideology that corresponded to its now higher level of social and economic development. The Islamization of the Volga Bulğars became the source of a number of conversion tales that are noted by the twelfth century and continued to be reworked into the ninteenth century. See DeWeese, Islamization and Native Religion, pp. 74–78; A.J. Frank, Islamic Historiography and ‘Bulghar’ Identity Among the Tatars and Bashkirs of Russia (Leiden, 1998).
22 See E.A. Xalikova, Musul’manskie nekropoli Volžskoj Bulgarii X-načala XIII v. (Kazan’, 1986), pp. 43 ff.
23 See T. Tekin, Volga Bulgar Kitabeleri ve Volga Bulgarcası (Ankara, 1988); M. Erdal, Die Sprache der wolgabolgarischen Inschriften (Turcologica 13, Wiesbaden, 1993).
24 See G.M. Davletšin, Volžskaja Bulgarija: duxovnaja kul’tura (Kazan’, 1990); Xužin, Volžskaja Bulgarija, pp. 116–119; A. Temir, “Kuzey Türk Edebiyatı” in Türk Dünyası El- Kitabı (2nd ed., Ankara, 1992), 3, pp. 705–706.

Manichaeanism had been moving along the Silk Route finding devotees among the Soġdians (as had other religions such as Nestorian Christianity) and those peoples with whom they closely interacted, especially the Turkic nomads. Manichaean missionaries at the Uyğur court soon encountered opposition from the Uyğur aristocracy which Bögü for a time overcame. Whether the religion spread much beyond the Uyğur elite is unclear. It did, however, become the state religion,25 surviving the assassination of Bögü Qagan in a coup led by anti-Manichaean aristocrats in 779.26 The Soġdian adherents of this faith living among the Uyğurs are noted as niġošaklar and sartlar.

The former term denotes, literally, the “listeners,” i.e. the Manichaean rank and file. Sart, a term ultimately of Sanskrit origin (sârtha “caravan”) (särtha “merchant”), meant “merchant” and subsequently in Turkic came to denote the urban (usually Iranian) (usually Sogdian and Indian, later also Iranian traders) populace as a whole, with some pejorative connotations.27 Religion and commerce were often inseparable along the Silk Route.28

The Uyğur conversion was known to the medieval Muslim geographers and historians 29 and an-Nadîm portrays them as ready to retaliate against the whole of the Muslim community in their lands, should any of their coreligionists be harmed in the Islamic lands.30 Clearly, a closer identification with the new faith was taking place.

Much later, in the Mongol-era, Juvainî gives us an important if somewhat garbled account of the conversion itself with Buddhism (much more widespread among the Uyğurs in his day) substituted for Manichaeansim.

Buddhism - Mongkol

25 S.N.C. Lieu, Manichaenism in the Later Roman Empire and Medieval China. A Historical Survey (Manchester, 1985), pp. 189–194; H.-J. Klimkeit, Gnosis and the Silk Road. Gnostic Texts from Central Asia (San Francisco, 1993), pp. 364–368.
26 See C. Mackerras, The Uighur Empire According to the T’ang Dynastic Histories (Canberra, 1972), pp. 88–89, 152–153. Bögü had adopted the title zahag-i mani “emanation of Mani,” changing, thereby, the ideological underpinnings of his rule.
27 See also W. Bang, A. von Gabain, “Türkische Turfan Texte, II, Manichaica” Sitzungsberichte der preussischen Akademie der Wissenschaften, Philosoph. — historische Klass, Berlin, 1929), pp. 412, 414, line 16; Clauson, ED, p. 846; E. De la Vaissière, Histoire des marchands sogdiens 2nd ed. (Paris, 2004), p. 73.
28 See Liu, Xinru, Silk and Religion. An Exploration of Material Life and the Thought of People A.D. 600–1200 (Oxford-Delhi, 1996).
29 See V.F. Minorsky, “Tamîm ibn Bahr’s Journey to the Uyghurs” Bulletin of the School of Oriental and African Studies, 12 (1948), pp. 279/283; al-Mas‘ûdî, Murûj, ed. Pellat, I, p. 155, and al-Bîrûnî, Atâ' r al-Baqiyya ‘an al-Qurûn al-Hâliyya, ed. E. Sachau (Leipzig, 1923), p. 204.
30 an-Nadîm, ed. Flügel, I, p. 337; al-Nadîm/ Dodge, II, pp. 801–802.

The account is filled with shamanic references to tree cults, holy mountains, the miraculous birth of rulers and dreams that precede the conversion and concludes with a religious debate, all familiar elements in Turkic conversion narratives and found in the Khazar accounts as well.31

Why did the Uyğurs convert to this much-persecuted faith? As in other instances, it may have been a very public way of proclaiming their ideological independence from China and to sharpen distinctions with their rivals, the Qarluqs and Qırğız, among whom Nestorian Christianity had made some headway.32

The Uyğur conversion to Manichaeanism did not leave a legacy. Manichaeanism faded during the post-imperial diaspora, while Buddhism became dominant, alongside smaller Nestorian Christian communities, both eventually supplanted by Islam (except for a small Buddhist community).33 A distant observer, the Zoroastrian Dênkart (last redaction in the tenth century) provides a curious footnote to this. Surveying the impact of non-Zoroastrian faiths, it remarks that “thus it is clear that the false doctrine of Yišô in Rome (Hrôm) and that of Môšê among the Khazars and that of Mânî in Turkistan took away their might and the valor that they once possessed and made them feeble and decadent among their rivals...”34

With this as background we may now ask what do we know of the indigenous belief system or systems that were present among the Khazars? 35 What were they converting from?

31 'Alâ ad-Dîn Atâ Malik Juvainî, Tarîh- i Jahân-Gušâ, ed. M. Qazwînî (Leiden-London, 1912, 1916, 1937), I, pp. 40–45, E.J. Boyle (trans.), The History of the World-Conqueror (Cambridge, Mass, 1958), I, pp. 55–60. See the excellent discussion in DeWeese, Islamization and Native Religion, pp. 282–286, 511–514. Bögü (Turkic “wise”) is here called Buqu Khân. After being visited by a tutulary spirit who takes him off to a holy mountain (Aq Tağ) and other shamanic types of initiation, he gains victories. He then has a dream of a holy stone, a dream shared by his vizier. This prompted a campaign that resulted in conquests and the founding of the city of Balasağun/Quz Balıq. The Uyğurs were still shamanists, having experts in the “science of magic” whom they call qam (the Turkic term for “shaman”). Buqu then convened a religious debate between the qams and the toyins (Turk. “Buddhist monk” < Chin. daoren [tao-jên], see Clauson, ED, pp. 569, 625)) from China (Hitâi). The Buddhist monks won the debate and the Uyğurs were converted to that faith.
32 Khazanov, “The Spread of World Religions” in Gervers, Schlepp (eds.), Nomadic Diplomacy, pp. 19–21.
33 Lieu, Manichaeism, pp. 199–201; E. Pinks, Die Uiguren von Kan-chou in der frühen Sung-zeit (960–1028) (Asiatische Forschungen, 24, Wiesbaden, 1968), pp. 108, 113–116.
34 M. Molé, La légende de Zoroastre selon textes pehlevis (Paris, 1967), pp. 236–7.
35 R.A.E. Mason, “The Religious Beliefs of the Khazars” The Ukrainian Quarterly, LX/4 (1995), pp. 383–415, attempts to assemble much of the data on Khazar religious practices, drawing heavily, as one must, on analogies with the Türks and North Caucasian Huns. This useful study, however, omits a discussion of qut and its significance for Khazar sacral kingship.

Khazar Tengriism

According to Ibn Rusta, aside from the Judaized ruling stratum, the “rest of the Khazars profess a religion similar to that of the Turks.”36 Gardîzî, who drew on the same sources, compares the Khazar native religion to that of the Oğuz Turks.37 It is quite likely that Khazar native religion was much like that of the North Caucasian Huns and other Turkic peoples. We have already mentioned the tree cult and how it was transformed by the Christian missionaries to their own purposes. Mention is also made of a god called K’uar38 (not otherwise attested as the name of a deity). Those persons or objects struck by “flashes of thundering fiery lightning and ethereal fire” are considered sacrifices to him.39 The principle god is Tangri Xan, the familiar supreme celestial god Tengri of the Inner Asian peoples.40 To this fearsome “gigantic savage monster,” according to Bishop Israyêl, they sacrificed horses. They also made offerings to fire, water, the moon, to “all creatures considered in their eyes to be in some way remarkable” and to “certain gods of the road.”41 This is probably a reference to the Old Türk yol tengri,42 probably a god of fortune. The North Caucasian Huns also beat drums and whistled over corpses, cut themselves as part of their mourning ritual, engaged in naked sword fights at the graves, wrestled and raced their horses this way and that as some were occupied with “weeping and wailing and others in games of diabolical fury.”43 Israyêl also mentions the “royal graves of the thunder” (čopayk’) and “the tall idols and the čopayk’ with the filthy skins of the altars.”44

36 Ibn Rusta, Kitâb al-A‘lâq an-Nafîsa, ed. M.J. De Goeje (Leiden, 1892), p. 139. Ad- Dimašqî, Kitâb Nuhbat ad-Dahr fî ‘Ajâ’ib al-Barr wa’l-Bahr, ed. M.A.F. Mehren (St. Petersburg, 1866), p. 263, says that they “knew not religion (milla), like the Turks.”
37 Gardîzî, Ta’rîh-i Gardîzî, ed. ‘Abd al-Hayy Habîbî (Tehran, 1363/1984), p. 580.
38 If not a corruption of Iranian Xwâr “sun,” this may be from Turkic *köğer < kök “sky” + er “man,” see Golden, Khazar Studies, I, p. 259.
39 Dasxuranci/Dowsett, p. 156. On the thunder and lightning cult among the pagan Uyğurs, see the report in the Wei-shu, N. Ja. Bičurin, Sobranie svedenij o narodax obytavšix v Srednej Azii v drevnie vremena (1851, reprint: Moskva-Leningrad, 1950), I, pp. 215–216; P.W. Schmidt, Der Ursprung der Gottesidee, III, Die Religionen der Hirtenvölker, Bd. IX (Freiburg, 1949), pp. 42–43; see also Roux, La religion, pp. 121–122.
40 U. Harva, Die Religiösen Vorstellungen der altaischen Völker (Folklore Fellows Communications, LII, No. 125, Helsinki, 1938), pp. 140 ff.; A. İnan, Tarihte ve Bugün Şamanizm (Ankara, 1954), pp. 26–29; Roux, La religion, pp. 110–121, 122–124.
41 Dasxuranci/Dowsett, p. 156.
42 Noted in the Irq Bitig, see T. Tekin, Irk Bitig. The Book of Omens (Turcologica, 18, Wiesbaden, 1993), pp. 8 (#2), 20 (#48).
43 Dasxuranci/Dowsett, p. 156.
44 Dasxuranci/Dowsett, pp. 165–166.

The term is still found in the North Caucasus today. In Osetin, coppay and in Čerkes čoppa denote a “ritual dance and singing performed around a person who has been struck by lightning, the refrain repeated during this ritual...”45 In Qaračay-Balqar, it is the name of a fertility deity of thunder and lightning, second only to Teyri (= Tengri) “to whom a kid was sacrificed.” Prayers to Čoppa were accompanied by ritual dancing.46 There are depictions, in aristocratic grave complexes in Khazaria, of scenes of ritual combat, dancing of naked warriors armed with spears and wearing masks. There is also evidence of the worship of oak trees to which animal sacrifies (wild boars) were brought and of totemism.47 Of Khazar native religious practices we have few direct notices.

Funerary practices, as evidenced by the sites investigated in Khazaria by Russian and Soviet archaeologists, indicate a belief in another world that was much like the one they had left. Warriors were buried with their horses, weapons and food, items they would need in their journey to the next world. Sometimes, in the graves of well to do there were human sacrifices of young women and children (probably slaves or servitors).

The skeletons were often wholly or partially destroyed to render the deceased harmless and perhaps to cleanse him or free him from worldly concerns. These and other pagan spiritual concepts were widespread throughout Khazaria, regardless of location or the ethnic affiliations of the deceased.48

The Khazar Hebrew conversion accounts mention that the Khazar ruler Bulan drove out the “magicians” (qosmîm) and idolators.”49 Presumably, this is a reference to Khazar qams (shamans). The sun amulets worn by the North Caucasian Huns, perhaps a part of the Tengri cult, are found all over Khazar territory.50

45 V.I. Abaev, Istoriko-étimologičeskij slovar’ osetinskogo jazyka (Moskva, 1958), I, p. 314.
46 M.Č. Džurtabaev, Drevnie verovanija Balkarcev i Karčaevcev (Nalčik, 1991), pp. 126–128; M.D. Kareketov, Iz tradicionnoj obrjadovo-kul’tovoj žizni Karačaevcev (Moskva, 1995) which has a detailed analysis of this cult.
47 S.A. Pletnëva, Očerki xazarskoj arxeologii (Moskva-Ierusalim, 1999), pp. 213–214.
48 Pletnëva, Očerki, pp. 207–210.
49 P.K. Kokovcov, Evrejsko-xazarskaja perepiska v X veke (Leningrad, 1932), Hebrew text, p. 21/Russ. trans. p. 75.
50 Magomedov, Obrazovanie, p. 155.

In the Vita Constantini, the famous “apostle to the Slavs,” the Khazar ruler who summoned him to participate in a religious debate at the Khazar court says “from the first (isprva) we know of one God who is above all and to Him we bow towards the east and observe some of our shameful (stoudny) customs.”51 At a dinner with the Qağan, the latter raised his goblet saying “we drink in the name of the One God who created every living thing (tvar’).”52 These are probably references to the Tengri cult.

There is clear evidence of a cult of ancestor worship and human sacrifice connected with it. The death of a Khazar tudun, in 710–711, occasioned the killing of the Byzantine official who was with him together with three hundred soldiers as part of the funeral observances.53 Ibn Fadlân tells us that after constructing over a river a multi-room mausoleum, termed “Paradise,” for the deceased Qağan, they decapitated those who built it.54

Al-Istahrî notes the reverence with which the Khazar royal grave was approached. Anyone riding towards the tomb had to dismount, prostrate himself before the tomb and then continue on foot until he was at a suitable distance from the holy site.55 The Qağan, of course, given his possession of qut (the heavenly mandate/good fortune to rule)56 was accorded extraordinary respect. Al-Istahrî, and others, remark on his sacral character. He rarely appeared in public.

51 T. Lehr-Spławiński, Żywoty Konstantyna i Metodego (Obszerne) (Poznań, 1959), pp. 26–27.
52 Lehr- Spławiński, Żywoty Konstantyna i Metodego, pp. 34–35.
53 Theophanes, Chronographia, ed. C. de Boor (Leipzig, 1883, reprint: Hildesheim, 1963), I, pp. 378–379. Theophanes remarks that this was done εις δοχεν (lit. “for the entertainment”) of the Tudun. This is hardly correct. Dieter Ludwig, noting the other ms. readings (δογην, δογην, δουγεν) correctly connects this term with the Old Turkic *doğ > yoğ “funeral feast,” see Ludwig, Struktur und Gesellschaft, I, pp. 356–357; Clauson, ED, p. 895.
54 Ibn Fadlân/Togan, Arabic, p. 44/Germ. trans. pp. 99–100. After the death of the Qağan, “they build a great hall (dâr kabîra) for him with twenty rooms (bait) and they dig a grave in each room for him. They crush stones until they become like kohl and it is spread about and lime (nûra) is thrown over this. Under the hall is a river. The river is a large one that flows rapidly. They make the river flow over the grave and say (this is done) so that no devil (šaytâ n), or man, or maggot (dûd) or reptiles (huwâm) may reach it. When he is buried, they strike the neck of those who buried him so that no one will know where his grave is among those rooms. They call his grave “Paradise.” They say he has entered Paradise. They spread across all the rooms silk brocade woven with gold.”
55 Al-Istahrî, Kitâb Masâlik al-Mamâlik, ed. M.J. De Goeje (2nd ed., Leiden, 1927), p. 224.
56 See A. Bombaci, “Qutluγ Bolzun!” Ural-Altaische Jahrbücher, 36 (1965), pp. 284–291; 38 (1966), pp. 13–43.
57 The title Išad derives from the Soġd. Išxâd' which in Khazar became *Ihšad > Išad, see P.B. Golden, “Khazarica: Notes on Some Khazar Terms” Turkic Languges 9/2 (2005), pp. 212–213.

When the Beg/Išad 57/Yilig (the deputy ruler who ran the actual affairs of state) approached him, he prostrated himself, “rolling himself in the dust” (tamarraġa fî at-turâb) and then waited until he was summoned.58 Ibn Fadlân reports that the deputy ruler daily “enters (into the presence of the Great Qağan, humbly (mutawâdian), showing humility (ihbât) and calmness (sakîna). He only enters before him barefoot and in his hand is a piece of firewood (hatab). When he greets him, he ignites this piece of firewood between his hands and when it is all burned up, he sits together with the king on his throne, on his right side.”59 The use of purifying fire so that one might be admitted to the royal presence was well-known in the steppe world.60 These rites must also be considered part of the pre-conversion Khazar religion.

These or similar types of funerary practices involving human sacrifice were widespread in the steppe. Al-Muqaddasî (al-Balhî)61 says of the Turks that “among them are those who bury with the deceased their slaves and servants, (leaving them) alive, in the grave mound, until they die.”62 Other examples can be cited for groups ranging from the Scythians, Xiongnu (Huns), Hephthalites, Qitans, Qıpčaqs, Mongols, Jurčens and Ottomans.63 The Khazar grave structure described by Ibn Fadlân is an example of the qorığ “a royal enclosure,”64 which is clearly linked with Old Türk burial customs.65

58 Al-Istahrî, ed. De Goeje, p. 224.
59 Ibn Fadlân/Togan, Arabic, p. 43/Germ. trans. p. 99.
60 For example, Zemarchus, the Byzantine envoy to the Türks and his baggage, were subjected to a shamanic, purifying ritual involving fire when he entered the Türk-controlled lands of Soġdia, see The History of Menander the Guardsman, ed. trans. R.C. Blockley (Liverpool, 1985), pp. 118/119. The Rus’ prince, Mikhail of Černigov, was required to walk between two fires before he was to be brought before the Činggisid Batu. When he refused to bow before idols of Činggis, he was executed, see Die Mongolengeschichte des Johannes von Piano Carpini, ed. trans. J. Gießauf (Graz, 1995), Latin text, pp. 89–90/Germ. trans. pp. 135–146; see also M. Dimnik, Mikhail, Prince of Chernigov and Grand Prince of Kiev 1224–1246 (Toronto, 1981), pp. 130–135 and Roux, La religion, pp. 222–224.
61 I. Ju. Kračkovskij, Arabskaja geografičeskaja literatura in his Izbrannye Sočinenija (Moskva-Leningrad, 1955–1960), IV, p. 195 identifies the author of this work with Mutahharb. Tâhir al-Muqaddasî.
62 Al-Balhî, Kitâb al-Bad wa’t-Ta’rîh, ed. Ch. Huart (Paris, 1899–1916), IV, p. 22.
63 E. Tryjarski, Zwyczaje pogrzebowe ludów tureckich na tle ich wierzeń (Warszawa, 1991), pp. 205–208; De Weese, Islamization and Native Religion, pp. 263–267; S. Vryonis, “Evidence of Human Sacrifice Among the Early Ottomans,” Journal of Asian History, 5/2 (1971), pp. 140–146; Golden, “Religion Among the Qıpčaqs” CAJ, 42/2 (1998), pp. 194–195. According to Marco Polo, some 20,000 people were slaughtered with the burial of Möngke (d. 1259), see Marco Polo, The Description of the World, ed. trans. A.C. Moule, P. Pelliot (London, 1938), I, pp. 167–168.
64 Clauson, ED, p. 652 qorığ “an enclosure, enclosed area, particularly one enclosed by a ruler.” In Qarkhanid Turkic it denoted “the private property of chiefs etc., any enclosed space is called qorığ.” In the Činggisid era Qıpčaq, qoruğ or qoru could mean both “royal estate” and “private property.”
65 See discussion in DeWeese, Islamization and Native Religion, pp. 181–183, 188. Connected with these customs, perhaps, is the strange account in at-Tabarî, according to which in the aftermath of the Arab defeat suffered at the hands of the Khazars at Balanjar in 32/652, the body of the slain commander of the Arab forces from Bâb al-Abwâb/Darband, ‘Abd ar-Rahmân b. Rabî'a, was placed in a basket (safat) and they “ask for water by means of it and ask for victory by it up to today.” Zaxoder saw in this a possible reflection of the rain-cult and rain-stone that was so well-known among the Turkic peoples, see at-Tabarî, Tarîh at-Tabarî, ed. M. Ibrâhîm (Cairo, 1967–1969), IV, pp. 304–305; B.N. Zaxoder, Kaspijskij svod svedenij o Vostočnoj Evrope (Mosvka, 1962, 1967), I, pp. 148–149. On the rain cult and rain stone, see İnan, Şamanizm, pp. 160–165; Roux, La religion, pp. 95–98.

Archaeologists have found many ashpits in maritime Daghestan, an early area of Khazar habitation. These are connected with sun or fire cults that may have been part of Khazar worship.

In addition, there are finds of the teeth of boars, dogs and wolves, some of them perforated, in both Daghestan and Saltovo-Majackaja culture areas, that, very likely, had some cultic functions or were used as amulets.66

Monotheistic world religions were already penetrating Khazaria, in particular Christianity.67 Archaeologists have unearthed evidence of a sizable number of Christian churches in maritime Daghestan, a region of Khazar influence. In particular, this was the work of Armenian/Albanian missionaries operating out of Č’or (Darband/Bâb al-Abwâb or a locale near it)68 long active in missionary efforts among a variety of “Hunnic” peoples. There is also evidence of Georgian Christian missionary activity elsewhere in the region.69 In this connection we may note the Georgian Life of St. Abo (< Arab. Habîb) who was martyred in 786. An Arab Muslim who had taken service with the Georgian Prince Nerses, he fled with his master to the Khazars, ca. 779–780. The latter are described as “wild men, fearsome of face, savage in character, drinkers of blood, without religion except that they recognize a god the creator.”

66 Magomedov, Obrazovanie, pp. 155–158.
67 Pletnëva, Očerki, p. 214, suggests that “the intellectual elite of Khazar society” was already moving to “the adoption of world religions.” The evidence for Zoroastrianism is very thin, see Ludwig, Struktur und Gesellschaft, p. 318 and D.M. Dunlop, The History of the Jewish Khazars (Princeton, 1954), p. 189.
68 Also Č’ołay / Čoray/ Sûl/Τζουπ, Dasxuranci/Dowsett, pp. 5n.3; J. Marquart, Osteuropäische und Ostasiatische Streifzüge (Leipzig, 1903, reprint: Hildesheim, 1961), pp. 444, 489.
69 Magomedov, Obrazovanie, pp. 5, 158, 164–171. See also discussion in Ludwig, Struktur und Gesellschaft, pp. 318–318. Remains of two churches have been found in the necropolis of the Verxnečirjurtovskoe gorodišče. Large numbers of Christian artifacts and cultic objects are also present.

In other words, the Khazars, at this time, were still largely followers of the Tengri cult. Here, Abo, converted to Christianity.70 According to al-Muqaddasî, writing in the late tenth century, the bulk of the inhabitants of Samandar, one of the early Khazar centers in the North Caucasus, were Christians.71 Al-Istahrî, however, mentions that there are Muslims and mosques in Samandar, but notes that “their king is a Jew” and a relative of the Khazar ruler. Ibn Hawqal (who took much of his information from al-Istahrî or their common source) says that “the Muslims and people of various other faiths and pagans (Tabaqât ahl al-milal wa’l-wata'niyûn) inhabited this country and adds that there are mosques, churches and synagogues in Samandar.72

The Crimea, a region in which there was often an uneasy Byzantine- Khazar condominium, was another source of ongoing Christian influences emanating from the Crimean Goths and the Black Sea Christian communities.73 When the Khazars extended their influence over Ap’xazet’i/Abxazia-Western Georgia (whose ruler ca. 780, Leon, was the grandson of the Khazar Qagan and who had asserted his independence of Constantinople with Khazar aid), an anti-Khazar revolt broke out (780s, perhaps 790s) in Gothia. Although there is no direct evidence to tie Byzantium to it, it is indicative of the friction that had developed in this region.74 These conflicts form part of the background to the Khazar conversion. There is also some evidence for the creation of a larger Church structure in the region in the form of a number of episcopal seats, subject to the metropolitanate of Doros (Gothic Crimea) and covering the lower Volga, North Caucasian and Crimean zones.75

70 The sources of his conversion, apparently, were the “many towns and villages in that northern land which by the grace of the Holy Ghost abide securely in the Christian faith,” see D.M. Lang (ed. trans.), Lives and Legends of the Georgian Saints (London, 1956, reprint: Crestwood, NY, 1976), pp. 118–119.
71 Al-Muqaddasî, Ahsan at-Taqâsim fî Marîfat al-Aqâlîm, ed, M.J. De Goeje (Leiden, 1877, 2nd ed. 1906), p. 361. On Samandar and the difficulties of its geographical location, see Ludwig, Struktur und Gesellschaft, I, pp. 246–248.
72 Al-Istahrî, ed. M.J. De Goeje, pp. 222–223; Ibn Hawqal, Kitâb Sûrat al-Ard (Beirut, 1992), p. 333.
73 For an overview of Byzantine Christian activities in Khazaria, see F. Dvornik, Les légendes de Constantin et de Méthode vues de Byzance (Praha, 1933, 2nd ed. Hattiesburg, Mississippi, 1969), pp. 157 ff.
74 K’art’lis C’xovreba, ed. S. Qauxč’išvili (T’bilisi, 1955), I, p. 251; see discussion in
Golden, Khazar Studies, I, p. 66; Artamonov, Istorija Xazar (Leningrad, 1962), pp. 252–258.
75 These included Χοτζηρων (perhaps “of the Khazars”), ο Αστηλ (Atıl), ο Χουαλης (the Khwârazmian elements of the Khazar state, cf. Rus’ Õâàëèñû) ο Ονογυρον (the Onoğurs), ο Ρετιγ, ο Ουννων (the North Caucasian Huns?), Ταματαχα (Rus’ Òìóòîðîêàíü); for the list see G.I. Konidares, “Αι Μητροπολεις και Αρχιεπισκοπαι του οικουμενικου Πατριαχειου και η ταζις αυτον” Texte und Forschungen zur byzantinsichneugriechischen Philologie, 13 (1934), p. 100; see also J. Moravcsik, “Byzantinische Mission im Kreise der Türkvölker an der Nordkküste des Schwarzen Meeres” in J.M. Hussey et al. (eds.), Proceedings of the XIIIth International Congress of Byzantine Studies (Oxford, 1967), pp. 21–24; Ludwig, Struktur und Gesellschaft, p. 322.

The dating of this organizational proposal and its actual impact are uncertain (last quarter of the eighth century?).76 Clearly, there was a serious effort to strengthen the Christian/Byzantine position in Khazaria at this time. Why? Was this a response to the growing Jewish influence, to the conversion of the Khazar ruler to Judaism noted in al-Masûdî or a factor contributing to that conversion?

Marwân forced Qagan to convert to Islam

Equally important as a background factor was the protracted struggle in which the Khazars and Arabs had been engaged for control of the Caucasus.77 In 737, the Umayyad general (and subsequently last Umayyad Caliph) Marwân broke through into the Khazar core lands on the Volga, captured the Qagan and forced him to convert to Islam. According Ibn Atam al-Kûfî, “and with him many people of his house became Muslims and people of his country.”78

There is very little evidence to indicate that the Qagan remained a Muslim. The quick retreat of the Arab armies and the growing disorder in the Umayyad Caliphate, which was toppled in 750, left little political pressure to remain Muslim. There is one notice, however, which may indicate that Islam had continued among some elements of the Khazar ruling clan.

76 It is preserved in a fourteenth century Greek manuscript. Obolensky would place it earlier, between 733–746, see Artamonov, Ist. Xazar, pp. 258–261; D. Obolensky, The Byzantine Commonwealth. Eastern Europe 500–1453 (London, 1971), pp. 174–175. J. (Gy.) Moravcsik, “Byzantine Christianity and the Magyars in the Period of Their Migration” The American Slavic and East European Review, 5 (1946), pp. 40–41; Dvornik, Les légendes, p. 164.
77 On the Arabo-Khazar war, see the excellent treatment in Dunlop, History, pp. 41–87.
78 Ibn Atam al-Kûfî, Al-Futûh (Beirut, 1412/1992), III, pp. 252–255. A briefer account of these events is given by al-Balâdurî according to whom the “chief of al-Khazar” (azîm al-Hazar) terrified by Marwân’s might, quickly accepted Islam when Marwân offered him the choice of conversion or war, see al-Balâdurî, Futûh al-Buldân, ed. M.J. De Goeje (Leiden, 1895), p. 208 or the later edition by R.M. Radwân (Cairo, 1959), pp. 209–210.

Al-Istahrî (writing in the middle of the tenth century, although much of his data is drawn from earlier sources) reports that the Qağanal office was barred to a very able member of the royal clan (who sold fish in the market) because he was a Muslim, “for they only summon to the Qaganate one who professes Judaism.”79 The story, if true, would indicate that some members of the Qağanal house had either retained their allegiance to Islam or had been subsequently converted.

In any event, there were still Muslims within the ruling clan. It is hard to imagine, however, given the high positions held by Muslims in Khazar society and government, that this young man was reduced to fish-mongering because of his religion.

According to Ibn Fadlân the legal affairs of the Muslims who reside in or come to Khazaria for trade are handled by a Muslim slave (ġulâm) of the Khazar king who is called Hz (cf. Khwârazmian χž “pleasant”?)80 “and no one else hears their cases or passes judgments among them.”81 Al-Masûdî, in the 940’s, reports that the wazîr of the Khazar king was a Muslim named Ahmad b. Kûya and notes that Muslims “are predominant (ġâlib) in” the capital Atıl/Itil, “because they constitute the army of the king. They are called in this town Ursiyya.”82 This standing army of the king, perhaps the royal comitatus, al-Masûdî is only talking of the troops within the capital, was staffed by an immigrant community of Muslims from the Khwârazm region who had left their homeland, “after the appearance of Islam,” and had taken service with the king of the Khazars. Ahmad b. Kûya was from this community.83 When the king warred on Infidels, he was accompanied by 7000 of them. They stood aside, however, in wars with Muslims.84 We can see that non-Judaic religious allegiance did not preclude access to the highest levels of power.

79 Al-Istahrî, ed. De Goeje, p. 224. See Dunlop, History, pp. 97–98 for a slightly different translation.
80 W.B. Henning, A Fragment of a Khwarezmian Dictionary, ed. D.N. MacKenzie (London, 1971), p. 49.
81 Ibn Fadlân/Togan, Arabic, p. 45/Germ. trans. 102.
82 That is Ors/Urus, this is an old Iranian name going back to the Auruša “white,” cf. the Ἀορσοι (Aorsoi), cf. Osetic Ors/Urs, see J. Harmatta, Studies in the History and Language of the Sarmatians (Szeged, 1970), p. 85; P.B. Golden, “Cumanica III: Urusoba” in D. Sinor (ed.), Aspects of Altaic Civilization III. Proceedings of the Thirtieth Meeting of the Permanent International Altaistic Conference, Indiana University, Bloomington, Indiana, June 19–15, 1987, Bloomington, 1990), pp. 33–46.
83 Omeljan Pritsak has suggested that Ahmad b. Kûya was of a family/clan of hereditary wazîrs in Khazaria and that the family name, Kûya (< kaoya of old Iranian origin from the sacral Iranian ruling dynasty of the Kaway) lies at the root of the toponym Kiev, see N. Golb, O. Pritsak, Khazarian Hebrew Documents of the Tenth Century (Ithaca, 1982), pp. 53–54. For a critique of that view, see P.P. Toločko, “K voprosu o xazaro-iudejskom proisxoždenii Kieva” Xazarskij al’manax 2 (Kiev-Kharkov-Moskva, 2004), pp. 99–108
84 Al-Masûdî, Murûj, ed. Pellat, I, p. 213.

In Turkic, Ursiyya < Urušoba = uruš  “battle, clash, strong” + oba “habitat”, i.e. “Land of Fighters” or “Land of Strongmen”, aka Karabulgar “West Bulgar”, “Strong Bulgar”, and some other meanings, usually translated as “Black Bulgar”. No need to cite Persian or any other homophones, present in nearly every other language, in English urus is an ox, but that does not make Urušoba the “Land of Oxen”. The attempts to substantiate an Iranian linguality of the Sarmats conflict with the attestations of the contemporaries, and are driven by notions other than the science. A chain of recitations of a faulty premise does not make it right.

The people Kayi, termed in the Russian rendition Kaitags “Mountain Kayi” are one of the oldest known Türkic ethnonyms, they were an “old” maternal dynastic tribe of the Eastern Huns, then a paternal dynastic tribe of the Qians (Qiangs), and a dynastic line of the early Ottomans, and the Kayi are still numerous and prominent in Turkey and Pakistan. That Ahmad ben Kûya came form a Kayi tribe is indicated in his name, probably he was an offspring of a Kayi mother and a dynastic father, it was common to add the origin of the mother to the names of their offsprings, thus there could be Ahmad ben Alan, Ahmad ben Ars, etc., all children of the same father. Again, there is no need to cite Persian or English.

The name of the Kyiv is utterly transparent, Ky is Türkic “White”,  iv is “house, estate”, its Slavic name was Belgorod “White City”, an estate on one of the Kyiv hills, another form was Askel “White Fort”, Slavic Belgorod “White City”. Again, there is no need to cite Persian or English.

The line of recited misrepresentations, combined with the absence of the native etymologies and in lieu of them looks quasi-scientific, but does a disservice to the objectivity of the analysis.


Conversion to Judaism

With these remarks we have come to the question of the conversion itself. What do we know of the circumstances of the conversion, its setting and dating? The historicity of the conversion narratives has been explored by Dunlop, Pritsak and others.85 Pritsak has termed them “epic narratives” and there are certainly elements that fit within the general mold of Turkic conversion tales. The dating remains problematic. It is, perhaps, most useful to examine our sources chronologically or according to the era from which our source has drawn its information.

As has long been noticed, none of the contemporary sources emanating from their immediate neighbors make direct mention of the Khazar conversion. The earliest source is the passage in Christian of Stavelot’s (also known as Druthmar of Aquitaine) commentary on Matthew (Expositio in Matthaeum Evangelistam) the extant manuscripts of which date to the tenth century.86 The sources for the notice are dated to sometime after the death of Charlemagne and before the conversion of the Balkan Bulğar ruler Boris to Christianity in 864 — although a recent study suggests that it might have stemmed from South Italian or Roman sources of the 860’s or 870’s. The text says “We are not aware of any nation under the sky that would not have Christians among them. For even in Gog and Magog, the Hunnic people who call themselves Gazari, those whom Alexander confined, there was a tribe more brave than the others. This tribe had already been circumcised, and they profess all dogmata of Judaism (omnem Judaismum observat). However, the Bulgars, who are also from those seven tribes, are now becoming baptized.”87

The Gazari are, presumably, the Khazars although this term or the “Kozary” of the perhaps nearly contemporary Vita Constantini (Constantine/Cyril was in the Khazar state ca. 861) could have reflected any of a number of peoples within Khazaria.88 The Vita Constantini (written perhaps in the early 880s, if not immediately after Constantine’s death in 869)89 tells of a religious debate held at the court of the Khazar ruler.

85 Dunlop, History, chaps. v–vi; Pritsak, “The Khazar Kingdom’s Conversion to Judaism” Harvard Ukrainian Studies, II (1978), pp. 261–281.
86 L.S. Chekin, “Christian of Stavelot and the Conversion of Gog and Magog. A Study of the Ninth-Century References to Judaism Among the Khazars” Russia Mediaevalis, IX/1 (1997), p. 15.
87 See text Christian Druthmar, “Expositio in Matthaeum Evangelistam,” Patrologiae cursus completus, series latina, ed. J.-P. Migne (Paris, 1864), t. 106, c. 1456 and the lengthy analysis of the text by Chekin, “Christian of Stavelot” Russia Mediaevalis, IX/1  (1997), pp. 13–34, Latin text, and Eng. trans. pp. 17–18.
88 Chekin, “Christian of Stavelot” Russia Mediaevalis, IX/1 (1997), pp. 29–30; Pritsak, “Conversion: HUS, II, (1978), p. 271.
89 The earliest mss. however stem from the fifteenth century, see Lehr-Spławiński, Żywoty Konstantyna i Metodego, pp. xxiii–xxiv; C. Zuckerman, “On the Date of the Khazar King’s Conversion to Judaism and the Chronology of the Kings of the Rus Oleg and Igor” Revue des Études Byzantines, 53 (1995), p. 243. On the numerous problems associated with the Žitie Konstantina, see S.B. Bernštejn, Konstantin-filosof i Mefodij (Mosvka, 1984), pp. 23 ff.; O. Pritsak, “Turkological Remarks on Constantine’s Khazarian Mission in the Vita Constantini in E.G. Farrugia et al. (eds.), Christianity Among the Slavs: The Heritage of Saints Cyril and Methodius; Acts of the International Congress Held on the Eleventh Centenary of the Death of St. Methodius, Rome, October 8–11, 1985 (Rome, 1988), p. 298.

According to this account the debate took place because “the Jews are counseling us (ustjat ny) to accept their faith and usages (detel’) and the Saracens, on the other hand, offering peace and many gifts, are pulling us to their faith.” Hence, the appeal to the Byzantine emperor, “because of our old friendship and love,” to send “a learned man” to present the Christian position and “should he defeat the Jews and Saracens we will adopt your faith.”90 From this it would appear that the question of Khazar religious orientation, at the highest levels, was still undecided ca. 861, although local partisans of Judaism clearly held high, influential positions at the court and were shaping the policy debate within the government. In other words, Judaism was by this time a powerful presence within the Khazar state. Constantine’s mission was fundamentally political in nature and could not have been otherwise. Political issues were, in any event, inevitably and inextricably intertwined with proselytizing efforts.91 Success would have altered the Judaizing course of the Khazar government and brought the Khazar Empire fully into the “Byzantine Commonwealth.” Although it did not succeed on the religious plane, Constantine, according to the Vita, was able to bring home some Byzantine prisoners who had been in Khazar captivity and firm up the Khazaro-Byzantine entente.92 The Vita indicates more than hints that not all had been well in Khazar-Byzantine relations. More importantly, the Vita appears to have distorted the actual situation at the Khazar court. The Khazar elite had already converted to Judaism (see below).

90 For the Vita Constantini, see Lehr-Spławiński, Żywoty Konstantyna i Metodego, pp. 27 ff.; Constantinus et Methodius Thessalonicenses, Fontes, eds. F. Grivec, F. Tomšić in Radovi Staroslavenskog Instituta (Zagreb), IV (1960), pp. 109 ff.
91 Dvornik, Les légendes, p. 176 writes “Le but de l’ambassade de 860 était donc plutôt politique, contrairement à ce qu’en dit la Vie de Constantin.” See also Obolensky, The Byzantine Commonwealth, pp. 176–177.
92 Indeed, the Qağan wrote to the Emperor that “we are all confederates and friends (druzi i prijateli) of your empire and are ready to serve you wherever you may request,” Lehr-Spławiński, Żywoty Konstantyna i Metodego, pp. 58–59.

Somewhat more problematic is the account of Eldad ha-Dani, a mysterious Jewish traveler and tale-teller, who is said to have visited Spain ca. 880 and wrote of the presence of the tribe of Simeon and the half-tribe of Manasseh in the land of the Khazars. The authenticity of the texts has been called into question.93 He is, perhaps, the figure claiming Danite descent who visited Spain “in the days of our fathers” mentioned in the letter of Hasdai ben Šaprût, the Jewish courtier of the Spanish Umayyads who initiated the Khazar correspondence.94

Ibn Hurdâd bih (d. ca. 912) was the master of the post in al-Jibâl (Media) and a man who was well informed about the lands of the ‘Abbâsid Caliphate and its neighbors. His Kitâb al-Masâlik wa’l-Mamâlik, which underwent two redactions (846–7 and 885–6, not all of which survives), has a very valuable notice on the Jewish merchants, the Râd âniyya, who came to the Khazar cities as part of their trans-Eurasian itinerary and appear to have been supplanted by the Rus’ by the time of the second redaction.95 The surviving text says nothing about Khazar Judaism. Ibn al-Faqîh, however, who drew liberally on Ibn Hurdâd bih has more information on this, as we shall see. Pritsak has suggested that the thirteenth-century compiler, Yâqût, in his Mujam al-Buldân, who remarks that “their (the Khazar) king” (malikuhum) is a Jew,96 took this part of his text from Ibn Hurdâd bih. This would be the earliest Muslim reference to Khazar Judaism.97 But, Yâqût himself says that he took his information from Ibn Fadlân and it is more widely accepted that the former and al-Istahrî are the primary sources for the Mujam’s notices on the Khazars.98

93 See discussion in Dunlop, History, pp. 140–141, 168; Marquart, Streifzüge, pp. 197–198 and n. 3. On Eldad, see S.W. Baron, A Social and Religious History of the Jews, III (New York, 1952–1983), III pp. 116–117, 208, VI, pp. 220–221.
94 Kokovcov, Perepiska, Heb. text, p. 18/Russ. trans.pp. 69–70.
95 Ibn Hurdâd bih, Kitâb al-Masâlik wa’l-Mamâlik, ed. M.J. De Goeje (Leiden, 1889), pp. 153–155. On Ibn Hurdâd bih, see the comments in T. Lewicki, Źródła arabskie do dziejów Slowiańszczyzny (Wrocław-Kraków, 1956–1988), I, pp. 43–63; A.P. Novoselcev, Xazarskoe gosudarstvo i ego rol’ v istorii Vostočnoj Evropy i Kavkaza (Moskva, 199), p. 10; Mihály Kmoskó, Mohamedán írók a steppe népeiről. Földrajzi irodalom, (Budapest, 1997), I/1, pp. 40–42; H. Göckenjan and I. Zimonyi, Orientalische Berichte über die Völker Osteuropas und Zentralasiens im Mittelalter. Die Gayhânî-Tradition Veröffentlichungen der Societas Uralo-Altaic, Bd. 54 (Weisbaden, 2001), pp. 29–30.
96 Yâqût, Mujam al-Buldân (Beirut, 1957), II, p. 368.
97 Pritsak, “Conversion” HUS, II (1978), p. 279 n. 76.
98 Dunlop, History, pp. 100 ff.

There is a sizable corpus of Muslim historians who take note of the Khazars (given the prolonged warfare between the Arabs and the Khazars they could hardly have failed to do so), but they, too, are silent about Khazar Judaism. They include such important ninth and early tenth century authors as: al-Ya'qûbî, al-Balâdurî and at-Tabarî. Ibn Atam al-Kûfî (d.926) noted the conversion of the Qagan to Islam, but has nothing to say regarding other religions in Khazaria.

The situation changes, however, with Ibn al-Faqîh whose Kitâb al-Buldân was probably written ca. 902–903 (he died in the 950s)99 and largely drew on Ibn Hurdâd bih. He reports, however, that “all of the Khazars are Jews. But, they have been Judaized recently.”100 Lewicki was of the opinion that this notice was taken directly from Ibn Hurdâd bih and hence dates to the 840’s or 880’s.101 It might also, however, be an updating of the latter’s material by Ibn al-Faqîh, reflecting the large-scale Judaization of the Khazars that had occurred by the time of his writing.

The Kitâb al-Alâq an-Nafîsa (written ca. 903–912) of Ibn Rusta, only one book of which has survived, dates from about the same time as Ibn al-Faqîh’s work. Basically a compiler, like so many of the Islamic geographers, Ibn Rusta drew on a number of sources, both contemporary merchants and travelers and written works from the mid- to late-ninth century.102 He says of the Khazars that “their supreme chief professes Judaism as does also the Κâ[d] and those of the leaders and great ones who sympathize with his inclinations. The rest of them profess a religion similar to that of the Turks.”103

Gardîzî whose Zayn al-Ahbâr dates from ca. 1050, drew from many of the same sources used by Ibn Rusta. He reports that the Qagan and Išâd are Jews as “are all who are inclined (meil dârad) to (these) two from among the generals (sarhangân) and great men. The rest of them have a faith (bar dînî and) which resembles that of the religion of the Oğuz Turks (ba-dîn-i turkân-i ġuzz mânad).”104

99 See discussion in Lewicki, Źródła, II/1, pp. 9–10.
100 Ibn al-Faqîh, Kitâb al-Buldân, ed. M.J. De Goeje (Leiden, 1885), p. 298.
101 Lewicki, Źródła, II/1, pp. 40–41 and comments p. 121.
102 These included Ibn Hurdâd bih, the lost work of al-Jaihânî (d. ca. 892–907?), the famous essayist al-Jahîaz, Ibn Qutayba, Hârûn ibn Yahyâ (who as a prisoner spent time in Byzantium and later in the Balkans, Venice and Rome) and other sources. On the complexities and dating problems of Ibn Rusta and his sources, see Lewicki, Źródła, II/2, pp. 7–17; Göckenjan and Zimonyi, Orientalische Berichte, pp. 33–35.
103 Ibn Rusta, Kitâb al-Alâq an-Nafîsa, ed. De Goeje, p. 139.
104 Gardîzî, Tarîh, ed. Habîbî, p. 580. On his sources, some of which, with regard to the Turkic world, date to the latter part of the eighth century, see K. Czeglédy, “Gardîzî on the History of Central Asia (746–780 A.D.)” Acta Orientalia Academiae Scientiarum Hungaricae, 27 (1973), pp. 257–267 and Göckenjan and Zimonyi, Orientalische Berichte, pp. 36–42. The anonymous Hudûd al-Âlam, trans. V. Minorsky (London, 1937, 2nd ed., 1970), pp. 161–162, in its truncated and somewhat garbled version of this tradition makes no mention of Khazar Judaism.

Ibn Fadlân, one of our few sources who actually set foot in the region (in 921–922), in his capacity as secretary to the Caliphal mission to Volga Bulğaria, reports that “the Khazars and their king are all Jews” (wa al-hazar wa malikuhum kulluhum yahûd). This section, however, is not found in the Mašhad manuscript, but preserved in Yâqût’s occasionally jumbled compilation.105 It is clear, nonetheless, that by his time the core element of the state, the Khazars, were Judaized.

Roughly contemporary to Ibn Fadlân was Saadiah Gaon (892–942) who makes several references to the Khazars and Khazaria, but says nothing explicitly about their Judaism. The latter might be implied from his mention of a certain Isaac Bar Abraham of Iraq who went to Khazaria and settled there.106 But, it could be argued that the Khazar economy attracted men of a variety of religions and Saadiah’s comment might only indicate that there was a Jewish community there. Saadiah’s lack of explanation about Khazar Judaism might also indicate that it was so well-known to his audience that there was no need to belabor the obvious.

The Qaraite scholar, al-Qirqisânî, writing ca. 937, in his comments on Genesis 9:27, mentions that “some other commentators are of the opinion that this verse alludes to the Khazars who accepted Judaism.”107 Again, there is no explanation, but rather the sense that the audience knew well what the reference was. Certainly, by this time, the association  of Khazaria and Judaism in the Jewish world was an established fact, not requiring further commentary. Zvi Ankori who examined these and other statements emanating from the Qaraite communities, concluded that the general tenor of al-Qirqisânî’s remarks displayed a certain lack of enthusiasm for these converts.

105 Yâqût, Mujam al-Buldân (Beirut ed.), II, p. 369. Earlier, II, p. 368, he had only noted “their king is a Jew” (wa malikuhum yahûdî); Ibn Fadlân/Togan, Arabic text, p. 45/Germ. trans. p104. The edition by Sâmî ad-Dahân, Risâla Ibn Fadlân (Damascus, 1379/1959) omits this part.
106 See citations in Dunlop, History, pp. 220–221.
107 Text and translation cited in Z. Ankori, The Karaites in Byzantium (Jerusalem- New York, 1959), pp. 67–68; see also Dunlop, History, p. 132n.44. Genisis: 27 reads: “God shall enlarge Japheth and he shall dwell in the tents of Shem; and Canaan shall be his servant.”

Other Qaraite commentators, such as Yafet b. 'Alî of Basra (fl. 950–980), maligned them with the charge of bastardy. Ankori, on the basis of Qaraite hostility towards the Khazars, considers it quite unlikely that the Khazars would have been converts to Qaraism.108 This appears to find further confirmation in the genuine Khazar documents. In particular, the letter which is believed to have stemmed from the Khazar community in tenth century Kiev, shows no traces of non-Rabbinical Judaism.109 Al-Masûdî, in his description of the Khazar capital, Atıl/Itil, writes: “In this city there are Muslims, Christians, Jews and pagans. As concerns the Jews, they are the king (malik, by which he later notes he means the Qağan), his entourage and the Khazars of his tribe (min jinsihi). The king of the Khazars converted to Judaism during the caliphate of (Hârûn) ar-Rašîd (reg. 786–809, pbg). Some Jews joined him, arriving there from various Islamic urban centers and from Byzantium (Rûm). This was because the king of Rûm, in our time i.e in 332/943 it is Armanûs (Romanos Lekapenos, reg. 920–944), converted those Jews who were in his kingdom to the Christian religion, using coercion on them...” He goes on to note that many Jews fled from Byzantium to Khazaria as a consequence of this. He also writes that he has discussed the conversion of the Khazar ruler in another work.110 Unfortunately, this work has not come down to us. This is, however, the clearest statement we have for the conversion to Judaism at the highest levels in Khazar society and for an influx of Jews from abroad.

It is quite likely that we have an abridged version of the conversion account in ad-Dimašqî, writing ca. 727/1327, who attributes the notice to Ibn al-Atî r and has clearly jumbled, other, imperfectly understood, sections from Ibn al-Miskawaih. He also places it in the time of Hârûn ar-Rašîd.111

108 Although some tenth century Palestinian Qaraite authors show evidence of messianic expectations from Khazaria, see Ankori, The Karaites, pp. 64–78.
109 Golb, Pritsak, Khazarian Hebrew Documents, pp. 30–32. The interpretation of the names of the letter’s signatories is not without problems, see A.N. Torpusman, “Antroponimiia i etnicheskie kontakty narodov Vostochnoi Evropy v srednie veka” in M. Chlenov (ed.), Imia — étnos — istoriia (Moskva, 1989), pp. 48–53. See some of the questions raised regarding the place of composition of this letter in Erdal’s contribution to this volume.
110 Al-Masûdî, Murûj, ed. Pellat, I, p. 212. On the persecution of Byzantine Jews by Romanos, see A. Scharf, Byzantine Jewry. From Justinian to the Fourth Crusade (London, 1071), pp. 97–99.
111 Ad-Dimašqî, Nuhbat ad-Dahr, ed. F. Mehren, p. 263; Ibn Miskawaih, Tajârub al-Umam, ed. H.F. Amedroz, trans. D.S. Margoliouth (Oxford, 1920–1921), II, p. 209; Ibn al-Atî r, Al-Kâmil fî’t-Tarîh, ed. C.J. Tornberg (Leiden, 1851–1876, reprint: Beirut, 1965–1966 with differing pagination), VIII, p. 565.

Of the other surviving works attributed (albeit with some considerable uncertainty) to al-Masûdî, the ‘Ajâib ad-Dunyâ and the Ahbâr az-Zamân we find the brief notice that “as concerns the Turks..., they possess no religion. Some of them profess the faith of the Majus and others are Judaized.”112 This tells us little.

Al-Istahrî (ca. 951, but most probably harkening back to an earlier time) reports that “their king is a Jew” and that “the Khazars are Muslims, Christians and Jews and among them are idolaters. The smallest group are the Jews, most of them are Muslims and Christians, except the king and his people of distinction (who) are Jews. The morality of the pagans prevails among them.”113 Ibn Hawqal (writing in the 970’s) repeats essentially the same notice as does also the thirteenth century compiler, Yâqût (ca. 1229).114 This theme limiting Judaism to the Khazar ruling elite is also reflected in the mid-eleventh century author, Ishâq ibn al-Husain and in the anonymous twelfth century Risâla fî’l-Aqâlîm.115

From about the mid-tenth century also stems the famous Khazar Correspondence between Hasdai ibn Šaprût, the Jewish courtier of the Spanish Umayyads and the Khazar king, Joseph. Stemming from this same era is the “Letter of an Unknown Khazar Jew” or Cambridge Document” (previously called the “Schechter Document”) found among the treasures of the Cairo Geniza and dated, perhaps, to ca. 949.116 The issue of the authenticity of the Correspondence has a long and mottled history which need not detain us here. Dunlop and most recently Golb have demonstrated that Hasdai’s letter, Joseph’s response (dating perhaps from the 950s)117 and the “Cambridge Document” are, indeed, authentic.

112 For the ‘Ajâ’ib ad-Dunyâ (Bursa Hüsein Çelebi Kütüphanesi, Ms. 746, ff. 63b– 64b), see R. Şeşen (ed. trans.), El-Cahiz, Hilâfet Ordusunun Menkibeleri ve Türkler’in Fazîletleri (Ankara, 1967), p. 32; Ahb âr az-Zamân wa man abâdahu’l-Hidtâ'n, ed. H. 'Asî (Beirut, 1386/1966), pp. 98–99. See also discussions in D.M. Dunlop, Arab Civilization to A.D. 1500 (New York, 1971), p. 110; T. Khalidi, Islamic Historiography. The Histories of Masûdî (Albany, 1975), pp. 154–155.
113 Al-Istahrî, ed. De Goeje, p. 220.
114 Ibn Hawqal (Beirut ed.), p. 330; Yâqût, Mu'jam (Beirut ed.), II, p. 367.
115 See V.F. Minorsky, “The Khazars and the Turks in the Âkâm al-Marjân” Bulletin of the School of Oriental Studies, (1937), p. 142: “their supreme king professes the Jewish faith.” For the Risâla see Şeşen, El-Cahiz, pp. 33–35: “their supreme king professes Judaism... Most of them practice Islam.”
116 See Zuckerman, “On the Date” REBs, 53 (1995), p. 240. On the relationship of the “Cambridge [Schechter] Document” to Hasdai’s correspondence, see Golb, Pritsak, Khazarian Hebrew Documents, esp. pp. 94–95

What we have are copies (as Golb suggests), perhaps from an 11th century codex of Hasdai’s diplomatic correspondence.118 The “Cambridge Document” has many controversial points, which we need not consider at this moment, focusing, instead, on the conversion narrative contained within it. Jews are said to have arrived in Khazaria from or via Armenia at some unstated time (perhaps as early as the period of persecution unleashed by the Byzantine Emperor Heraclius (610–641) in the early 630s).119 In Khazaria they intermingled with the Khazars while preserving, apparently imperfectly, elements of their ancestral faith. Subsequently, one of the prominent Jewish military leaders, coaxed by his devout wife and father-in-law, began to espouse more vigorously a form of Judaism more closely adhering to traditional Jewish norms. This angered the Byzantines and the Muslims. It was at this stage that this “great chief ” (ha-sar ha-gadol) called for a religious disputation to settle the roiling religious question. In its aftermath (the conclusions are nowhere clearly stated), the Khazar officers called for Jewish books which had been kept in a “cave in the plain of Tyzwl (תיזול). These were produced and the explanations offered by the “sages of Israel” proved critical. “Then,” we are told, “returned Israel, with the people of Qazaria, (to Judaism) completely” and Jews began to emigrate to Khazaria from Iran, Iraq and Byzantium. This strengthening of religious faith also led to centralization of a hitherto diffused form of government (“there was no king in the land of Qazaria; but rather whoever would achieve victories in war would they appoint over themselves as chief officer of the army”).

117 It survives in a Long and a Short Redaction, the manuscript of the Long Redaction (Firkovič Collection) is from the thirteenth century and the Short Redaction [Christ Church, Oxford] from the sixteenth century, see Kokovcov, Perepiska, p. xi; Golb, Pritsak, Khazarian Hebrew Documents, p. 76.
118 See Dunlop, History, pp. 116 ff.; Golb, Pritsak, Khazarian Hebrew Documents, pp. 75–95 and the recent discussion by Zuckerman, “On the Date” REB, 53, (1995), pp. 239 ff. Yehudah b. Barzillai, ca. 1090–1105, in his Sefer ha-Ittim notes that he had seen Joseph’s letter. He offers, however, a disclaimer, stating that “we do not know if the letter is genuine or not and if it is a fact that the Khazars, who are Turks (lit. “sons of Togarmah”), became proselytes...” see Kokovcov, Perepiska, Heb. text, pp. 127–128/Russ. trans., pp. 128–131; Dunlop, History, p. 157. Golb (Op. cit., p. 77) suggests that all these texts may well have come from the Cairo Geniza or other “repositories of old Hebrew manuscripts in the same city.”
119 The Jews were blamed for the earlier fall of Jerusalem to the Sâsânids. On Heraclius’s anti-Jewish policies (including forced baptisms), see W. Kaegi, Heraclius. Emperor of Byzantium (Cambridge, 2003), pp. 29, 79–80, 216–218; Zuckerman, “On the Date” REB, 53, (1995), p. 241.

Now, they “appointed over them one of the sages” who is called kagan (בגן) i.e. Qağan.120 Constantine Zuckerman, who has most recently analyzed this text, suggests that the notion of a “return” to Judaism was a means of getting around the potential divisions between Jews and converts.

It was, “a practical way to save the cohesion of the Khazar people.” Moreover, the primitive, “bookless” Judaism that had existed up to this time finds parallels in the characterization of Khazar beliefs found in the Vita Constantini,121 although Khazar religious beliefs described briefly in that text could just as easily be a reference to the Tengri cult.

Joseph’s response to Hasdai’s letter contains another version of the conversion narrative. In it, the Khazar king (melek') Bulan, “a wise and God-fearing” man drove out the “sorcerers and idol-worshippers.” Having received two heaven-sent dreams as a result of his strivings and having then convinced this heavenly apparition to appear to the “great chief ” (ha-sar ha-gadol, if Bulan is the Beg, then obviously this would be the Qağan)122 to win him over, he then gathered together his “princes, slaves and his entire people” and they converted to the new faith. A third angelic visitation produced a request that he build a temple and assured him military success in a raid on Azarbayjan as a means to gather the necessary funds. Byzantium and the Muslim rulers, having learned of this (clearly there were pro-Byzantine and pro-Muslim factions at the court), sent him gifts and tried to win him over. A religious disputation followed in which Judaism prevailed. The conversion is said to have taken place 340 years before Joseph’s time. Some generations after Bulan, during the reign of Obadiyah, the kingdom was “renewed” and the faith strengthened according to traditional Jewish norms.123

120 See text and translation in Golb, Pritsak, Khazarian Hebrew Documents, pp. 106– 114.
121 Zuckerman, “On the Date” REB, 53, (1995), pp. 242, 244.
122 Zuckerman, “On the Date” REB, 53, (1995), p. 252.
123 Kokovcov, Perepiska, Heb. text, pp. 21–24, 28–31/trans. pp. 75–80, 92–97. The dating is only noted in the Long Redaction. See also Pritsak, “Conversion” HUS, II (1978), pp. 272–276. Zuckerman, “On the Date” REB, 53 (1995), pp. 249–250, has suggested that Obadiyah may be a later interpolation. He is not mentioned in Jehuda b. Barzillai’s summary. I have also previously (“Khazaria and Judaism” Archivum Eurasiae Medii Aevi, III (1983), pp. 147–148) expressed the opinion that the Obadiyah episode may be nothing more than a “pious topos.” On the other hand, his activities do fit into the Eurasian conversion pattern, see below.

Al-Muqaddasî (writing ca. 985) describes Khazaria as “a broad district beyond the (Caspian) sea, (a land of) squalor and woe, of many sheep, honey and Jews.”124 The Khazar state was mortally wounded by the Rus’ campaigns (in alliance with the Oğuz) of 965–969. In this notice, if it is a reflection of the situation at the time of al-Muqaddasî’s writing, we find evidence that there were still large numbers of Jews in Khazaria. On the other hand, he also notes in his description of the “Khazar,” the Khazar capital on the Volga, that its inhabitants had earlier moved to the sea coast (i.e. the Caspian?) and then returned. “They converted to Islam since they had earlier been Jews.”125 This clearly refers to the time after the Rus’-Oğuz overrruning of the city. Muslims had, in all likelihood, comprised the majority — or at least plurality — of the population of the capital. After the disasters of 965–969, many Jews and Judaized Khazars had undoubtedly fled the city. Those that remained appear to have converted to Islam. The continuing existence of Khazar Jewish communities, however, may be seen in the account of the Rus’ chronicles of the series of religious interrogations conducted by the ruler of Kiev, Vladimir I, in 986 who declined Volga Bulğar Muslim, Khazar Jewish and German Catholic invitations to accept their faiths before converting to Orthodox Christianity.126

An-Nadîm, writing ca. 987–988, in a notice on the script systems used by the “Turks and those related to them,” comments that “the Khazars write Hebrew.”127 An echo of this theme is found in Fahr ad-Dîn Mubârakšâh (ca. 1206) who, in a notice clearly conflated from several sources, states that the “Khazars also have a script which is derived from that of the Rûs, a branch of the Rûm who live near them. They write in this script and are called Rûm-Rûs. They write from left to right, the letters are not joined. There are twenty-two letters all together (and no more). The greater part of these Khazars who use this script are Jews.”128

124 Al-Muqaddasî, ed. De Goeje, p. 355.
125 Al-Muqaddasî, ed. De Goeje, p. 361. On the dating of the Rus’ campaigns, see I.G. Konovalova, “Padenie Xazarii v istoričeskoj pamjati raznyx narodov” Drevnejšie gosudarstva Vostočnoj Evropy 2001 (Moskva, 2003), pp. 171–190.
126 Polnoe sobranie russkix letopisej (Moskva-St. Peterburg/Petrograd/Leningrad, 1843– 1995), I, cc. 84–85. Judging from the text these were not local, Kievan Jews, but Jews who came from the Khazar lands.
127 An-Nadîm, Fihrist, ed. Flügel, I, p. 20, Dodge trans. I, pp. 36–37.
128 Ta’rîkh-i Fakhru’d-Dîn Mubârakshâh, ed. E. Denison Ross (London, 1927), p. 46; see also V.V. Bartol’d (Barthold), “O pismennosti u xazar” in his Sočinenija, V, p. 466.

A Sunnî qâdî, ‘Abd al-Jabbâr b. Muhammad al-Hamdânî of Rayy (Iran) in a polemical work directed against the Ismâ'îlîs and other Shî'ites as well as the Christians written ca. 400/1009–1010, cited the Khazar conversion, which he placed “recently in the days of the ‘Abbâsids and during their rule” as an example of a bellicose, violent people who were converted by a single proselytizer using the power of argument rather than military force.129 The individual is left unnamed.

Al-Bakrî (scr. 1086, d. ca. 1094) has a longer notice for which he, obviously, drew on a number of sources: “in general, the Khazars are Muslims and Christians. Among them are (also) idolators. The smallest of the groups among them is the Jews.130 Their king professes the Jewish faith.” Having been first a pagan (majûs) and then a Christian, he “saw the wrongness of his what he believed” and convened a religious disputation “between the three Peoples of the Book (ash âb al-kitâb).” A learned local Jew bested the Christian bishop in debate and then had someone poison his Muslim opponent. Thus, “the Jew won the king’s favor (istimâla) to his religion and he became a Jew.”131

Interestingly enough, it is one of our latest sources, Yehudah Halevi, who, in the opening lines of his famous defense of Judaism written in Arabic in 1140 (the Kitâb al-Hujjâh wa’d-Dalîl fî Nasr ad-Dîn ad-' D'alîl or more popularly Kitâb al-Hazarî, it was translated into Hebrew, in 1167 by Yehudah Ibn Tibon, the Sefer Ha-Kuzarî), says that having been asked to “provide refutations” to defend Rabbinical Judaism, he remembered “the arguments of the rabbi who studied with a Khazar king, who converted to Judaism some four hundred years ago.”132 His contemporary, Abraham ibn Daud, writing in the 1160’s confirms the interest in Khazar matters among Iberian Jewish intellectuals. He mentions the Khazar Jewish community as part of the larger Rabbinical, as opposed to Qaraite, community, knows of the Khazar Hebrew correspondence and even remarks that “we have seen some of their descendants in Toledo, scholars who informed us that their legal practice conforms to Rabbanite usage.”133

129 See the Kitâb Tatb'ît Dalâ’il Nubuwwat Sayyidinâ Muhammad and discussion in S. Pines, “A Moslem Text Concerning the Conversion of the Khazars to Judaism” Journal of Jewish Studies 13 (1962), pp. 45–55, Arabic text (in transcription), p. 55, Eng. trans. p. 47. The notice reads: “One of the Jews undertook the conversion of the Khazars (who) are (comprised of) many peoples, and they were converted by him and joined his religion.”
130 Al-Bakrî’s original text has: wa ahl al-firq minhum al-yahûd which the editors (see below) have restored to wa aqall al-firaq minhum al-yahûd (“and the smallest of the groups among them is the Jews”) from Al-Istahrî (ed. De Goeje, p. 220): wa aqall al-firaq al-yahûd.
131 Al-Bakrî, Kitâb al-Masâlik wa’l-Mamâlik, ed. A.P. van Leeuwen, A. Ferre (Beirut, 1992), I, pp. 446–447.
132 Yehuda Halevi, The Kuzari. In Defense of the Despised Faith, trans. N.D. Korobkin (Northvale, New Jersey-Jerusalem, 1998), p. 1.
133 Abraham Ibn Daud, Sefer Ha-Qabbalah. The Book of Tradition, ed. trans. G.D. Cohen (Philadelphia, 1967), Eng. trans. pp. 92–93/Heb. pp. 67–68.

The Khazar Hebrew correspondence is also noted by the somewhat older Yehuda ben Barzillay of Barcelona (ca. 1090–1105), who, however, was less certain about its authenticity and indeed of the fact of the Khazar conversion.134

Despite the overwhelming evidence of the written sources, archaeological evidence for Judaism is hard to come by. One of the brick buildings in Sarkel (a fortress-trade center on the Don built with Byzantine assistance in 840–841)135 may have been a synagogue, but this is by no means certain. There are no other traces of Judaic or Christian belief in that complex that have been uncovered thus far.136 Warrior graves with Jewish symbols are found at Čelarevo (70 km. from Belgrad) in Yugoslavia.

These may have been Qabar. Given the absence of Jewish, Christian or Muslim religious paraphernalia in the Khazar sites investigated thus far, Pletnëva concluded that paganism remained the “unifying religion” of Khazaria.137 There can be little doubt that paganism was an important element in Khazar culture and probably a significant substratal element in the religious beliefs of those that converted to one or another of the monotheistic religions. But, one must ask, are we to expect much religious
paraphernalia in a relatively recently converted steppe society? Do the Oğuz, in the century or so after their Islamization, present much physical evidence in the steppe for their new faith? These conclusions must be considered preliminary.

134 Kokovcov, Perepiska, Heb. text, pp. 127–128/Russ. trans., pp. 128–131; Dunlop, History, p. 132.
135 On the dating of the construction of Sarkel, see now C. Zuckerman, “Two Notes on the Early History of the Thema of Cherson” Byzantine and Modern Greee Studies 21 (1997), pp. 213–214.
136 There is some evidence of human sacrifice, obviously associated with pagan cults, see S.A. Pletnëva, Sarkel i “Šëlkovyj Put’ ” (Voronež, 1996), pp. 78–80.
137 Pletnëva, Očerki, pp. 215–217. On the Čelarevo finds, see also Vl. Petrukhin, “The Decline and Legacy of Khazaria” in P. Urbanczyk (ed.), Europe Around the Year 1000 (Warsaw, 2000), pp. 112–113. A recent summation of the Čelarevo complex by R. Bunardžič, “Čelarevo — nekropol’ i poselenie VIII–IX vekov” in Xazary, ed. V. Petruxin, V. Moskovič et al. in Evrei i Slavjane 16 (Ierusalim-Moskva, 2005), pp. 522–531, dates the finds to the late eighth-early ninth centuries, notes the Judaic objects in the finds, connects them with one of the three groupings represented there, a people with “northern Mongolian” physical characteristics, but is not prepared to declare them Khazar. It seems unlikely, however, to view them as “late Avar” with which they may correspond chronologically as we have no evidence regarding the spread of Judaism to the Avars.

Archaeology has not yet revealed all of its secrets, as Petruxin has noted in his “afterword” to Pletnëva’s book.138  Most importantly, the Khazar capital remains undiscovered.139

Dating and Impact of the Conversion

The dating of the conversion remains a matter of scholarly dispute.

Dunlop, after an exhaustive analysis of the Arabic and Hebrew sources, concluded that “sometime before 112/730, the leading Khazars may have come under the influence of Judaism.” In the aftermath of a religious debate, ca. 740, the Qağan accepted “a modified Judaism” and in 800, moved to full Rabbinical Judaism.140 A cautionary note should be sounded here. In 733, when Constantine Copronymus married Čiček, the daughter of the Qağan, the Byzantine sources make no mention of
Khazar Judaism (although, as was noted previously, they consistently ignored it) nor do we find reference to it in the Georgian Life of St. Abo who was in Khazaria in 786 or in the Arab accounts of the marriage of al-Barmaqî, governor of Arminiyya and Ad arbâyjân in 798/799, to a daughter of the Qağan.141 Pritsak also viewed this as essentially a three-staged process, the first movement towards Judaization taken ca. 730–740, with a second stage coming during the era of Hârûn ar-Rašîd, ca. 799–809 and a final stage ca. 837–843. The second stage he associates with the reforms of Obadiyah noted in Joseph’s letter which he terms “the official conversion of the beg.” The third stage came at the conclusion of the Khazar civil war (the revolt of the Qabars, discussed below), which forced the Qagan to accept Judaism.142

138 Pletnëva, Očerki, pp. 227–230. Petruxin attributes the “elusiveness” of Khazar Judaic monuments to both the incompleteness of the Khazar archaeological record and to the “thin stratum” of Jewish adherents.
139 Al-Istahrî, ed. De Goeje, p. 220, says that the Khazar capital had some thirty mosques for its Muslim inhabitants.
140 Dunlop, History, chaps. v–vi, conclusions on p. 170.
141 See I. Sorlin, “Le problème des Khazares et les historiens soviétiques dans les vingt derniéres années” Travaux et mémoires (Centre de Recherche d’histoire et civilisation byzantines), 3 (1968), p. 441.
142 Pritsak, “Conversion” HUS, II (1978), pp. 278–280.

Dieter Ludwig, following these same stages of Judaization, concluded that by the time of Constantine’s mission (860) the preponderance (Übergewicht) of Jews at the Khazar court indicates that the conversion of the Khazar elite to Judaism had been accomplished, probably sometime before 835–840.143 Artamonov also viewed this as a “long, sometimes unnoticed process” in which the intermingling of Jews with Khazars played a greater role than preaching. The process began, he suggests in Daghestan, one of the early centers of the Khazars. The “culprits” (vinovniki, an interesting choice of words) in this process were local Daghistanian Jews.

He also points to the old Jewish settlements in the Crimea and Taman peninsula (Phanagoria) whose numbers had grown thanks to Byzantine persecutions under Heraclius and Leo III (717–741). A Daghestanian setting for the conversion, however, is complicated by the absence, thus far, of any Jewish cultic objects in the various Khazar sites that have been examined in that region. It should also be added that there are no contemporary traces of Islam either. Local chronicles, toponyms and popular legends, however, do point to old, Pre-Judeo-Tat Jewish settlements and these may go back to Khazar times.144

Following al-Masûdî, Artamonov points to the late eighth-early ninth century as a particular turning point. In his interpretation, the Vita Constantini clearly shows that the Khazars were Judaized. He further argues that the conversion to one of the acceptable monotheisms of Western Asia, given Khazar involvement with both Byzantium and the Caliphate, was a necessity in the aftermath of the events of 737. Faced with either Christianity or Islam, they chose Judaism as a middle way out.

It gained them “entry into the circle of medieval civilization and at the same time secured them an independent status” between the competing Christian and Muslim states. There were also domestic needs, a new ideology to meet the requirements of “new forms of socio-economic relations.” Although a good choice as a demonstration of independence, Artamonov says that Judaism, with its non-proselytizing character was poorly suited to compete with the younger faiths, Christianity and Islam.145

143 Ludwig, Struktur und Gesellschaft, I, pp. 161–163. Since he dates (pp. 328–332) the letter of Christian of Stavelot to a period ca. 835–840 or “a little later,” the conversion of the Khazars must have occurred before that.
144 Magomedov, Obrazovanie, p. 173. There is nothing but conjecture to sustain the thesis that the Khazars converted to the Maġâriyya sect of Judaism in the Caucasus, see D. Lang, “A Kazárok zsidósága” Magyar Nyelv, XLIV (1948), pp. 37–42.
145 Artamonov, Istorija Xazar, pp. 262–273.

The notion that Judaism provided a neutral status between Byzantium and the Islamic World as well as securing commercial relations across the Mediterranean is an old one, going back to the Russian Eurasianist school.146 But, if Judaism here was non-proselytizing, how were the Khazars converted?

In point of fact, conversions of elites or large-scale conversions to Judaism were not very common, but not unknown, e.g. the conversions of the Idumaeans and the Ituraeans in the late second century BCE, the Judaization of the ruling elites of Adiabene (in present day Northern Iraq) in the mid-first century C.E. and of the Himyârî kings in Yemen (perhaps as early as the fourth century C.E.) and the incorporation of Berber elements into North African Jewry in medieval times.147 In Medieval Europe, Jewish policies regarding proselytism varied with time and place, depending on local and external conditions.148 A frontier zone like Khazaria, at the interstices of the great Eurasian and West Asian trade routes and the cultures and religions that traversed this zone borne by merchants and others, was precisely a region in which Judaism, unfettered by Christian or Muslim overlords who prohibited Jewish proselytizing, could freely compete.

Novoselcev dates Khazar interest in monotheistic faiths to the seventh century (e.g. the mission of bishop Israyêl noted earlier). He discounts the notice of the Qağan’s conversion to Islam in 737, noted only by Ibn Atam al-Kûfî and al-Balâdurî, as reflecting a not very reliable oral tradition that was not repeated by more discriminating authors such as at-Tabarî, al-Masûdî and Ibn al-Atî r among others. The Muslim sources fix Judaism as the state religion in Khazaria by the 850– 870’s. Novoselcev considers al-Masûdî’s notice as the only one “worthy of belief.” Beyond that, the sources do not permit a more exact dating of the event.

146 See the comments of G. Vernadskij (Vernadsky), Opyt istorii Evrazii (Berlin, 1934), pp. 51–52.
147 A. Schalit (ed.), The Hellenistic Age: Political History of the Jewish People from 332 BCE to 67 BCE (The World History of the Jewish People, 1st series, vi, New Brunswick- Jerusalem, 1972), pp. 217–224; J.R. Rosenbloom, Conversion to Judaism from the Biblical Period to the Present (Cincinnati, 1978), pp. 94–98. On Jewish proselytism in Antiquity see L.H. Feldman, Jew and Gentile in the Ancient World (Princeton, 1993), chap. 9. On the Jews of Arabia, see G.D. Newby, A History of the Jews of Arabia From Ancient Times to Their Eclipse Under Islam (Columbia, South Carolina, 1988), esp. chap. 4; H.Z. Hirschberg, “The Problem of the Judaized Berbers” Journal of African History, 4 (1963), pp. 313–339. The origins of Ethiopian (Falasha) and Chinese Jewry (the now nearly completely absorbed Kai-feng community) are inexplicable without reference to converts.
148 See brief discussion of the differing schools with reference to the Khazar question in Golden, “Khazaria and Judaism” AMAEe, III (1983), pp. 132–134. On conversions in Western Asia and Europe, see P. Wexler, The Ashkenazic Jews. A Slavo-Turkic People in Search of a Jewish Identity (Columbus, Ohio, 1993), pp. 181 ff.

He attributes the process of conversion to the Išâd/Beg who forced the Qagan to accept Judaism. Islam and Christianity were excluded because of political complications whereas, in his view, the Jews, on whom the Carolingians and Spanish Umayyads looked with special favor, “monopolized European trade,” controlling the transit trade between Europe and Asia. Khazaria attracted Jews fleeing persecutions in Byzantium and emigrants from the Caliphate. The Jewish colonies in the Crimea were also significant. Nonetheless, based on a selective reading of the Muslim sources, Novoselcev concludes that Jews comprised the smallest grouping of those espousing monotheistic faiths in Khazaria. Judaism and Jewish culture (including literacy in Hebrew) had some impact, but so did Islam which in the tenth century was gaining ground and was the religion of the royal army.149

Most recently, Constantine Zuckerman has argued for the primacy of the Vita Constantini in dating the conversion. He dismisses al-Masûdî’s notice as a “somewhat confused resumé of a more detailed account in a work which is no longer preserved,” but holds open the possibility of rehabilitating him in light of the revival of Jewish traditions noted in the Cambridge Letter.150 Why al-Masûdî should be confused about his own work and what elements of confusion are to be found in his account are not spelled out. Rather, Zuckerman says that since only one religious debate took place (the Cambridge Document and the Vita Constantini are describing one and the same event), in 861, the conversion must have occurred not long after the debate. Conversion was in one stage, at the initiative of Bulan, the Beg/Military chief. It was probably prefaced by some individual conversions (this may be the background to al-Masûdî’s notice). Zuckerman contends that “the spread of Judaism among the Khazars was, in reality, more gradual and slower than the Cambridge Document would admit (though in no way limited to the upper class). He concludes, however, that it had only a “limited penetration” among the Khazars.151 Finally, he suggests that “the link between the conversion and the new mode of government, as established in the Genizah Letter, appears to be wholly justified.”152

149 Novoselcev, Xazarskoe gosudarstvo, pp. 144–154.
150 Zuckerman, “On the Date” REB, 53 (1995), pp. 246, 250.
151 Zuckerman, “On the Date” REB, 53 (1995), pp. 244–245, 250.
152 Zuckerman, “On the Date” REB, 53 (1995), pp. 252–253; see also his “O proisxoždenii dvoevlastija u xazar i obstojatel’stvax ix obraščenija v iudaizm” Materialy po arxeologii, istorii i étnografii Tavrii IX (2002), pp. 521–534. J. Shepard, “The Khazars’ Formal Adoption of Judaism and Byzantium’s Northern Policy” Oxford Slavonic Papers, n-s. XXXI (1998), pp. 13–14 essentially follows Zuckerman’s dating.

The linking of the dual kingship in Khazaria with Judaization has appeared over the years in a number of variants. Artamonov placed the emergence of the dual kingship in the early ninth century and connected it with the activities of Obadiyah, to whom he attributed the imposition of full Rabbinical Judaism and a coup d’état that reduced the Qagan to a largely ceremonial position.153 This produced the revolt of the Qabars.154 The Qagan was also compelled to convert to Judaism. This form of governance did not derive from Judaism, but was, he suggests, an innovation resulting from the replacement of one ruling dynasty by another.155

Pletnëva identified Obadiyah with the unnamed Khazar ruler (the Qagan in her view) mentioned in al-Masûdî’s notice as converting to Judaism. The consequence of this Judaization of the Qağan, Beg and Itil aristocracy, she argues, was to alienate the ruling clique from the rest of the Khazar aristocracy which resulted in a power struggle between the non-Judaized provinces and the capital. Pletnëva concluded that Obadiayah was among those who perished in this “Khazarian Fronde” and the state itself was weakened.156

Pritsak also placed the emergence of the Beg to sometime after 799 but before 833 when the Beg 157 is shown playing an important role in foreign affairs (the building of Sarkel). The Beg, who, he conjectures, is the major domo from the Iranian *Barč/Warâz/Bolčan clan, is identified with Obadiyah and the Khazar ruler who converted during the reign of Hârûn ar-Rašîd. This Beg forced the Qağanal clan to convert to Judaism, producing the “Judaization of the institution of the Khaghanate,”
according to which only a Judaized Khazar could hold that position (cf. al-Istahrî). Pritsak dates the Qabar revolt (a reaction in his view to Judaization) to between 833 and 843 (or 835), by which time the Qagan “had lost all power” and had, indeed, been compelled to Judaize as well.158

153 Cf. Ibn Fadlân/Togan, Arab text, pp. 43–46/trans. 139–140; al-Mas‘ûdî, Murûj, ed. Pellat, I, pp. 212, 214–215; Al-Istahrî, ed. De Goeje, p. 224 etc.
154 Constantine Porphyrogenitus, De Administrando Imperio, ed. Gy. Moravcsik, trans. R.J.H. Jenkins (Washington, C.D., 1967), pp. 174/175: “The so-called Kabaroi were of the race of the Chazars. Now, it fell out that a secession was made by them to their government, and when a civil war broke out their first government prevailed, and some of them were slain, but others escaped and came and settled with the Turks in the land of the Pechenegs...”
155 Artamonov, Istorija Xazar, pp. 275, 280–282., 324 ff.
156 S.A. Pletnëva, Xazary (2nd ed., Moskva, 1986), pp. 62–66.
157 See Constantine Porphyrogenitus, DAI, pp. 182/183: ο πεχ Χαζαπιας
158 Pritsak, “Conversion” HUS, II, (1978), pp. 278–280 and his “Turkological Remarks on Constantine’s Khazarian Mission in the Vita Constantini ” in E.G. Farrugia et al. (eds.), Christianity Among the Slavs. The Heritage of Saints Cyril and Methodius in Orientalia Christiania Analecta, 231 (Roma, 1988), pp. 295–298.

As for the much-debated question of the dating of the conversion, we now have some new evidence that gives added weight to al-Masûdî’s notice. It has become clear that the Khazars in the early ninth century minted coins, perhaps in response to a decline in Muslim minting in the 820s. The Khazar coins were invariably imitations of Islamic dirhams.

However, they included, in some instances, the inscription ard al-hazar (“Land of the Khazars”), Turkic tamğas and some dirhams with the most striking formula (for imitation Islamic coins): Mûsâ rasûl Allâh “Moses is the Messenger of God,” a clear substitution for the Islamic Muhammad rasûl Allâh. The five “Moses dirhams” uncovered thus far have been dated to 837/838 on the basis of die-chains.159 There can be little question that, at the least, the governing strata of Khazaria had been Judaized by this time — perhaps relatively recently so (as al-Masûdî indicates) and wanted to dramatically and visibly assert this new religious profile in the turbulent 830s. Coins were important ideological symbols and in the medieval “Age of Faith” projected the official religion of the state or at least that of its ruling group.

As I have noted elsewhere, the institution of a dual kingship, often encompassing a sacral king as well, is a widespread phenomenon in Eurasia — indeed worldwide.160 There is nothing but conjecture to connect it with the reforms of Obadiyah, the further evolution of Khazar Judaism or the Qabars (who may have had Judaic elements in their midst). The fact is we do not know when, precisely, the Khazar system of dual kingship emerged. It could not have come ex nihilo. It was not present in the early stages of Khazar history. Given the Old Türk traditions of the Khazar state (described in al-Istahrî and mirroring the same investiture ceremonies depicted in the Chinese sources for the Türks)161 and the overall institutional conservatism of steppe society, one must exercise great caution here.

159 R.K. Kovalev, “What Does Historical Numismatics Suggest About the Monetary History of Khazaria in the Ninth Century? — Question Revisited” Archivum Eurasiae Medii Aevi 13 (2003), pp. 106–114 and his “Creating Khazar Identity through Coins: The Special Issue Dirham of 837/8” in F. Curta (ed.), East Central and Eastern Europe in the Middle Ages (Ann Arbor, 2005), pp. 220–253.
160 See Golden, “Khazaria and Judaism” AEMAe, III (1983), pp. 147–149 and the literature cited there.
161 Al-Istahrî, ed. De Goeje, p. 224; Liu, Die chinesischen Nachrichten, I, p. 8.

Clear evidence for it is relatively late (the latter part of the ninth century perhaps and more probably the tenth century — although it was probably present by the first third of the ninth century. Iranian influences via the Ors guard of the Qağans may also have been a factor.162 These were societies that were not given to political
experimentation or innovation. Finally, why would Jewish “reformers” create a sacral monarchy still laden with pagan elements?163 Judaization, however, would certainly play a role in how the Qaganate presented itself to Jewish audiences (such as Hasdai b. Šaprût) and the Islamo-Christian world. Over time it also became internalized.


Pritsak has suggested that Joseph’s letter represents the “official version of the proselyte dynasty” and the Cambridge Document is an “unofficial version of indigenously Jewish circles.”164 DeWeese, however, cautions us that “Khazar converts might be just as likely to assert their ancestral links with Jewish tribes as immigrant Jews would have been to assert their primacy in bringing the religion to the Khazars.” Indeed, these two accounts show a creative synthesis of the two traditions, Khazar paganism and Judaism.165 Zuckerman contends that the Khazars created, in effect, a myth of a Khazar “return” to Judaism, so that the various stages of Judaism could be presented as a revival and the unity of the converts and Jews maintained.166 One would hardly expect otherwise. This is an important element in the internalization of the new religion. Clear evidence that the Khazars were doing this can be found in the Kievan Khazar Hebrew letter. Turkic shamans (qam), it would appear, could become kôhêns and Levites, as Golb has suggested. This was part of what he terms “the construction or gradual evolution of a mythic substratum justifying and explaining the conversion.” The “sacerdotal metamorphosis”

162 See P.B. Golden, “The Khazar Sacral Kingship” Pre-Modern Russia and Its World, ed. K. Reyerson, Th. Stavrou, J. Tracy, Schriften zur Geistesgeschichte des östlichen Europas (Wiesbaden: Harrassowitz Verlag, 2006), pp. 79–102
163 See also the comments of V. Ja. Petruxin, “K voprosu o sakral’nom statuse xazarskogo kagan: tradicija i real’nost’ ” Slavjane i ix sosedi 10 (2001), pp. 73–78, who suggests (p. 77) that Judaization may have “desacralized the status of the earthly ruler — he   became the embodiment of the Law, but not a Divinity.” In any event, Judaism did not shape or alter the qağanal office.
164 Golb, Pritsak, Khazarian Hebrew Documents, p. 132.
165 DeWeese, Islamization and Native Religion, p. 305.
166 Zuckerman, “On the Date” REB, 53, (1995), pp. 241–242.

The “sacerdotal metamorphosis” of individuals or groupings from the old shamanic element that had previously conducted sacrifices for the community perfectly accords with this new world view.167

Another manifestation of this synthesis of old and new can be seen in the cave motif in the conversion narratives. The ethnogonic myth of the Türks centered on the ancestral cave in which the Ashina were conceived from the mating of their human ancestor and a wolf ancestress. The Ashina-Türks, from whose western ruling branch the Khazar ruling house most probably derived, continued to offer sacrifices at the ancestral cave.168 As DeWeese has noted, this myth and ritual complex became highly politicized during the period of the Second Türk Empire and it would be surprising “not to find echoes of this complex among the Khazars.” Moreover, “the complex of mountains, sea and cave” of the Cambridge Document mirrors that of other Inner Asian ethnogonic myths.169

Clearly the notion of a “return” to Judaism had been internalized and is reflected in the conversion narratives. Joseph, while placing the genealogy of himself and his people, as descendants of Japheth, in a context that would be familiar to Hasdai, also notes that the eponymous “Khazar,” was one of the ten sons of Togarmah from whom Jews of the Middle Ages understood the Turks to derive.170

The “reforms” of Obadiah, if they are not a pious interpolation, also fit into the familiar pattern of conversion and internalization. After several generations (in some groupings considerably longer), the new faith has not only supplanted and covered over the more obvious elements of the old religion, but has, inevitably, produced those learned enough in its precepts to desire to bring it into full conformity with the “proper” laws and traditions of the faith. We are moving, using Eaton’s model, from inclusion to displacement.

Some Conclusions

Conversion probably took place in stages. It began, as so often was the case, at the top, with the ruling house, perhaps episodically as early as the mid-eighth century (certainly after 737), but not extending at that stage much beyond the ruling strata. Some important stage must have been reached by the era of Hârûn ar-Rašîd, as reported by al-Masûdî.

167 Golb, Pritsak, Khazarian Hebrew Documents, pp. 27–29.
168 Liu, Die chinesischen Nachrichten, I, p. 5; Chavannes, Documents, p. 15.
169 DeWeese, Islamization and Native Religion, pp. 300–305.
170 Kokovcov, Perepiska, Heb. text, p. 19/trans. 72, where he titles himself “the Turkic
king” (ha-melek' ha-togarmî), pp. 20–21/74.

The “Moses coins” clearly reflect the growing weight at the top political levels of society. This was also a period of conflict with Byzantium which was trying to strengthen the Christian presence in Khazaria. The now overt movement towards Judaism may have represented a critical shift in foreign policy. Perhaps, the Byzantine contribution to the building of Sarkel (ca. 840–841) was not only to deal with threats in the steppe (Hungarians and perhaps more distantly, Pečenegs), but also to win favor with the Khazar governing elite? The conversion of the Khazar ruling elite to Judaism could not have been welcome news in Constantinople.

Conversion at the top does not necessarily mean immediate mass conversion. We may note here the example of the Činggisid Khan Berke (1257–1267), ruler of the Jočid ulus who converted to Islam before his accession and was followed by other Islamized rulers, but the movement of important, critical sectors of this ulus to Islam occurred only after the conversion of Özbek Khan (1312–1341). At the time of Constantine’s mission, the Qağan may have still been willing to entertain other religious systems or at least appear to do so, although Jews and Judaized elements within the immediate ruling strata, were most probably already guiding the internal dialogue on this question. By the latter part of the ninth century and early tenth century, Judaism, at least within the core tribes, was becoming more widespread, hence the notices in Ibn al-Faqîh (“all of the Khazars are Jews, but, they have been Judaized recently”) and Ibn Fadlân (“the Khazars and their king are all Jews”). In the history of religious conversion in the steppe this is what one would expect to find.

Moreover, these sources directly contradict the assertion by Artamonov and others that Judaism was limited only to the ruling elite and “never became the religion of the Khazar people or more precisely those tribes that formed Khazaria.”171 On the contrary, Khazar Judaism may have also begun to reach subject or allied peoples. Thus, Seljukid tradition accords Old Testament names (Mîkâ’îl, Isrâ’îl, Mûsâ and Yûnus) to the sons of Seljuk, whose father Toqaq Temür Yalığ, according to some
accounts, was in the service of the Khazar Qağan.172

171 Artamonov, Istorija Xazar, p. 266.
172 C. Cahen, “Le Malik-Nameh et l’histoire des origines Seljukides” Oriens, II (1949), pp. 31–65. See also Dunlop, History, pp. 258–261.

The conversion narratives give differing versions regarding the actual agents of conversion, the Khazar Jewish accounts emphasizing its “internal” sources, i.e. stemming from Khazar Jews who have recovered their ancestral faith. Pritsak has suggested that the Khazars, like their contemporaries, the Uyğurs, were introduced and converted to these universalistic faiths (Judaism and Manichaeanism respectively) by long distance merchants.173 With the Uyğurs this is certainly true. Like the Türks before them, they had developed a symbiotic, political, cultural and economic relationship with the Soġdians, the great culture-bearing merchants along the Silk Route. Did Jewish merchants, who appear to have been outnumbered in the Khazar capital by the large Muslim community, play a similar role? Were the local Jewish merchants affiliated with the Râd âniyya? Our sources provide no clear-cut answer.

Trade and the spread of religions and the cultural paraphernalia associated with them (e.g. script systems) do seem to be connected across Eurasia,174 and it is not implausible to posit a similar kind of connection in Khazaria. But, this is not an absolute certainty. These powerful royal courts invariably attracted “holy men” of non-commercial origins who proselytized because they believed they were doing the bidding of their God. Later Jewish sources, beginning in the thirteenth century (Nachmanides) mention the still elusive figure of Isaac Sangarî who is credited with converting the Khazar king.175 Of course, in any number of instances, the holy man and merchant could be one and the same. The tradition of debates or royal interrogations is also well established in the Inner Asian conversion paradigm.176

173 Pritsak, “Conversion” HUS, II (1978), pp. 280–281.
174 Cf. the studies by Bentley, Old World Encounters. On the interplay of religion and commerce, see Xinru Liu, Silk and Religion and P. Risso, Merchants and Faith. Muslim Commerce and Culture in the Indian Ocean (Boulder, Colo., 1995).
175 Those who contend that the Khazars converted to Qaraism maintain that Sangari was a Byzantine Qaraite, see discussion in Dunlop, History, pp. 122–125; Ş. Kuzgun, Hazar ve Karay Türkleri 2nd ed. (Ankara, 1993), pp. 151 ff.; S. Szyszman, Le Karaïsme (Lausanne, 1980), p. 71; R. Freund, Karaites and Dejudaization, Acta Universitatis Stockholmiensis, Stockholm Studies in Comparative Religion, No. 30 (Stockholm, 1991), pp. 38–39. Recent archaeological investigations have completely undermined this theory, see D. Shapira, “Yitshaq Sangari, Sangarit, Bezalel Stern, and Avraham Firkowicz: Notes on Two Forged Inscriptions” Archivum Eurasiae Medii Aevi 12 (2002–2003), pp. 223–260
176 DeWeese, Islamization, pp. 169–173.

The “reform” movement inaugurated by Obadiyah, if, indeed, true, similarly follows the conversion paradigm of Inner Asian peoples: nominal conversion, backsliding followed by a renewal and deepening of the faith along with bringing it more into conformity with the norms of that faith, the pattern, as DeWeese terms it, of “summons, consent, test and decisive affirmation.”177 The Khazar conversion, seen in its Eurasian context, does not, in fact, appear so exceptional.

The impact of the Khazar conversion in terms of Khazaria’s relations with its neighbors is also difficult to assess. It has been claimed that after the conversion, Constantinople devoted more efforts to the Caucasus and the Rus’ danger. Indeed, Judaization, never mentioned in the Byzantine chronicles, in effect ended the Khazaro-Byzantine entente. Others contend that no real change occurred.178 It is clear from Constantine Porphyrogenitus’s De Administrando Imperio that in his day Byzantium was relying more on the Pečenegs as an instrument of policy in the steppe. This, however, may have been the normal consequence of the Khazar decline. There is no overt indication of a religious issue, the Byzantines remaining remarkably silent on this question. This silence may itself tell us much about the Byzantine reception of the growing Judaization of the Khazars. The Byzantines had failed to bring the Khazars into Christendom. This was a foreign policy defeat and undoubtedly the source of some resentment in Constantinople. As for the Islamic world, with which the Khazars had a contentious early history, we have only the report from Ibn Fadlân that the Khazar king destroyed the minaret of the Friday mosque in Atıl in retaliation for the destruction of a synagogue in Dâr al-Bâbûnaj. The king is alleged to have said that he would have done worse if not for fear of Muslim retaliation against Jews.179

177 DeWeese, Islamization, pp. 302, 314–315.
178 G. Huxley, “Byzantinochazarika” Hermathena, 148 (1990), p. 85. J. Shepard, “The Khazars’ Formal Adoption of Judaism and Byzantium’s Northern Policy” Oxford Slavonic Papers, n.-s., XXXI (1998), pp. 11–34, posits a mass conversion of the Khazars taking place after 860 which was “disappointing and humiliating,” a “spectacular blow to the Byzantine establishment’s ambitions for the north-east of the Black Sea,” all of which led to persecutions of the Jews under Basil I (867–886). The Jewish persecutions of the tenth century by Romanos Lekapenos grew out of domestic, not foreign political needs — with unpleasant consequences for Byzantium in the North. Nonetheless, Constantinople continued to cooperate with Khazaria when it was in its interests to do so. Shepard concludes that the conversion did not reorient Byzantine policy. Nonetheless, Khazaria’s increasingly Judaic profile, perhaps strengthened by the presence of Byzantine Jewish refugees at the qağanal court “came to have a significant, if oblique, impact on the Byzantine programme of defensive imperialism.”
179 Ibn Fadlân/Togan, Arab. p. 45/trans. 102–104.

The persecutions of Jews by the Byzantine emperor, Romanos Lekapenos, resulted in attacks on Christians in Khazaria.180 Clearly, the Khazar ruler, by this time, saw himself as a defender of the Jews.

Centuries after the fall of Khazaria as a major power (965–969), there are obscure references in some documents from the Cairo Geniza to messianic ideas or movements associated with Khazaria.181 “Jewish” Khazaria retained some symbolic value, but had long since ceased to be a reality.

180 Al-Mas‘ûdî, Murûj ed. Pellat, I, p. 212. On the complexities of this era, see N. Golb, O. Pritsak, Khazarian Hebrew Documents of the Tenth Century (Ithaca, 1982), pp. 80, 90, 104, 114–115.
181 See Dunlop, History, pp. 254–256.


In Russian
Contents Huns
Contents Tele
Contents Alans
Ogur and Oguz
Tengri, Khuday, Deos and God
Crescent And Star
Türkic Islam

Buddism, Nestoriansm, Islam
Alan Dateline
Avar Dateline
Besenyo Dateline
Bulgar Dateline
Huns Dateline
Karluk Dateline
Khazar Dateline
Kimak Dateline
Kipchak Dateline
Kyrgyz Dateline
Sabir Dateline
Seyanto Dateline
Ðåéòèíã@Mail.ru  “” “” “”θδğŋɣşāáäēəð ï öōüūû“” Türkic Türkic –