Are We All Khazars Now?

Geneticist Eran Elhaik’s article on the Khazar ancestry of Ashkenazi Jews made a stir from the moment it appeared. Oxford University Press immediately notified the scientific community of its publication in the journal Genome Biology and Evolution through the American Association for the Advancement of Science’s “EurekAlert!” website. The story was soon picked up by ScienceDaily, and not long after Elhaik was the subject of somewhat breathless articles in Ha’aretz and the Forward. According to Elhaik’s website, it has been discussed on more than 50 news sites and at least 18 blogs. It is, in fact, now one of the most-read articles ever published in Genome Biology and Evolution. However, there has been little critical discussion of it outside the scientific community.

Most historians have assumed that the Jews of Eastern Europe are the descendants of Central European Jews who moved eastward in the Middle Ages or shortly thereafter. In 1976, Arthur Koestler popularized an alternative hypothesis. In The Thirteenth Tribe, he argued that most Ashkenazi Jews are descended from the Khazars, a Central Asian people who ruled a large kingdom on the Black Sea and apparently converted to Judaism in the 8^th century. This hypothesis has been taken up more recently by Shlomo Sand in a book called The Invention of the Jewish People. Koestler, one of the oddest and most extraordinary public intellectuals of the 20^th century, wanted to weaken anti-Semitism by demonstrating that many Jews weren’t Semites at all. Sand, a self-avowed post-Zionist who teaches at Tel Aviv University, is apparently driven by the desire to prove that Ashkenazi Israelis are interlopers in the Middle East. 

The Khazars certainly existed, though not much is known about them. However, the story (or myth) of their conversion to Judaism has seized the imagination of generations of writers, from Judah Halevi, whose 12^th-century classic The Kuzari is a philosophical dialogue between a Khazar king and the rabbi who convinces him of the truth of Judaism, to Michael Chabon, who considered calling his 2007 novel Gentlemen of the Road “Jews with Swords.”

Neither Arthur Koestler nor Shlomo Sand based their iconoclastic, politically driven conclusions on serious research. Eran Elhaik, however, is an accomplished scientist who has apparently come to the same conclusion through sophisticated statistical analysis of the salient genetic data. In his view, this proves that most contemporary Jews are descendants of the Khazars. He also thinks that it solves a demographic puzzle. How, he asks, are we to explain “the vast population expansion of Eastern European Jews from fifty thousand (15th century) to eight million (20th century),” particularly given “the severe economic restrictions, slavery, assimilation, the Black Death and other plagues, forced and voluntary conversions, persecutions, kidnappings, rapes, exiles, wars, massacres, and pogroms” to which they were subjected? I shall return to this puzzle, but first let us examine Elhaik’s solution.

Elhaik’s article has become rather popular in some anti-Semitic and anti-Zionist circles, but this proves nothing about its truth. However, Elhaik’s media appearances make it clear that he is far from naïve about the uses to which his findings have been put. Elhaik himself has summarized these findings quite clearly in the article’s abstract:

The question of Jewish ancestry has been the subject of controversy for over two centuries and has yet to be resolved. The “Rhineland hypothesis” depicts Eastern European Jews as a “population isolate” that emerged from a small group of German Jews who migrated eastward and expanded rapidly. Alternatively, the “Khazarian hypothesis” suggests that Eastern European Jews descended from the Khazars, an amalgam of Turkic clans that settled the Caucasus in the early centuries CE and converted to Judaism in the 8th century. Mesopotamian and Greco-Roman Jews continuously reinforced the Judaized empire until the 13th century. Following the collapse of their empire, the Judeo-Khazars fled to Eastern Europe. The rise of European Jewry is therefore explained by the contribution of the Judeo-Khazars . . . We applied a wide range of population genetic analyses to compare these two hypotheses. Our findings support the Khazarian hypothesis and portray the European Jewish genome as a mosaic of Near Eastern-Caucasus, European, and Semitic ancestries.

Elhaik’s arguments did not go unchallenged. In a detailed review in the Proceedings of the Russian Academy of DNA Genealogy, Anatole A. Klyosov dismissed much of his analysis as mere acrobatics. However, since this article appeared in Russian, it got little attention. Recently, at least two studies have come to similar conclusions. A scientific team led by M. Metsapalu announced that it has found “no indication of Khazar genetic ancestry among Ashkenazi Jews” (the paper is forthcoming). Meanwhile another team led by M. Costa has argued both that there is strong evidence of the admixture of European women in the ancestry of Ashkenazi Jewish women and that there is no evidence for significant Khazar ancestry. On his website, Elhaik has argued that neither paper disproves his thesis. A third team, led by Doron Behar, has a paper coming out in the journal Human Biology whose title announces “No Evidence from Genome-Wide Data of a Khazar Origin for the Ashkenazi Jews.” But Elhaik will, no doubt, maintain his position.

Can a non-scientist enter into this debate? Let us return to Elhaik’s paper, which turns on comparing the genomes of individuals, especially males. “The complete data set,” he writes, “contained 1,287 unrelated individuals of 8 Jewish and 74 non-Jewish populations.” This is impressive, but it says nothing about the number of Eastern European Ashkenazi Jewish males whose Y chromosomes are central to Elhaik’s analysis. If one searches Elhaik’s website, it turns out that there were exactly 12 Eastern European Ashkenazi Jews in the data set. How many were male? To find out, I had to turn to the Institute of Molecular and Cell Biology at the University of Tartu. It turns out that there were eight males in the sample. As small as this is, however, it turns out to be huge compared to the data set on Khazar DNA.

The trouble with obtaining Khazar DNA is that no population group today is recognized to have descended from the Khazars. Elhaik acknowledges this difficulty and deals with it efficiently. According to him, “Caucasus Georgians and Armenians were considered proto-Khazars because they are believed to have emerged from the same genetic cohort as the Khazars.” He bases this claim on “Polak 1951; Dvornik 1962; Brook 2006.” This appears quite convincing unless one is familiar with the names cited—and the ones missing. Polak and Dvornik were important scholars, but their work is a half-century old and outdated, while Kevin Brook is a talented but amateur Khazar enthusiast who has no first-hand knowledge of Central Asian studies. In fact, no contemporary scholarship supports this claim. Moreover, elsewhere in the article Elhaik himself refers to a study by Balanovsky et al., but fails to mention that it concludes that of all the national groups in the Black Sea region, the Georgians and Armenians were the least likely to have absorbed significant populations from other national groups. In other words, while there was DNA from eight Ashkenazi males in Elhaik’s study, there was no Khazar DNA at all. This makes it a bit difficult to come to significant conclusions about the Khazarian ancestry of Ashkenazi Jews. The problems, however, do not end there.

Following Koestler (whose The Thirteenth Tribe Elhaik has told interviewers he read as a child), Elhaik says that after the downfall of the Khazarian empire, “Some Judeo-Khazars were left behind, mainly in the Crimea and the Caucasus, where they formed Jewish enclaves surviving into modern times . . . [a] vestige of the Khazar nation is the Mountain Jews in the North Eastern Caucasus.” Unfortunately, Koestler had no evidence for this whatsoever. But there is a more serious problem, at least for Elhaik’s argument. If the “Mountain Jews” are a “vestige of the Khazar nation,” why bother with the Georgians and the Armenians? Elhaik could have just gone directly to these “descendants” and compared them with Ashkenazi Jews. The only answer I can see is that this is a case of the dreaded academic syndrome “Cut and Paste Disease.”

Unfortunately for the readers of Elhaik, there are more problems with his research. One of the methods of genetic analysis that he employed is known as Principal Component Analysis (PCA). For present purposes, it is not necessary to describe the method. What is interesting is how Elhaik tested it:

To assess the ability of our PCA-based approach to identify the biogeographical origins of a population, we first sought to identify the biogeographical origin of Druze. The Druze religion originated in the 11th century, but the people’s origins remain a source of much confusion and debate (Hitti 1928). We traced Druze biogeographical origin . . . Half of the Druze clustered tightly in Southeast Turkey, and the remaining were scattered along northern Syria and Iraq. These results are in agreement with Shlush et al. (2008) using mtDNA analysis.

The founder of the Druze religion, Hamza ibn ’Alī ibn Ahmad, was of Persian descent and was active in Egypt. His missionaries had their greatest success in present-day Lebanon and Syria. The religion never expanded into Turkey, a point made explicitly by Hitti, who is Elhaik’s source. What can one say about a method that identifies the origin of the Druze in Southeast Turkey? To be fair, the developers of PCA warned that poor sample sizes can yield problematic results.

In his “Identity by Descent” analysis of the Ashkenazi DNA, Elhaik concluded that “the maternal analysis depicts a specific Caucasus founding lineage with a weak Southern European ancestry . . . whereas the paternal ancestry reveals a dual Caucasus-Southern European origin.” Elhaik explains these lineages as the result of “ancient migrations from Southern Europe toward Khazaria (6th–13th centuries) and more recent migrations from the Caucasus to Central and Southern Europe (13th-15th centuries) (Polak 1951; Patai and Patai 1975; Straten 2003; Brook 2006; Sand 2009).” Precisely none of the cited authors had or has direct familiarity with the primary sources of the history of the region. In fact, neither migration ever took place.

Elhaik might have been thinking of the claim of the 10^th-century historian Ali al-Mas’udi that many Jews fled from Byzantium to the Khazar lands during the reign of the Byzantine emperor Romanus Lacapenus, but there is no evidence for this in Byzantine sources. Moreover, a refugee population would have included similar numbers of men and women, which would have generated equal male and female levels of Southern European ancestry, giving Elhaik more problems rather than less. And there is no evidence whatsoever for any “more recent migrations from the Caucasus to Central and Southern Europe.” Nor is it likely, or even possible, that “Mesopotamian and Greco-Roman Jews continuously reinforced the Judaized empire until the 13th century. Following the collapse of their empire, the Judeo-Khazars fled to Eastern Europe.” The last Khazar capital was destroyed by Sviatoslav of Rus before 970, and the empire never recovered.

Not only is there no evidence for the claim of significant migration to Poland from Khazaria, but such a claim flies in the face of basic facts that we know about Eastern European Jewry. The settlement patterns of the Jews in Eastern Europe suggest that large-scale settlement began in western Poland and not the parts closest to Khazaria. Moreover, there is no evidence of influence of Byzantine Jewish liturgy and customs on Eastern European Jewry and none of Central Asian languages on Yiddish.

The main appeal of a theory of migration is that it seems to explain the demographic puzzle I mentioned at the outset: How did we get so many Jews in Eastern Europe? How did we go from a population of 50,000 in the 15^th century to eight million in the 20^th, especially given all the depredations visited upon the Jews over those five centuries? In fact, Elhaik’s litany of woe (“economic restrictions . . . assimilation, the Black Death . . . conversions, persecutions, kidnappings, rapes, exiles, wars . . . and pogroms”) is a drastic overstatement of the experience of Eastern European Jews during those centuries, and their growth in numbers is not really that mysterious.

Populations do not grow arithmetically, they grow—not unlike credit card debts—exponentially. The Afrikaners in South Africa started from a group of about 2,000 settlers who came in the late 17^th century. Today, roughly 13 generations later, they number about three million. A little over three hundred years ago, five thousand French immigrants came to Quebec; their descendants now number about 6.5 million. What needs to be explained is not why the Jewish population in Eastern Europe grew exponentially in the modern period but rather why the Jewish population of Central Europe did not grow. But that is a different question, and the Khazars are of no help in solving it.

How did a distinguished journal from Oxford University Press publish an article like this? Usually, it is difficult to check the work of the scholarly referees a journal employs. However, in this case, Elhaik posted the referees’ reports on his website (they have since been taken down). The first referee was aware that the paper would arouse controversy, predicting that it would be “highly cited,” but apart from expressing some doubts about Elhaik’s observations on Druze origins, he remarked only—and inaccurately—that Elhaik “has been more thorough than most (if not all) previous studies on the issue of Jewish ancestry.”

The second referee recommended that M.I. Artamonov and his book History of the Khazars and L.N. Gumilev’s The Rhythms of Eurasia should be cited. These two books (both in Russian) are exceedingly odd recommendations. Artamonov’s was written under severe Soviet censorship, which, as current research has shown, prevented him from writing what he really thought. Gumilev’s book is discussed in Vadim Rossman’s Russian Intellectual Antisemitism in the Post-Communist Era under the rubric “Antisemitism in Eurasian Historiography.” It is a good example of contemporary Russian populist nationalism, but it is less than useful for unbiased research. The referee also noted that “currently Druze do not live at the border of Khazaria. They seem to have migrated. This should be explained.” Indeed. Neither Elhaik nor his referees are apparently familiar with the work of Vladimir Petrukhin, who is the dean of Russian scholarship on the Khazars. There are also serious books in English that Elhaik ought to have cited, such as Dunlop’s History of the Jewish Khazars or the works of Peter Golden.

Judging by the comments, neither referee seems to have been very familiar with the literature on genetics of the Jews or the issue of the genetic background of Ashkenazi Jews. The referees did not compare Elhaik’s thesis to the findings of other researchers, nor did they seem to notice the internal problems raised here. They certainly did not ask about the size of the data set. In general, their comments tended to focus on style and presentation rather than technical details or the overall cogency of the argument.

What happened here? I doubt that there was an overt political agenda on the part of the editor. The key probably lies in the first reviewer’s prediction that the paper would be “highly cited.” Nonetheless, it remains to be explained how it seems to have evaded any critical scrutiny at all before being published. There is at least one virtue of Elhaik’s article: It is a valuable reminder that despite sophisticated-looking methods, peer review, publication in a prestigious scientific journal, not to speak of media coverage, there is no alternative to critical reading.

When all is said and done, the accepted wisdom is still acceptable. There is no evidence that Ashkenazi Jews are descended from Central Asian “Jews with swords,” and there is every reason to think that they simply came from Central Europe. The findings of other genetic researchers that the DNA of most Jews seems to link them with other Jews more than with any other group has not been disproven.