DNA and the Origin of the Jews

In premodern times, the question of where Jews come from had an obvious answer: The Bible tells the story of Israel’s origins beginning with the Patriarchs and Matriarchs, moving on to Moses and the exodus from Egypt, and continuing on the conquest of Canaan, the judges, the monarchy, the exile, and so on. Modern scholars have come to challenge that narrative, however, just as scientists began to challenge the creation story in Genesis, looking beyond the biblical account for an explanation for how the Jews came to be.

Whereas the earlier parts of the biblical origins narrative, such as the Patriarchs or the conquest, have been increasingly understood as mythic among contemporary Bible scholars, the latter parts of the biblical account, namely the existence of the kingdoms of Israel and Judah, the exile to Babylon and the reestablishment of Judea in what is known as the Second Temple Period are generally accepted in broad outline as historical facts. And yet, even the connection between Jews and Judea has been challenged over the past few decade.^[1]

Despite all that they have learned about biblical history, scholars are still seeking to understand the ancient origins of the Jews. Did their Israelite ancestors originate in Mesopotamia as the book of Genesis suggests? Did they emerge from within the indigenous population of ancient Canaan as many biblical scholars now argue? In addition to textual and archaeological arguments, advances in population genetics over the last few decades have paved the way for a biological approach to the question of Jewish origin, of when, where, and from what earlier population the Jews arose.

Jewish Genes?

The refined ability of biologists to decipher DNA has led some scientists to attempt to pinpoint the ancestry of Jews and to place those ancestors geographically. This kind of research has been part of an explosion of genetic studies of the Jews in the last two decades, much of it driven by medical questions but some of it focused on historical and genealogical questions. Genetic testing companies such as Ancestry.com, Gene by Gene, and 23andMehave made this kind of research famous, and it has also involved certain ambitious research projects aimed at illuminating the genetic history of the Jews, such as:

The Jewish HapMap Project, co-founded by Harry Ostrer and Gil Atzmon, which aims to construct a genetic map of Jewish diaspora populations around the world.

The Ashkenazi Genome Consortium, founded by 11 labs in NY and Israel, which sequenced the genome of 128 Jewish individuals of Ashkenazic ancestry (as of 2013).

These initiatives are meant to aid in the study and treatment of genetic disease—but they also represent new historiographical resources that allow us to learn information about Jews unavailable in textual and archaeological sources.

1. The Kohanim Study

One of the earliest and most famous of these studies is the study of kohanim, i.e., Jewish males who identify as members of Judaism’s priestly clan.^[2] According to rabbinic-Jewish law, membership in the Jewish community is based on maternal descent, from mother to child, but priestly descent is passed down paternally, from father to son. The Torah traces the kohanim back to Moses’ brother, Aaron, who passed down the status to his male descendants.

We do not know if Aaron actually existed, but there is evidence from Josephus and rabbinic sources that priestly status was transmitted from father to son in the time of the Second Temple and the following centuries.

If priestly status has been consistently transmitted through the paternal line, in theory, there should be mutations that have accumulated on their Y chromosomes that sons inherit directly from their fathers.^[3]

Discovering the Cohen Modal Haplotype (CMH)

In 1997 geneticists Michael Hammer, Karl Skorecki, and their colleagues tested self-identified kohanim from Israel, the United States, and Britain to determine whether they shared a distinctive genetic inheritance that tied them to one another as co-descendants of a paternal lineage. The focus of their study was an array of haplotypes found on the Y chromosomes of their subjects.

A haplotype is a group of genes within an organism inherited from a single parent. Haplotypes on the male side tend to mutate at a relatively rapid rate, which means that when two males share a distinctive haplotype in their Y chromosomes, that is a sign that they share a common ancestor on their male side in the relatively recent past. Hammer, Skorecki, and their team discovered that a distinctive haplotype, which they called the Cohen Modal Haplotype (CMH), was present in about half of the self-identifying kohanim they tested, showing that they shared a common ancestor on their paternal side.

There was no way to identify who this ancestor was, but since the relevant genetic signature showed up among both Ashkenazic and Sephardic kohanim, the authors of the study reasoned that this common ancestor must have lived before the split between the two populations (that occurred sometime in the mid-first millennium C.E.).

A later study that was trying to estimate when the mutation that distinguishes the Cohen Modal Haplotype occurred, placed this ancestor between 2,100 and 3,250 years ago, admittedly, not a very precise dating. Nevertheless, the existence of the CMH along with its dating range were suggestive enough to lead the public to mistake the genetic findings as scientific evidence that kohanim were descendant from Aaron himself.

Problems with the Study

Almost twenty years later, we know that these early studies can be challenged; even their authors have significantly revised their conclusions. The latest study that I have been able to locate from 2014 calls the entire project into question, challenging how the earlier studies dated the mutation, and cautioning against the use of the haplotype as a marker of any kind of Jewish ancestry.^[4]

Despite such criticism, there is no denying the impact that the original studies had on the public’s perception of such research, and on geneticists’ own hopes for what this kind of research could reveal about the Jews. More recent research has shown that the ancestry of the kohanim themselves is more complex than the earliest studies suggested, reflecting many different lineages that emerged at different points in history—and there is still debate about how to interpret the genetic evidence.

But as murky and convoluted as the genetic ancestry of the kohanim has proven to be, the effort to uncover that ancestry in the late 1990s proved pioneering nonetheless, unleashing curiosity about the ancestry of the Jews in general. As a result, there has now emerged a distinct new subfield of Jewish history, the genetic history of the Jews, that reaches beyond what we can learn from biblical and archaeological sources about the earliest ancestors of the Jews.

2. Jews Are from the Middle East

One of the most ambitious of studies to date involved the analysis of whole-genome samples from fourteen different Jewish communities representing about 90% percent of the world’s Jewish population. The research was published in 2010 by the Israeli geneticist Doron Behar together with twenty other contributors.^[5] The samples used were compared with genetic samples from sixty-nine neighboring non-Jewish populations, producing the most comprehensive and precise genetic profile of the worldwide Jewish population to date.

The study found that, genetically, most of the Jews sampled fell into three distinct sub-clusters that were distinguishable from another because each had intermixed with the local population among which it lived. At the same time, it also found that almost all these different populations had connection to Middle Eastern populations. This suggests that despite being dispersed around the world, Jews do indeed have ancestries that connect them to the Near East, as Jews themselves have long believed.

This research suggests that it is possible to use a living population’s genetic profile to trace its ancestry to a remote place and time. But despite this evidence, not everyone accepts that contemporary Jews go back to a Middle Eastern population.

Jews and Khazar Theory

A few years after the publication of Behar’s study, the molecular biologist Eran Elhaik argued quite differently, that the data actually supported the conclusion that Ashkenazic Jews came from the Caucasus Mountains. This seemed to confirm, he argued, what has come to be known as the Khazar theory,^[6] that Ashkenazic Jews descend from a Turkic people known as the Khazars who inhabited the region between Europe and Asia and converted en masse to Judaism during the Middle Ages.

The theory goes back to the 19^th century, but was made popular in the United States by Arthur Koestler in a 1976 book called The Thirteenth Tribe. It continues to have advocates to this day, including the Ku Klux Klan Leader David Duke, who only recently began to question the theory when he realized that its earlier advocates included communist Jews like Koestler. Historians, relying on textual sources, have exposed many weaknesses in the Khazar theory,^[7] but with Elhaik’s research, it seemed to find a new scientific footing, the data suggesting that Ashkenazic Jews do not descend from Middle Eastern founders but from a Turkic population in the Russian steppe, as argued by the Khazar theory.

Where Was Khazaria? Behar’s Response

Behar and his associates seized on Elhaik’s analysis as a chance not just to defend their original analysis but to deepen their research.^[8] Behar pointed to a methodological problem with Elhaik’s analysis: There are no more Khazars alive today, no known Khazar descendants, and no Khazar skeletal remains from which to draw DNA evidence. Hence, there is no way to draw a genetic connection between Ashkenazic Jews and a Khazar population.

To overcome this problem, Elhaik drew his samples from proxies— Armenians and Georgians from what he (erroneously) took to be the region of Khazaria—and it was on the basis of these samples that he found that 70 percent of European Jews and almost all Eastern European Jews cluster with these populations, and by extension their supposed Khazar ancestors.

Behar and his colleagues argued that the use of living Armenians and Georgians as genetic proxies for the Khazars was a mistake, as they hail from the southern part of the Caucasus whereas the Khazar kingdom was located farther to the north. They sought to correct for this problem by comparing a large collection of Jewish samples with the largest available genome-wide sample set of the various non-Jewish populations from the regions where Ashkenazic Jews may have originated, including people from Western Europe, the Middle East and individuals living in the (actual) Khazar region, north of where Elhaik took his sample.

Their approach led to a very different result, calling Elhaik’s conclusions into question by showing that Ashkenazic Jews share the greatest genetic ancestry not with people from the Caucasus but with those from Europe and the Middle East. Elhaik has recently published a study to expand on his views, now describing the supposed founding population of Ashkenazic Jews as a “Slavo-Iranian confederation,” but he has not overcome the findings of Behar or of historians like Shaul Stampfer who dismisses this research as “nonsense.”^[9]

Is Genetic Research Racist?

Even as the genetics research has taken great strides in the last 20 years, we need to take note of a critique of Jewish genetic research that has developed among historians and anthropologists of science, who contest how the evidence is being interpreted, question the motives of those doing this kind of research, and worry that it has dangerous implications as a rationale for new forms of racism.

As geneticists themselves are the first to acknowledge, the data, however impressive, does not speak for itself. It has no intrinsic meaning but has to be interpreted in order to tell a story. As is true of biblical interpretation, the process of interpreting genetic results can yield very different understandings of the evidence.

Thomas Jefferson’s Slave Children?

Consider as an example another sensational genetics study from the same period as the kohen study: the use of genetics to connect descendants of Thomas Jefferson’s family with descendants of Sally Hemings, the slave with whom he was rumored to have had several children. Employing the same kind of analysis of Y-chromosome haplotypes used to uncover a common ancestor for contemporary kohanim, DNA testing showed that descendants of Hemings through her son Eston were related to Jefferson, which led to the reports that genetics had revealed Jefferson to be the father of Eston.

But the genetic evidence was more ambiguous than the media made it out to be—Jefferson might have been the father but so too could any of his male kin. Thus, while the Thomas Jefferson Memorial Foundation, which runs Monticello, accepted the President as the father, a competing organization, the similarly named Thomas Jefferson Foundation, objected that the test only established a strong probability that Eston was fathered by one of the more than two dozen Jefferson men known to be in Virginia at that time. This latter group pointed to Randolph Jefferson, the president’s brother, as the more likely father.

The point is that the data itself does not tell a single story. Genetic research was able to uncover a connection between the descendants of Hemings and the Jefferson family but it is still ambiguous enough to allow for multiple interpretations, some sympathetic to Jefferson, some critical. The same is true for the genetic evidence bearing on the origin of the Jews.

Telling a Story with Data

This example illustrates why, from the perspective of a critic such as the American anthropologist Nadia Abu El-Hajj,^[10] there is nothing truly objective about Jewish genetic research. To become a genetic history, the data must become part of a narrative and there will always be more than one way to tell the story. According to Abu El Haj, bias, simplification, and distortion of the genetic data are built into the very act of converting it into a historical account.

Although Abu El Haj’s critique of the field is unfair in some ways, my own brief interaction with genetics research confirms in a small way that the data, while capable of illuminating the ancient ancestries of living peoples, can be made to fit within different historical narratives. A good illustrative example of the problem is recent research into the origin of the Samaritans.

Samaritans, Jews, and Kohanim

In 2013, my Stanford colleague, Noah Rosenberg, and I co-edited a special volume of the journal Human Biology (85.6) that included a genetic study of the Samaritans, a small community in Israel today who claims descent from the northern tribes of Israel.^[11] The study revealed a genetic connection between Samaritans and Jews, especially Jewish kohanim. This association suggests that Samaritans and Jews shared a common ancestry in antiquity, just as the Samaritans had long claimed by tracing themselves back to the ancient Israelites.

Josephus: Kohanim Ran to the Samaritans to Avoid Divorce

When I reviewed the Samaritan study, I pointed out to one of its authors, the Stanford geneticist Marc Feldman, that the genetic connection between Samaritans and Jews of kohanic descent was open to a different explanation. Josephus, the first-century historian, in a narrative that appears in Jewish Antiquities (9.302-347), describes how the Samaritans acquired their priesthood. He claims that in the time of Alexander the Great (ca. 332 B.C.E.), the elders of Jerusalem rose up against a high priest in Jerusalem named Manasses for marrying the daughter of the Samaritan governor, insisting that the priest either divorce his Samaritan wife or give up his priestly role. Manasses chose instead to defect to the Samaritans, accepting appointment as their high priest of their newly built temple on Mount Gerezim. According to Josephus, Manasses wasn’t alone in his defection: many Israelites and priests joined him and married Samaritans.

Feldman and his colleagues believe the genetic connections between Samaritans and Jews affirms the Samaritans’ self-understanding as co-descendants from the ancient Israelites. Josephus’ report suggests another potential explanation for the genetic connection between Samaritans and Jewish kohanim—intermarriage in the Second Temple period rather than common descent from ancient Israelite ancestors. Feldman acknowledged that this theory offered another potential explanation for the genetic data he and his colleagues uncovered.

I am not arguing that Josephus’ description of the Samaritans is correct, and I want to be especially careful here because the Samaritans are a tiny religious minority in Israel today and have had legal run-ins with Jewish religious authorities because of the ambiguity surrounding their origins. The Samaritans self-perception as Israelites may be accurate, or some Samaritans may be connected to ancient Israelites and others connected to Second Temple Jews. Perhaps others will suggest additional hypotheses to explain that data.

My point is that the historical sources, as few as they are, allow for more than one way of narrating the genetic history of the Samaritans.^[12] Such a situation gives one pause about drawing conclusions about Samaritan origins from the genetic data.

The Jewish Genetic Narrative

The same may well be true of what genetics can tell us about the origin of the Jews. Genetic history is a developing field, and like most science, a self-correcting one, and perhaps someday, scientists will be able to resolve the ambiguities we have noted here. But even then, geneticists will always need to rely on non-genetic evidence to make any historical sense of the data—written texts, oral traditions, and interviews with people about where their ancestors come from.

It is impossible to turn the testimony of DNA into a definitive account of the past. The process of assemblage, dot-connecting, and interpretation means there will also always be some degree of imagination involved in the construction of genetic history, and choices to make about which story to believe.