« Readings | Gene Expression Front Page | Celebrity baby boom »
September 11, 2004

Indians aren't Aryans (or Dravidians)?

There is a important new paper (you can view the full PDF if you follow the link) out that surveys the genetics of South Asians viewed from the angle of mtDNA, that is, the direct female lineage. If you follow this stuff, you won't be surprised to find out that the authors conclude that "Since the initial peopling of South and West Asia by anatomically modern humans, when this region may well have provided the initial settlers who colonized much of the rest of Eurasia, the gene flow in and out of India of the maternally transmitted mtDNA has been surprisingly limited." Why surprising you say? Well, about 3/4 of South Asians speak Indo-Aryan languages which have an affinity with other branches of the Indo-European family, and likely derive from the environs of the eastern Black Sea. The "conventional model" was that such a linguistic dominance implied a volkswanderung where the fair skinned Aryans enslaved the dark skinned natives (Dasas), and the broad spectrum of coloration that expresses itself in some concordance with caste was the result of intermarriage between the two groups.

The problem with the model in light of the data above is that most of the South Asian mtDNA pool seems to have diverged from the lineages of Iran before 10,000 years ago, with many coalescence times on the order of 30,000-50,000 years. Philological analysis makes South Asia an implausible ur-heimat for Proto-Indo-European (PIE), so that seems to scratch out the possiblity of a volkswanderung outward. But even a possible expansion of "Dravidian" farmers during the Neolithic from the Middle East (some linguists postulate an affinity between Dravidian and an ancient language of Southwestern Iran, Elamite) seems to have been moderate in its impact.

The work of Toomas Kivisild, one of the primary authors of this paper, is the linchpin of the argument presented by Stephen Oppenheimer in The Real Eve that India is the mother of all Eurasian lineages. Oppenheimer's argument demands very little mixing between subpopulations since the Last Glacial Maximum (LGM).

I did note that Oppenheimer seemed to ignore research headed by Spencer Wells which pointed to a strong exogenous element in the Indian genetic structure that displays a Northwest-Southeast cline. Additionally, he did not publish late enough to read a more recent paper that makes the argument for the expansion of a supermale lineage in Northern India 3,500 years ago. These papers focused on the Y chromosome, which seems to show more signatures of the "invasion" and "immigration" that is characteristic of histories of the subcontinent, but seeing as this is the male lineage and disruptive migrations are likely "male mediated," this would not be surprising. Unlike the mtDNA papers, there are those who dispute the exogenous origins of the haplogroups in question, so I tend to think that a predominantly indigenous (that is, lineages extent in South Asia prior to 10,000 years ago) narrative is more probable (that is, the case for mtDNA being indigenous is stronger than that of the Y being exogenous).

Addendum: The authors note that Cochin Jews carry mtDNA lines that are predominantly Indian. Unpublished data indicates that the Bene Israel Jews of India have a high frequency of the Cohen modal haplotype, that is, they share ancestry with other Jewish groups. Traditionally the Cochin Jews have been more Jewishly aware than the Bene Israel and maintained their contacts with the world Jewry, so I would be surprised if their Y line was non-Jewish where the more outwardly non-Jewish Bene Israel retain the Cohen signature.

(via Dienekes)

Posted by razib at 12:27 AM