I’m too busy to really blog today, but I thought of putting up a post, the gist of which was actually expressed in Ian’s comment below:
When I was younger, I thought of human races as archetypes, and the variation between them a product of mixing. I blame it on the fact that I read Coon when I was about 14. Still, as a (half)Indian, it’s hard to see reconcile the reality of a billion people in the subcontinent with models that try to classify people into 3-5 races. As I learned more biology, I came to the conclusion that human variation was clinal, and race was really an artefact of where you chose to sample along the continuum…as a plant ecologist, I think about things like that a lot. (I’m also somewhat skeptical of ecozones.)
Thanks to a number of convergent strands (of which Razib’s blogging has been a key element), I have come to a rather different conclusion. Race, in my opinion, is more a feature of agriculture than evolution.
Consider two possible models of race: Model 1, in which sharp distinctions existed before the Neolithic, and have been maintained and enhanced as certain groups adopted agriculture and displaced their hunter-gatherer neighbours; and Model 2, in which variation was clinal prior to the Neolithic, but that the immense demographic expansion of certain groups expanded THEIR specific points on the continuum, and brought them into contact (or nearly into contact) with other expansionist agriculturalists.
To me, the Model 2 seems more plausible than Model 1. Is that an argument against race? No, but it does suggest that races shouldn’t really be seen as “locally adapted optima” and rather, should be seen more as transient phenomena produced by historic contingency. Whether this means that race is “real” or not is, to me, a little beside the point. But I’m not convinced by Coyne’s argument that these differences represent the “accumulation of genetic differences between isolated populations”.
Isolation-by-distance is a powerful null model. But if Reconstructing Indian History is correct then it is a poor description of the recent history of the Indian subcontinent. Since there are ~1.5 billion South Asians I think that that is a major objection. And I don’t think it is just South Asians. Even 10,000 years ago I suspect that a clinal isolation-by-distance model would be confronted by several clusters of demes which were operationally allopatric (e.g., West Eurasia & North Africa vs. East Eurasia, Austarlasia, the New World, Sub-Saharan Africa, and likely South Asia). This probably had to do with the fact that human occupation is not, and was not, evenly distributed.
Image credit: Wikipedia