Earlier this year, I presented a post that offered the following data:

From the book On Blondes, a survey of Jewish school children in Nazi Germany suggested that the following was the hair color breakdown:

10% blonde.
50% “mixed.”
40% black.

David B also forwarded me this data on Jews from 19th century London:

English Ashkenazim: Blue 11.1 Grey 30.1 Brown 58.8.
English Sephardim: Blue 21.3 Grey 11.9 Brown 66.8.

My own personal opinion is that these data points can be explained by admixture from the surrounding non-Jewish populations (and likely a non-trivial amount). Some, like researcher Michael Hammer, argue for a predominantly Middle Eastern origin for Ashkenazi Jews (see his paper), often on the basis of NRY data. Seeing as how the Bene Israel Jews of India display a prevalence of the Cohen Modal Haplotype and an Indian physical appearence, it seems that predominant Middle Eastern Y lineages and a high rate of mixture with non-Jews can coexist (through intermarriage with local women).

But what if the admixture level was low? What if Hammer & co. are vindicated by a large number of studies on neutral markers over the next 10 years? What could explain the relatively high frequency of blue eyes and blonde hair among Ashkenazi Jews (in comparison to their putative source population).

Perhaps the introgression of Sub-Saharan genes into Arab populations might have resulted in a decrease in the expression of these phenotypes from their earlier norms. On the other hand, Yemeni Jews, who have not been nearly as heavily impacted by this genetic flow are, to my knowledge, a rather swarthy population. Additionally, the Anatolian population has not been as deeply impacted by Sub-Saharan gene flow, but I am skeptical that the frequency of blue/grey eyes is as high as 10%.

OK, cutting to the chase, I wonder, could the shift in phenotype be the result of selection? Greg & Henry’s paper focused on IQ, but there has been a lot of work done recently on human color variation. One hypothesis holds that when homonids lost their body hair they became heavily pigmented (this Nick Wade article sketches out the interconnected theses). After “sweeps,” where selection operated upon the MC1R locus and forced any given allelic form to fixation, humans spread out of Africa, and the constraint (the negative impact of high levels of radiation on exposed skin) was “relaxed.” Not only did the level of radiation decrease (both latitude and greater cloud cover could be a factor), but the need for clothing in more frigid climes was likely an important factor.

Some have argued that sexual selection played an important role in the prominence of blondism in northern European (sexual selection could play a role if environmental selection was no longer a limiting constraint). Additionally, there is the old theory that light skin facilitates the synthesis of Vitamin D in regions characterized by low levels of sunlight (so, positive environmental selection).

Ashkenazi Jews were predominantly an urban people, and traditionally do not have “outdoors” reputations. One could posit that selective constraints against a expression of fair phenotypes might have been relaxed as Jews made a transition from being farmers in the Levant to merchants in North-Central Europe. Perhaps a “natural” human tendency to prefer fair skin, or, more precisely, for human males to prefer females fairer than the population mean, was given free reign in the context of an urbanized culture living in a region where sunlight was no longer an overabundant danger?

Ultimately, I do think admixture is the primary causative agent for the relative fairness of the Ashkenazi in comparison to Middle Eastern populations. My personal impression is that many Ashkenazi Jews also reflect Central European populations in traits of facial morphology. But, it is something to wonder about.

Addendum: One point people bring up is the relative darkness of Australian Aboriginals even though much of the continent is in the “temperate” zone. It could be fairness (or the genetic variants that result in it) is not in the “genetic” background of this population, but blondeness among Aboriginals of the deep desert seems to falsify this thesis, and one supposes that the original populations that exited northeast Africa were also rather small and lacking in genetic diversity.

But, Australia is mostly dry, and so rather sunny, in comparison to Europe. Only Victoria is really in a temperate maratime climate. And, to my knowledge, no group of Aboriginals wore very much clothing.

A more interesting case is that of Tasmanian Aboriginals. This group, now extinct in an un-admixed state, was very dark skinned, and had been isolated on Tasmania for about 10,000 years. Tasmania does have a cool maratime climate like much of Europe. My impression is that Tasmania aboriginals were not totally naked. For me, this is a strong argument against the idea that relaxed constraint on the MC1R locus always leads depigmentation because of sexual selection, possibly due to sensory bias or sexual dimoprhism and fecundity indicators (that is, lighter skinned women within a population are more fertile).

Posted by razib at 12:23 AM