Caste in India

Share on FacebookShare on Google+Email this to someoneTweet about this on Twitter

David Reich says “There are populations that have lived in the same town and same village for thousands of years without exchanging genes.” Sounds familiar.

If this was indeed the case, and if fitness payoffs differed significantly between castes – then there has almost certainly been genetic adaptation. Those whose ancestors lived a particular kind of life for a long time, with very low inward gene flow, should on average have traits that better fit them to that (past) way of life. This doesn’t necessarily mean that they would better at the supposed formal role or purpose of that caste, more that they would be more successful (in a reproductive sense) in that niche . For example, depending on the reward structure, the soldier with greatest fitness might well be one who avoided combat.

Reich said that average inward gene flow in castes appears to have been less than than 1 in 30 per generation: that’s low enough to allow this.

16 Comments

  1. I doubt that India is quite as stable as that. China makes a fetish of stability and ancient tradition, but large areas have been depopulated and repopulated multiple times during the last 3,000 years. (I don’t know anything about India specifically.)

  2. follow up to john’s comment: 
     
    1) indus valley + doab would seem similar to iraq and much of iran and the north china plain insofar as political stability was probably essential in opening up space for public works projects which would keep up primary production to support a large population. no irrigation = no farming (i know that dry land farming is doable in china and northwest india, but the productivity is presumably lower, right?) 
     
    2) the eastern gangetic plain, basically bengal, especially the eastern half, was cleared and settled within historical memory. this would be a region of demographic expansion and population growth from founder populations. it should exhibit different values then of Fst based on this model…though we don’t have those populations sampled very well (the ones in the east in reich’s paper are atypical tribal ones i think).

  3. If you think the castes are younger, argue with Reich, not me.

  4. @Razib: Rice-cultivation needs water, monsoons are enough for other grains (jowar, barley, etc.). My impression (could be wrong) is that irrigation really took off after 18th century.  
     
    @gcochran: 
    In particular, you have the brahmins and a constellation of trading castes called banias that played a social role similar to the Jews and overseas Chinese: 
     
    http://www.livemint.com/2009/08/27220957/When-will-the-BrahminBania-he.html 
     
    There’s a north/south divide on this too. Trading castes dominate trade in the north, while the relative lack of bania groups in the South leads to greater domination by other groups. 
     
    Note Reich also finds a high presence of regressive diseases, for which they blame founder effects. I’m not aware of any Tay-Sachs equivalents that have beneficial side effects, but it would be worth checking into.

  5. If I understand correctly, Reich _predicts_ an unusually high rate of genetic diseases in castes, based on the idea that they experienced severe founding bottlenecks. But that is not the same as finding them.

  6. I think one interesting area to look for in caste studies would be Kerala in India’s South West. Razib, you have already alluded to the Namboodiri‘s (the brahmins of Kerala) who have maintained a systematic hierarchy. They have also contributed genetics (mostly outward) to almost all the other castes because they allowed the male members to maintain relationships but not the females.

  7. Oops..sorry I thought this post was written by Razib.

  8. I’m not aware of any Tay-Sachs equivalents that have beneficial side effects, but it would be worth checking into. 
     
    Has it been confirmed yet that Tay-Sachs et. al. do in fact have beneficial side effects, or does this remain speculative?

  9. Aren’t the Brahui renowned wanderers? Their language has words from Indo-Aryan only after 1000 AD, indicating that they wandered to what is now Pakistan, not that they’re a remnant of a once more pervasive Dravidian culture. Though I would say that the Dravidians were once dominant in Northern India, for Sanskrit has retroflex consonsants, which the Dravidian languages have, but other ancient Indo-European languages lack.

  10. Their language has words from Indo-Aryan only after 1000 AD, indicating that they wandered to what is now Pakistan, not that they’re a remnant of a once more pervasive Dravidian culture. 
     
    do you have a source on this sid? i’ve looked around about the brahui, but stuff seems thin on the ground.

  11. Razib – I read that from this article. 
     
    http://www.people.fas.harvard.edu/~witzel/IndusLang.pdf 
     
    He cites this article. Unfortunately, I couldn’t find a link to it. 
     
    Elfenbein, J.H. A periplous of the ‘Brahui problem’. Studia Iranica 16, 1987, 215-233

  12. Caste is an imprecise term. Varna and jati are clearer, but stuill religiously loaded. Let’s try socially-stratified occupation-based endogomous groups — SSOBEG — a nice neutral term. 
     
    SSOBEGs in one village for 1000s of years? Indian settlement patterns aren’t quite that stable. Bengal and the lower Brahmaputra were stteled post 1100. Most of western Punjab was settled after 1848 (google: canal colonies).  
     
    You can have stable long-term SSOBEGs even with recent settlement and significant migration. But it mucks things up.

  13. Regarding caste, does anyone have any good books or articles that they would recommend?

  14. thanks. however I think you accidentally pasted the same link twice.

  15. http://www.people.fas.harvard.edu/~witzel/IndusLang.pdf actually says Dravidian language was not present in North India, an Austro-Asiatic language was, based on loanwords in the Rig Veda.

a