« Minority Rule in Argentina | Gene Expression Front Page | Born to be bad »
April 25, 2003

soc.history.what-if & blasts from the past

I used to be a participant in soc.history.what-if a few years back (especially 4-5 years ago) and wrote up the "Manzikert Timeline" with Andrew Reeves. I saw a referral from this blog which has links to SHWI participants that I haven't thought of in a while! I know Ikram Saeed remembers me from SHWI and I get an email now & then from people who are familiar from that forum. Anyway, I just wanted to welcome the SHWI people (smart & often as esoterically inclined as I) to the blogosphere! And I didn't even connect Jonathan Edelstein of Head Heeb with the Jonathan Edelstein of SHWI who I riffed off of a few times.

Finally, for GNXPers, looking for the Manzikert file I found some other data I had thought I had lost. I told Steve Sailer a few years ago that I decided to send out an email to a bunch of the profs at various Indian Institute of Technology campuses and ask what their mother tongue was to figure out if Tamil's were overrepresented. Here is the data:

Raw data
Resorted data
Comparison with Census (1990)

I did this survey in the winter of 2000 by sending out a mass email :) Haven't had time to do anything about it, but it was interesting, and confirmed my suspicion that Tamils are overrepresented in the Indian technocrati.

Posted by razib at 09:15 PM

Fun alternate history! Could have used some more details for China and the New World though.

Posted by: Aaron at April 26, 2003 08:45 AM

Oh my, the Tamils are absolutely dominating IIT.

While Tamils are 6.3% of the Indian population, they account for 25.7% of profs at IIT. What do you believe accounts for this enormous overrepresentation of Tamils amongst the Indian technocrati, and what does this bode for Tamil independence prospects in Sri Lanka?

Let the Tamils carve their own nation out of Tamil Nadu and northern Sri Lanka/Jaffna and watch it become the first brown lion...

Posted by: Sen at April 26, 2003 09:03 AM

a north indian would black lion :) and a lions are indigenous to western india. the tiger is the big cat of most of the subcontinent (this is how historians presume that dynasties centered around patilaputra had western origins-they used lions on their blazons).

Posted by: razib at April 26, 2003 10:25 AM

Interestingly, for the Southern States of India - Andhra Pradesh(Telugu), Tamil Nadu(Tamil), Karnataka(Kannada) and Kerala(Malayalam)- there doesn't seem to be a correlation between literacy rates and representation in the IIT faculty.

Literacy rates from the 2001 Census

Andhra Pradesh : 61.11 %
Karnataka : 67.04 %
Tamil Nadu : 73.47 %
Kerala : 90.92 %

Spearman's rank correlation here works out to r = 0 .

This probably means that the size of the Brahminical/high-caste elite in individual States determines their representation in the IITs, irrespective of the levels of literacy of the remaining population there. So, even if Tamil Nadu's literacy rate were shoot up significantly to say 95%, I'd wager its representation wouldn't change much.

Anyway, I might be wrong if this article on low-caste weavers in the extremely backward (heck, even by Indian standards) Northen state of Bihar is any indication :


On a related topic: I did an informal survey a couple of years ago on the top 100 IIT entrants - incredibly, newspapers here in India list the 100 or so highest ranked IITians each year. 17 happened to be Tamil Brahmins. Quite mind-boggling considering they're only around .4% of the Indian population.

Posted by: king kong at April 26, 2003 12:13 PM

tejas-where did you get that number (% of tamil brahmins)??? i thought they haven't done a caste survey since 1931-but do they do one on the state level?

Posted by: razib at April 26, 2003 02:20 PM

Razib :

Here's the link, the author only happens to make a casual remark about the Tamil Brahmin population but he's rather well regarded in India. So, I don't have any reasons to doubt the veracity of his claim. I've heard similar numbers being tossed around in other places, so this is probably a close enough approximation.


Scroll down to this part :

"But for all its pretentions to the contrary, the Dravidian Movement was much more a movement for social liberation than secession. The target was the Brahmin and the values of the caste system the Brahmin represented. It was, in a democracy of numbers, an unequal battle. For, unlike in northern India where the upper castes constitute a significant percentage of the population, the upper caste population of Tamil Nadu was about five per cent, of which the Brahmins constituted two per cent, if that."

Do the Math after considering Tamil Nadu's population as a percentage of India's and the two percent statistic for that state would constitute much less than even .4% of India's population. I've taken the uppermost approximation to be on the safe side.

Here's another link to an article written by a Sikh author/historian where he claims the Brahmin population in India is 3.5% :


The highest number i've ever come across was in the article GNXP linked to :


There's no Census data of course, which makes it hard to decide on a suitable number.

Posted by: king kong at April 26, 2003 02:59 PM

On a side note, India illustrates what happens when the government quits collecting crucial communal information in the census (caste), but continues to discriminate on the basis of that crucial communal category.

A model for California, post racial Privacy Initiative (RPI. Though I'm not sure if it is a success story of a cautionary tale.

Posted by: Ikram Saeed at May 1, 2003 09:34 PM