Several weeks ago, I posted a message by a commenter named Rec1Man, who has been doing a lot of work to come up with a model of India's potential average IQ. This is one of the most important questions for predicting the course of the 21st Century.

The first edition of Lynn and Vanhanen's IQ and the Wealth of Nations reported four studies from India from 1959-1986, with averages varying only across a rather narrow range of 78-82.

On the other hand, as I've pointed out since 2002, India is the most complicated country in the world, so it's hard to get a nationally representative sample.

Further, the higher IQs registered by Indians in their diaspora, like the higher IQs recorded by African Americans and West Indians compared to Africans, suggests that better nutrition, health, education, and so forth could raise the relative standing of Indians. After all, they've been plagued by extreme poverty, which is starting to abate. For example, India is particularly beset by lack of iodine in the diet, which is a relatively easy IQ-lowering problem to deal with.

Rec1Man's goal has been to come up with a model that's simple enough to comprehend. Lots of critics in the comments responded that his eight part categorization of Indian demographics was too simple. As Einstein said, Science should be as simple as possible, but no simpler.

Still, we need somewhere to start, so Rec1Man's model, after being refined over the last few weeks, seems like a decent starting point.

First, here is fellow commenter Nsam's verbal summary of Rec1Man's approach, followed by Rec1Man's latest quantitative summary:

Here is my attempt at a summary of rec1man's analyses. Edit as required.

This eight-segment breakdown of the Indian population provides a more accurate picture of the human resource potential of the Indian population, compared to simplistic unimodal population model with mean IQ of 85 and SD =15.

The Indian population can be grouped in terms of the four traditional caste super-categories :

1. Brahmin 2. Merchant, 3. Peasant, 4.1 Dalit ("Untouchable") 4.2 Scheduled Caste 4.3 Low caste converts to Islam

cross-cutting with a

A. North vs. B. South

geographical distinction.

Note: The usage of the 4 caste supercategories is not literal. For the purposes of estimating national average IQ, Syrian Christians or Parsees would be lumped in with the Brahmins, and the few peasant castes with a literate tradition would be moved up to the merchant class.

The "Brahmin" and "Merchants" have had multiple generations of a literate tradition and also engaged in professions that required greater cognitive ability (memorization, calculation), compared to the "Peasants" (who formed the ranks of the warrior class and had/have the most political power) and Dalits (who also have the least amount of central-west asian-european admixture and have been engaged in menial tasks on the fringes of society). [As in the Cochran-Harpending model of the evolution of high average Ashkenazi Jewish IQs, under Malthusian pressure, these occupational slots are theorized to have selected for higher IQs. — Steve]

This model posits an IQ penalty for North Indian groups as a result of Islamic invasions/genocide of north Indian elites over a millennium.

Improvement in living conditions, nutrition, and access to education is expected to increase mean IQ by 7.5 points by the end of this century, relative to the Anglo-American average. India has the human resource potential to be to be a developed society with advanced technological capabilities by the end of this century. The cognitive elites of India will primarily come from the Brahmin and Merchant super-categories and numerically compare favorably with the other Asian superpower China.

And here's Rec1Man's latest summary table:

Current IQ

Southern Brahmin, 1% @ 120 IQ Northern Brahmin, 4% @ 115 IQ Southern Merchant, 3% @ 110 IQ Northern Merchant, 12% @ 105 IQ Southern Peasant, 8% @ 93 IQ Northern Peasant, 32% @ 88 IQ Muslim, 15% @ 75 iQ Southern Dalit, 5% @ 75 IQ Northern Dalit, 20% @ 75IQ

Avg = 86.10 IQ Current Flynn for India is +7.8

Hypothesized Flynn effect for India in 21st Century: Brahmins and merchants = +5 Peasants = +7 Muslims and Dalits = +10 The Flynn for Africa = +10

Future Indian IQ

Southern Brahmin, 1% @ 125 IQ Northern Brahmin, 4% @ 120 IQ Southern Merchant, 3% @ 115 IQ Northern Merchant, 12% @ 110 IQ Southern Peasant, 8% @ 100 IQ Northern Peasant, 32% @ 95 IQ Muslim, 15% @ 85 iQ Southern Dalit, 5% @ 85 IQ Northern Dalit, 20% @ 80IQ

Avg = 94

After Flynn High End Indian IQ Over 120 IQ = 55 mil Over 130 IQ = 20 mil Over 140 IQ = 5 mil

I certainly can't comment on the validity of the empirical details that underlie this model — India being so complicated — but the general approach seems not unreasonable to me. The point is to put a framework out there for the public to consider.

One potential problem that jumps out at me with Rec1Man's approach, though, is the potential for disparate fertility interfering with improving environmental conditions. Are fertility rates the same across all groups, or are lower IQ groups growing faster? The Bombay Parsis, who were the most affluent and educated group in India for much of the 20th century, are said to be imploding demographically. If IQ, education, and declining fertility are correlated, then dysgenic / dyscultural effects could undermine some of the gains from an improved environment.

Granted, we haven't seen much hard evidence of a dysgenic / dyscultural relative effect yet. But, then, we haven't seen much hard evidence of a positive Flynn relative effect. We've seen lots of evidence of an absolute Flynn effect, but little evidence of countries shifting relative average IQs, as I documented here. We've probably seen more evidence of relative changes in height than relative changes in IQ.