Tuesday, August 15, 2006

Hot, smart brown girls   posted by Razib @ 8/15/2006 10:50:00 PM

[Originally posted by agnostic, reposted by Razib since blogger mystifies his assophilic self]

Sometimes you notice patterns that don't fully crystallize consciously until some precipitating event really "wakes you up" to the reality. I was riding the metro home tonight and across from me was a girl who was engrossed in Kafka on the Shore -- apparently a high-brow fiction book, not like The Da Vinci Code or something (I don't keep up to date on fiction) -- and who got off at the extremely affluent Friendship Heights neighborhood. So you get the rough idea of her IQ. Unusually for someone that smart & well-to-do, at least in my experience, she was smoking hot. She looked South Asian, maybe Persian. Then it clicked on a conscious level: for those of you who are say, aged 18-40 -- have you ever noticed how many hot, smart brown girls there are in your age group?

To put things in perspective, one estimate of South Asian Americans puts them at ~2 million, or about 2 orders of magnitude less frequent than Euro-Americans. So, assuming no difference across groups, the absolute number of their hot, smart girls should be about 1% the absolute number of hot, smart white girls. Now, "good looks" isn't as easy to research quantitatively, but even impressionistically, there's no way I've seen that low of a number of hot, smart brown girls. If I had to guess, I'd say they're overrepresented by about 1 order of magnitude.

Now, the stereotype I grew up with, and which I didn't see violated overall, was that looks and brains were pretty independent. In a recent post (and also here), Darth mentioned Jensen & Sinha's view that the correlation between height & IQ might reflect cross-assortative mating segregating alleles for both height & IQ into the same families. But, hypothetically, imagine a population that is already highly stratified by IQ / social status and has been for at least roughly 1000-2000 years. And imagine in this population, the males are at least or more likely than elsewhere to select red-hot trophy wives rather than intellectual equals (again, initially, say beginning 1000 or so years ago). This would tend to produce an even stronger correlation between good looks and smarts on the (sub-)population level.

The first assumption is met by the South Asian caste system (though history scholars can correct my time estimate), and the second is met by any geographical area that's wracked by pathogens, such as the Subcontinent. Buss & Gangestad (1993) found that there was a strong correlation between pathogen prevalence and emphasis that people from that area placed on "good looks" when evaluating a potential mating partner. In India, in particular, according to Buss' (1989) massive survey of cross-cultural mating preferences, males rated "good looks" as 2.03 (from 0 -- not important at all -- to 3 -- essential). Skimming the other countries, that appears at or above median. As an aside, Indian females rated this variable as 1.97 -- one of the few countries where there was no statistically significant sex difference in emphasis on good looks in a partner (along with Poland and Sweden). Assuming these two features have been present for even 1000-2000 years, that's 50-100 generations. Devlin et al (1997) found a narrow-sense heritability of cognitive ability amounting to 0.35; no one's quantified it for physical attractivenss, though a twin study by McGovern et al (1996) found a broad-sense heritabilty of 0.64, and this mostly reflects additive variance, as the MZ concordance rate was almost exactly twice the DZ rate.

Thus, the outcome of 50-100 generations of brainy, high-status males snatching up the prettiest girls in the area (where the looks of most are defiled by bugs) seems to be a (sub-)population where the common stereotype of the independence of looks & brains no longer holds. If you're looking for a girl who's both hot and smart, upper caste South Asians aren't a bad bet. Now, this leaves out personality traits, which may or may not complicate the search (informally, I've noticed that hot, smart girls tend to be more princess-y -- what a shock!). Yet, the violation of expectation here isn't due to wispy cultural factors -- whenever I've talked to people about the brains-looks relationship, the common response is that ugliness causes one to become smarter, since there aren't many ways to gain influence, and if looks ain't your niche, brains it is; whereas the hot person decides to coast on appearance rather than invest time in making themselves smarter. In reality, it seems the only way to un-do the common stereotype is cross-assortative mating between brainy guys and hot girls: kinda like how Dr. Strangelove imagines the re-population of the world will take place after nuclear holocaust. A world where girls are smart and hot -- the apocalypse wouldn't be so dystopic after all!

Open question for future research: does the same hold true for upper caste South Asian males? Intelligence in males & females is roughly the same thing, but "good looks" may be spring from distinct sources in males & females, or they may reflect a sex-blind source like strengthened immune system. As I discovered going through the hotcaptcha for judging guys, I don't have a reliable hunch for which guys are hunky or not, so I'll let any female readers offer their impressions.


Buss (1989). Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures. Behavioral & Brain Sciences, 12, 1-49. Link (Ctrl+F "bbs" for free pdf)

Buss & Gangestad (1993). Pathogen prevalence and human mate preferences. Ethology and Sociobiology, 14, 89-96.

Devlin, Daniels, & Roeder (1997). The heritability of IQ. Nature, 388, 468-71.

McGovern, Neale, & Kendler (1996). The independence of physical attractiveness and symptoms of depression in a female twin population. Journal of Psychology, 130, 209-219.
[Originally posted by agnostic, reposted by Razib since blogger mystifies his assoliphic self]