I've been reading about the AEI seminar about IQ in Ashkenazic (of European ancestry) Jews with some interest (ScienceBlogling Razib raises a really good point that I hadn't thought of too). A few years ago, I was asked to review a paper that dealt with this issue.
Let me state that I think IQ as a measure of intelligence doesn't mean all that much, except when it is extremely low or extremely high. Nonetheless, it is a trait that we can measure--we should just be very careful about how much importance we place on IQ.
OK, back to the paper I was asked to review. It basically made the same argument that Entine makes. Ashkenazic Jews, because they had very low introgression from non-Ashkenazic populations, had a population structure where adaptive mutants for higher IQ could arise and be maintained given the appropriate selection pressure. The result is that Ashkenazic Jews have relatively higher frequencies (0.5-2.0%) of alleles (genetic variants) of genes that result in unusual neurological development (e.g., Tay-Sachs, Gaucher's syndrome, other sphingophospholipids).
These alleles haven't swept to fixation (i.e., occur at really high frequencies) because when they occur in the homozygous state (i.e., you have two copies of the allele) the effects are very bad--in Tay-Sachs, the child dies at a very young age. If heterozygous parents are too common (those who possess one allele), they will marry each other relatively frequently, and one-quarter of their offspring will die (this would be an example of negative-frequency dependent selection).
A corollary of this is that, if there were selection for alleles that confer higher IQ but confer fitness costs when homozygous, we would expect to see lots of loci with variant alleles at low frequencies, instead of one locus with a variant allele (with many different loci, the odds of a heterozygote marrying someone who is a heterozygote at the same locus are quite small). We do observe quite few loci with rare alleles with deleterious effects in Ashkenazic Jews.
What I've found frustrating in all of this is that these are testable hypotheses: we can screen for Tay-Sachs heterozygotes and give them IQ tests and compare them to homozygotes (obviously, you would have to control for demographic factors). Ideally, we would like to do twin studies, but I imagine, at this point, the few Ashkenazic twins out there are sick and tired of being enrolled in twin studies.
So, is anyone going to do these tests?
An aside: As an Ashkenazic Jew, I would love to be your genetically superior overlord (kidding), but as Dana Milbank put it:
Left unchallenged was the question of whether Jews are indeed smarter than others -- even though it would have only required a walk down the hall to the office of new AEI visiting fellow Paul Wolfowitz, whose leadership on the Iraq war and conflicts of interest as head of the World Bank demonstrate that Jews are capable of questionable judgment.
- Log in to post comments
"Intelligence" is too amorphous for a comparative study of different ethnic groups, and sfaik the Jews, having wandered all over the map, are not a sufficiently isolated population to be a good study group.
I propose researchers look for groups which have faced consistent selection pressures over many centuries, with little admixture from migration, etc, and see what genetic patterns they find. For example, the Irish and the inland Chinese have been small-scale agriculturalists (aka peasants) for millennia, whereas, say, Amazonian Indians have been hunter-gatherers for even longer. I'd suggest that the stationary populations might show greater resistance to certain infections, while the latter might have better eyesight.
The above pre-supposes that we can identify specific genes "for" specific traits, which is at best wildly premature - but it still seems more realistic than IQ-centric research (or proposals from Wolfowitz).
I really hate to bring this up, but please understand that I'm a merely a student with an interest in these things, and not a person involved in this field in any way. I'm not trying to cause a commotion or anything.
I do have to ask, though:
Why is it that tests such as these are considered viable and not racist, while James Watson's recent comments landed him in so much hot water?
Isn't it viable that if Jews' isolation gave rise to a higher than average intelligence, than Africans' isolation could have done the opposite?
Is it more the manner in which Mr. Watson gave his statement, coupled with his history of such remarks, that caused what he said to be so outrageous?
Basically: is there a scientific reason why it's acceptable to speculate why Jews may be smarter, but not to speculate why Africans may be less intelligent, or is it more related to the knee-jerk "racism! racism!" reaction western society has when discussing racial differences?
Again, I'm not defending Mr.Watson's remarks, but simply trying to understand the situation more clearly.
I'm sure I'm even less credentialed than you are, but it seems to me:
A) Watson hit a sociopolitical sore spot: he wouldn't have been in half as much trouble if he'd made a purely scientific blunder, such as claiming that humans had 64 chromosomes.
B) Africa's "isolation" is _only_ a sociopolitical factor: viewed genetically, the human population there has much greater variation than that of the rest of the world combined (and interbreeding among subgroups seems to have been widespread). Though studies of African genomes will surely tell us a great deal, the mother continent may well be the least appropriate place to look for intra-group genetic research.
As to Watson -- as he himself conceded, what he said about Africans was not at all scientific -- it was his opinion colored by his own personal experience, which, valid or invalid, ain't science. So, yes, some of the outrage was because he sounded racist against Blacks (something any civilized European indeed ought to be cautious of, given the history of racist exploitation.) Watson has made a great reputation as a scientist, then used that reputation to propound views that many consider repugnant. (Not to mention that others have had very different experience -- see Guns Germs and Steel, for example)
As a reminder of what IQ might be and might not be, I like to point to Showa cloths, which were made by the BaKuba people. These weavings had intricate patterns that typically merge four (sometimes more) designs into each other in subtle and interesting ways. (For an example, see http://www.grainsofafrica.com/catalog/product_info.php?products_id=95 ) The Belgians, however, used to say that the cloths showed inferior African intelligence because, in the Belgian view, the weavers were unable to continue the same pattern for the whole cloth. So, by Belgian measures of "intelligence" the Bakuba were inferior. This was just about as "scientific" as what Watson said.
Which brings me to the posting. Given all the various needs for scientific studies, why would someone prioritize a study of intelligence variation and genetic markers amongst Askenazic Jews? The posting itself concedes that IQ isn't really a very important subject. It's very hard to define what's being measured except self-referentially. If we single out Jewish or Blacks or Tibetan or Plutocratic IQ to study, it reinforces the whole idea that IQ actually is meaningful. If we study IQ across cultures without accounting for cultural differences, the results are questionable at best. (Which is why there isn't any scientific evidence to prove or refute Watson's racial comments.) So, if there's some correlation between "intelligence" on one test or other and something genetic among a small subpopulation, why does it matter enough to take resources away from other sorts of studies that aren't being funded these days?
The subject of Ashkenazi Jews brings up a point that occurred to me when the notorious book "The Bell Curve" was published. If it were actually documented that group X had lower average IQ than group Y, it doesn't actually mean anything about any intrinsic genetic difference between Xs and Ys. In particular, it does not imply that group Y has some kind of "mutation" that leads to higher-IQ that group X lacks. It just means that whatever genes are involved in influencing IQ may have different frequencies in the two groups. And that difference could very well be the result of cultural differences.
Just as an example: Suppose that in group Y, it is expected that highly intelligent individuals will dedicate their lives to becoming celibate monks contemplating the mysteries of the universe. In group X, the highly intelligent individuals are encouraged to have lots and lots of children. If such cultural differences lasted long enough, it could lead to differences in frequencies of smart people in the two groups. But there would be nothing intrinsic about the differences. If the cultures changed (and they always do, given enough time), then the genetic frequencies would shift again.
I think it is possible that Ashkenazi Jews may have for many generations participated in cultural practices that led to increased frequencies of smart people. But for modern, assimilated, secular Jews, those cultural practices are mostly gone. So we should expect that in the future, Jews will be as dumb as the rest of us.
Why is it that tests such as these are considered viable and not racist, while James Watson's recent comments landed him in so much hot water?
Well, it may be because "blacks" are not a homogenous genetic group, but "Ashkenazi Jews" (a subset of "Jews") are. "Race" is a social, not biological, construct.
Like Icelanders, the Amish, etc, Ashkenazi Jews have been mostly introgamous, so you have the possibility of investigating genetic links to intelligence, among other things hereditary. I'm not convinced that any group has any particular genetic advantage in the intelligence line, because I don't believe that any group has been under enough selective pressure to produce it, and we don't really have a decent handle on what constitutes "intelligence" outside of a specific subset of intellectual abilites.
With crap like the Watson comments floating around, it would be nice to be able to pinpoint any factors that may exist, though.