Gifted Issues Discussion homepage Forums Parenting and Advocacy School funding leaves gifted students behind

The Bell Curve debate is always emotionally charged, so thanks to everyone for keeping it reasonable so far. A couple of points.

First, Catalana quotes from Block's article:


Val thinks this is a mischaracterization of the book:


I agree with Val that Herrnstein and Murray hedge their bets. They say, for example, that certain factors "suggest, without quite proving, genetic roots" for differences in IQ between ethnic groups (p. 270 of the version available on Google books). On the other hand, it is certainly the goal of their book to argue that the heritability of IQ is at least in part genetically determined. They do so by arguing, for example, that environmental differences are not sufficient to explain the 15 point gap between blacks and whites in IQ scores (see the section entitled "Are the differences in overall black and white test scores attributable to differences in socioeconomic status", pp. 286-289); by arguing that the gap is not attributable to other factors such as biased questions in the tests (see p. 282); and by arguing that the noted convergence between black and white test scores over the last several decades is not likely to end in equality (see pp. 289ff). In addition to arguments that non-genetic differences are not sufficient, they also give a variety of arguments to the effect that the IQ difference between blacks and whites is genetically determined. (See the section entitled "Reasons for thinking that genetic differences might be involved", pp. 299ff.) In all, then, I think Block is giving a fair account of the book when he says that it argues (with respect to IQ at least) that "Black Americans are genetically inferior to Whites".

The real issue around the book, though, is not what it argues but what its reasons for arguing it are. The claim that Block and others make is that H&M go too quickly from the claim that IQ differences are heritable to the claim that they are genetically determined. To say that IQ is 60% heritable within whites, as H&M and Block agree, is to say that there is a 60% correlation between variance in genes and variance in IQ. But this leaves completely open the question whether the genetic variance directly determines the IQ difference or indirectly determines it. In Block's example, earring wearing was highly heritable in American society during the 1930s: there was a strong correlation between variance in the sex chromosome and variance in earring wearing practices. But intuitively, at least, it would be wrong to say that whether you were an earring wearer or not was a genetically determined trait. The reason is that there are a number of different possible reasons for the correlation between genotype and practice, and in this case the correlation seems to have come about largely as a result of the way the environment treated the genders differently, not as a result of the genetic influence of the sex chromosome itself.

The nice paper that Dandy links to by Sesardic draws out these different kinds of genotype-environment correlation. It seems to me exactly on target on the facts, but to leave Block's criticism of H&M completely untouched. The reason is that Block draws exactly the same distinctions that Sesardic does, but then goes on to argue that in the case of IQ heritability in particular, there is simply no way of knowing which kind of genotype-environment correlation is in effect. If only some of these kinds of correlation justify the move from heritability to genetic determination, and if we don't know which one is in play in the IQ case, then the move that H&M make from IQ heritability to the genetic determination of IQ is simply not justified.

I'd be fascinated to hear what others think. Perhaps we won't get Dandy's dream of a live debate between Block and Sesardic. But we could try to figure out for ourselves how the debate would go.

BB