>It's pretty clear he doesn't. In the book tour for Bell Curve, he
>stated a number of times that a hereditability ratio for intelligence
>of .60 means that 60% of a person's intelligence is due to genetics
>and 40% is due to the environment. This is incorrect. The hered.
>ratio is a group level statistic that estimates the percentage of
>group level variation in a characteristic that can be accounted
>for by genetic variation. The percentage of a person's intelligence
>that is due to genetics cannot be quantified in any meaningful way.
Not only do heritability estimates not allow the attribution of some portion of any one's intelligence to genetics, they also only apply only to a particular population in a particular environment.In one environment, heritability could be 100%; in another environment heritability could be zero perent (see examples in Vandermeer Reconstructing Biology, pp. 1440-48). Heritability thus does not enable a measure of some essential or inherent attributes of organisms.
The deepest discussions of the heritability concept seem to Lewontin's and Elliot Sober's of which I still have quite an imperfect understanding.
yours, rakesh