search
Wednesday, 2002 May 1, 12:18 — humanities, psychology, sciences

exegesis of intelligence

Stephen Jay Gould, in one of his Natural History essays, wrote:

I believe that any solution to this key puzzle in Darwinian biography must begin with a proper exegesis of intelligence – one that rejects Charles Spearman’s old notion of a single scalar quantity recording overall mental might (called g or general intelligence, and recently revived by Murray and Hernnstein as the central fallacy of their specious book, The Bell Curve – see the second edition of my book The Mismeasure of Man). Instead, we need a concept of intelligence defined as a substantial set of largely independent attributes.

(I have read neither of the books mentioned, but when does that stop anyone?)

The thesis of The Bell Curve, I gather, is that the races (whatever that means) have different distributions of g, though these differences are dwarfed by the variation within each race: the area under the two bell curves overlaps except at the fringes.

Current enlightened thought says that intelligence is a composite of several uncorrelated talents, and ‘general intelligence’ is about as meaningful as the sum of your age and your shoe size. But is that a fatal fallacy? If ‘average sum of age and shoe size’ is found to differ between two groups, isn’t that a sign that their ages and/or their shoe sizes differ? At most we might find that if the components were weighted differently (say by using European rather than American shoe sizes) the ranking would change.

If g is a composite, I’d expect the gaps to be bigger (though not all in one direction) if the components are measured separately – rather than vanish, as I assume TBC‘s denouncers want us to think.* But Gould says it’s fallacious to measure g at all; implying that g is not even a composite, but measures something completely meaningless.

What gives? <whine> Do I gotta read another book? </whine>

*This sentence rewritten after discussion with an old friend. The original version implied that all the spreads are in the same direction, i.e. that the median White is much brighter than the median Black in each of the seven-or-so ways.

No comments yet.

Subscribe without commenting

RSS feed for comments on this post. TrackBack URI

Leave a comment