Tuesday, October 30, 2007

Dust Settles on Watson

As the dust settles on the James Watson “racism” affair (see entry for Friday Oct 19th) what conclusions can we draw?

1. There was no stomach in the scientific community for a fight.

Dawkins and Blakemore had to respond, as well-known ‘public defenders’ for science. Their response was to denounce the witch hunt and insist that there are no questions off-limits to science. However, they did not initiate any further debate on the substantive issue nor refer to any of the research in psychometrics or evolutionary psychology which underpinned Watson’s statement. Honestly, I can’t say I blame them: they have careers and reputations and mortgages too.

2. The “Standard Social Science Model” is far from dead.

Those scientists who were brought out to denounce Watson were from the social science community justly attacked by Pinker in The Blank Slate. These are people who will willingly subscribe to evolution in the abstract, but who will not apply it to their own discipline - New Scientist had one such last week. It was similarly dispiriting to hear Craig Ventnor – who should know better – parroting that classic of the SSSM “race is just a social construct” - on Newsnight. Still, we all know his attitude to Watson dating from the Human Genome Project, and he’s perhaps more salesman than scientist.

3. Decision-makers do know where the truth lies here.

In “The g Factor” Arthur Jensen explained carefully the emergence of a ‘general intelligence’ attribute from factor analysis of batteries of tests measuring different facets of cognitive abilities. IQ tests are useful insofar as they strongly correlate with g, and g itself is important insofar as it strongly correlates with performance in cognitively demanding tasks. Otherwise, why are we bothering?

If one is not allowed to use IQ tests for political reasons, one can create ‘highly g-loaded’ proxy tests which one can pass off as task-specific aptitude tests. This is how the US military (and others) screen soldiers who are too dim to fight in a modern army. Racial differences in ‘aptitude’ emerge here too, as Jensen observed.

Screening in other public policy areas (e.g. immigration) through qualifications or vocational quotas is another way of proxying for intelligence and perhaps certain personality attributes. Interestingly, one would expect such filters to preferentially select people of Ashkenazim or east-Asian descent – tough on the Caucasian chauvinists.

Finally, I just love this “race is a social construct” thing. We’re quite similar to hamsters genetically, so probably 'species' is a social construct. The evolutionary argument is that humans, as they geographically dispersed out of Africa, radiated into subspecies with distinct adaptations, some of which included intelligence and personality adaptations. On top of these well-defined biological differences cultures have erected social categories which determine how they understand and treat representatives of such different races, not all of them pleasant. How hard is that to understand?