It's not all that common that I find myself agreeing with Peter Singer on a controversial ethical issue, but I was reading a section of his book on the moral status of animals and came across a passage where he discusses racial differences, and I found his discussion refreshingly honest in a way that would come across as politically incorrect in many circles.
He complains that the primary opposition most people have toward racism seems to come from being able to see that people in different races aren't all that different, and thus discrimination against people of a certain race merely because of their race is arbitrary and morally unfounded. I've just started reading one of his articles directly on that topic, and I may well have something further to blog about it once I'm done, but in the animals discussion Singer registers some worries about this approach that I've long had myself, worries I don't see very many people expressing.
A great number of trees have been killed to try to defend the claim that there are no racial differences that might be remotely connected to anything morally significant. At any suggestion that part of the cause of the IQ gap has to do with something besides current racism, sociologists and psychologists do all sorts of empirical work trying to show that black students score lower when they know their race is being recorded with their score and such things. It's as if everything anti-racist hangs on being able to establish that the only possible cause of racial differences in test scores might be racism itself.
Singer points out that this assumption is actually the problem. If you assume that overcoming racism requires it to be a fact that there are no significant differences in intelligence between two given races, then the racist with some reason to see a difference will seem more excusable. If anti-racism rests on the assumption that there will never be a gap in intelligence between two races, what happens if you discover that, at least with respect to one method of registering differences, there is such a gap? What if part of the intelligence test gap actually comes from some biological differences in brain capacity that might make it harder to do certain tasks? Would that then make it all right to discriminate against people of that race? So basing the argument on empirical facts that might turn out to be false isn't the best idea.
The further thought that I've had is that, whenever you take an average on anything, you're bound to have averages that differ. It's almost overwhelmingly guaranteed that one of the averages will turn out to be closer than the other to the goal in question, even once you adjust for environmental factors. You're simply not going to be able to establish the view that there are no differences that lead to slightly higher averages on some measure with white people than with some non-white group. Such a result isn't just some remote possibility that we hope isn't true. It's almost certainly going to be true statistically speaking. If you manage to get a test that does fairly well at testing for skills of a certain sort, it's overwhelmingly likely that some racial groups will test better at it, because those people who happen to belong to certain races are not likely to have exactly the same average than those of every other race. One will turn out to be better on average, and that result would be a pretty poor excuse if someone wanted to use it to justify racism. If you took any random sampling of humanity and tested them, then took a different random sampling and tested them, it's extremely unlikely that they'd have the same average. If the average for blacks turned out to be lower than the average for whites, what would that tell us? Absolutely nothing of any consequence, since we would expect that the numbers couldn't be the same, and when you're talking large numbers of people with variable scores it might be reason to suspect divine intervention if you got exactly the same result for each group.
So I'm in full agreement with Singer about those who resist tooth and nail any possibility that there might be a lower average among one race when it comes to a particular measure of intelligence. It's a fruitless quest, and the factual discovery such people so strongly want to resist isn't really going to lead to enough of a morally-significant difference to justify the strong resistance.