Last Saturday, I had the pleasure of reading A. K. Dewdney's Yes, We Have No Neutrons, describing eight celebrated ?scientific? blunders, results of earnest but deeply flawed scientific method. The failures behind cold fusion, Freudian psychology, and Biosphere 2 are entertaining until you realize that similar mistakes are endemic to science; science may be reliable, but the scientists are only human.
One of Dewdney's case studies is J. Phillipe Rushton's infamous treatise correlating race with intelligence: Mongoloids are (supposedly) more intelligent than Caucasoids, who are in turn (again supposedly) more intelligent than Negroids. I don't defend Rushton's Race and Behavior; as Dewdney points out, it's bad science. On the other hand, it bothers me that Rushton was defeated in the court of public opinion, booed off Geraldo Rivera's stage, rather than broken on scientific grounds. Scholars and the daytime-TV rabble alike rejected Race and Behavior, not because it was poorly reasoned, but because they didn't want to hear the conclusions. ?I've already made up my mind; don't confuse me with the facts.?
There is nothing inherently implausible in the idea that different races may have different brain sizes, different tendencies toward violence, different degrees of sexual drive, or even different intelligence ? whatever that means ? on a broad average. We can see that behavioral as well as physical traits can be carried in the genes by looking at dogs. Golden retrievers, for example, are widely praised in dog books for their sweet tempers. Collies who have never so much as smelled a sheep, nor seen a farm, nonetheless herd small children as they would sheep; I've seen it happen, myself. German shepherds are notably more intelligent than the average breed, bulldogs notably less. Whatever mechanism can code behaviors into dogs can code it into humans, too.
But before my friends panic and think I've gone racist, I need to point out some big obstacles in actually discovering what the racial differences are.
One: while foot length, melanin count, and brain weight are easily measured, but more nebulous concepts like intelligence, aggression, and sexual desire are so poorly defined as to defy measurement. (The history of the intelligence quotient is another chapter in bad science.)
Two: humans make poor test subjects for the scientific method. You can drop a billiard ball from a given height onto a given surface and see how high it bounces hundreds, even thousands of times in an afternoon. You cannot grow a child from birth hundreds of times; one shot is all you get. Setting up a ?control? for comparison purposes is horribly difficult for most of the interesting questions, and doing so would, 99% of the time, be horribly unethical. Want to keep a thousand people isolated in blank rooms for twenty years to ensure they all experience the same environmental stimuli?
Three: humans are intelligent. We react to their environment in enormously complex ways, thereby clouding the data a scientist might hope to gather, often to the point of white noise. Worse, we can recognize that we are being studied and ? perhaps deliberately, perhaps unconsciously ? behave differently than we would unobserved. Physicists think they have it rough, with observation affecting the observed subatomic particle; psychologists have to contend with a subject that may willfully attempt to ruin the experiment.
So the jury is still out, and is likely to remain so for a long, long time. I'll listen respectfully to the hypothesis that whites are, on the average, fractionally smarter than blacks, but I'll listen with equal respect to the hypothesis that blacks are, on the average, fractionally smarter than whites. I've seen some suggestive evidence that primitive humans are more intelligent than civilized ones. Soft living preserves our stupid genes, while nature kills off stone-age morons. That's another trait we share with dogs, who are dimmer than their wolf ancestors.