The Economist’s Will Wilkinson has a rather perplexing article about the Jason Richwine affair — and the reactions of various conservatives including yours truly. There are so many different threads here that the piece is difficult to rebut, but here are a few things that stood out to me.
For starters, Wilkinson misrepresents a key argument made in Richwine’s dissertation:
[Richwine argued] that Hispanic immigrants are less intelligent than non-Hispanic white Americans, that this gap has a genetic basis, and that immigration policy should discriminate against less intelligent groups of people.
As I wrote in my article, the last item on that list is not true: Richwine did not argue for excluding entire groups of people based on the average IQ of the groups. To the contrary, he suggested admitting high-IQ people from all groups, largely to benefit immigrants “who lack educational access in their home countries,” as he put it in the paper’s abstract.
Wilkinson then responds to an argument that I made — that if we want to maintain a stigma against racism, we should not define racism to include hypotheses that may be true:
If scientists are to ferret out even uncomfortable truths, they cannot be made to feel that they will be punished for it. Yet racism has always been predicated on falsifiable hypotheses about racial inferiority. No one has defined racism to include the assumption of hereditary racial inequality; that’s simply an assumption racists tend to have.
I’m not sure that testing a hypothesis in an academic setting is really comparable to assuming another group’s blanket inferiority as a way of dehumanizing them, but there is a point here: Racists often do believe things about other races that are in some way testable.
Ridiculous lies — the blood libel, etc. — can be corrected with truth if indeed racists really believe them. But in other cases the problem with racism isn’t the assumptions but the conclusions. Even well-established facts, such as racial differences in crime rates or a bad personal experience with a member of a certain group, can form the basis for racist attitudes. The solution isn’t to ignore the facts — that just lets racism fester — but to point out that the statistical trends, individual experiences, and so on do not prove that entire categories of people are bad or unworthy of respect.
Wilkinson seems to approach the question of assumptions, conclusions, and racism quite differently:
If Mr Richwine’s view “turns out to be correct”, what we are to do is to acknowledge that the racists were right all along — that racism has, to some extent, a valid scientific basis. People are understandably a bit touchy about this possibility. However, the subject is not fraught because “the left” has loaded it with toxic racial politics. It’s fraught because the scientific validation of hereditary racial inequality would imply that there’s something to be said for the racist convictions that made America’s brutal history of slavery, apartheid, and colonial genocide possible.
I find this flabbergasting. First of all, twin studies have established that IQ differences between individuals are largely genetic in nature — does this mean there’s “something to be said” for the convictions that led to the forced sterilization of the mentally disabled? Of course not. The problem with sterilization was that it took low IQ as a legitimate reason to violate human rights. The question of whether the people who were sterilized truly had genes that cause low IQ is utterly irrelevant.
But even that analogy doesn’t fully reveal the problem here: Sterilization, a policy based on the assumption that people with low IQs don’t have rights, actually targeted people with low IQs. Denying rights to an entire racial group because of a difference in average IQs makes no sense even with that horrifying assumption. It’s like putting every individual man in prison because men in general are biologically more inclined to be violent. Worse than that, even: The gender gap in violence is much more severe than any racial gap in intelligence could conceivably be.
The hypothesis that some statistical gaps might have a genetic basis should mean nothing for individual interactions; properly understood, it’s merely a way of explaining broad sociological trends. I am under no illusions that everyone will properly understand it, and it’s certainly reasonable to worry about how such an idea could be used. But there is no such thing as a “valid scientific basis” for racism, because racism is a moral failing.
Wilkinson then argues that Richwine’s work was so shoddy that we can conclude he was building toward a specific conclusion rather than testing a hypothesis in good faith. But the only evidence Wilkinson has for this is the opinion of the political scientist Daniel Drezner, who wrote a scathing if vague critique after having “perused parts of” the document. And the most impressive section of the dissertation — in which Richwine breaks out data on Hispanics from the National Longitudinal Survey of Youth and looks to see whether IQ scores improve from generation to generation — is purely empirical.
As I said in my piece, I am myself skeptical of claims that racial IQ gaps have a genetic basis — they rely on controversial, complicated science. I expect that most of the literature on this topic will soon be obsolete, replaced by research conducted directly on the human genome. I have no idea what that research will say. But whatever it is, I hope thought leaders will not interpret it as a “valid scientific basis” for denying rights based on race.