Yesterday we looked at evidence for a genetic theory of racial differences in IQ. Today let's look at some of the arguments against it. Again, I'm drawing heavily on a recent exchange of papers published by the American Psychological Association.
One objection is that IQ tests are racially biased. This is true in the broadest sense: On average, African and Asian kids have different advantages, and IQ tests focus on the things at which more Asian kids have the edge. But in the narrower sense of testing abilities that pay off in the modern world, IQ tests do their job. They accurately predict the outcomes of black and white kids at finishing high school, staying employed, and avoiding poverty, welfare, or jail. They also accurately predict grades and job performance in modern Africa. The SAT, GRE, and tests in the private sector and the armed forces corroborate the racial patterns on IQ tests. Kids of different backgrounds find the same questions easy or hard. Nor do tests always favor a country's ethnic majority. In Malaysia, Chinese and Indian minorities outscore Malays.
If the tests aren't racist, some critics argue, then society is. That's true, in the sense that racism persists. But that alone can't account for the patterns in IQ scores. Why do blacks in the white-dominated United States score 15 points higher than blacks in black-dominated African countries, including countries that have been free of colonial rule for half a century? And why do Asian-Americans outscore white Americans?
Another common critique is that race is a fuzzy concept. By various estimates, 20 percent to 30 percent of the genes in "black" Americans actually came from Europe. Again, it's a good point, but it bolsters the case for a genetic explanation. Black Americans, like "colored" South Africans, score halfway between South African blacks and whites on IQ tests. The lowest black IQ averages in the United States show up in the South, where the rate of genetic blending is lowest. There's even some biological evidence: a correlation between racial "admixture" and brain weight. Reading about studies of "admixture" is pretty nauseating. But the nausea doesn't make the studies go away.
My first reaction, looking at this pattern, was that if the highest-scoring blacks are those who have lighter skin or live in whiter countries, the reason must be their high socioeconomic status relative to other blacks. But then you have to explain why, on the SAT, white kids from households with annual incomes of $20,000 to $30,000 easily outscore black kids from households with annual incomes of $80,000 to $100,000. You also have to explain why, on IQ tests, white kids of parents with low incomes and low IQs outscore black kids of parents with high incomes and high IQs. Or why Inuits and Native Americans outscore American blacks.
The current favorite alternative to a genetic explanation is that black kids grow up in a less intellectually supportive culture. This is a testament to how far the race discussion has shifted to the right. Twenty years ago, conservatives were blaming culture, while liberals blamed racism and poverty. Now liberals are blaming culture because the emerging alternative, genetics, is even more repellent.
The best way to assess the effects of culture and socioeconomic status is to look at trans-racial adoptions, which combine one race's genes with another's environment. Among Asian-American kids, biological norms seem to prevail. In one study, kids adopted from Southeast Asia, half of whom had been hospitalized for malnutrition, outscored the U.S. IQ average by 20 points. In another study, kids adopted from Korea outscored the U.S. average by two to 12 points, depending on their degree of malnutrition. In a third study, Korean kids adopted in Belgium outscored the Belgian average by at least 10 points, regardless of their adoptive parents' socioeconomic status.
Studies of African-American kids are less clear. One looked at children adopted into white upper-middle class families in Minnesota. The new environment apparently helped: On average, the kids exceeded the IQ norms for their respective populations. However, it didn't wipe out racial differences. Adopted kids with two white biological parents slightly outscored kids with one black biological parent, who in turn significantly outscored kids with two black biological parents. The most plausible environmental explanation for this discrepancy is that the half-black kids (in terms of their number of black biological parents) were treated better than the all-black kids. But the study shot down that theory. Twelve of the half-black kids were mistakenly thought by their adoptive parents to be all-black. That made no difference. They scored as well as the other half-black kids.
In Germany, a study of kids fathered by foreign soldiers and raised by German women found that kids with white biological dads scored the same as kids with biological dads of "African" origin. Hereditarians (scholars who advocate genetic explanations) complain that the sample was skewed because at least 20 percent of the "African" dads were white North Africans. I find that complaint pretty interesting, since it implies that North Africans are a lot smarter than other "whites." Their better critique is that the pool of blacks in the U.S. military had already been filtered by IQ tests. Even environmentalists (scholars who advocate nongenetic explanations) concede that this filter radically distorted the numbers. But again, the complaint teaches a lesson: In any nonrandom pool of people, you can't deduce even average IQ from race.
Other studies lend support to both sides. In one study, half-black kids scored halfway between white and black kids, but kids with white moms and black dads (biologically speaking) scored nine points higher than kids with black moms and white dads. In another study, black kids adopted into white middle-class families scored 13 points higher than black kids adopted into black middle-class families, and both groups outscored the white average.
Each camp points out flaws in the other's studies, and the debate is far from over. But when you boil down the studies, they suggest three patterns. One, better environments produce better results. Two, moms appear to make a difference, environmentally and biologically. (Their biological influence could be hormonal or nutritional rather than genetic.) Three, underneath those factors, a racial gap persists. One problem with most of the adoption studies is that as a general rule, genetic differences in IQ tend to firm up in adolescence. And in the only study that persisted to that point (the one in Minnesota), kids scored on average according to how many of their biological parents were black.
The best argument against genetics isn't in these studies. It's in data that show shrinkage of the black-white IQ gap over time. From these trends, environmentalists conclude that the gap is closing to zero. Hereditarians read the data differently. They agree that the gap closed fractionally in the middle decades of the 20th century, but they argue that scores in the last two to three decades show no improvement.
I've been soaking my head in each side's computations and arguments. They're incredibly technical. Basically, the debate over the IQ surge is a lot like the debate over the Iraq troop surge, except that the sides are reversed. Here, it's the liberals who are betting on the surge, while the conservatives dismiss it as illogical and doomed. On the one hand, the IQ surge is hugely exciting. If it closes the gap to zero, it moots all the putative evidence of genetic barriers to equality. On the other hand, the case for it is as fragile as the case for the Iraq surge. You hope it pans out, but you can't see why it would, given that none of the complicating factors implied by previous data has been adequately explained or taken into account. Furthermore, to construe meaningful closure of the IQ gap in the last 20 years, you have to do a lot of cherry-picking, inference, and projection. I have a hard time explaining why I should go along with those tactics when it comes to IQ but not when it comes to Iraq.
When I look at all the data, studies, and arguments, I see a prima facie case for partial genetic influence. I don't see conclusive evidence either way in the adoption studies. I don't see closure of the racial IQ gap to single digits. And I see too much data that can't be reconciled with the surge or explained by current environmental theories. I hope the surge surprises me. But in case it doesn't, I want to start thinking about how to be an egalitarian in an age of genetic difference, even between races. More on that tomorrow.
Will Saletan covers science, technology, and politics for Slate and says a lot of things that get him in trouble.