People vary in their abilities based in part on genetic differences. Suppose these differences at the individual level sometimes add up to differences in average ability between people of one race and people of another. Should we say so?
Here are three perspectives on the question. On Wednesday, the New York Times ran the following story:
'No Child' Law Is Not Closing a Racial Gap
The achievement gap between white and minority students has not narrowed in recent years, despite the focus of the No Child Left Behind law on improving the scores of blacks and Hispanics, according to results of a federal test considered to be the nation's best measure of long-term trends in math and reading proficiency.
On Thursday, I raised a question about the Times story:
Why categorize and measure students by race? Aren't there better ways to organize the data? … [Parts of the test report] organize the data by factors that can help us target and adjust educational policy: kids with low scores, kids in public school, kids in high school, kids whose parents didn't graduate. … But race? Does that category really help? And what message does it send to kids when headlines assert a persistent "racial gap"?
The reason people all over the world and of all different ideologies can't help but be interested in race is [that] a racial group is, fundamentally, an extended family. So, race is about who your relatives are, which is an inherently interesting topic.
Saletan has been arguing that we should just group people by looking at one gene at a time. (Of course, on average, individual gene differences will tend to follow racial lines.) But, more fundamentally, what he doesn't get is that racial groups have an existence independent of genetics. They are fundamentally genealogical entities—who begat whom. Unsurprisingly, when you stop and think about it, the genes tag along with the begats.
Sailer, like the Times, is embracing racial averaging of test scores. But unlike the Times, he's doing so in the belief that differences in the resulting averages are in large part genetic. He's arguing not just that some people do better than others based on inherited ability (the genetic question) and that this ability is more prevalent among people of one race than among people of another (the distribution question), but that this is how the data should be aggregated, averaged, and compared (the framing question).
It's important to separate these three questions. We know that genes influence many abilities. We also know that some of these genes vary considerably in prevalence between ethnic groups. One example is the RR variant of ACTN3, a gene that affects fast generation of muscular force and correlates with excellence at speed and power sports. The opposite variant of the gene is called XX. Tests indicate that the ratio of people with RR to people with XX is 1 to 1 among Asians, 2 to 1 among European whites, and more than 4 to 1 among African-Americans.