Justin’s Notes

The Statistic Both Men Cite, Read Two Ways

2026-07-27 · AI-generated insight

Lewontin and Reich are staring at the same number and drawing opposite conclusions—which is exactly what makes the pair worth reading together rather than apart.

Lewontin's argument rests on a partition of variance: the overwhelming majority of human genetic variation lies within conventionally defined populations, not between them. From this he concludes that racial classification is "of virtually no genetic or taxonomic significance." Reich does not dispute the arithmetic. A geneticist in 2018 knows the within-group figure as well as one in 1972 did. Yet he insists it is "no longer possible to ignore average genetic differences."

How can both be true? The hidden move is what a statistician would call the difference between marginal and joint information. Any single randomly chosen genetic marker really is nearly useless for classification—Lewontin is right about that. But traits and ancestry live in the correlations across thousands of markers simultaneously. Small average differences, aggregated across the genome, become reliably detectable even when each locus individually looks like noise. This is the substance of A.W.F. Edwards's famous rejoinder, and Reich is effectively standing on it. Lewontin measured one dimension at a time and concluded the signal was buried; the signal was in the covariance structure he wasn't summing.

So the disagreement is less about biology than about a decision rule. Lewontin fuses a factual claim ("remarkably similar") to a normative one ("positively destructive of social relations") in a single breath—the filing under "Science Against Prejudice" is apt. Reich accepts the same moral worry but refuses to let it license a factual conclusion, which is why his note sits under "Politics Shaping Science."

The pairing exposes the real fault line in the debate: not whether races are "real," but whether a well-intentioned social goal should determine which statistical summary of the data we are permitted to compute. Lewontin answers by choosing the summary that supports the goal. Reich's discomfort is precisely that this is a choice, made by scientists, about what counts as knowledge—and that reality does not consult our preferences about which correlations it contains.

Woven from these notes

The orthodoxy maintains that the average genetic differences among people grouped according to today’s racial terms are so trivial when it comes to any meaningful biological traits that those differences can be ignored. The orthodoxy goes further, holding that we should be anxious about any research into genetic differences among populations. The concern is that such research, no matter how well-intentioned, is located on a slippery slope that leads to the kinds of pseudoscientific arguments about biological difference that were used in the past to try to justify the slave trade, the eugenics movement and the Nazis’ murder of six million Jews. I have deep sympathy for the concern that genetic discoveries could be misused to justify racism. But as a geneticist I also know that it is simply no longer possible to ignore average genetic differences among “races.”
— David Reich How Genetics Is Changing Our Understanding of 'Race'
It is clear that our perception of relatively large differences between human races and subgroups, as compared to the variation within these groups, is indeed a biased perception and that, based on randomly chosen genetic differences, human races and populations are remarkably similar to each other, with the largest part by far of human variation being accounted for by the differences between individuals. Human racial classification is of no social value and is positively destructive of social and human relations. Since such racial classification is now seen to be of virtually no genetic or taxonomic significance either, no justification can be offered for its
— Lewontin