Justin’s Notes

The Statistical Version of Not Reading the Book

2026-07-23 · AI-generated insight

Read together, these notes stage a single drama about how a conclusion can outrun its examination.

The Coon anecdote gives the crude version: a room of scientists votes to denounce a book almost none of them have read. Condemnation without examination. What makes it damning is that it happens under the banner of scientific authority—the very authority the public trusts precisely because it assumes evidence was weighed.

Winegard and Carl expose a subtler, more respectable version of the same failure. Lewontin did examine the evidence—locus by locus—and found within-group variation swamping between-group variation. But by looking at each characteristic in isolation, he committed the analytic equivalent of judging a face by nose size alone. The information lives in the correlations across many variables, exactly as we experience real faces and real sexes. Lewontin's error wasn't laziness; it was choosing a frame that guaranteed the answer before the data arrived.

And notice how Lewontin's own note behaves. It slides, in two sentences, from an empirical claim ("remarkably similar... based on randomly chosen genetic differences") to a normative verdict ("no justification can be offered for its continuance"). The word "randomly" is doing enormous quiet work—it is the assumption that pre-decides the case, the statistical equivalent of never opening the book.

Gould's "skin deep" fits the pattern too, but with the polarity reversed. He warns that visible traits fool us into overestimating difference—our perception is biased upward. Winegard and Carl warn that single-locus analysis fools us into underestimating it. Both are claims that naive perception misleads; they simply disagree about which direction.

What unites all four is a lesson about method rather than race: the frame you adopt can manufacture your conclusion, and the more that conclusion flatters your prior moral commitment, the less you feel any need to examine it. The one honest man in Coon's meeting raised his hand not because he agreed with Putnam, but because he had actually looked.

Woven from these notes

obvious: As long as most scientists accepted the ancient division of races, they expected important genetic differences. But the recent origin of races… squares well with the minor genetic differences now measured. Human groups do vary strikingly in a few highly visible characters (skin color, hair form)—and this may fool us into thinking that overall differences must be great. But we now know that our usual metaphor of superficiality—skin deep—is literally
— Gould

In brief, Lewontin ignored the fact that differences among human populations are correlated. Although population variation at any one genetic locus tends to be small, global population structure becomes clear if one examines correlated differences across loci. For a simple analogy, consider men and women’s faces. If one takes any particular characteristic, say nose size, there is likely to be more variation within a sex than between the sexes (and it would be nearly impossible to classify faces by sex with any accuracy using only nose size). However, if one considers all facial characteristics together (which is, after all, how we actually experience human faces), then sex differences become sufficiently clear that an observer can guess the correct sex more than 95 percent of the time.

— Bo Winegard, Noah Carl Superior: The Return of Race Science—A Review
Carleton Coon was one of the greatest anthropologists of the 20th century and a champion of the study of racial differences. In 1961 he was elected president of the American Anthropological Association. It was during this year that Carleton Putnam's infamous book Race and Reason came out. Putnam's book argued that there were significant differences between the races and that this had important political implications. A few anthropologists became upset when they saw the book gain popularity and wanted the AAA to intervene. But they knew that Coon supported the book and that, as a result, having the AAA take action against it would be difficult. So they organized a secret meeting behind Coon's back in order to issue a statement denouncing Putnam's work. Coon found out about the meeting and went to stop it. Upon entering the meeting Coon asked all the association members who had actually read Putnam's book to raise their hands. Only one rose. He then asked for the hands of everyone that had even heard of the book prior to that meeting. Only a few hands rose. None the less, the statement against the book passed. Coon, disgusted by the actions of his peers, resigned from the AAA (1). This incident stands out in the history of science as a particularly clear example of scientists not living up to what people expect of them. People often take what scientists say for granted. They trust that scientists have looked at the evidence rationally and are giving the public as accurate an account of all the relevant facts as they can. When explaining science to laypeople researchers are expected to only make authoritative statements on topics they are knowledgeable of, to not lie about scientific evidence, and to not omit obviously relevant facts. People trust scientists to do these things and so feel comfortable taking what they tell them for granted. (See
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 continuance.
— Lewontin