Manufactured Consensus, Real Consequences
2026-08-07 · AI-generated insight
These notes together tell an uncomfortable story about how the postwar consensus on race was actually built — and the Coon anecdote and Duedahl's archival work quietly undermine the clean picture the Lewontin and Diamond quotes present.
Lewontin offers what looks like a straightforward empirical finding: most genetic variation is within groups, therefore racial classification has no taxonomic significance. Diamond builds a whole macrohistory on the premise that biology plays no role. Both read as science speaking truth to prejudice. But Duedahl's account of UNESCO shows the consensus was, in part, engineered: the experts were "carefully-chosen," social scientists were preferred over physical anthropologists precisely because they'd deliver the desired conclusion, and only once in seventeen years was a known critic invited. The Coon incident is the same machinery operating at the level of a professional association — a book denounced by a room in which one person had read it.
The genuinely hard question this raises: what is the epistemic status of a consensus assembled backwards from its moral conclusion? One tempting answer is that it doesn't matter — the conclusion was substantially correct, and Duedahl documents real payoffs, including expert testimony that helped end American segregation. On that view, UNESCO's statements were less science than a successful moral technology, and complaining about the process is pedantry.
But notice what's lost. Lewontin's within-group variation argument was itself later challenged (correlation across many loci can classify populations reliably even when each locus varies mostly within groups — the very point filed under "correlation across many variables"). A consensus built by excluding critics can't easily detect its own weak arguments, because the weak arguments were never stress-tested. The anti-racist conclusion may survive scrutiny; specific pillars supporting it may not — and when a pillar cracks publicly, the whole edifice looks like advocacy, handing ammunition to the people the project was meant to defeat.
Duedahl notes that critique of UNESCO's publications "only added to public awareness" — suppression and denunciation amplified the message. That worked in 1955. It's less clear it works in an era when the sealed rooms have windows.