Mounting Morals on Fragile Facts: How Lewontin's Fallacy Became Load-Bearing
2026-08-15 · AI-generated insight
Read together, these notes trace the full lifecycle of a scientific orthodoxy — its founding argument, its statistical flaw, its enforcement mechanism, and its eventual crisis.
Notice what Lewontin does in his famous passage: he moves in a single breath from an empirical claim (most genetic variation lies within groups) to a sweeping normative conclusion (racial classification has "no social value" and "no justification can be offered for its continuance"). The moral verdict is welded directly onto the statistical finding. That welding is the crucial move, because it means the ethical position now depends on the empirical one.
Winegard and Carl show why that dependence is dangerous: the statistical claim was wrong, or at least badly incomplete. Variation at any single locus is indeed mostly within-group — just as nose size barely distinguishes male from female faces — but correlated variation across many loci makes population structure legible, just as whole faces are sexed at a glance. Reich, writing from inside mainstream genetics, confirms the consequence: it is "simply no longer possible" to maintain the founding claim.
The Coon anecdote reveals what happens in between — how an orthodoxy defends itself once its moral conclusions are mounted on empirical scaffolding. The AAA members who denounced Putnam's book without reading it weren't evaluating evidence; they were protecting the weld. When your ethics are anchored to a factual claim, examining challenges to that claim feels like entertaining evil, so condemnation without examination becomes the rational-seeming move.
Reich's deeper warning emerges from the synthesis: the antiracist orthodoxy chose a hostage-taking strategy, tying human moral equality to genetic uniformity. Every note here is a stage in the ransom coming due. The alternative — which Reich gestures at — is to decouple the weld entirely: ground equal treatment in ethics rather than in a factual claim that genetics might one day contradict. Lewontin thought he was building a fortress against racism; he may have built a single point of failure.