[lexcorvus]

This is exactly right. Unfortunately, the fallacy—often called Lewontin's Fallacy [1], after the Harvard biologist who most famously committed it—seems nigh ineradicable. Perhaps there should be some sort of Godwin's Law for it: "As an online discussion of human biological differences grows longer, the probability of someone committing Lewontin's Fallacy approaches 1."

[1]: https://en.wikipedia.org/wiki/Human_Genetic_Diversity:_Lewon...

[sweeneyrod]

But you've got to be careful not to make the mistake of incorrectly accusing someone of committing Lewontin's fallacy. Saying that race isn't real is clearly wrong, in the sense that people reliably fit into racial categories.

On the other hand, the other claim (race has no discernible effect on behaviour) isn't fallacious (at least if Lewontin's results are accurate). If there is much greater variance within populations than between them, then it is foolish to make decisions about people based on their race, as it gives very little information about them.

[lexcorvus]

If there is much greater variance within populations than between them, then it is foolish to make decisions about people based on their race, as it gives very little information about them.

All you say is true, but it won't necessarily save you from the Thought Police. For example, the variation in strength within male and female populations is bigger than the variation between them. It may therefore be foolish to make decisions based on gender rather than on strength when hiring, say, dockworkers. But when people complain about how few female dockworkers there are, what will you tell them? Mutatis mutandis.