I found this characterization of Jared Diamond’s ‘Lactase-positive Race’ position compelling:
Jared Diamond is perhaps the best-known example (Diamond 1994). He maintains that it possible to base a genuine racial division on the basis of invisible features such as possession (or absence) of anti-malaria genes or the enzyme lactase alone. He speaks of Swedes as belonging with Fulani in the “lactase-positive race,” and most African “blacks,” Japanese, and American Indians as belonging together in the “lactase-negative race.” Diamond’s rhetorical genius lies in the subtlety with which he teaches the reader a new, non-standard way of projecting the word ‘race’. He introduces the novel language game of applying the word ‘race’ to divisions based on any arbitrarily. chosen characteristic, without alerting the reader to this fact. And he surreptitiously invites readers to slide from the sound idea that the (application of the) concept of race is arbitrary (in many contexts) to the deeply erroneous idea that the concept itself is completely arbitrary in the sense that the features that go into the determination of the racehood of groups is completely arbitrary.
He obviously doesn’t take it too seriously. Consider that the year before, he was telling us (Natural history Nov. 1993):
There are also practical reasons for interest in Jewish genes. The state of Israel has been going to much expense to support immigration and job retraining of Jews who were persecuted minorities in other countries. That immediately poses the problem of defining who is a Jew.
And later (1994) was conjecturing that:
A second hypothesis is selection in Jews for the intelligence putatively required to survive recurrent persecution, and also to make a living by commerce, because Jews were barred from the agricultural jobs available to the non-Jewish population.