If I follow this argument, it essentially says that since “races” aren’t very definable genetically (let’s agree to permit that point for now rather than belabor it), it makes no sense to suspect that different races (now a “purely social construct”) would have differences other than superficial ones.
This is the argument I do not understand, because it seems to ignore the history of human migration and evolution, which is reasonably well-accepted.
Here’s the history of humans I accept:
We started in africa. We developed a boatload of diversity there (all groups L3 and earlier if we are using mtDNA haplogrouping as a reasonably convenient marker), and some population(s) made it out of africa somewhere around 70kya. Since humans began, geographic, migration and cultural influences have led to all kinds of populations all over the place; clinal to some degree but not still a homogenous pool, and particularly not smooth where separation times have been in excess of tens of thousands of years with only minor admixture across those (typically geographic and cultural) boundaries.
OK; so we cannot define a “race.” I mean, it’s a matter of definition. No problem.
But people do sort themselves, and one such sorting is into SIRE groups the US uses; these roughly reflect continent of origin. The history of migration and evolution has affected average gene pools for those groups. Gene frequency measurements within those groups cluster by SIRE groups, and there is a very plausible explanation for that clustering:
Advantageous genes produced by evolution will achieve high penetration within descendant groups from the point where the gene evolved. If that descendant group is separated from a different group by geography and culture, then the new gene will cluster for the descendant group. The descendant group will have a different average gene pool from ancestral groups (themselves evolving their own new gene pools).
We have good examples of this. The SIRE group of “white” has lighter skin than the SIRE group of “black” because the source population from which it derived has–on average–a different gene pool. It is not the case that there is a completely homogenous gene pool out there, out of which the “whites” and “blacks” are arbitrarily categorized. There is a real and well-accepted migration/evolution explanation for it, and it’s at the point of the L3/M-N split; post this split, there is a different average gene pool for “whites” than there is for “blacks.”
So I don’t understand the argument that we need to decide first what defines a group genetically. That’s an argument of lumpers and splitters and linguists. Instead, we approach the question of genetic differences from the opposite side of the coin. IF you self-identify as “black” and IF we take large numbers of self-identified blacks, then the average gene pool (Navin Johnson notwithstanding) of the black pool is different for melanin handling than is the average gene pool for that same characteristic for whites.
And the thing that makes that average gene pool difference is not that we arbitrarily plucked out one outcome from a million possible characteristics that could have been used for taxonomy. The thing that drives that difference is a well-accepted, well-defined, measurable divergence of populations at the L3/M-N out of africa migration split.
So getting rid of “races” does not get rid of this problem: Why would nature somehow have manipulated only genes for “superficial” characteristics after anatomically modern humans appeared 200 kya? Why would not every population develop new genes driven by evolution, and why would not there be average differences in gene pools for descendant populations at every splitting point? Why would not a “significant” outcome difference in populations that represent ancient splitting point divisions be reasonably ascribed to genes just the way we ascribe physical appearance differences to the same evolutionary changes affecting average genes in the same descendant populations?
I believe that is a serious scientific mistake to try and get rid of the “genes” argument for outcome differences by simply claiming that “there is no reasonable biological definition of race.”