John responds to the "race" response from Matt & I. I'm not interested in making a point-by-point response to the response because I don't think the "objective" difference in opinion is that great, rather, it seems to be that we are clashing in the turbulent waters of nominalism.
First, I will respond to what I believe is the perception by John that I am conceiving of race as an essential and fundamental taxonomical unit. I don't hold to that. I've rejected the Platonic conception of race before. The problem that "race based public policy" often has is that the legal system is deterministic, and probablistic entities like subpopulations don't fit into neat categories. I don't deny that human populations exhibit spatial and temporal gradients. I stated before that my attitude toward population substructure is instrumentalist. I am happy to treat human beings as fundamentally the same, having equal rights before God & Nature. This the basal starting point, but, we often just stop there.
Consider a situation where I, a South Asian American, am in need of a bone marrow transplant. If we simply assume the basal model, that we are all fundamentally the same, the finite tissue matching dollars should be ethnicity insensitive. But the reality is that my chances of a tissue match are far higher within my "ethnicity" than without, so one should target the search appropriately. Similarly, drug testing trials should not be absolutely generalized across ethnicities, and when possible it might be best to optimize the diversity of the subject population. If you are a white American this doesn't matter much since the majority of Americans are white, so not taking ethnicity (genetic substructure) into account is of little importance in terms of your own potential mortality. But, if you are a member of a small minority the "we are all the same" attitude can be problematic and fatal.
This doesn't mean that I think race is a metaphysical concept of sacral value. I could care less, I just don't want my risk of personal extinction elevated because of an excessive emphasis population interchangeability. One can change the name from "race" to something else, I don't care, as long as pragmatically oriented people don't disregard the correlations of alleles which map onto geographical races.
In any case:
To be races, I would think that you'd need at the least to have long-term cladogenesis of the type found in haplotype groups of geographically isolated populations of other species.
Well, then I guess by that definition I don't believe in races. We can use some other word, I don't particularly care, but we are a diverse species characterized by genetic substructure, and the 85/15 intra vs. intergroup variance number on one locus tends to elide that in the public imagination.
The problem for me isn't that people reject the word race, it is that when they reject that word they seem to default to a position which assumes that interpopulational differences are "superficial" and "unimportant." I don't think this is so. A trivial example can be illustrated with food, in Some like it Hot, conservation biologist Gary Nabhan chronicles the confusions that abound in terms of cuisine transported across cultures. Some "diets" really don't work well outside some populations, and giving Native Americans "free milk" is really dumb when you consider this group can't digest milk well. On a less trivial level, the peoples of the northern Andaman Islands were decimated by diseases introduced by immigrants from the Indian mainland (they were moved to an isolated island in the 1960s and their numbers have bounced back from a dozen to perhaps 50 individuals). Today the Indian government has quarantined North Sentinel Island in part because it fears that outside diseases will result in the decimation of this group. The Eurasian pathogenic environment has resulted in the reality that some groups isolated from this evolutionary cauldron simply die when exposed to the disease load that is endemic to most Eurasians!
Wilkins' mistake appears to stem predominately from his understanding of the Tang et al and Rosenberg et al papers. The Contingency Table's review of this mistake is good, but I thought I'd point it out again.
So, given that there are genetic and phenotypic substructures, the only really significant dispute between us is whether the "folk taxonomy" categories match up to any biological reality. I think I'll stick with my claim that there isn't.
The concordance of "folk taxonomy" categories to biological categories inferred from genetic data alone is precisely the question asked by Tang et al., and the conclusion is that they are highly concordant. Rosenberg asks a similar question regarding the "folk taxonomy" that people can be described by their continent of origin, and the conclusion is similarly robust.
This is also an issue when giving drugs for anti-rejection after transplants. In the book "Black and White and Red All Over" Warren Brown discusses the fact that he was treated as an African American, who are more likely to reject transplants than caucasians. He, coming from New Orleans, is much more of a mixture, and this meant he rejected his first kidney.
Eventually, he was able to accept a kidney donated by his female white colleague, Martha McNeil Hamilton. Which is what the book is all about and just goes to show that "race" isn't all face value.
Which is what the book is all about and just goes to show that "race" isn't all face value.
the key is that it isn't "all" or "none." in a place like brazil racial taxonomies are all out of wack because of generations of admixture + assortative mating, so that ancestry and phenotype are decoupled. in the USA you have cases where hypodescent has resulted in predominantly (ancestrally) white individuals identify with their black ancestors and so are defined as black. but the key is that though about 20% of the african american genome is european, there is variance across black american populations. those which are mixed in origin, like the creoles of louisiana, should be treated as such. those who aren't, such as the people of the georgia coastal islands, should be treated as such.
and of course, it is all about probabilities. just because the chances of a within-race match are higher, that doesn't mean that there isn't going to be an out of race match. the utility function has to be sensitive to conditions, if your own ethnic group is thin on the ground then trying to find matches from within that group by spending a lot of $$$ on search and recruitment might not be as useful as just putting all the funds into getting matches in the general population.
The Contingency Table has an excellent overview of the issues (for that matter, you should bookmark the blog as it is)
ah ha, so this is where all those people are coming from :)
thanks for the mention.
To me the semantic arguments about the word 'race' are largely a waste of time. The interesting question is how the observed patterns of geographical variation have actually evolved. At present, the observed variation is approximately continuous, except where there are major geographical barriers or historically recent migration between distant areas. So is the observed continuity the result of clinal evolution in situ, or of the secondary merger of populations previously more sharply distinct?
Perhaps we could apply the term "variety" to animals as well as plants. I agree that we may be arguing over semantics, but this is another issue in science that when exposed to the general population leads to more confusion than explanation. We need to convey the facts of the various genetic differences between populations while enforcing the greater consanguinity of humanity.