On the Ethnic Kinship Coefficient

by Dr Swaggins

Here’s an excerpt from my upcoming rebuttal of this oddity of a piece, a collaboration with RaceRealist at the NotPoliticallyCorrect blog. In these few paragraphs I sum up some data and math relating to the finding that you’re about as similar to someone of your own race as you are to a first cousin.

Rebutting Jayman’s denial of the ethnic kinship coefficient requires an explanation of the concept of relatedness as a whole. How, for example, can I be 50% identical to my father if I’m around 65% identical to chickens? The answer is that I am not 50% identical to my father; rather, I am 50% identical to my father by comparison to the baseline level of relatedness of all living humans. If all living humans are 99.8% genetically identical then I’m 99.9% identical to my father. Jayman’s argument that two random coethnics aren’t related fails to factor this into account: a calculation of relation needs a baseline level of relatedness for comparison. So he’s correct in stating that two coethnics are not similar to one another- but only by comparison to the baseline level of relatedness of their entire population.

Since the ethnic kinship coefficient has been worked out to the equivalent of first cousins, it may be useful to frame the issue in those terms. If I am 12.5% identical to my first cousin by comparison to any other coethnic, it is because there is an eighth of my genome that I share with my cousin due to our common descent. Specifically, our mutual descent from our mutual grandparents gives us a specific combination of genes that nobody else is likely to have. 12.5% of my genome is 100% identical to his alleles of the same genes and the other 87.5% is as similar to his as it is to any other coethnic, but taken as an average across my entire genome, any given allele is 12.5% more likely to be shared with him than it is everyone else in our race.

The ethnic kinship coefficient works in an uncannily similar way. Instead of inheriting those 12.5% identical genes from recent common ancestors, the two coethnics inherit the same genes due to the fact that people of their race usually have those genes (think melanin, keratin, microcephalin, EDAR, HERC2, or any other gene for which the frequency of alleles differs over population). In spite of that difference in the origin of ethnic vs familial similarity, the mathematics are shockingly similar: according to Henry Harpending in his review paper Kinship and Population Subdivision, “Many studies agree that Fst [genetic distance between populations] in world samples of human populations is between ten and fifteen percent,” with “a conservative general figure” being 12.5%. What’s more, Fst “is computed for each allele at each locus, then averaged over all loci.” A statistically average gene is 12.5% more likely to be shared with a first cousin than with a coethnic and 12.5% more likely to be shared with a coethnic than with anyone else.

One such study finds, for example, that a Frenchman is 12% identical to another Frenchman if your baseline for comparison is the genetic similarity between the French and Japanese.

This is the inevitable implication of the central tenet of HBD: that the various races of the world are genetically different from one another. It is also the inevitable implication of Lewontin’s famous finding that 15% of all human genetic variation is racial; if it were 100% then all coethnics would be identical and it were 0% then race wouldn’t exist at all. If it were 15%, though, then that 15% would be composed of genes whose alleles vary in frequency across populations; these are genes you share with coethnics much more often than you share with anyone else. If you’re more likely to share a lot of genes with coethnics than you are with anyone else, then you’re more genetically similar to coethnics than you are anyone else. When they sequence the genes of people of different races and compute the odds of similarity locus for locus, you’re much more likely to share some genes (ABCC11, MC1R, etc) with coethnics than you are others, but taken as an average across the entire genome, it’s 10-15%. Apply those odds to the 20,000 or so genes in the human genome and the result will be consistent with the data that members of a given race are 10-15% identical by comparison to members of other races.

Here’s a thought: since you’re actually 56.25% identical to your kids, assuming they’re of the same race, I should go back and revise my numbers from Fecundity is Immortality.

Edit: Upon studying the Fixation Index more, I’ve been reminded of the fact that Fst and similar values are calculated via odds of heterozygosity, which reaches its maximum at 0.5 in a population assuming that there are two alleles in play. Moreover, assuming that conditions for most genes fall within Hardy-Weinberg expectations, we find 50% of a population’s variation within each of its individuals due to the diploid nature of the human genome. This means that Fst values, like kinship coefficients in general, will be half of the actual relatedness (or lack thereof) between individuals. I may make a post about this later; in the meantime I leave you with the appendix of this Frank Salter paper, written by Henry Harpending himself.

On the Ethnic Kinship Coefficient

One thought on “On the Ethnic Kinship Coefficient

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s