Human Races

quote:

---

I've never understood why so many people think that 'races' in the human species should be qualitatively different from 'subspecies' in other animals.

---

I don't have any real opinion on whether human population structure is unusual or not. I'm just of the opinion that use of the term "race" for anything biological about humans at this point produces more confusion and less communication than any possible gain it brings.

quote:

---

See my not-very-well-written paper, "The Race FAQ" at Just a moment...

---

Interesting material, especially that comparing human variation with other species. Much of that is new to me, since I don't know much about non-human genetics (having somehow neglected to take a course in genetics at any point in my schooling). One thing that make me scratch my head, though: it's becoming clear from autosomal SNP data (vast quantities of autosomal SNP data) that chimps really do have at least twice the diversity of humans -- but this rather significant difference doesn't show up at all in the tables. No doubt (as usual) more data will clarify the situation.I think there are a few distinctions that you slide over too quickly, however, for a FAQ that's addressing the question of race in humans. You treat folk definitions, anthropological definitions and genetic definitions of human races as if they were interchangeable. They're not. In the U.S., folk definitions of races are such that any detectable amount of African ancestry makes a person "black". Since such a person may be genetically and morphologically more similar to the average European, this means that the folk category does not align well with any biological category. This may not matter much if you're interested only in whether humans can be described as having subspecies, but it matters a heck of a lot if you actually want to use race as a category in the real world.Similarly, anthropological definitions are not the same as genetic ones. "Sub-saharan African" is not one of the traditional, widely used racial categories; "Negroid" is. The latter category includes Africans, Filipinos, Andaman Islanders, and Melanesians, groups that don't form any kind of meaningful genetic cluster. Using racial terminology again serves only to confuse things, for no obvious purpose.There's a more subtle problem in the use of "clade" (and not just by you). Despite Nei and Roychudhury's language, you can't really form a phylogenetic tree for human populations, or for any interbreeding diploid populations; or rather, you can't form a unique, non-reticulating tree. Genes have phylogenetic trees; populations form webs. For many purposes a phylogenetic tree is a reasonable approximation, but when one uses it it's easy to forget that it is only an approximation, and that the population you're talking about may be an average that doesn't correspond to any actual individuals. I point this out mostly because talking about Africans as a clade bugs me. It seems very likely that the Out of Africa migration(s) came from a source population in East Africa, much of which remained in Africa; their descendents are still there. If you're thinking in terms of phylogenetic trees, that branch together with the rest of the African population do not form a clade; rather, the East Africans form a clade with the rest of the world's populations. When you cluster people by gene frequency, however, you will find that the East Africans' descendents within Africa do form a "clade" (a cluster, really) with other Africans, at least at most loci, thanks to continuing gene flow within Africa.

quote:

---

But eye colour also is a problem for me.

​

From people I know with mixed-raced parents, they all seem to have
the dark eyes and tight curls of that side of the family.
If that is the case all of the time it would suggest that
the genetic controlling this feature are dominant (blue eyes
are a recessive feature I believe).

​

Why have recessive features become dominant in some
populations?

---

In this case, probably a secondary result of selection for lighter skin.

quote:

---

No, you demonstrate that neither the Bantu-speeaking not Khosian
speaking villages are representative of the races in question.

​

The history of this relationship, in terms of inter-tribal
marriages, changes the situation.

​

The genetic distance between one Khosian speaking village
and another (assuming one is relatively isolated) provides
a measure of 'out breeding'.

---

So if the Khoisan "race" consists of twenty villages, each situated near a different Bantu-speaking village, which village is representative of the race? It's quite possible for each of the 20 villages to be closer genetically to its immediate neighbor than to any of the other Khoisan-speaking villages, and for the 20 groups still to share a common cultural identity and distinctive genetic and phenotypic traits (all of which they would share to some extent with other Khoisan populations). By your definition they're a race, but by your argument here, it's a race without any representatives. And while this is an extreme case, it is an extreme case of the normal situation in human genetics.

quote:

---

You cannot make ANY arbitrary geographical demarcation and find
a genetic correlation. To take an example from sfs's reply to the
same post that you are replying to) look at a city.

​

Say Dublin, or London, or Paris.

​

There is NO genetic marker available for such a population
because there is a very wide range of lineages that originate
in different geographioc locations.

---

I chose my analogy fairly carefully. Cities vary in their composition. Some, like London, are highly cosmopolitan, with genetic input from many sources. Others, like, I don't know, Lagos maybe, or Osaka, draw their population primarily from one region; such cities will share all of the genetic characteristics of their region, and will also have their own (quite faint) distinctive genetic signature because of inbreeding within the city boundaries. Both kinds of city are analogous to culturally defined ethnic groups, some of which have highly heterogenous composition and some of which don't.

quote:

---

Take a city population and study their genetic diversity and
it will be all over the map. Sub-divide the same group based
upon assumed racial origin and you'll start to see correlations.

​

If this were not the case genetic distance would not correlate
to geographic distance.

---

Look again at what you just wrote. Genetic distance correlates with geographic distance. Cities are compact geographically. What conclusion can you therefore draw about genetic distances within a city?

quote:

---

To be honest it's not an aspect of the debate that interests me.

---

Then you are not someone who should be writing a FAQ on human races. Write a FAQ on human diversity or on human subspecies, but if you're not interested in the issues involved in the use of the term "race", just leave the subject alone.

quote:

---

Supposedly there's all this confusion over what race means to the man on the street or the college freshman or whomever. It's not something I've encountered but I'm not in a position to know if it's out there or not.

---

It's not that there's so much confusion about what race means to the man on the street, or to an anthropologist, or to the geneticist. It's that they each means something different. You, unfortunately, contribute to the confusion by writing as if the term carried only a single meaning. It doesn't.

quote:

---

I mean, do you actually know somebody who would consider a European-looking person black because he or she had 1% African ancestry? Maybe you do, but just because a term or concept can't be idiot-proofed doesn't mean it has to be discarded.

---

No, but I know many, many people who consider a European-looking person black if he has any stereotypical African features, especially if he is known to have some African ancestry. All the people who considers Colin Powell an African-American (including himself), for example, or all the people who call Tiger Woods a "black athlete". I am also aware that someone else who has an identical amount of African ancestry but who was from Latin America would be quite unlikely to be considered black. That's how race actually works in the U.S.