Here is a quote from C. Loring Brace in the Journal of Forensic Science “Skeletal analysis provides no direct assessment of skin color, but it does allow an accurate estimate of original geographical origins. African, eastern Asian, and European ancestry can be specified with a high degree of accuracy”
Dr. Risch published a paper in the journal Genome Biology “on the basis of existing surveys that testing a person’s DNA at 100 random sites along the genome, or at 30 specially chosen ones, would be sufficient to distinguish the major racial groups”
So you can tell a person’s race by looking at their skeleton or looking at their DNA. That means that race has a biological component.
An oversimplification, of course.
For example, by DNA analysis, that just means you can differentiate say an East Asian from somebody from the Indian subcontinent or somebody of Yoruba descent.
It doesn’t mean that everybody from Asia has the same skeletal structure or DNA. Or that everybody from Africa has the same DNA or skeletal structure.
That’s rather the point.
You can’t divide people into 3-5 simple races using these criteria. You’d have to group people who are unrelated genetically into a common race, even if they share a closer genetic legacy with somebody from a different putative “race”.
I figured out a better example may make the point more clear.
Genetically, we can differentiate lions, zebras, tigers, llamas, gorillas, orangutans, capybara, etc.
Clearly, we can use genetics to determine which of these animals are from Africa, Asia, and America. It’s not really even that hard for a layperson.
The concept of a single “black” race is like grouping lions, gorillas, and zebras into a single animal “family” because they’re all from Africa, even if lions are more closely related to tigers than to gorillas or zebras.
It makes more sense to group the African lion with the Asian tiger. Or the African gorilla with the Asian orangutan.
Sure, we can use DNA to identify which animals came from Africa, Asia, and America, just as we can identify which human populations come from Africa, East Asia, and the rest of Eurasia.
But that doesn’t mean “race” is anything more than an artificial grouping we made based on geography, rather than actual genetic commonalities.
So we should group people into genetic commonalities instead of geography, but if we did that then race would be a biological concept?
The good news is that someone already did that and they found "In one of the most extensive of these studies to date, considering 1,056 individuals from 52 human populations, with each individual genotyped for 377 autosomal microsatellite markers, we found that individuals could be partitioned into six main genetic clusters, five of which corresponded to Africa, Europe and the part of Asia south and west of the Himalayas, East Asia, Oceania, and the Americas "
Serious question. So what are the preferred ways of grouping populations?
Can you link to this source? According to mitochondrial DNA analysis here, many African populations are actually closer, genetically, to Asian and European populations than to many other African populations. For example, the Bamileke of Cameroon have a more recent common ancestor with (and are therefore more closely related to) Europeans then they do with the Kikuyu of Kenya.
Here is the link:
Your link goes to a dna analysis of a hominid fossil, is that what you are trying to link to?
I think it has to be both. Otherwise one could move between races as one chose to. You know, like in high school you could be a jock, a hippy, a nerd, whatever and then later you could become a middle class business person. Those are social constructs based on changing values systems and behaviors.
I lived in a Black community for a while but I can guarantee that no matter how I adjusted my behavior or values I would have never been considered Black.
My link contains a phyogenetic tree that shows the most recent common ancestor of a varied selection of mitochondrial DNA from many different populations around the world. The figure shows that many African populations (like the Bamileke) that are actually closer, genetically, to European and Asian populations than they are to many other African populations (in my example, the Kikuyu).
Thus, “black” is not a distinct genetic grouping, because some “black” populations are closer to non-“black” populations than they are to some other “black” populations.
What racial group are the people living in Mexico? How about in Ethiopia? Yemen? Tajikistan?
OK, let’s consider height. If height means anything to race, then surely we should be able to say which race is tallest? Well, that’s easy. The Bantu are the tallest population of people in the world, so clearly the black race is the tallest race. Likewise, we surely should be able to say which race is shortest, right? Again, that’s easy. The Mbuti are the shortest population of people in the world, so clearly the black race is the shortest race.
Wait, what’s that? The shortest population and the tallest population are both the same race? But how can that be, if height is a racial trait?
No, it means that being Sweden has no basis in biology.
Unless you are seriously arguing that the borders of Sweden are themselves determined by some sort of biological factor, then this is a textbook fallacy of composition.
The average resident of Sweden has a greater height than the average resident of Vietnam.
Height is based upon biology.
Therefore Sweden and Vietnam are based upon biology.
The average resident of Vietnam is more religious than the average resident of Vietnam.
Religion is based upon imagination.
Therefore Sweden and Vietnam are based upon imagination.
Exactly the same argument, and both equally erroneous. You can not ascribe the properties of the parts as a property of the whole. That is a fallacy of composition.
No, you don’t. The first part of what you aid is correct. Human population are clinal, and geographic barriers can produce zones of greater discontinuity. But there are no barriers that correspond to races. So there is no “thus”. That is an total non sequitur.
No, this is another fallacy of composition.
If we can’t group people into genetic commonalities instead of geography then race can’t be a biological concept.
That does not allow you to conclude that if we can group people into genetic commonalities instead of geography then race will be a biological concept.
In case you still don’t spot your logical error:
If Jim doesn’t have any money then Jim can’t buy my house.
Agreed?
That doesn’t allow you to conclude that because Jim does have some money Jim can buy my house.
Does it?
Jim might not have *enough *money to buy my house? Or he might be a foreign citizen, and thus disqualified by law from buying my house and so on and so forth.
What you have done is focussed upon a single flaw that invalidates your classification scheme, and conclude that if *that *flaw was removed the scheme must b valid.
Nuh uh. you can’t do that.
To be even given serious consideration as biologically valid a classification scheme needs to meet two basic criteria:
It needs to be objectively verifiable based upon biological characters. A scheme that can group people into genetic commonalities would meet such a standard.
But it also needs to be revealing. What that means is that placing an organism into a category it needs to tell you something other than the very concept that you used to define that category.
This is where all the racial schemes I have ever seen fall down. It is certainly possible to group people together based upon skin colour, but such a scheme won’t tell you a damn thing about any other trait that those people share.
The article that you reference here to has the same problem. By using selecting 993 genetic markers, out of literally billions of potential markers, they managed to construct a grouping that corresponded, not to race, but to geographic boundaries.
That’s great, but what does such a scheme tell us? If I tell you that a person falls into the Oceania geographic grouping, what can you tell me about that person? Can you tell me, with >95% certainty, anything about his skin colour? About his height? About his genetics?
No, you can not. Having assigned someone to one of these geographic races, you can do absolutely nothing further with that information. The classification scheme is perfectly self-referential. And it is thus logically invalid and, being illogical, it is biologically invalid.
Look, nobody ever doubted that it was possible to assign people to races based upon a fine enough analysis of DNA. That was obvious as soon as DNA was determined to be the mechanism of inheritance. At the finest scale, we can classify people into tiny clades based upon who their grandparents were. I can simply look at someone’s family tree, see the name of their grandparents, and assign them to grandparent group 1. then I look at the next person and assign them to Grandparent group 2, and so forth. Cousins will of course all fall into the same Grandparent group.
So if I categorise everybody based upon their grandparents, I can then arrange those grandparent groupings so that they correspond with geography, can’t I? I can say that 95% of the people in Europe have grandparents from Groups 1, 697, 9987672, 11 and so forth. By doing this i would obtain a 95% correlation between grandparent group and geography. Do you agree?
Since who your Grandparents were obviously has a huge effect on your genetics, this is a surrogate for genetics. I have effectively produced a genetic scheme with about 2 billion markers, rather than the 1, 000 used in the Rosenberg study. But again, nobody disputed that this was not only possible, it was inevitable. By using enough genetic markers, it had to be possible to categorise people into geographic zones. The only question was how many markers it would take. We knew from the grandparent example that it had to b possible with 2 billion markers. Using great, great grandparents we know that it had to be possible with half a billion.
Rosenberg et al have proved it can be done with a s few as 1, 000 markers. That’s interesting, but it does not make the scheme a racial scheme because, as it stands, it is entirely self referential. By assigning people based upon the distribution of those thousand markers, all that you learn about them is where they fall within the distribution of those thousand markers. You don’t even learn anything about the person’s genetics, because people in those groups don’t actually have all 1, 000 markers. The system relies upon combinations of those thousand markers. So any two people within a grouping may actually have 0% genetic commonality. All we know is that the genetic pattern that they do have causes them to cluster together under this system.
To see what I mean, try to answer this question: using another 1, 000 markers, would it be possible to produce an equally accurate distribution scheme that lumped Asians and Europeans into the same race and lumped Africans and Amerindians? And the answer is, yes, it almost certainly would be possible, just as it would be possible to do so based upon grandparents.
So, since the same technique could produce a classification scheme that produced totally different grouping, why should we favour *this *scheme? What is the utility of this scheme over those others? And of course the answer is that there is no reason to favour it. It’s completely arbitrary and self-referential, and thus biologically invalid.
Another fallacy of composition. A scheme having a biological part does not mean that the scheme is biological. You can not ascribe the properties of the parts to the whole.
I have toenails, that does not mean that I *am *a toenail.
And that would be a straw man. What is your definition of race? Does it apply to other species? As noted above by Jerry Coyne, if you use a standard definition from evolutionary biology then it seems there are human races (it would be a bit odd if there weren’t really).
You get mixed groups, or “zones of intergradation” in sub-species and races. After all, they’re not separate species. They interbreed. Risch et al comment:
Sigh. Same old nonsense from Chen.
“Clines, not clades” - it’s really that simple.
They’re not even subspecies- all humans are far more closely related than any two Chimp subspecies.
You’re saying that the Bamileke are a mixed group? And the Mandenka? That phylogenetic tree shows Most Recent Common Ancestors- which shows that the ancestors of Kikuyu and Bamileke separated before the ancestors of the Bamileke and Eurasians separated. Not that they’re mixed, but that the Bamileke share a more recent history with Eurasians then they do with Kikuyu of Kenya.
Note that with the “out of Africa” theory this is not surprising- populations separated many times in African pre-history, and it makes perfect sense that the group that eventually left Africa shares a more recent history (and more recent common ancestor) with some African populations than either do with others (for example, population X splits into Xa and Xb, Xb splits into Xba and Xbb, and Xbb leaves Africa- so Xba is closer to Eurasians then it is to Xa).
Of course, that is simply is one aspect of the semantic games that the pro-race people play. They can point, (legitimately), to hundreds of human populations and can use the word “race” to label them. Then when the the word “race” has been accepted in that context, they go back to talking about the Three Races that cannot be legitimately so described. Classic bait and switch.
(NDD does this, regularly and you have been known to get very close to the same nonsense. He spouts off about the Three Races all the time, then, when challenged on the “superiority” of “Mongoloid” intelligence, he falls back on a claim of “Oriental” intelligence. Even ignoring the obvious fact that his claims for “Oriental” intelligence are not supported by real world evidence, he is changing the definition of “race” from one post to the next.)
-
Is there a chimp benchmark used for determining races and subspecies across species? Consider this example of fst values for various other subspecies. Is it much different to that between human races? There appear a number of examples you can look at of animal subspecies where the level of genetic differentiation is similar to that of humans.
-
That may well be the case for the Bamileke. Does that mean there aren’t human races as defined above by Coyne?
I don’t know- what races does Coyne claim? Does he use the common “three races”, or something else?