Quote:
Originally Posted by Drolefille
Essentially humans are too young evolutionarily and have mixed too much to make race meaningful as a biological construct. Skin color or eye shape are very tiny variables within a much broader diversity of genes. Would you assume that a black man from South Africa, a black man from Northern Africa, an indigenous Australian and an African-American are necessarily more similar than four people of different races from the same geographic area? Had Europe and Africa branched off many millions of years before they did, and then stayed separated due to continental shift or some other reason, we might have two different human subspecies today.
It's quite possible I may have explained it wrong but I'll try to do this anyhow. Essentially 85% of genetic variation occurs within a population, whether that is Japanese people, British, whatever. This number has been very consistent over the years. About 6-9 percent is between different groups within the same race. Japanese and Chinese, British and French. The rest is between populations.
And for the record, I'm not saying race isn't a real social construct. But it's one that can be traced to our desire to classify people like we did with animals, atoms, plants, etc. during the scientific revolution. Race isn't completely useless as a way to distinguish people, it's just not genetically accurate.
I pulled these sources out of Wiki articles on race because the articles themselves are huge and provide more than anyone here probably wants to read. However they do contain some of the actual data to back up the other articles.
http://en.wikipedia.org/wiki/Race
See particularly the footnotes for these and other articles. The most interesting were pdfs but I can't link them because I'm on a Mac at the moment and I can't figure out how to capture the link.
http://raceandgenomics.ssrc.org/Lewontin/ Lewontin is big in this area.
http://raceandgenomics.ssrc.org/Goodman/ |
I agree with all of this, but this does not really support your counterintuitive claim that "random white/black comparisons will be more similar than random white/white comparisons" - since the variation is within populations, and it is more likely the person (regardless of race) will be outside of your population, there seems to be no reason why there would be any difference in variation. Scanning the Lewontin article yields no support for your claim, and seems instead to back up my intuition.
Do you have a specific citation that says different racial groups are more likely similar than within a racial group? Or was that misstated?
I understand completely the social construct model of race - I don't understand the specific "fact" you quoted.