Dienekes,

The point of Sailer's definition was to take "race" out of the top-down Linneaen mold, and make it scaleable -a bottom-up approach. W/o a replacement method of classification you don't make it clear, for instance, why the Amish shouldn't be considered a race, you just assume it's self-evident why they are not. Zooming out, perhaps we could also consider them members of the German race, and then the Caucasoid race.

That is because these groups are inbred populations that have had enough time to evolve group specific genetic profiles. That is what makes a race.

The Amish don't have any sort of genetic profile? How specific does the profile have to be? Where are the lines drawn?

The point of the definition is to be flexible with a fuzzy concept. Races are just statistical collections of genes. The problem is that no top-down approach has yet been submitted to tell us the exact quotient of genetic clustering that is worthy of being "racial" (or why a top-down approach is even preferable), or if something besides the clustering needs to be considered for the defnition, and for what reason.

. . .or for semantic clarity would you like to reserve the word "race" for the continental sized clusters?

On an interesting note, Steve Pinker used Sailer's def. in "the Blank Slate" (though attributed it to Vincent Sarich)

>> The point of Sailer's definition was to take "race" out of the top-down Linneaen mold, and make it scaleable -a bottom-up approach.

Being top-down, or bottom-up does not necessarily imply scalability. Indeed, biologists have managed to separate based on a number of markers the five continental populations, and also to further split Africans into four populations, etc. Hence, the top down approach seems to work just fine.

>> W/o a replacement method of classification you don't make it clear, for instance, why the Amish shouldn't be considered a race, you just assume it's self-evident why they are not.

The Sailer definition confuses race with isolated population and with ethnic group. Past generations have worked hard to disentangle these concepts.

Race is a population that has been isolated for a sufficiently long time to make it fairly distinctive. We may assign with a high degree of confidence individuals to the Caucasoid or Mongoloid race by simple observation, or by running a genetic test. We can't do the same for groups like Germans and British, even though they are also inbred to some degree.

The distinction between a population inbred in a historical time scale vs. a population inbred in an evolutionary time scale is not unimportant.

>> The Amish don't have any sort of genetic profile?

As I alluded above, the Amish have whatever genetic profile their ancestral group had. Allele frequencies will have changed, as well as a few new mutations would have accrued. Over time, they might evolve into a population that is genetically distinct from their ancestors.

>> The point of the definition is to be flexible with a fuzzy concept.

I realize that this is the intention. But, the definition trivializes race by allowing all sorts of groups to be re-classified as races.

>> or for semantic clarity would you like to reserve the word "race" for the continental sized clusters?

Not necessarily. The continental sized clusters are the obvious candidates whose existence, known from physical anthropology is increasingly validated by genetics.

But, the continental sized clusters can be subdivided further to second-order clusters. Indeed, it has already been established that two groups of pygmies, the San, and Negroids are genetically identifiable subclusters within the African cluster.

">> The point of Sailer's definition was to take "race" out of the top-down Linneaen mold, and make it scaleable -a bottom-up approach.

Being top-down, or bottom-up does not necessarily imply scalability. Indeed, biologists have managed to separate based on a number of markers the five continental populations, and also to further split Africans into four populations, etc. Hence, the top down approach seems to work just fine."

No, I was considering the taxonomical approach to be the top-down one, not the cladistic one you are referring to- which is based quanitatively on shared genes (i.e bottom-up). But that works for shared ethnic and family history as well. Steve wasn't proposing a new way to study race, anyhow (we're all in agreement that the current methods of population gentics is fully adequate for quantifying the genetic distance between people), nor was it an argument over termonology (which is what it seems you consider it) the point of the article was a new way to think about race (to put 'race' in perspective). If you would like to reserve the word 'race' for things that happen over "evolutionary time", that's fine, but it is largely semantic ("evolutionary time" is usually used for much larger time frames than the trivial amount it took for the human races to split, though), relatively speaking ethnic groups and isolated populations are just smaller versions of the exact same phenomenon of divergence+isolation+inbreeding.

I realize that this is the intention. But, the definition trivializes race by allowing all sorts of groups to be re-classified as races.

Whatever you call someone won't change the genetic distance between them and other people, Dienekes, that is why this disagreement is semantic. I agree that different terms are necessary for distinguishing continental races (i.e races), ethnic groups, isolated populations, etc., and I'm pretty sure the point of Steve's definition wasn't that we should drop all distinctions of degree- the point was that we can successfully think about race in terms of shared genes.

Sailer thinks that the study which discovered the genetic distinction between the five continental populations to be a top-down approach. He doesn't seem to approve:

Personally, I'm not that enthusiastic about these top-down attempts to lump humanity into a small number of continental-scale races.

>> the point of the article was a new way to think about race (to put 'race' in perspective).

Well, he does make a big deal about his "fool-proof" definition of race. He also claims that race is "nothing more" than an extended family that is inbred to some degree.

The Sailer definition can lead to many misinterpretations of race. It justifies the notion of "national races". It doesnt' distinguish between inbred recent mixed-race populations and races. A race is indeed a partially inbred population. But not only that.

>> If you would like to reserve the word 'race' for things that happen over "evolutionary time", that's fine, but it is largely semantic

"An extended family that is inbred to the degree that its members can be distinguished with high confidence from the rest of mankind" is a better definition of race. Thus, "Sub-Saharan African" is good candidate for a race. We can classify Sub-Saharan Africans correctly with high confidence (say, more than 95% of the time). "German" is not a good candidate to be considered a race. There does not exist at present (and it is unlikely that there will ever be) a test that can distinguish e.g., Germans from Austrians with confidence.

>> Whatever you call someone won't change the genetic distance between them and other people, Dienekes, that is why this disagreement is semantic.

I guess I want to be more precise. Sailer's definition is too imprecise. It generalizes the concept of race too much. It lets it be used for the chance assortments of peoples in historical time, rather than the temporally deep and geographically clear splits between different branches of mankind.

I've found a site that stated that Negroids on average have an IQ of 88 I'm an African American and I have an IQ of at least 150 and most of the Africans that I know are quite intelligent. I'd like to know if this statistic is true or not