[Perkwunos]

23andMe's racial classification is based on haplogroups

Not solely, to my knowledge. Cf. https://isogg.org/wiki/23andMe

[Pants-of-dog]

Then why didn't you ask me earlier?

I did.

I'm arguing that the idea of race as social construct is barely tenable in light of clear Ebuhdins™ that it is possible to discover / classify racial categories from genetic data. It's only "socially constructed" in the sense that any such cluster analysis is, but that doesn't mean what "race is a social construct" people want it to.

Well, the actual claim is that races are not biologically distinct, and that the diversity within any "race" is larger than the diversity between races. Thus, the races are weak proxies for genetic diversity.

But races still exist insofar as we can walk down the street and see white people and black people and indigenous Americans, etc., and thus we have racism, but most of that is social behaviour. This is why we call it a social construct.

Now, it is possible for there to be more genetic diversity within races than between races, and still build these clusters from the cited study. All you would have to do is focus on the small portion of genetic diversity that occurs between races.

[Perkwunos]

Well, the actual claim is that races are not biologically distinct

But they are.

But races still exist insofar as we can walk down the street and see white people and black people and indigenous Americans, etc., and thus we have racism, but most of that is social behaviour. This is why we call it a social construct.

But xenobiologists from another world would also be able to recognize these differences. And they'd probably be more willing to entertain the notion that there are aggregate differences in behavioral traits along racial lines because presumably a race of beings rational enough to master interstellar travel wouldn't accept "THASS WRONG, ACUZ THASS RAYCISSISS" as a reason to believe or deny something. (And they wouldn't have a dog in the race.)

Now, it is possible for there to be more genetic diversity within races than between races, and still build these clusters from the cited study. All you would have to do is focus on the small portion of genetic diversity that occurs between races.

Man, it would be really great if I didn't have to keep seeing Lewontin's fallacy.

[Pants-of-dog]

But they are.

I don't think so.

But xenobiologists from another world would also be able to recognize these differences. And they'd probably be more willing to entertain the notion that there are aggregate differences in behavioral traits along racial lines because presumably a race of beings rational enough to master interstellar travel wouldn't accept "THASS WRONG, ACUZ THASS RAYCISSISS" as a reason to believe or deny something. (And they wouldn't have a dog in the race.)

Yes, the differences that we base race on are simplistic surface traits that even a socially inept person can pick up on. That doesn't make race a good proxy for genetic diversity, nor does it contradict the fact that vast majority of the phenomena associated with race are social phenomena.

Man, it would be really great if I didn't have to keep seeing Lewontin's fallacy.

Perhaps you could explain exactly what the fallacy is and why it is a fallacy.

[Perkwunos]

I don't think so.

You've already been given Ebuhdins™.

Yes, the differences that we base race on are simplistic surface traits that even a socially inept person can pick up on. That doesn't make race a good proxy for genetic diversity, nor does it contradict the fact that vast majority of the phenomena associated with race are social phenomena.

Not if, say, there are aggregate differences in intelligence.

Perhaps you could explain exactly what the fallacy is and why it is a fallacy.

It focuses on loci separately rather than together. Most genetic information about race is correlational. And that information is very substantial, as demonstrated by the accuracy with which it is possible to cluster racial groups.

[Pants-of-dog]

You've already been given Ebuhdins™.

The evidence shows that there is some genetic diversity between races, but that there is more genetic diversity within races. Also, these clusters can still exist and there could still be no discrete races.

Not if, say, there are aggregate differences in intelligence.

Please write in compete sentences so that I know to which part of my post you are replying. Thank you.

It focuses on loci separately rather than together. Most genetic information about race is correlational. And that information is very substantial, as demonstrated by the accuracy with which it is possible to cluster racial groups.

What does? The fallacy focuses on loci separately?

[anasawad]

This is relevant.

Their sources are in the description.

[Perkwunos]

Also, these clusters can still exist and there could still be no discrete races.

I can imagine how fuzzy rather than crisp clustering could be used for the race issue but I'm not sure it would be generally useful. It might make sense, for example, with clustering Russians into European and Asian clusters.

Please write in compete sentences so that I know to which part of my post you are replying. Thank you.

What I'm saying is that at least some of the observed racial differences in intelligence testing might be due to heredity.

What does? The fallacy focuses on loci separately?

Lewontin's reasoning only focuses on loci separately, when it shouldn't. That's why it was branded "Lewontin's fallacy".

[mikema63]

What I'm saying is that at least some of the observed racial differences in intelligence testing might be due to heredity.

It might also have something to do with any of a dozen random correlations you could probably find with IQ. Levels of chocolate consumption and homosexuality are things that have actually been linked to IQ. This does not mean that these things actually have anything to do with it. IQ tests are on shaky ground outside their narrow application in the school system which is what they were developed for. It is not a tool invented by or for biological study.

From the point of view of the field of biology you wouldn't use a psychologists test for intelligence you try to understand intelligence biologically before you explain it biologically. Using an IQ test to explain biology is like trying to read palms to find exoplanets.

Lewontin's reasoning only focuses on loci separately, when it shouldn't. That's why it was branded "Lewontin's fallacy".

Actually if you wanted to measure the meaningfulness of the variation of gene loci you would use a proportion of variation approach. This is essentially a comparison of the overall genetic variation within a group compared to the variation between groups. The guy who coined the fallacy your citing was talking about using correlations certain loci with identifying the geographic region they come from. Using more known corelates you can be more certain. This does not however mean that these correlations are significant and this critique does not actually overthrow the conclusions of lewtonin's paper.

[Pants-of-dog]

I can imagine how fuzzy rather than crisp clustering could be used for the race issue but I'm not sure it would be generally useful. It might make sense, for example, with clustering Russians into European and Asian clusters.

Well, the concept of race as an jndicator of genetic diversity is also not useful.

What I'm saying is that at least some of the observed racial differences in intelligence testing might be due to heredity.

Yes, they might be. But many things might be.

Lewontin's reasoning only focuses on loci separately, when it shouldn't. That's why it was branded "Lewontin's fallacy".

How does this disprove the notion that most diversity occurs within racial groups?

[Suntzu]

Nobody believes in the concept of race . . . well, except for the government.

[Truth To Power]

It might also have something to do with any of a dozen random correlations you could probably find with IQ.

"Random" correlations? If such correlations are statistically significant and replicable, they are unlikely to be random.

Levels of chocolate consumption and homosexuality are things that have actually been linked to IQ. This does not mean that these things actually have anything to do with it.

But they almost certainly do, as both homosexuality and chocolate consumption are related to brain chemistry.

IQ tests are on shaky ground outside their narrow application in the school system which is what they were developed for. It is not a tool invented by or for biological study.

But they are the best measure we have of a quality that is biologically rooted and scientifically interesting.

From the point of view of the field of biology you wouldn't use a psychologists test for intelligence you try to understand intelligence biologically before you explain it biologically.

No one has said IQ tests enable understanding of biology, any more than a tape measure enables understanding of what makes people tall or short. They are a measuring tool. They measure something that is rooted in biology, but the explanation has to come from elsewhere.

Using an IQ test to explain biology is like trying to read palms to find exoplanets.

Does a tape measure explain height, or just measure it?

[mikema63]

"Random" correlations? If such correlations are statistically significant and replicable, they are unlikely to be random.

Theres a statistically significant correlation between the divorce rates in Utah and suicides using explosive materials. This isn't because they actually have something to do with one another it's because there's enough data to find statistically significant correlations between anything and something stupid.

But they almost certainly do, as both homosexuality and chocolate consumption are related to brain chemistry.

The fact that I have to point out that coorelation isn't causation is tedious enough without you whining about how maybe chocolate either causes or is a sign of high IQ. Despite my enjoyment of chocolate it seems that I'm not smart enough to understand how typing the words brain chemistry just explains everything.

But they are the best measure we have of a quality that is biologically rooted and scientifically interesting.

It's highly disputable whether or not IQ tests measure some generalizable factor of the brain or just a couple of random abilities that people assume all come from a single g factor.

No one has said IQ tests enable understanding of biology, any more than a tape measure enables understanding of what makes people tall or short. They are a measuring tool. They measure something that is rooted in biology, but the explanation has to come from elsewhere.

A tape measure is a far different tool than an IQ test, that's a false equivalence. A tape measure tests an extremely simple and straighforward factor, physical height. IQ tests use a battery of tests across a couple different skill sets and a psychologists assessment of how those skills are related to an unknown g factor that they don't even know if it exists as a single generalizable factor and then norms that against the average score people get.

It would only be equivalent if you were using a tape measure to measure the size someones front door and trying to figure out their height.

[Suntzu]

I'm guessing you don't do well on I.Q. test. Interestingly, Isaac Asimov was a Mensa member.

[mikema63]

Guess whatever you like, as it's not relevant. Much like Isaac Asimov.

[JohnRawls]

The existance of Race is a matter of belief. Race has no clear definition. Race is a postulate which you either believe in or you don't.

[Perkwunos]

Well, the concept of race as an jndicator of genetic diversity is also not useful.

I don't know about that, but it is a medically and forensically relevant variable.

Yes, they might be.

I am already aware of at least some evidence this is the case:

http://www.sciencedirect.com/science/ar ... 9615001087

How does this disprove the notion that most diversity occurs within racial groups?

It doesn't, it just disproves that that is the only relevant consideration.

It's highly disputable whether or not IQ tests measure some generalizable factor of the brain or just a couple of random abilities that people assume all come from a single g factor.

I don't think the mere existence of g is hugely controversial. It doesn't account for all of the variance in IQ but it doesn't have to.

[Pants-of-dog]

I don't know about that, but it is a medically and forensically relevant variable.

As I said before, race is a weak proxy for genetic diversity since most genetic diversity occurs within races.

I am already aware of at least some evidence this is the case:

http://www.sciencedirect.com/science/ar ... 9615001087

Maybe. But I am not here to discuss the race and IQ debate, but whether or not races exist.

It doesn't, it just disproves that that is the only relevant consideration.

How so?

[Perkwunos]

Maybe. But I am not here to discuss the race and IQ debate, but whether or not races exist.

OK. Good. They exist.

[Drlee]

Ok. So there is emerging research to show that IQ differences can be hardwired. Believe it or not research on Einstein's brain was recently published. I can't post the article because it is in a password protected website. It is probably available on psychology today but I don't have time today to ferret it out. A quick google just gave e some newsy things.

We are back to the essential argument and have been sidetracked by looking for some obscure method of proving that race biologically exists. Clearly it does. It may not follow skin color as closely as we may have thought before but it does exist.

In the end though this argument is between racialists and non-racialists. The societal effects of being black in America are even more profound that those of having a genius level IQ. This is not due to IQ. It is due to an inescapable physiological trait that causes the human environment to react differently based upon skin color. Beyond recognizing that what is the point of trying to prove some further biological determination of an irrelevant distinction?

Racists or racialists would like to proves some explanation for black behavior in our society that at least in part relieves them from some responsibility to alleviate a problem as obvious as the elephant in the room. Conversely, those who wish to assert a completely level playing field outside of societal pressures constructed some notion that there is no biological difference between races. There clearly is.

We have a race problem in the US and much of the world. That is not a theory. It is a fact. It has to be considered at a personal level. Is there a coorelation between high IQ and success? Yes. Is it well documented? Yes. It is inescapable. But the fact is that high, average or low IQs within our community, being black exerts extreme downward pressure on ones likelihood of success. We will not benefit from trying to explain it away (even in part) to some biological disadvantage and distract from the absolute fact that their (blacks) difficulties in achievement are entirely societal. Entirely.