« Flat earthers | Main | What to do? What to do? »

Ugly questions

29 Nov 2007 03:05 pm

Best post by far on the IQ/Race debate.

I am clearly not qualified to deliver a final opinion on the actual merits of the race/IQ debate. But I think that our social reaction to it is disturbing. And I do mean "our". I'm as creeped out as any latte-sipping liberal when people start arguing that blacks have genetically lower IQs than whites do. But hysterical revulsion is not the correct response to what is basically an empirical question.

In part, this is a justified reaction to the fact that so many of the people advancing these theories in public are clearly racists who have seized on a theory that validates their priors. But only in part. After all, the fact that any discussion of the possibility is greeted with hysterical revulsion guarantees that only two types of people will take the "pro" side in public: fearless iconoclasts who do not care what anyone thinks of them; and racists.

If there is a difference, and that difference is genetic, I assume, in my classical liberal way, that we are better off knowing than not knowing. But my sense is that it is currently not possible to examine the question in any rigorous way right now, because almost no one will touch the subject with a ten foot pole.

And yet, the question matters. We gauge the success of our social policy by looking at macro results under the assumption that everyone (in aggregate) starts off with the same basic genetic endowment. If this is not in fact true, that would alter how we should look at that data.

To be sure, I am not clear on how one entirely overcomes the deep entwinement of society, environment, and genes in this case. I was recently talking to a friend who was mourning the way he watched the girls get dumber as they hit puberty, lose their interest in math and science, in a way that seemed much less likely to be the result of estrogen on their math receptors than the result of social conditioning about what men should find attractive in women.

Nonetheless, I'm pretty sure we could be doing better than we are now, except that people on all sides are terrified of finding any answer that doesn't confirm what they already believe.

TrackBack

TrackBack URL for this entry:
http://meganmcardle.theatlantic.com/cgi-bin/mt/mt-tb.cgi/17848

Comments (124)

"But hysterical revulsion is not the correct response to what is basically an empirical question."

Did the Holocaust really happen? Thats an empirical question. I'm going to hysterically react to anyone who argues that it didn't happen.

A) the historical facts of the holocaust are an empirically settled question; this is not

B) Hysterical reaction gets you nowhere, and leaves the festering sore advancing his rotten theory to infect other people, as he legitimately claims that he is being shut out of the debate. The correct approach is to calmly point out that he's a nutjob.

Personally, I believe the truth is better for humanity than ignorance, lies, or wishful thinking. But a lot of people seem to disagree.

The issue is not the black-white IQ gap, per se, it's how the taboo zone _around_ the entire issue that makes realism career-threatening, which therefore undermines so much sensible public policy. How else do you think we arrived at President Bush and Senator Kennedy coming up with a No Child Left Behind act that legally mandates that every student in the country score above average by 2014!

Realism was off the table in designing NCLB because realists eventually get around to admitting that they don't have a clue how to narrow the white-black IQ gap, which implies it's partly genetic, which gets you fired. So, it's best for your career to not even start down the path of realism. So, there's a wide No-Fly zone that extends far, far out from the central taboo, and it does terrible damage to much of our public policy.

I agree that "hysterical revulsion" is not the appropriate response. How about a yawn instead? I'm not sure that the question is all that interesting, or even sufficiently well-posed to permit scientific investigation.

Race is just not a well-defined concept. it is based on phenotype rather than genotype and must therefore be used with caution. As the American Anthropological Assoication puts it: "With the vast expansion of scientific knowledge in this century, however, it has become clear that human populations are not unambiguous, clearly demarcated, biologically distinct groups."

If race is not a well-defined concept, then what sense does it make to spend time thinking about IQ's and race. Add to that the fact that IQ itself is rather dubious as a measure of innate intelligence, and he whole question of race and IQ becomes so muddled as to be hardly worthy of consideration.

Surely we have better things to do with our time.

Unfortunately, as Saletan conceded today, J. Philippe Rushton is a nutjob and is one of the scientists propagating and disseminating these theories (most famously in The Bell Curve). Now I think he is a nutjob, but many people do not. Go read the comment's at Ross's site everytime he mentions the issue. There are plenty of people who think blacks are dumber than whites, and now they have the benefit of people like Andrew Sullivan, William Saletan, Ross Douthat, and Charles Murray lending their credibility to the likes of Rushton and Flynn. The result is that believe that black people are dumber than whites--by their very nature--now has some credibility (or at least more credibility than it did have) and isn't a verboten idea to express in public discourse. Plenty of people--serious, thoughful people--have debunked the claims made in the Bell Curve (e.g., Thomas Sowell). Nevertheless, despite people like Jim Mazni announcing that there is not conclusive evidence to show that the genetics of blacks makes them dumber than whites, the meme out there is that there is conclusive evidence, and all serious criticisms are motivated by PC Dogma.

The journal that publishes much of this tripe, Mankind Quarterly, had an associate of Dr. Mengele on its board. These are not good people.

I'm a experimental scientist by training and profession, so I'm heavily biased towards empirical answers to questions. But I simply don't see how that's presently possible unless we start putting diazepam in the drinking water for everyone involved in the discussion. Too many people are simply incapable of behaving calmly and rationally.

Look at Larry Summers. All he did was postulate that there are more men in science & engineering because of differences in the sigma for aptitude; not the mean -- the width of the distribution, and he got run out of town on a rail.

Now put the mean in play, and add complications of class, discrimination, etc...it's hopeless.

Right, but you're validating my point. The only people investigating the question--which clearly is an open question--are mostly nutjobs, precisely because we've made it a career killer to come up with the wrong answer.

Dear Megan,
Please stop talking about race.
Signed,
Everyone

"we've made it a career killer to come up with the wrong answer."

Have we? I keep hearing this assertion, but I've never seen any evidence for it. Or are you saying that Hertstein died BECAUSE he wrote the Bell Curve? That surely ended his career.

Those that complain about PC hysteria regarding this issue act as if there isn't a library full of detailed, cogent criticism of these racialist ideas. There are.

These questions don't really bother me, in the same way that Larry Summers' comments on women in science and math didn't bother me (I'm a white female engineer), for a few reasons:

1) We're talking about averages, not individuals. If there are more smart white people than smart black people, or more male scientists than female scientists, I don't really care as long as no students are discouraged from excelling at school/math/science/whatever due to race or gender.

2) Raw intelligence is only one part of the "success" equation - other factors like motivation, social connections, access to quality education, and societal/familial pressure to succeed academically matter a great deal. I suspect (and I have no data to back this up here) that IQ matters a lot less (especially in the mid-range) than whether or not your family pushes you to do well in school, and whether or not your friends are serious about school.

Race is just not a well-defined concept.

It doesn't have to be well-defined to be meaningful and useful. Lots of meaningful and useful categories are not well-defined. Read Jim Manzi's post.

>Right, but you're validating my point.

As was my intent. I never said I was disagreeing...

There are a lot of reasons why we want to settle this question.

If it is true that Asians are nearly 1 SD above the mean in IQ compared to the US population (and have about the same SD), it means that China and India will be outstripping the US in any field where intelligence is required - science, engineering, medicine, etc. - once they get their educational system and general infrastructure up to Western standards. If scientists that make major breakthroughs are 1 in a million, then today the US could have 300, Europe 729, India 1120, and China 1320.

In addition, when the genes for IQ are identified, it is not unreasonable to assume that techniques for modifying these genes will be developed. Totalitarian governments have already shown that they will do harmful medical procedures on citizens to win Olympic medals. (see East Germany) Would they not do the same to produce Nobel prize winners and the economic advantages from their discoveries?

Scientific American had an article on "gene doping" in athletics that said that given what was known about gene therapy as of 2004, the 2008 Beijing Olympics could have one or more competitors that have had gene therapy to enhance their ability. Experience should tell us that if there is enough money or prestige at stake, there is someone that will try to get an advantage. If the balance of economic or military power is at stake, what country wouldn't try to enhance the population?

Add to that the fact that IQ itself is rather dubious as a measure of innate intelligence, and he whole question of race and IQ becomes so muddled as to be hardly worthy of consideration.

Um. "g", which IQ tests measure, is well known to be significantlt inheritable. It's also positively correlated with income and education. Now, what it measures includes many of the aptitudes needed for knowledge-based jobs, it's pretty important.

One item that flows from this is the crummy job US schools do at educating "talented and gifted" students. They have dropout rates comparable to the general school population.

Nevertheless, despite people like Jim Mazni announcing that there is not conclusive evidence to show that the genetics of blacks makes them dumber than whites, the meme out there is that there is conclusive evidence, and all serious criticisms are motivated by PC Dogma.

On the contrary, the prevailing meme in public discourse seems to be that the hypothesis of genetic racial differences in intelligence has been thoroughly debunked and that the consensus of expert opinion is that any apparent differences in intelligence between racial groups are purely environmental. But that is in fact a misrepresentation of the state of expert opinion, which certainly does not hold that the hypothesis has been discredited. See this article, for example.

IQ is far from the only determinant to future success. For one thing, we should remember that it's a quotient (the "Q" part) not an absolute, "you're that way forever," take on someone's abilities.

Differences in sub-cultures, & the residual effects of slavery, segregation & racism can not be denied either. "Get over it, the Civil Rights Act was signed 40 years ago" is not an adequate response, as Ms. McA. indicates when bemoaning the state of "poor" or "inner-city" schools.

Are there statistics comparing poor rural Euros to upper middle class & better Euro suburbanites/city dwellers? (Hell, drop the whole "Euro" qualification, just go w/ poor rural people compared to wealthy urbanites. Then, of course, we'd get the old "poor people are poor because they're stupid.") Might be interesting.

"It doesn't have to be well-defined to be meaningful and useful. Lots of meaningful and useful categories are not well-defined."

I don't see how this particular classification is useful here. Racial classifications arose from outward traits like skin color that have no causal link to intelligence. A useful racial classification would have to be based upon genotype, not phenotype.

If researchers want to spend time on this question, they sould be free to do so without abuse, but I doubt very much that their inquiries will yield any important results. I doubt even more that their research could have any important policy implications.

Suppose that people with one skin color have slightly higher native intelligence than people with another skin color. What does that imply about public policy? My opinions about the major policy debates of the day would be unaffected. But maybe I'm missing something.

I'm a person with a high IQ, a member of Mensa, and a Phi Beat Kappa graduate of an Ivy; but that is not who I have become at 66 years of age.

IQ is not important except at the ends of the tails. The difference in means between 'races' may be statistically significant, but one would be a fool to categorically overlook a person who 'looks' to be of African origin in favor of a person who looks European. There is a non-trivial chance that the person from the group with the lower average IQ has a higher IQ than the person from the group with the higher average IQ. There is an even better chance that the difference in IQ between the two individuals is within the confidence interval.

g is a measure of intelligence so-called. It is not a measure of character or good sense or ambition or skill. The problem lies in dealing with people as representative of genetic subgroups. It is a natural tendency perhaps in a world of small bands of hunter-gatherers to favor those whose genes are closest to one's own; but we are thousands of years beyond the hunter-gatherer world. We follow laws written by long-dead Englishmen. We buy manufactured goods made thousands of miles away by people who cannot read our script as we cannot read theirs. We ask other people thousands of miles away to write the programs that run the computers to which we entrust our economy.

In America, early colonists came up with 'Black Codes' because most bound labor came from Africa whilst most free labor came from Europe. It was then a short step to treating Black people as inferiors. Nowadays, hoever, it is well-meaning "progressives" who perpetuate invidious racism of this sort. Progressives keep telling us that African Americans do poorly in school and we must ignore their lack of educational accomplishment and hire and promote them by quota. This is a message to Black kids and their families that says don't bother trying to do well in school, you're going to fail anyway. This is a message to white racists that says, yes, you're right, Blacks are dumber than white people.

With this nonsense one can understand the anger of Clarence Thomas.

Except that, "we must ignore their lack of educational accomplishment and hire and promote them by quota" isn't true.

Suppose that people with one skin color have slightly higher native intelligence than people with another skin color. What does that imply about public policy?

Well, it suggests that obsessive racial bean-counting and trying to force every profession, company, and college to have the exact same racial makeup as the general population is a bad idea.

I'd think it was a bad idea either way, of course.

" ... Racial classifications arose from outward traits like skin color that have no causal link to intelligence ..."

Let's say hypothetically that my ancestors spent the past 20,000 years paddling around the tropical Pacific in open canoes while your ancestors spent the same time chasing wooly mammoths over frozen icecaps. Doesn't it stand to reason that you & I are going to have very different mental strengths & weaknesses -- for the same reason that we have different chemicals in our skins?

The brain is a physical bodily organ, just like the skin. Why shouldn't it also be subject to the pressures of differential evolution?

rwe,

I don't see how this particular classification is useful here.

It is useful--essential, actually-- to the study of racial discrimination, for example. You obviously can't determine whether or to what degree such discrimination is present without racial classifications.

Racial classifications arose from outward traits like skin color that have no causal link to intelligence.

So what? Yes, "skin color" may have no "causal link" to intelligence, but other genes that are correlated with genes for skin color may.

A useful racial classification would have to be based upon genotype, not phenotype.

Not at all. Since genotype is an expression of phenotype, it could be based on either.

If researchers want to spend time on this question, they sould be free to do so without abuse, but I doubt very much that their inquiries will yield any important results.

Why do you doubt it?

Suppose that people with one skin color have slightly higher native intelligence than people with another skin color. What does that imply about public policy?

If the difference is only slight, it may have no clear policy implications. But if the difference is significant, it may have implications for all sorts of public policy, from education to affirmative action.


brad wrote: Dear Megan, Please stop talking about race. Signed, Everyone

Dear Brad,

I have consulted with all of the mice in my realm, and we have determined two very important things:

1) That you do not speak on our behalf;
2) That we are capable of speaking on our own behalf.

Next time, would you please sign your class actions as "Everyone, except for anony-mouse and his minions"?

Thanks!

"If the difference is only slight, it may have no clear policy implications. But if the difference is significant, it may have implications for all sorts of public policy, from education to affirmative action."

Like what, exactly? Suppose that black Americans have lower average native intelligence than white and East Asian Americans (which I do not believe, though many commenters seem to). What does that imply about education policy and affirmative action?

My policy choices would be the same regardless. I would want to improve the schools in poor neighborhoods and would oppose affirmative action just as I do now. Native intelligence we can't change, but education and skill levels we can. I think this whole line of inquiry is just a distraction.

Like phrenologists and astrologers, a lot of people want to make wild causal inferences. Africans fared much worse than Norwegians over the centuries, therfore, ipso facto, they must be geneticaly inferior. But that doesn't follow.


"Since genotype is an expression of phenotype, it could be based on either."

I think you have that reversed. But you probably just mistyped it.

Like what, exactly? Suppose that black Americans have lower average native intelligence than white and East Asian Americans (which I do not believe, though many commenters seem to). What does that imply about education policy and affirmative action?

It might mean that we would devote additional educational resources to black children, for example, to try and compensate for the genetic disadvantage. Or it might mean that we abandon race-based affirmative action policies in employment justified on the hypothesis that discrepancies in racial representation are the result of discrimination rather than qualification.

In any case, I don't believe the scientific study of human racial differences should rest on any supposed policy implications. It's worth doing regardless of its implications for policy, just as the scientific study of human nature in general is.

Like phrenologists and astrologers, a lot of people want to make wild causal inferences.

So you keep saying. To which my response is: Like flat-earthers and creationists, a lot of people want to ignore scientific inquiry that threatens their entrenched ideology.

I think you have that reversed.

I do, yes. I meant to write that phenotype is an expression of genotype.

Mixner-

You're missing an important point--The scientific inquiry, and the results that said scientific inquiry produces, do not "threaten their entrenched ideology" (their being liberals). The reason is, the evidence is inconclusive.

This is from Stephen Jay Gould, whose persuaded me that this whole line of inquiry is not likely to be fruitful. Here he makes the same point I was making about phenotypes (or rather "folk taxonomies") and genotypes:


My favorite trivia question in baseball is, "Which Italian American player for the Brooklyn Dodgers once hit 40 home runs in a season?" Nobody every gets it right, because the answer is Roy Campanella, who was as Italian as he was black. He had an Italian father and a black mother, but he's always classified as black. You see, American racial classification is totally cultural, and it's based on the unfortunate and sad legacy of racial distinction based on this ridiculous metaphor, the purity of blood. You're identifiable as having black ancestry because we can see it. I mean, who's Tiger Wood, who's Colin Powell? Colin Powell is as Irish as he is African, but we don't classify him as that. No, we have a really screwed up classification. To think it's biological is just plain wrong. It's based, flat-out, on the legacy of racism and the metaphor of the purity of the blood. It's a very troubling issue.

Maybe Gould and I are wrong. Maybe this is really a good subject for research. I certainly think people ought to be free to inquire into this, but resources are scarce. I would much rather funds go into cancer research or research on the efficacy of school vouchers.

The whole interview with Gould is here.

Like what, exactly? Suppose that black Americans have lower average native intelligence than white and East Asian Americans (which I do not believe, though many commenters seem to). What does that imply about education policy and affirmative action?

Seriously? Everything. It may go as far as to implicate entire teaching methods as being inadequate to meet the dominant learning style within a group. It may mean that affirmative action really does set people up to fail by placing them above their skillset on the basis of a false premise.

If there are differences that are real and significant, then that knowledge needs to be on the table when determining how to best acommodate that group. If the differences have some genetic basis, then a permanent rethink of means and methods may be required. Or, it may be that the basis is primarily or entirely due to a systematic lapse in early environmental factors. But even then, that's also useful information which will help with addressing the problem.

"It may go as far as to implicate entire teaching methods as being inadequate to meet the dominant learning style within a group."

So then what? At the beginning of class separate each racial group? Have a 'Jew' textbook and a 'black' textbook and an 'asian' textbook, each advocating different methods? What happens to adopted kids? Kids who are Jewish only through the matriarchy? Ambiguous groups? Mixed-race? White people born in Africa?

Lebanese?

Part of the problem is in the definition of IQ. The test has been engineered so that both men and women score, on average, 100 points. We know that women and men are better at different tasks, and the test is designed so that these differences average out. It could easily be tweaked to favour either sex.

When it comes to race (another leaky concept: why is someone with a black mother and a white father black? Why do we adopt the "not one drop" criterion), it is easy to imagine that, even if there are ability differences between the races (in itself, hard to show), the test could be adjusted to give anyone an advantage.

rwe,

Maybe Gould and I are wrong.

You are wrong. The claim that "racial classification is totally cultural" is obvious nonsense. You yourself just mentioned that racial classifications are based on, among other things, the biological trait of skin color.

I'd be cautious about citing Gould on anything. He was once a great scientist who did groundbreaking work, but towards the end of his life he became a polemical nut. As John Maynard Smith put it a few years ago, when Gould was still alive, Gould "is giving non-biologists a largely false picture of the state of evolutionary theory."

I certainly think people ought to be free to inquire into this, but resources are scarce. I would much rather funds go into cancer research or research on the efficacy of school vouchers.

So do you also oppose all scientific research with no clear practical applications or significance, or only research on racial differences in intelligence?

You're missing an important point--The scientific inquiry, and the results that said scientific inquiry produces, do not "threaten their entrenched ideology" (their being liberals). The reason is, the evidence is inconclusive.

The evidence is inconclusive so far. They fear that further research will produce evidence of genetic racial differences in intelligence, a conclusion they don't want to be believe is true. Hence all their efforts to discourage and disparage scientific research on this issue.

" ... American racial classification is totally cultural, ..." --- Stephen Jay Gould, quoted by 'rwe'.

That's probably true as far as it goes, but the fact that Americans' concept of race is seriously screwed up doesn't prove that races are just a figment of the imagination.

Looking at the complete Gould interview that was cited, he makes the absurd statement that ~100,000 years was not enough time for the development of genetic variation except for " ... very superficial features like skin color and hair form...." That is a nonsensical claim: Just look at the amount of intra-species variation that has taken place in domesticated animals in only ~10,000 years.

Surely no one would argue that the brain of a Border Collie (for example) is wired the same way as that of a Labrador Retriever or a Terrier -- and yet they are all descended from the same wolfish ancestor within 20,000 years or less.

Gould was such a radical ideologue, I'd take everything he said with a large grain of salt. (He was a darned entertaining writer, though.)

Maybe Gould and I are wrong.

Actually, I think you're wrong in citing that Gould quote in defense of this point, or if Gould was addressing the same point at the time, he was doing the same kind of tip-toeing that prevents an honest assessment of it now.

If you have a specific genetic quirk in a family line -- for example, the breast cancer gene, baldness, or red hair -- it tends to follow that family line. The more dominant it is, the more it expresses itself in the purer bloodlines where copies are received from both parents.

The citation of Woods, Campanella, and Powell exactly proves the point. Yes, these are all identified as "black" because any recent black ancestrage in the bloodline tends to produce strong phenotypical traits. On the other hand, people with very fair skin and unusually dark or unusually red hair are often identified as Irish; skin folds that give the illusion of slanted eyes quickly marks someone as having East Asian ancestry; and Middle Eastern and Mexican feature sets are also very strong and take a couple generations of interbreeding to thin down. So I'm not sure what that statement proves.

Meanwhile, if there WERE a genetic component to intelligence, and African, Western European, and East Asian genetics respectively trended low, median, and high, then the observed results would be compeltely consistent: bloodlines that have maintained "purity" would also maintain the trend, while mixed bloodlines would tend to produce averages.

That, of course, continues to assume that environment isn't the dominant factor, which it may well be. Being raised in a positive early childhood environment under a Confuscian or a Judeo-Christian value system may tend to produce significantly higher intelligences irregardless of race, while having a crack-whore mother, an absent father, and poor childhood nutrition may set a person up for failure, again irrespective of race -- with the results tending to follow racial lines due to historical factors and obsolete taboos on intermarriage.

"They fear that further research will produce evidence of genetic racial differences in intelligence, a conclusion they don't want to be believe is true."

I don't really see whats wrong with that. I do, however, thing something is terribly wrong for the opposite desire to be held.

I don't really see whats wrong with that.

There's nothing wrong in itself with hoping that an empirical proposition is false rather than true. There is something very wrong with discouraging or disparaging scientific research into that proposition because you fear it will be confirmed rather than falsified. I can think of a few reasonable exceptions to this for particular individuals in particular situations, but it certainly doesn't apply to the scientific study of racial differences.

Being raised in a positive early childhood environment under a Confuscian or a Judeo-Christian value system may tend to produce significantly higher intelligences ...

Ha ha ha. Good one. As Bertrand Russell said, "So far as I can remember, there is not one word in the Gospels in praise of intelligence." In America today, it's hard to think of a community more hostile to reason and learning than devout Christians...

Mixner,
I think it would be better to say fundamentalist Christians and not devout Christians. Notwithstading such things as Creationism, I think being raised under Judeo-Christian value system does help create the needed skills for learning.

No, it's not only "fundamentalist" Christians. It applies to devout Christians in general. Perhaps its most common manifestation is the nonsensical assertion that religious faith is a "way of knowing" that complements or supercedes scientific and rational inquiry. The confusion of belief with knowledge, of wishful thinking with evidence, is at the core of "the Judeo-Christian value system."

Wow. So Mixner hates Christians and he's a racist. Who are you, Tom Buchanan?

"Actually, I think you're wrong in citing that Gould quote in defense of this point, or if Gould was addressing the same point at the time, he was doing the same kind of tip-toeing that prevents an honest assessment of it now."

I think you would profit from reading Gould's thoughts on the subject. He was a sharp critic of precisely the sort of race and IQ research that many on this thread seem to be advocating. And his reasoning was the same as mine: first, the usual racial classifications are problematic and second, IQ is a dubious measure of innate intelligence. Thus reasearch that relies the usual, flawed racial classifications and on IQ as an indicator of innate ability is so muddled from the outset that it isn't likely to produce anything valuable.

I think Gould also shared some of rickm's concerns. This was, after all, a line of thinking the Nazis found appealing: "Blacks fell behind not due to bad luck but due to their own inherent inferiority, making them an 'inferior race,' best suited to slave labor." Anything the Nazis found so congenial is at least suspect.

As for racial classifications being "totally cultural" I imagine Gould was exaggerating a little. I think what he meant was that they are not realiable indicators of underlying genetic differences. So Colon Powell gets classified as "black" even though he has a mixed heritage. Indeed, it would be hard to think of a sensible racial classification scheme that would find a place for him at all.

Megan, to the extent that reputable science shies away from these questions, it's not because of societal pressure. It's because they understand that we don't now have the tools to make meaningful measurements that can distinguish genetic and environmental effects.

We don't know what IQ is measuring. Is it intelligence, or some culturally determined ability to succeed in society?

We don't know whether g, a statistical artifact and not an observed variable, exists. Thurstone showed that g disappears with only a few very basic alterations to the statistical model. Shalizi showed that g can be interpreted as consistent with a single-variate intelligence model or a multi-variate one.

We can't explain the Flynn effect. While a disparity of perhaps 15 points between average white and black IQ has held constant, average IQ of everyone has risen by much more than 15 points over the last century.

The significant point, however, is that we don't have effective ways of answering these questions. Society could line up 100% behind intelligence researchers, but it wouldn't make there results any more meaningful.

By the way, I think the writer of the post you linked oversimplifies the notion of heritability. Here's a more thorough discussion.

Being raised in a positive early childhood environment under a Confuscian or a Judeo-Christian value system may tend to produce significantly higher intelligences irregardless of race

Hindus score higher than Christians. Though not higher than Jews. Perhaps being raised in a Jewish household causes higher intelligence than being raised in a Christian one? Or will you say that's just genetic, since secular Jews also score high? Or maybe being raised in a secular household causes higher intelligence? Or maybe having higher intelligence causes one to be secular?

Data mush. All conclusions will merely reflect prior prejudice. Move to topic where meaningful things can be said.

On the one hand, there are theoretical and empirical reasons to believe group differences in measured IQ are both genetic and important. On the other hand, there were theoretical and empirical reasons to believe similar ideas in the past that turned out not be the case. In addition, some of the most fervent advocates for the claim are would-be totalitarians, which makes adopting the idea much riskier than skepticism.

The above paragraph also applies to global warming.

I really, really hate this discussion.

1) Realistically, it's going to be impossible to separate social and environment influences from genetic ones.

2) From what I've read, there's some indication that the gap between blacks and whites has been closing within the last few decades. How does that square with the notion that the gap is the result of 100,000 years of evolution making white people smarter?

3) Even if you accept that, say, whites are on average 5% smarter--so what? Does anyone here seriously think they could craft a public policy that would somehow take that 5% into account and actually work better than a policy which ignored it?

4) Likewise, does genetic influence somehow obviate the effects of racism? (Of course, I'm sure plenty of people think it's evidence that racism itself has no effects anymore, and that every woe befalling African Americans is the result of the fact that they as a group score a bit lower than whites on IQ tests.)

5) Frankly, this whole fuckin' discussion strikes me as essentially racist. I mean, we're basically debating whether we can employ racial classifications in some way to optimize public policy around intelligence, while ignoring the fact that if such optimization were really our goal, we'd cut all this slightly smarter on average asians vs. still pretty smart whites vs. maybe a little dumber blacks crap and just make the distinction of fucking stupid people vs. really smart people. It's ridiculous. We're not talking about making policy for any other collection of 40 million people that average mid-90s IQ scores--no, conveniently, it's just the collection of 40 million that have been abused for centuries on account of their supposed inferiority. That, I think, was Gould's point. We make the racial distinction because it's been made in the past, not because it's particularly useful for predicting intelligence.

This is from Thomas Sowell's critique of "The Bell Curve":

Perhaps the strongest evidence against a genetic basis for intergroup differences in IQ is that the average level of mental test performance has changed very significantly for whole populations over time and, moreover, particular ethnic groups within the population have changed their relative positions during a period when there was very little intermarriage to change the genetic makeup of these groups.

When IQ shifts around as much as it has, it seems clear that it must be heavily influenced by enviroment and not merely by natural ability. Sowell points to the IQ's of Jews--less than a hundred years ago they were so low "as to cause Carl Brigham, creator of the Scholastic Aptitude Test, to declare that these results 'disprove the popular belief that the Jew is highly intelligent.'" Now, of course, Jews score well above the average. All of this makes it fairly clear that any research that uses IQ as a measure of inate ability is specious, as Tom T and Matt argue above.

"5) Frankly, this whole fuckin' discussion strikes me as essentially racist. I mean, we're basically debating whether we can employ racial classifications in some way to optimize public policy around intelligence, while ignoring the fact that if such optimization were really our goal, we'd cut all this slightly smarter on average asians vs. still pretty smart whites vs. maybe a little dumber blacks crap and just make the distinction of fucking stupid people vs. really smart people. It's ridiculous. We're not talking about making policy for any other collection of 40 million people that average mid-90s IQ scores--no, conveniently, it's just the collection of 40 million that have been abused for centuries on account of their supposed inferiority. That, I think, was Gould's point. We make the racial distinction because it's been made in the past, not because it's particularly useful for predicting intelligence."
--Matt

"On the one hand, there are theoretical and empirical reasons to believe group differences in measured IQ are both genetic and important. On the other hand, there were theoretical and empirical reasons to believe similar ideas in the past that turned out not be the case. In addition, some of the most fervent advocates for the claim are would-be totalitarians, which makes adopting the idea much riskier than skepticism.

The above paragraph also applies to global warming.


Posted by Joseph Hertzlinger | November 30, 2007 1:48 AM

Yet another way to divide the flock, we should wake up. Hopefully, before we get back to nose length( see: Hutu-Tutsi/Rwanda ).

Sailer's point, here: "That’s not the structure of the controversy. The structure of the controversy is: research group A, resourced by fund X with bias M, saying this is so, while a mighty host of journo-school grads, law-school grads, and liberal-arts department heads — yes, and even a few careerist, tenure- or office-seeking biologists and money-seeking, PC-compliant pop-science authors — shriek YOU MUSTN’T TALK ABOUT THAT! YOU ARE BAD PEOPLE! That’s the structure of the controversy."
-applies far beyond the simple bounds of this discussion..
jbb's anecdote: "The last time I was asked to speak I was on a panel, all educators but me, to 200 kids in the program. I spoke last. My statement was very straightforward- "The average car salesman in the region makes 70k and the highest paid makes 250k. None have a degree. They have drive, ambition and the ability to relate to and communicate with all types of people. There are thousands of jobs like that where a degree is not needed. A skilled machinist can make 80k easily and a degree is not needed. Explore other options. " I was never asked to speak again."
-is hardly singular, the last thing these entrenched bureaucracies are looking for is any kind of objective truth..


Public policy currently takes race into account.

The existence of a genetic difference in IQ between races is really an argument for not taking race into account when making public policy.

Many people in this discussion believe that people should be treated as individuals based on their ability and not based on their race. Current public policy DOES NOT do this.

We have many race-based programs and policies that are based on the assumption that all differences in outcome are due to discrimination. If there are real genetic differences in intelligence among the races, regardless of what they are, then these policies are not helping and possibly doing harm.

The truth is that individuals should be treated as individuals and public policy should do this. Opportunity should be equalized, not outcome.

I would dearly love to see all government forms eliminate the "race" entry. Private employers are forbidden to ask about race. Government agencies routinely ask about race.

EI

Matt said:
I really, really hate this discussion.
1) Realistically, it's going to be impossible to separate social and environment influences from genetic ones.

Why can't we separate some of the effects? Are you saying we will never find any genes related to intelligence? We have already found some.

2) From what I've read, there's some indication that the gap between blacks and whites has been closing within the last few decades. How does that square with the notion that the gap is the result of 100,000 years of evolution making white people smarter?

Be quantitative, a small amount of closing is not inconsistent with evolution. Also if you confident that we can accurately measure the small amount of closing, you should be very confident in the measurement of the large gap.

3) Even if you accept that, say, whites are on average 5% smarter--so what? Does anyone here seriously think they could craft a public policy that would somehow take that 5% into account and actually work better than a policy which ignored it?

1 Standard deviation is a lot more than 5%. I think its better to think about the difference in percentiles, ie if someone were at the 50th percentile of the white distribution, were would they be on the black or asian distributions, and vice versa. Several public policy ideas, stop associating differences in outcomes with proof of discrimination. Realize complicated tax laws, with incentives to shape behavior may have disparate impact.

4) Likewise, does genetic influence somehow obviate the effects of racism? (Of course, I'm sure plenty of people think it's evidence that racism itself has no effects anymore, and that every woe befalling African Americans is the result of the fact that they as a group score a bit lower than whites on IQ tests.)

Of course racism will still exist, but be quantitative, how much is racism affecting the outcome. Every woe befalling an individual African Americans is not due to racism.

5) Frankly, this whole fuckin' discussion strikes me as essentially racist. I mean, we're basically debating whether we can employ racial classifications in some way to optimize public policy around intelligence, while ignoring the fact that if such optimization were really our goal, we'd cut all this slightly smarter on average asians vs. still pretty smart whites vs. maybe a little dumber blacks crap and just make the distinction of fucking stupid people vs. really smart people. It's ridiculous. We're not talking about making policy for any other collection of 40 million people that average mid-90s IQ scores--no, conveniently, it's just the collection of 40 million that have been abused for centuries on account of their supposed inferiority. That, I think, was Gould's point. We make the racial distinction because it's been made in the past, not because it's particularly useful for predicting intelligence.

I would prefer to make policy for everyone, there are plenty of poor low IQ whites who could benefit from policy that doesn't assume we are all equal.

I would prefer to make policy for everyone, there are plenty of poor low IQ whites who could benefit from policy that doesn't assume we are all equal.


Posted by benp | November 30, 2007 11:53 AM

"I would prefer to make policy for everyone"

how telling...

I wish everyone would get off the hobbyhorse of black=leftendofthebellcurve.

The problem, ultimately, is that there *is* a left end of the bell curve. It doesn't matter if the people involved are black, white, or purple with pink polka dots.

It is foolish to advocate giving everyone a college education. The left end of the curve needs more skills training, while the right end needs more intellectual education. The mix changes with the student's abilities. The mantra of "there are no wrong answers" helps encourage inquiry, but is dangerous when learning how to pour concrete, weld structural steel, wire or plumb buildings, and so on.

We've lost sight of this fact with our "everybody is the same" idiology. We lose the real contributions people can make when we apply one-size-fits-all policies like NCLB. The quality of our plumbing and wiring deteriorates, too.

My daughter is quite bright. My son is autistic, and we may never know how bright the light is under that bushel basket. We do *not* treat them the same. That would be a crime.

We have many race-based programs and policies that are based on the assumption that all differences in outcome are due to discrimination. If there are real genetic differences in intelligence among the races, regardless of what they are, then these policies are not helping and possibly doing harm.

Yeah, the Sully argument.

It's kinda bullshit though. You're just arguing that our efforts at gauging the effects of racism may be off by a couple degrees, owing to the fact of a genetic intelligence gap between races. Unless you think that the intelligence gap accounts for the entire difference in outcome, it's sort of a moot point. And I don't really see how you could argue that in good faith at this point, considering how flakey IQ is (as someone else pointed out, the Flynn effect suggests a strong environmental component to IQ scores), how wide the achievement gap is compared to the supposed IQ gap, and the fact that we're barely 40 years out of the Jim Crow era.

I have two comments on this thread.

1: When Erica posted an allusion to Larry Summers' comments on women in elite math, she may have thought that he's claiming that women are less capable on the average, but he was claiming that we should investigate the question as to whether men have a higer standard deviation. Or maybe she did understand the distinction. It makes sense that men should have a higher standard deviation in measures that may lead to extreme reproductive success. A gene that gives a male a 5% chance of getting him inside any woman's pants to the tune of fathering a hundred children, and a 95% chance of turning him into a schlameil who is essentially certain to die childless, is a gene that will spread in the population. There is no analogous possibility for women, who can never have more than perhaps a dozen children even if they are among the most reproductively fit women ever.

2: In this episode of This American Life, act 2, the protagonist, a sex reassignment patient, claims to have developed a sudden interest in math when she/he started receiving testosterone. It's an anecdote, I'm skeptical, but it's amusing.

-dk

The Flynn effect may be an illusion. There are skills involved in taking standardized tests. As more and more people take more and more standardized tests they will naturally get better at them.

Whether one race is genetically more intelligent than another race is a question that should be decided by level-headed scientists.

Whites--and others--should be in favor of segregation no matter what the answer to the racial IQ question is. Finding an appropriate mate for oneself is an absolutely essential condition of happiness for almost everybody. Since I have never met a white man who is attracted to black females, the demise of white females as a result of integration will mean that the pool of potential eligible mates will continue to shrink for white men. While most white men I talk to are shocked by such racist revelations, none of them are willing to even go out on a date with a black female.

Jeff-

Judging that the men you meet don't get much, i'd say they are being picky.

The Flynn effect may be an illusion. There are skills involved in taking standardized tests. As more and more people take more and more standardized tests they will naturally get better at them.

Right, that's one of the environmental components I was referring to.

You know, I hate it when I agree with Rickm.

"Since I have never met a white man who is attracted to black females, the demise of white females as a result of integration will mean that the pool of potential eligible mates will continue to shrink for white men. While most white men I talk to are shocked by such racist revelations, none of them are willing to even go out on a date with a black female."

Man, you really don't get out much, huh?

Matt,

1) Realistically, it's going to be impossible to separate social and environment influences from genetic ones.

No it isn't. Twin studies and adoption studies, for example, provide a useful method for distinguishing environmental from genetic influences. This is true not just for the question of differences in intelligence between racial groups, but for the issue of genetic vs. environmental effects on human traits and behaviors in general. And once we identify genes that affect intelligence, we could measure the incidence of those genes in different groups.

2) From what I've read, there's some indication that the gap between blacks and whites has been closing within the last few decades. How does that square with the notion that the gap is the result of 100,000 years of evolution making white people smarter?

Yes, this is the Flynn Effect. It's an interesting phenomenon, but it doesn't mean there is no genetic component to racial differences in intelligence.

3) Even if you accept that, say, whites are on average 5% smarter--so what? Does anyone here seriously think they could craft a public policy that would somehow take that 5% into account and actually work better than a policy which ignored it?

If the average difference is small, it may not have any strong implications for public policy. But if the difference is significant, it may have implications for many kinds of policy, as I and others have already mentioned. But regardless of policy implications, it's an interesting empirical question that deserves to be answered.

4) Likewise, does genetic influence somehow obviate the effects of racism?

No. But it may account for some of the socioeconomic differences between racial groups.

5) Frankly, this whole fuckin' discussion strikes me as essentially racist.

Frankly, that statement strikes me as utterly stupid. If you think the pursuit of empirical knowledge is racist, your conception of racism is ludicrous.

A growing mixed race generation will save us from ourselves.

rwe,

When IQ shifts around as much as it has, it seems clear that it must be heavily influenced by enviroment and not merely by natural ability.

This strawman yet again. I haven't seen anyone here argue that there is no environmental contribution to racial differences in IQ test scores. If anyone has expressed support for that position, I will cheerfully tell them that their assertion is wildly implausible. The issue is whether there is also a genetic contribution. The prevailing expert opinion among psychologists, sociologists, educators and cognitive scientists seems to be that there probably is a genetic contribution.

rwe,

And his reasoning was the same as mine: first, the usual racial classifications are problematic and second, IQ is a dubious measure of innate intelligence.

We've been over this. Yes, racial classifications are "problematic." Yes, the boundaries of racial categories are fuzzy. So are the boundaries of lots of other categories, even biological ones, like species and sex. The fact that a category is not well-defined does not mean it is not meaningful or useful. Do you believe the fact that racial classifications are "problematic" means that we should just abandon all attempts to detect, measure or understand racism and racial discrimination? Those questions obviously depend on "problematic" racial classifications just as questions of racial differences in intelligence do.

As for IQ test scores, as others have already pointed out, and as Jim Manzi explains in his piece (have you read it yet?), whether you call the trait it measures "intelligence" or something else, it seems to predict positive life outcomes in school performance, work performance, and health.

As I recall when I read "The Bell Curve", East Asians and European descended Jews generally had higher IQ's than the average white American. As an average white American, I'm not terribly offended.

The first step in the problem solving process, is to recognize the problem. If we refuse to recognize what is apparently a visible problem, it seems to me that we're acting in a racist manner. And, I suspect that the PC crowd wouldn't want to be described as racists.

George-

Those groups you mentioned did not have higher IQ's. They just scored higher on IQ tests. Big, massive, difference that those like Saletan like to gloss over. All that says is that East Asians, Europeans, and Jews are better at taking IQ tests than blacks and Latinos. Because this is a function of education when viewed in the aggregate, there should be no surprises.

"The prevailing expert opinion among psychologists, sociologists, educators and cognitive scientists seems to be that there probably is a genetic contribution."

That is certainly not the prevailing expert opinion among sociologists and educators. Among psychologists and cog sci folks, I'm pretty confident that the prevailing expert opinion is that there probably is a genetic contribution to intelligence. The idea that the "prevailing expert opinion among" among those two groups is that there are (partly) genetically based racial differences in intelligence - I am extremely dubious about this. Extremely. But you are welcome to make your case.

" If we refuse to recognize what is apparently a visible problem, it seems to me that we're acting in a racist manner."

Does everyone see what George is trying to do here?

"The prevailing expert opinion among psychologists, sociologists, educators and cognitive scientists seems to be that there probably is a genetic contribution."

Mixner, I don't object to that at all. I think its highly likely that genetics are an important determinant of intellectual ability (along with nutrition, education, etc...) I'm just arguing that IQ is not a good measure of genetic endowment--precisely because it is heavily influenced by environmental factors (as the dramtic changes in IQs of American Jews over the last century makes clear).

To figure out what correlation exists between race and inborn intelligence, one would have to get a good measure of inborn intelligence and, as even you now seem to be condeding, IQ ain't it.

Now, as for Manzi's article, I'm not sure why you keep citing it. He argues that there are basically two approaches to the question. One is to understand the underlying physical mechanism by which genes determine intelligence. And he makes clear that we are nowhere near understanding that.

The other is to use econometrics and "natural experiments" (to use IQ as a proxy variable and control for other relevant variables, if I understand him correctly) about which Manzi is not sanguine. As he puts it, "resolution will almost certainly require advances in understanding of the physical mechanism of intelligence, not more clever econometric analysis."

A lot of people posting above seem to be convinced that there is a large genetic difference between the races. Manzi essentially says there's no justification for such certainty. As he puts it, quoting the APA approvingly, “At present, this question has no scientific answer.”

Like Manzi, I don't know whether race and intelligence are linked in a causal way. Nor do I think it's a very important line of inquiry (no more than, say, asking whether redheads are more intelligent than blonds). Like Manzi again, I think researchers are nowhere near answering the question. And further I would add that, even if they did, their results would have no impact on my opinions on public policy. So with that, I withdraw from the field, though I will read any further criticism of my posts. It's been an interesting debate.

That is certainly not the prevailing expert opinion among sociologists and educators.

From the Wikipedia entry on race and intelligence:

A survey was conducted in 1987 of a broad sample of 1,020 scholars (65% replied) in specialties that would give them reason to be knowledgeable about IQ (but not necessarily about race; Snyderman & Rothman, 1987). The survey was given to members of the American Education Research Association, National Council on Measurement in Education, American Psychological Association, American Sociological Association, Behavior Genetics Association, and Cognitive Science Society.

One question was "Which of the following best characterizes your opinion of the heritability of the Black-White difference in I.Q.?" (emphasis original).[183] The responses were divided into five categories:

The difference is entirely due to environmental variation: 15%.
The difference is entirely due to genetic variation: 1% (8 respondents).
The difference is a product of both genetic and environmental variation: 45%.
The data are insufficient to support any reasonable opinion: 24%.
No response (or not qualified): 14%

In other words, the plurality of respondents, 45%, believe the causes of black-white racial differences in IQ are both environmental and genetic. This is three times the number who believe the differences are entirely caused by environmental factors.


Now.... Where's the actual evidence for this? That the gap is genetically determined AND heritable?

rwe,

I'm just arguing that IQ is not a good measure of genetic endowment

Yet another strawman. As far as I'm aware, no one here has asserted that "IQ is a good measure of genetic endowment." The hypothesis is that some part of the racial differences in IQ are genetic rather than environmental. We don't know how large the genetic component is.

To figure out what correlation exists between race and inborn intelligence, one would have to get a good measure of inborn intelligence and, as even you now seem to be condeding, IQ ain't it.

To repeat: Whatever you want to call the characteristic that IQ tests measure--"intelligence," "innate intelligence," "ability to pass IQ tests," or whatever else it may be--IQ test scores predict many life outcomes. Those who score higher tend to do better in school, better in work, have better health, and so on. Furthermore, some part of the differences in IQ scores seem to be genetic, so some of what causes whites to do better than blacks at school, at work, etc., seems to be genetic. It does not seem to be just a matter of environmental differences.

Now, as for Manzi's article, I'm not sure why you keep citing it.

Because he raises and responds to many of the talking points you have made here.

Mixter-
"Furthermore, some part of the differences in IQ scores seem to be genetic, so some of what causes whites to do better than blacks at school, at work, etc., seems to be genetic. It does not seem to be just a matter of environmental differences."

You're just asserting, or speculating, on this.


"seems to be" is not science.

You're just asserting, or speculating, on this.

I just cited a study of expert opinion that supports it.

"seems to be" is not science.

Nonsense. Every scientific theory is a statement of what "seems to be" true.

Seems I inadvertently posted an rorschach test and got a strong positive for personal prejudices. To review:

Mixner wrote: Ha ha ha. Good one. As Bertrand Russell said, "So far as I can remember, there is not one word in the Gospels in praise of intelligence." In America today, it's hard to think of a community more hostile to reason and learning than devout Christians...

Doomsday cults, neonazis, 9/11 Truthers, the Fundamentalist Church of Latter Day Saints, the rap music industry, and the Democratic Underground all slipped past your incisively judgmental nose, did they?

Interestingly, all I referred to was the "Judeo-Christian value system". You know, Thou shalt not kill, Thou shalt not bear false witness against thy neighbor, those kinds of things. Western Europe (thoughly highly secularized now) and the United States (though increasingly secularized) incorporated very large portions of Judeo-Christian tradition and principle into their law and culture, taught it to successive generations, and are very wealthy and properous today, in part, because of that tradition.

As for the Bertrand Russel quote, that's the kind of irrational non sequitur that one arrives at when a mind has been slowly consumed by a virulent prejudice. One might just as well thumb through The Joy of Cooking and then declare that "So far as I can remember, there is not one word in that book about how to check the automatic transmission fluid level in a '99 Buick