Wednesday, May 8, 2013

precisely how not to argue about race and IQ

So this Heritage Foundation immigration report, and its author's history when it comes to race and IQ, has prompted my politically-minded Facebook friends to give an absolute clinic on how not to respond on this issue. The response has been: this lunatic thinks that Hispanic Americans score lower on IQ tests and other standardized tests of intelligence than white Americans! How wrong! Only a racist could say that different races score differently on IQ tests. Which is playing precisely into the hands of the Steve Sailers of the world. The problem with people who argue for inherent racial inferiority is not that they lie about the results of IQ tests, but that they are credulous about those tests and others like them when they shouldn't be; that they misunderstand the implications of what those tests would indicate even if they were credible; and that they fail to find the moral, analytic, and political response to questions of race and intelligence.

Because this is true, folks: these instruments do find, according to their internal mechanisms, consistent racial differences. Take it from an official report (PDF) of the American Psychological Association— not, I hope we can agree, a racist organization. The report finds that standardized intelligence tests have found, consistently, that Asian Americans have a wider spread in IQ than other racial groups and a higher mean than other groups, that black Americans have a mean IQ that is a standard deviation below that of white Americans, and that Hispanic Americans lie in between the two. That is an accurate review of the extant literature. There is no more sense in denying perceived differences in IQ (not intelligence) between white Americans and Hispanic Americans than there is in denying a difference in high school drop out rates. The question at hand is the credibility of IQ itself.

My response is not to deny that these perceived differences are being accurately reported by the people who report them, but to say that the tests themselves are flawed and are not an accurate instrument for understanding real-world intelligence. I argue that the tests have a validity problem in that they don't correspond well to what we typically mean when we discuss "intelligence" in a lived sense, and a reliability problem in that the consistently perceived racial differences are not reflections of actual differences in intelligence but of systemic biases that render the metrics flawed. I also don't agree with the many ugly responses race realists have to the differences in IQ. But to say "there are perceivable differences in the results of IQ tests?" That's true, and it is treated as straightforwardly the case in peer reviewed literature in credible journals by responsible, non-racist researchers. Across a long time frame, between many different tests and many different administrations, those differences are perceived. What matters is that they aren't an accurate reflection of what human beings mean when they talk about real-world intelligence.

Racism thrives on conspiratorial thinking and the self-definition of racists as an oppressed group. When you say things that are true aren't, and especially when you do so in a way that treats the other point of view as forbidden, you play directly into their hands. I cannot imagine an easier way to give them fuel for their argument than to say that certain test results don't exist when they do. Perhaps it's easier to argue that way, and perhaps it's more emotionally satisfying, but it hurts the antiracist effort in the long run. More to the point: what are you so scared of? It amazes me how often I interact with white liberals who, despite being perfectly correct on the merits, talk about race in a state of absolute panic. I hate to cast aspersions but I sometimes suspect people I know secretly find the case against racism to be weak, and are afraid that if they have to argue, somehow, the racists will win.

Bullshit. The case against inherent racial inferiority is correct. The moral and analytic argument is on our side. You have to have the guts to confront the facts and make the case. Just as no one supposes that the racial achievement gap in grades, graduation rate, and college are somehow proof of racial inferiority, no one should mistake the perceived IQ gap as meaning something when it doesn't. Don't be afraid, and don't play their game. Stop getting panicky about race talk and engage. It's your moral responsibility.

30 comments:

JS Bangs said...

A part of this, I think, is that a lot of people don't know how to make the evidential argument against racism, or (worse) don't think that they should have to. They only know how to treat racism as axiomatically evil, and so even allowing the argument to be about evidence rather than moral absolutes is seen as somehow ceding ground to the racists.

L. Zambezi said...

Why define inferiority in terms of IQ? That's such a narrow definition. Instead of saying, IQ is invalid, say rather than IQ doesn't matter. IQ? So what? IQ doesn't matter. Asians are smarter than whites? So what. Defeat the Steve Sailers with a big so what. Likewise, don't try to stamp out every last trace of racism. Oh, you made a racist joke? I could hardly be bothered to notice. Like the blogger says, nothing ever gets repressed for long.

Derek said...

Arguing that IQ is totally uncorrelated with "real" intelligence is tantamount to saying that intelligence will remain forever impossible to measure. Virtually every standardized test ever given positively correlates with the results of IQ tests. For heaven's sake, Wechsler's itself is a battery of seemingly totally unrelated subtests (Vocabulary, Block Design, Digit Span) that ALL positively correlate with each other.

Moreover, the idea that the informal definition of intelligence used by ordinary people is superior to that used by psychologists is silly. The common sense definition almost always boils down to vocabulary and general knowledge of the world. Indeed, most people regularly conflate the meanings of "intelligence" and knowledge. What this means in practice is that we often underestimate the problem solving ability of quiet people who use language simply; people who could perform superbly on non-verbal IQ tests like Raven's Progressive Matrices. After a brief meeting with the two men, the great majority of people would claim that Christopher Hitchens was a smarter man than Terence Tao, but that just isn't so.

I don't really like getting theatrical and accusing my interlocutors of hypocrisy, but I am pretty sure that you, Freddie Deboer, regularly assess the intelligence of your peers using precisely the same method used by 99 percent of your fellow humans. If you do, it's kind of a problem, because your conventional assessment will almost certainly correlate positively with the results of IQ tests. If you use some unconventional method superior to all currently discovered, it's really your prerogative to bring it to the attention of psychologists, who have been trying and failing for decades to discover such a test.

66027fa8-b84b-11e2-af9f-000bcdcb8a73 said...

"I argue that the tests have a validity problem in that they don't correspond well to what we typically mean when we discuss "intelligence" in a lived sense"

I bet you a thousand Internet dollars that if you actually tested this out you would find it to be totally false. In fact, it would likely be a near perfect correlation.

Get a hold of a Wisc or WAIS (along with the scoring manual), and see what it really means to get a 120 or a 100 or an 80. If you're being honest with yourself, you'll likely come away thinking that there is no way anyone could score under a 90 and be considered intelligent in any real world sense of the term.

Unknown said...

Psychology is not yet a science. It would be cool if it was, but it's not. We have a long way to go.

For example: It is literally impossible for a human to create a culturally-blind test of any sort, let alone a test of intelligence--a term that Psychology has yet to define in a non-solipsistic fashion. Even tests that are supposedly non-verbal show significant cultural bias. (Psychology assumes that purely visual representations of objects are culturally neutral--they are not.)

Freddie is correct. Using standardized IQ tests as indicators of racial inferiority/superiority is something like saying "Leading Phrenologists Agree."

Thanks,

Ole

Unknown said...

Psychology is not yet a science. It would be cool if it was, but it's not. We have a long way to go.

For example: It is literally impossible for a human to create a culturally-blind test of any sort, let alone a test of intelligence--a term that Psychology has yet to define in a non-solipsistic fashion. Even tests that are supposedly non-verbal show significant cultural bias. (Psychology assumes that purely visual representations of objects are culturally neutral--they are not.)

Freddie is correct. Using standardized IQ tests as indicators of racial inferiority/superiority is something like saying "Leading Phrenologists Agree."

Thanks,

Ole

Unknown said...

...and, somehow, in spite of posting in an incompetent fashion, I managed to to miss the key point I wanted to make--that *even if these measures were significant*, they tell us nothing about how to treat other human beings.

Thanks,

Ole

Peter said...

I agree with the post, but I would go a little further and say that this is one area (the only one?) where I do not support free scientific inquiry. Even if the tests WERE valid and they WERE unbiased, I cannot see anything constructive coming out of this research. Ole's comment is right on:

"*even if these measures were significant*, they tell us nothing about how to treat other human beings."

...and given that, I think that while Freddie's right that questioning the methods is better in this case than denying the results, the best approach is just to refuse to play the game.

It's true that people who reflexively deny the results are implicitly endorsing the methods, but take it a step further: when you criticize the methods, you're implicitly endorsing the whole project of looking for racial differences in the first place. Why would you want to endorse that?

Peter said...

And just to add to my previous comment: I think we tend to reflexively accept certain research questions with race and gender that we would not accept with other categories. If someone said (in 2013) "look at my study showing that homosexuals have lower IQs," I suspect our first response would not be "your numbers are wrong" nor "IQ is a flawed concept" but rather "what's wrong with you that you even did that study?" I think we ought to apply that same reaction to race and gender.

Nathan Wright said...

Peter,

"I agree with the post, but I would go a little further..."

No, you completely missed the point of the post, because yours is exactly the kind of reaction that the post is pointing out.

Peter said...

Nathan, I don't think it is. If I understand the post correctly: Sailer et al. say there are innate racial differences in IQ, as shown by these studies. Standard liberal response is (A) the studies are flawed, the data is wrong

Freddie says (B) the studies are NOT flawed in that sense, but IQ itself is a biased measurement, and even aside from that it's not very useful.

I say ok, but we shouldn't even get that far because (C) it's not an acceptable topic of research, nothing good could possibly come out of it even if Sailer (and say Derek above) could make a persuasive case against Freddi here. So I'd rather just not engage at all.

Do you really not see a difference between A and C?

P. Adrian Frazier said...

Ole, if I am interpreting your argument correctly, it seems you are arguing that psychology is not a science because it has a problematic theory of intelligence (and perhaps other problematic theories). And yet theories in every science are problematic. Science progresses by discovering errors in theories and constructing new variations on older theories in an attempt to correct for them. Problematic theories do not make an inquiry non-scientific.

As for the "impossibility" of defining intelligence in a culturally neutral manner, even if that were true, such impossibility should, in principle, be explicable. While I have little doubt that much of what we take to be intelligence is very much cultural in nature, some account for the intelligence of pre-cultural humans (infants) and the intelligence of animals without the kind of robust social ontology of humans still has to be explained. I agree that the mainstream of intelligence research can be highly problematic, but I don't believe that, in principle, it cannot be explained in a culturally neutral manner, or at least in a manner that can account for the socially ontological nature of intelligence, if that is the case.

Lastly, I have to disagree with the claim that the research examining IQ scores across cultures is without value. For one thing, it generates a question—given whatever it is that IQ tests are measuring and its correlates, what can account for the differences? If the differences are "inherent," what does that mean and what are generative mechanisms that manifest such differences? If they aren't inherent, or are only partially inherent, then what else might participate in generating such differences? Seems like Freddie has a pretty good hypothesis regarding systemic inequality, but even that has to be explained, and I think the attempt to do so could be fruitful in terms of how we interact (in terms of, i.e., policy) with different groups. Group-averaged IQ scores, beyond the psychological constructs they are supposed to be measuring, might well serve as an indicator of a system working or not working for a group of people, and we could use the many correlates with IQ as clues about what to investigate if we are to determine how the system differs for one group versus another and investigate means by which to effectively alter the dynamics generating those differences.

P. Adrian Frazier said...

One more thought on this. Let's suppose that the differences in IQ are something "biological" in the sense that some people have the right kind of effectivities to manifest high scores on IQ tests, and IQ is measuring something that is only important to certain groups of people (namely, those who have high IQs). Or perhaps high IQ is a peculiar manifestation of particular cultures. That is, suppose that the utilization of high IQ scores in intelligence assessment is simply a matter of values rather than anything of "objective" import. Even if that were the case, investigating the manner in which such valuation itself generates differences in other domains (economic outcomes, for instance) can be useful if we are to understand how the dynamics of our society generate systemic differences.

Philosopher said...

Aren't you making the same mistake as your FB friends by disregarding the research on the validity and reliability of IQ? I used to study this, and there have been tons of studies on those. Here's one summary I found with a Google search: http://www.iq-tests.eu/iq-test-Practical-validity-800.html

Here's the real issue: the conflation of intelligence and moral value. Sure, intelligence is nebulous, but I challenge you to find a more rigorous construct in psychology than IQ. This data troubles us because it may seem to challenge egalitarianism. But was the case for equality ever premised on equal abilities?

Humans have equal value morally and should be treated equally under the law. Only in cases in which low intelligence impairs functioning should we treat individuals systemically differently. Otherwise, intelligence as measured (robustly but imperfectly) by IQ should have no bearing ethically.

Peter said...

Adrian: "I have to disagree with the claim that the research examining IQ scores across cultures is without value."

It's not that I think it's totally without value, and you're making a pretty good case for it. I guess I'm saying that the bad outweighs the good. And this:

"...could be fruitful in terms of how we interact (in terms of, i.e., policy) with different groups"

is probably the core of our disagreement. I cannot conceive of any policy decision that takes racial IQ differences into account in any way that wouldn't do more harm than good. The whole subject is so toxic, such a lightning rod, I just don't think we should study it.

Obviously reasonable people can disagree about this, and I don't like being against free scientific inquiry on any topic. But I think this one is just too fraught. The points that Philosopher is making (intelligence has no moral content, it's not the premise of egalitarianism) are certainly true, but I can't see them surviving the transition from the classroom to the real world. In the real world of politics and even just news, results showing racial differences in IQ -- on whatever metric, with whatever qualifiers -- the ways they could make things better seem a lot less likely than the ways they could make things worse.

quixote7 said...

Ole hits the nail on the head. Intelligence is an extremely complex property, and one whose desirable characteristics change depending on the kind of smarts needed for the task at hand. It's not even close to being neurologically characterized.

Like the old story about blindfolded people trying to describe an elephant, psychologists are groping around the elephant's foot. The fact that they may be doing an okay job on that foot, doesn't change the point that they haven't even completely defined intelligence yet, let alone figured out how to measure it.

The most important point is: "*even if these measures were significant*, they tell us nothing about how to treat other human beings."

All that said, it's still foolish to say "Don't even look at this. It's against human rights to do so." Facts are always worth studying. The facts can never damage human rights because they tell us nothing about how to treat other human beings."

sl4irl said...

'I agree with the post, but I would go a little further and say that this is one area (the only one?) where I do not support free scientific inquiry. Even if the tests WERE valid and they WERE unbiased, I cannot see anything constructive coming out of this research.'

What a thoroughly myopic view of research. You don't only perform research because you want to 'construct' something off the back of it, research has its own value.

Anyway. Freddie is correct. IQ is preposterous as a way to determine a concept which arose historically, (and is continuing to develop) in a fashion which elides being pinned down by social scientists.

However, I think it's worth mentioning that a lot of the behaviour noted in this post is tactical: sure racism comes out after two beers, but that's after a hefty struggle to stop it coming out immediately.

I'm not saying that that's some towering achievement, just that explicit racism being closed out of polite discourse is a front of the battle against racism in itself. A lot of that means flat out ignoring people, not engaging with them, and generally treating them as crazy outliers which can be forgotten about.

So they'll bubble about on Stormfront or whatever, or commenting on Steve Sailer blogs, but rarely appear in mass publications. It's a valuable enough tactic, and it can be applied to all sorts of things (atm in this country, for example, a loud young social democrat with not especially well developed politics has carved out a hefty niche by being basically the only person ready to stand for socialism and the poor...and that's because of explicit socialism basically vanishing from our discourse.

TGGP said...

The hypothesis that IQ tests are racially biased (bizarre considering the wide range of disparate tests even designed to do the opposite which keep giving the same sort of result) might expect you to believe that it underpredicts the performance of low-scoring groups and overpredicts the performance of high-scoring groups. So black students will do better in college than their test scores would lead you to expect, and asians will do worse than expected. In fact, the opposite is the case. The validity of IQ tests is something that can be empirically investigated, and Freddie cited no evidence for his interpretation. The anti-Bell Curve scholars like Flynn & Nisbett don't bother arguing that IQ tests are invalid measures (they know their predictive power), they argue over the source of group disparities. Flynn, for example, argues that socialism is necessary to correct for environmental disparities and has a study on the children of servicemen in Germany as evidence against Charles Murray.

I mentioned earlier that test scores systematically err in predicting the academic performance of some groups, and it's worth revisiting the point. IQ is certainly not the only thing, there are other traits like conscientiousness that seem important (James Heckman has focused on its importance and how to alter it in young children). But there's a huge gap between "only thing" and "nothing", it's something and that something is important. Its predictive power in a range of life outcomes is quite high for the social sciences. And that importance is why so many people have devoted so much effort to raising scores.

Fall in queue said...

@ Philosopher: Humans have equal value morally and should be treated equally under the law. Only in cases in which low intelligence impairs functioning should we treat individuals systemically differently. Otherwise, intelligence as measured (robustly but imperfectly) by IQ should have no bearing ethically.

It is telling that this is Peter Singer's argument for animal rights. The most you can get from this argument is that everyone is entitled to have their welfare taken into consideration---and this is all that Singer aimed to show with respect to animals. But this is a far cry from what egalitarian ideals require when it comes to people. It is far from obvious that things like equal rights of democratic participation, for example, do not presuppose roughly equal distributions of cognitive ability.

@Derek, @TGGP: One might perfectly consistently acknowledge that intelligence tests correlate with each other, and with academic tests, and yet deny that there is a single factor (general intelligence) that all these tests measure. See the wonderful (though technical) discussion by Cosma Shalizi here: http://bit.ly/10o0KBA

Bonus quote from the conclusion:

"The mythical aspect of g [general intelligence factor] isn't that it can be defined, or, having been defined, that it describes a lot of the correlations on intelligence tests; the myth is that this tells us anything more than that those tests are positively correlated. It has been known for almost as long as factor analysis has been around that positive correlations can arise in many ways which involve nothing remotely like a general factor of intelligence. It is still conceivable that those positive correlations are all caused by a general factor of intelligence, but we ought to be long past the point where supporters of that view were advancing arguments on the basis of evidence other than those correlations. So far as I can tell, however, nobody has presented a case for g apart from thoroughly invalid arguments from factor analysis; that is, the myth.

In primitive societies, or so Malinowski taught, myths serve as the legitimating charters of practices and institutions. Just so here: the myth of g legitimates a vast enterprise of intelligence testing and theorizing. They are supported only by the myth, and acceptance of the myth itself rests on what I can only call an astonishing methodological backwardness."

Fred Smith said...

One word I never hear the Steve Sailers of the world use when discussing race and IQ is "inherent" and it does seem like that word is used by his opponents to confuse or mischaracterize the position. The idea is that the groups of people that we conventionally think of to constitute different races have different average IQs. But any one individual from any race can have any IQ in the spectrum, and indeed high IQ people represent every race. This seems (to me) like a very different idea than "hispanic people are inherently less intelligent", which is how I've seen it characterized.

One other point to mention, several people above have argued that data about race/IQ cannot possibly lead to any good public policy. The Steve Sailers of the world generally argue that bad public policy is already in place, and that data about race/IQ can help to inform some of those problems. For example, the persistent "gaps" that that are measured in the performance of students, or discrimination lawsuits premised on disparate impact. Eliminating disparate impact as a theory of discrimination and lightening up on some of the attempts to stamp out "gaps" are about the only policy prescriptions I've seen from the likes of Sailer. Maybe those are the horrible ideas some of you are worried about, but they don't seem that horrible to me.

TGGP said...

Fall in Queue, I've done a text search through my comment and don't see where I even mentioned 'g'. I haven't read Richwine's dissertation, but it seems to depend on the simple finding that IQ results are predictive rather than whether it's a single or multiple-factored thing. As noted in the intro to "The Bell Curve" all those years ago, the issue of how many factors are involved isn't actually that important to many controversies.

The Human Varieties blog published a response to Shalizi recently, resulting in an interesting back and forth with the Noahpinion blog. They get into what implications (or lack thereof) there would be from a multiple-factored IQ that still gave the same empirical results we see.

Fall in queue said...

TGGP, thanks for the links. I really can't comment on the design of actual intelligence tests.

However, whether the single g theory or what this Dalliard guy calls 'sampling' is true matters a lot, both for how we conceptualize intelligence and for practical purposes. To paraphrase Tolstoy, if there is a single g then every low intelligence person must have the same defect, whereas if the sampling theory is true then every dumb person might be dumb in their own special way. This diversity would be masked by intelligence tests that cross-sample these modules.

Indeed, each such person might have a high level of ability in one or more areas, cancelled out by low levels of ability in others (your performance in each test being some linear combination of a bunch of these independent abilities). This would be a problem in practice, because people with this spread of abilities might be helped by narrowly targeted interventions, but if we rely only on general intelligence measures we would have no way of knowing that.

Dalliard's arguments against this seem pretty weak. First, his empirical arguments rely on tests that seek to explain intelligence using between 3 to 9 dimensions, all of them exceedingly general in themselves. I have no expertise on intelligence tests, but a truly modular picture of cognition would rely on many, many more modules, which would not necessarily map onto intuitive categories such as "verbal processing" or "visual processing".

And Dalliard's conceptual argument against sampling is just silly. He basically claims that even if something like sampling were the correct underlying picture of cognition, then we can still talk about general intelligence at the behavioral level. Well sure, if all you want is license to call people "smart" or "dumb" then go ahead and do it. The question is whether this corresponds to some underlying ability or not.

Alan said...

This area of inquiry truly is toxic and fraught with all kinds of troubling implications. However, I do not feel comfortable with the suggestions to sweep it under the rug.

I feel somehow compelled to establish some bonafides so as not to be lumped in with the slimier subterranean types that often inhabit the "race realist" side of the question. I voted for Obama, twice, and contributed money to his campaign both times. I consider Ulysses Grant our greatest historical president, and Thaddeus Stevens one of the greatest figures in American history, period. I only wish they had both had a longer reign in power to more fully crush the KKK and prevent Jim Crow.

I also agree that IQ tests, as presently composed, are woefully inadequate in summing up human intelligence. Gardner's multiple intelligences would be a good start in expanding from that "elephant's foot" on to other areas (including some where African Americans do excel); and surely there would be more profitable discovery in this area than what Gardner came up with in a sort of first stab.

However.

I am someone who scores very highly on IQ tests (and their cousins, the college board exams ACT and SAT). Highly enough, in fact, to qualify for membership in this organisation: http://www.triplenine.org/ As such, I can fairly be accused of being biased toward wanting to ascribe some legitimacy to the aforementioned test results. But while it is clear that acing these tests does not prove a person has command of every facet of superiour human cognition, it is the inverse proposition that I find more significant.

Which is to say: if you do well on such tests, you may or may not have intellectual deficits the test failed to detect. If however you can't do well on such tests, you are absolutely proven to have glaring intellectual deficits of an important kind, even if you may or may not have other hidden talents the test results failed to reflect.

I would argue that it is not only that, as a previous commenter noted, it would be very difficult to characterise someone scoring below 90 as someone we could reasonably call intelligent; but that even someone who scores around 120 (at the 90th percentile, roughly equivalent to scoring a 28 composite on the ACT) is intellectually mediocre in a pretty glaring way from my perspective. I can't even comprehend looking at the questions and failing to correctly answer a sufficient number of them to score that low. Not even if I were hung over, operating on very little sleep, etc.

I'm well aware this makes me sound like an arrogant jerk. But I do possess humility, believe me. A great love of mine is the game of tennis. I have played avidly for many years and am certainly better than the average schmoe. However, when I watch the Div. II college players practice on an adjacent court, it is painfully clear that I have never been anywhere near that good and never will be that good no matter how much I wish otherwise (and I do so wish, believe me). Some of us have exceptional abilities in one area, others in another. C'est la vie.

So I guess the point of this admittedly rambling comment is that it makes as much sense to dismiss differences in IQ as "meaningless" as to dismiss differences in running speed or tennis ability. For those people who are engaged in the activities measured by a tennis score, a track time, or an SAT score, the differences are persistent, real, and yes: meaningful.

Steve Sailer said...

http://isteve.blogspot.com/2013/05/coates-race-is-social-construct.html

Crispus Attucks said...

Even if we all agree that standardized intelligence tests have found consistently that black Americans have a mean IQ that is a standard deviation below that of white Americans, I still don't understand what conclusions we can draw about race and intelligence (or even IQ for that matter). What are the differences in IQ scores between African-Americans who descended from Slaves versus African Americans who did not? What are the differences in IQ scores between African-Americans and Black Europeans or Black Africans? And no need to limit it to black people. What are the IQ differences between Northerners and Southerners?

Would anyone be surprised if there were IQ score differences between people from Connecticut versus Mississippi?

If that is the case then what conclusions can I really draw regarding race and IQ?

Alan said...

Good questions, Crispus (nice historical reference). I followed Steve Sailerr's link and challenged him to look at other IQ disparities than just the white vs. black one. In particular, I challenged him to a thought experiment: what if only those above the 95th percentile in IQ were allowed to vote? i submit that this would not be a conservative electorate.

John Spragge said...

Derek, intelligence may or may not prove "forever" imposiible to measure. Neither I nor anyone else can predict that. I can tell you, however, that we cannot truly measure any aspect of human cognition without understanding it. To measure and understand what you measure, you need a model, a standard. And if we have a model that we clearly understand, we can encode it as a mathematical entity, and any aspect of cognition we can encode mathematically, we can program, anything we can program we can commodify, and then that aspect of cognition will cease to have a human value. How can any aspect of "intelligence" matter if your phone can duplicate it?

So I can't predict that we shall never measure intelligence, but I can tell you that it won't matter the way it does now if we ever succeed in measuring it.

Peter said...

I obviously didn't convince anyone in this thread of my position that research on racial/gender intelligence difference should simply be off-limits, but I'm dropping back in to link to John Horgan's latest column at Scientific American, which makes the same argument a little more articulately:

Should Research on Race and IQ Be Banned?

Alan said...

I don't agree with actually banning such research (and I'm not sure Horgan is truly advocating that); but I do agree with him that the people who do this research and publicise it are basically acting as bullies, rubbing salt in the wound for no real purpose.

Except that there is one area where this research has important real world implications. My wife is a special education teacher, and her profession is under siege, with education critics endlessly accusing teachers of failing their students and using poor test scores as their evidence. If an allowance were made that heavily minority schools are going to have lower test scores even with good teachers, that strikes me as being more fair to all the teachers out there who are doing their best.

Incidentally, what do you suppose Chomsky meant when he said the following?

"Surely people differ in their biologically determined qualities. The world would be too horrible to contemplate if they did not."

John Spragge said...

Peter: I wouldn't ban this research, since I don't believe banning anything works. But I do agree that any decent person or institution has an obligation to acknowledge the ethical taint to this field. Eugenics programs in Nazi Germany developed both the ideology and the technology of mass murder that reached its full expression in Auschwitz. Proponents of research on the relation between "race" and "intelligence" have an obligation to show an actual benefit that outweighs that history.