Mensa for Dummies

John Scalzi posted a grab of one of his tweets this morning:

It was prompted by the ongoing Sad Puppies vs. Social Justice Warriors "kerfuffles surrounding science fiction and its awards, there have been a couple of people (and their spouses, declaiming about their beloved) who have been slapping down Mensa cards as proof that they (or their spouse) are smart."  Scalzi explained, in his trademark style, why doing this tends to prove the opposite.  For example:

Your Mensa card does not mean you know how to argue. Your Mensa card does not mean you do not make errors or lapses in judgment. Your Mensa card is not a “get out of jail free” card when someone pokes holes in your thesis. Your Mensa card does not mean that you can’t be racist or sexist or otherwise bigoted. You may not say “I have a Mensa card, therefore my logic is irrefutable.”

Good enough.  The comments under his post are another matter, however.  They fell into two main groups. In one group, the commenter would mention that he or she had attended Mensa meetings, even joined for a while, and found the people in the organization to be mostly pretty nice people.  The other group declared that they'd never joined or gone to a meeting, but all the Mensa members they'd met were jerks.  I found this latter group fascinating, because despite their evident conviction of their own superior intelligence, they were making a fundamental logical mistake, one that Scalzi himself didn't: they were generalizing an entire group based on their experience of a few, probably unrepresentative, members.  Analogous stereotypes are "All the Christian fundamentalists I know are hypocrites," "Did you ever see a fag who wasn't effeminate?" (actual example), "All heroin addicts started out on pot, so smoking pot will turn you into a heroin addict."

(Just for disclosure's sake, I have never joined Mensa or gone to a meeting.  The Mensa members I know in person are quite nice and bright people, and the Mensa jerks I've encountered were all online, trying to establish their intellectual credentials by bragging about their IQ scores or their Mensa membership.)

Some of the discussion focused on IQ tests and SATs.  Several commenters pointed out the uselessness of IQ tests as a measure of intelligence.  One riposted:

IQ tests (what Mensa uses) are tests of aptitude. They are basically measuring how easily and quickly you will learn and absorb concepts of all types, and solve new problems. How accurate they are is almost beside the point because really they are irrelevant in most situations including arguments about topics.

How easily you could learn is not a measure of how much you know.
If two people sit down to learn a skill and one can attain expertise in 1 hour and the other needs 1.5 hours that is interesting. However if the first person never spends the hour learning the skill then the second person is absolutely the one you want around when you need that skill set.

IQ tests do not measure aptitude.  As far as I know, no one knows how to do that.  IQ tests mostly measure what you already know, or know how to do.  I last took an IQ test in high school, and I don't recall any part of it devoted to how quickly I could learn a skill; nor, from what I've read about the IQ controversies, has such an exercise become part of the test since then.

Similarly, the SAT, which was based on the Stanford-Binet IQ tests, was originally "called the Scholastic Aptitude Test, then the Scholastic Assessment Test, then the SAT Reasoning Test, and now simply the SAT."  The College Board. who owns the franchise, explains that it "tests the skills you’re learning in school: reading, writing and math. Your strength in these subjects is important for success in college and throughout your life," which sounds like what used to be called an achievement test.  It is not an aptitude test, and it's not even a very good predictor of college success, though that's its usual rationale.  This commenter's confident assertions are interesting; they seem to have no basis in fact, and I wonder where he or she got them.

Another commenter, a former Mensa member, wrote:

I studied rhetoric in school and my mom was a physicist; what I learned from this background is that the way to persuade people is to provide relevant and verifiable evidence.

I think this person may be confusing "is" and "ought."  I also value relevant and verifiable evidence, but I've learned to my disappointment that many, even most people, don't.  The way to persuade people in the real world appears to be to assert, as loudly as possible, that your opponent is fat or Republican or a libtard or a funditard or an asshole.  This approach is more "natural," and much easier.  It's also more effective, from what I see.

For example, this morning a liberal / progressive friend of a friend shared this meme on Facebook:

According to Snopes, Palin didn't say this and wasn't even on Hannity on that date.  I pointed this out in a comment on the Facebook post, exulting sarcastically that liberals aren't gullible or dishonest like Republitards.  Of course the person who'd posted was displeased -- she reacted exactly like the right-wingers I know react when I point out that they've posted something bogus, asking why I was on her timeline and getting indignant about my meddling.  Mockery is a very private thing, especially when you're posting it in public on Facebook.  One hears that social media are an echo chamber, that people want to engage only with people who share their politics; to a great extent that's true, as this person showed.  And I suppose we need places where we can find others who share our opinions and prejudices, but we also need to engage with people who don't, or the social and political changes this person hopes for will never happen.

Back at Scalzi's blog, the same commenter continued:

Anti-intellectualism is hardly the worst form of prejudice, but I know people who have been hurt. Also it’s like fat-shaming; we’re not a protected class and some people think it’s okay to show disrespect.

This lament was oddly off-topic.  The Sad Puppies clearly see themselves as intelligent, and intelligence of certain kinds as important and a sign of one's value.  They may well be anti-intellectual, since they associate what they call Social Justice Warriors with a kind of pointy-headed intellectualism that is widely devalued and mocked by people who think themselves intelligent.  "And let’s be honest — we all know someone who’s pretty book-smart and pretty life-stupid," wrote another commenter, providing an example of this distinction.  I can't recall where, but not too long ago I read something where the writer distinguished between being intelligent and being an intellectual.  I think of an intellectual as someone who works with more or less abstract ideas; an engineer or other scientist may be highly intelligent but no good at dealing with ideas, and dismissive of those who can.

As for the rest of his remarks: Being in “a protected class” doesn’t mean that others can’t “show disrespect” to you, nor should it. “Protected class” is a problematic legal term which means that the law will protect you from certain specified and more-or-less carefully defined forms of discrimination. But showing disrespect is fine, and hardly anyone really believes that it isn’t — except disrespect to themselves. For example, almost everybody wants respect for their religious affiliation, and discrimination based on religion is forbidden by Civil Rights law in certain spheres. But just about everybody has some religious class — liberals, fundamentalists, “Cafeteria Christians,” etc. — they love to mock and disrespect, and they’d be outraged if anyone told them not to. And the other part of the First Amendment guarantees our right to do so, as it should.

So sure, it’s perfectly okay to show disrespect to intellectuals, or to the intelligent.  It's not necessary to define bookish kids as a "protected class" to protect them from the bullying they too often face at school.  But kids who aren't "smart" also face bullying and contempt at school, including from their teachers, and they also need help from those around them.  If anything, they are probably more vulnerable than the smart kids: I know people who've been hurt.

I've mentioned before the graduate student I once knew who told me, sweetly and almost shyly, “I don’t say this to many people, but I think of you as my intellectual equal.” I thanked him, embarrassed, because I realized that though I hadn’t thought about it before, and don’t go around making such comparisons in the first place, I didn’t consider him my intellectual equal.  But, as Scalzi noted this morning, what he said revealed more about him than it did about me.

Credit where credit's due: I stole this post's title from another of Scalzi's commenters.

Traditional Intelligence Researchers Strike Back

Glenn Greenwald recently linked to a post by Andrew Sullivan lamenting liberals' lack of love for President Obama (hey, you "conservatives" can have him, Andy!). After I'd looked it over I noticed that Sullivan had written the day before about "The Study of Intelligence." I knew he'd had some stupid things to say about that in the past, so I decided to bring myself up to date.

Sure enough: The study of intelligence has, according to Sullivan, "been strangled by p.c. egalitarianism." Anyone who's still using "politically correct" seriously is a fool; in his boot-licking piece on Obama, Sullivan complained "If I hear one more gripe about single payer from someone in their fifties with a ponytail, I'll scream." Well, if I hear one more gripe about "p.c." from a balding right-winger with a goatee ...

Sullivan went on: "The reason is the resilience of racial differences in IQ in the data, perhaps most definitively proven by UC Berkeley psychologist Arthur Jensen", and quoted the article he'd linked on the subject:

"Jensen is still greatly respected by many traditional intelligence researchers," Garlick says. "By 'traditional intelligence researchers,' I mean researchers who still value IQ and continue to do studies that evaluate the effectiveness of IQ in predicting outcomes, or studies that examine possible mechanisms that may cause differences in IQ. However, due to the unpopularity of Jensen’s findings, this group of researchers is now very small.

Well, no, the reason isn't "the resilience of racial differences in IQ in the data", it's doubts about the validity of IQ as a concept. (A good place to start informing yourself on this would be Stephen Jay Gould's The Mismeasure of Man.) The writer of the Alternet article complains, "Somewhere along the way, the very idea of intelligence became politicized." Nonsense. The very idea of intelligence was politicized all along (and Sullivan hopes to continue the grand tradition), but especially since the original Binet intelligence tests were modified for use in the US by researchers at Stanford. "Its legitimacy as a field of study, as a measurable quality -- on par with height, eyesight and hand-and-eye coordination -- and as a concept came under fire." Well, there's your problem right there: you can't assume that intelligence is a trait like height, you have to prove it, and attempts to prove it have failed. Significantly, neither Sullivan nor the Alternet article even mentions The Bell Curve, a reminder that biological-determinist claims about intelligence and IQ could still get a rapturous reception in the US as late as 1994. Yet again, Andrew Sullivan shows that not knowing what he's talking about never inhibits him from taking a firm stand.

"The right response to unsettling data is to probe, experiment and attempt to disprove them - not to run away in racial panic. But the deeper problem is that the racial aspects of IQ have prevented non-racial research into intelligence, and how best to encourage, study and understand it," Sullivan concluded. Given the deranged lies Sullivan likes to tell about Noam Chomsky, I can't resist quoting what Chomsky wrote on this very subject forty-odd years ago:

In fact, it seems that the question of the relation, if any, between race and intelligence has little scientific importance (as it has no social importance, except under the assumptions of a racist society). A possible correlation between mean IQ and skin color is of no greater scientific interest than a correlation between any two other arbitrarily selected traits, say, mean height and color of eyes. The empirical results, whatever they might be, appear to have little bearing on any issue of scientific significance. In the present state of scientific understanding, there would appear to be little scientific interest in the discovery that one partly heritable trait correlates (or not) with another partly heritable trait. Such questions might be interesting if the results had some bearing, say, on some psychological theory, or on hypotheses about the physiological mechanisms involved, but this is not the case. Therefore the investigation seems of quite limited scientific interest, and the zeal and intensity with which some pursue or welcome it cannot reasonably be attributed to a dispassionate desire to advance science. It would, of course, be foolish to claim, in response, that “society should not be left in ignorance.” Society is happily “in ignorance” of insignificant matters of all sorts. And with the best of will, it is difficult to avoid questioning the good faith of those who deplore the alleged “anti-intellectualism” of the critics of scientifically trivial and socially malicious investigations. On the contrary, the investigator of race and intelligence might do well to explain the intellectual significance of the topic he is studying, and thus enlighten us as to the moral dilemma he perceives. If he perceives none, the conclusion is obvious, with no further discussion.

... The question of heritability of IQ might conceivably have some social importance, say, with regard to educational practice. However, even this seems dubious, and one would like to see an argument. It is, incidentally, surprising to me that so many commentators should find it disturbing that IQ might be heritable, perhaps largely so. Would it also be disturbing to discover that relative height or musical talent or rank in running the hundred-yard dash is in part genetically determined? Why should one have preconceptions one way or another about these questions, and how do the answers to them, whatever they may be, relate either to serious scientific issues (in the present state of our knowledge) or to social practice in a decent society? [from For Reasons of State, Pantheon, 1973, p. 361-362]

More or less Sullivan's position, you can see, only better informed and more intelligent. I also want to quote this passage from the same essay:

Similarly, imagine a psychologist in Hitler's Germany who thought he could show that Jews had a genetically determined tendency towards usury (like squirrels bred to collect too many nuts) or a drive towards antisocial conspiracy and domination, and so on. If he were criticized for even undertaking these studies, could he merely respond that “a neutral commentator ... would have to say that the case is simply not settled” and that the “fundamental issue” is “whether inquiry shall (again) be shut off because someone thinks society is best left in ignorance”? I think not. Rather, I think that such a response would have been met with justifiable contempt. At best, he could claim that he is faced with a conflict of values. On the one hand, there is the alleged scientific importance of determining whether in fact Jews have a genetically determined tendency towards usury and domination (an empirical question, no doubt). On the other, there is the likelihood that even opening this question and regarding it as a subject for scientific inquiry would provide ammunition for Goebbels and Rosenberg and their henchmen. Were this hypothetical psychologist to disregard the likely social consequences of his research (or even his undertaking of research) under existing social conditions, he would fully deserve the contempt of decent people. Of course, scientific curiosity should be encouraged (though fallacious argument and investigation of silly questions should not), but it is not an absolute value [360].

One major problem with Sullivan's claim is that it was researchers like Jensen, who "politicized" the study of intelligence by tying it explicitly to race. (I remember Sullivan himself speculating, in a piece for the New York Times Book Review, that East Asians might have higher IQs than whites, based on their test scores and school performance in the US. He had no evidence for the notion, he just liked the idea for some reason.) In a racist society like this one, such claims can never be innocent. As the Alternet article admits, there has been plenty of other research on intelligence in recent decades; I can only conclude that Sullivan doesn't like it because it's not politicized enough to suit him.

Sullivan also assumes that racial differences in IQ can be dealt with only by "disproving" them. As Chomsky indicates, though, they're not of any scientific importance, and they can only have political importance to racists. Political equality isn't tied to IQ scores or even a hypothetical better measure of intelligence. So, why is Sullivan so obsessed with the question?

From there I somehow found myself reading Michael Ruse's blog / column at the Chronicle of Higher Education. (You'll remember Professor Ruse as the man who believes that rape means "fucking the female if [a man] gets the chance," without asking another male's permission.) In a post on Darwin, Ruse wrote:

About thirty years ago, there was a huge debate among evolutionists about the applicability of the theory to social behavior, particularly to human social behavior. Arrayed on one side were a number of scientists, including our own David Barash. They argued that using Darwinian Theory casts considerable light on human behavior and society. Arrayed on the other side were many other scientists, including the late Stephen Jay Gould. They argued that Darwinian Theory does nothing but uphold the status quo – capitalist, racist, sexist. At least [Robert J. Richards] and I like each other and respect each other. Barash versus Gould put one a bit in mind of American politics. And I am not sure that ultimately they were (or would be now) any closer to settling their differences.

Ruse's characterization of Stephen Jay Gould is false; considering that Ruse lived through the period he refers to and participated in those debates, I don't think it's going too far to call it a lie. (Ruse and Gould also participated as expert witnesses for evolutionary theory against Creationism on at least one occasion.) The two positions Ruse sketches so casually aren't even really opposites. Gould and those unnamed "many other scientists" argued that their contemporaries who tried to apply Darwinian Theory to human behavior and society were misrepresenting Darwin, and doing bad science. I haven't read much of Barash's work, but Gould was a very civil debater, not like "American politics" at all. It was some of his adversaries, like Richard Dawkins, who acted like the Tea Party Movement. The same is true of Gould's colleague and ally, Richard C. Lewontin, as you can see by reading his articles from The New York Review of Books and the ensuing exchanges with his critics, collected in It Ain't Necessarily So: The Dream of the Human Genome and Other Illusions (New York Review Books, 2001); he concentrated on scientific questions while his opponents, like Stephen Pinker, simply red-baited him.

American Atheists have re-entered the lists for this year's War on Christmas season, with a series of billboards (h/t to JV). "37 Million Americans Know MYTHS When They See Them," reads one, with pictures of Poseidon, Jesus, Santa Claus, and Satan. Sigh. I'm a bit more careful in my use of words like "myth," but leave that aside for now. The point I want to make now is that scientists and secularists keep coming up with myths of their own, like Ruse's about the debates over evolution and social behavior. It's a myth in the strict sense: a story meant to express certain truths (or truthinesses) about a society, thereby promoting solidarity within the group. In the broader, less careful sense, it's a myth in that it's clearly and simply false. Sullivan is religious, of course, but his account of the study of intelligence, which appeals to science, is also a myth in both of those senses.

CNN: Like The Onion, Only Subtler

This article at CNN.com, reporting on a study which purports to show that American atheists and liberals have higher IQs than the religious and conservative, was noticed both by Roy Edroso and Avedon Carol. Edroso disliked it, appealing to the authority of PZ Myers to support him, while Avedon apparently liked it, at least "the part where religion is a feature of paranoia."

Myers dissects the study easily:

Show me the error bars on those measurements. Show me the reliability of IQ as a measure of actual, you know, intelligence. Show me that a 6 point IQ difference matters at all in your interactions with other people, even if it were real. And then to claim that these differences are not only heritable, but evolutionarily significant…jebus, people, you can just glance at it and see that it is complete crap.

At least the study is going to be published in a peer-reviewed journal, for what little that is worth. Myers might also have pointed out how easily average differences are absolutized, so that "atheists have higher IQs than believers" becomes "atheists are smart and believers are stupid." Edroso falls afoul of that mistake himself, brandishing "my experience of actual human beings, whom I have found fascinatingly varied in their abilities whatever their respective races, ideologies, etc." One of Avedon's commenters (sorry, no permalink) says the same thing: "In my church, there are many people with advanced degrees, including in the physical sciences. Everyone accepts the validity of evolution, seeing it as the tool by which God shapes things. Scripture is treated as sacred, but not literally inerrant." Well, duh! The trouble with statistics generally seems to lie with people who don't know how to read them. To say, for example, that men are on average taller than women does not mean either that all men are taller than all women, or that all men are the same height and all women are the same height. So a higher average IQ for atheists does not imply that there aren't believers with high IQs and advanced degrees -- or that there aren't plenty of unintelligent atheists out there. Statistics is a collection of methods for trying to find patterns in all that lovely variety; it doesn't eliminate the variation.

If not for the ScienceDaily story Myers linked to, though, I'd find it hard to believe that the CNN story isn't satire. Consider the "highlights" sidebar: "Behaviors may stem from desire to show superiority or elitism, which also has to do with IQ ... None of this means that humans are evolving toward a future where such traits are the default". Or the caption under the included photo: "The IQ differences are statistically significant, but experts say the data shouldn't be used to stereotype or make assumptions." Of course that's exactly how a lot of people are using the data, just as I would expect.

Who are those "experts," I hear you ask? Here's one:

"The adoption of some evolutionarily novel ideas makes some sense in terms of moving the species forward," said George Washington University leadership professor James Bailey, who was not involved in the study. "It also makes perfect sense that more intelligent people -- people with, sort of, more intellectual firepower -- are likely to be the ones to do that."

Bailey also said that these preferences may stem from a desire to show superiority or elitism, which also has to do with IQ. In fact, aligning oneself with "unconventional" philosophies such as liberalism or atheism may be "ways to communicate to everyone that you're pretty smart," he said.

Why CNN asked a "leadership professor" to comment on the study is hard to say, unless they were deliberately looking for uninformed, trivial, even irrelevant soundbytes. If so, Professor Bailey gave them what they were looking for. Species don't 'move forward' in Darwinian evolution, and as Myers pointed out, there's no reason to believe that either atheism or liberalism has any evolutionary significance. And that "neener neener neener, you atheists think you're so smart!" jeer is not just irrelevant but hilarious.

At least CNN talked to Satoshi Kanazawa, the author of the study.

Religion, the current theory goes, did not help people survive or reproduce necessarily, but goes along the lines of helping people to be paranoid, Kanazawa said. Assuming that, for example, a noise in the distance is a signal of a threat helped early humans to prepare in case of danger.

"It helps life to be paranoid, and because humans are paranoid, they become more religious, and they see the hands of God everywhere," Kanazawa said.

Wait a minute. According to Kanazawa, atheists aren't "paranoid" enough, apparently because of our higher IQs, and therefore not sufficiently wary of danger. (Avedon, you'll remember, liked the link to paranoia, but paranoia is not the right word for alertness to danger, especially when the environment really is out to get you.) To add to the weirdness, PZ Myers has some links that indicate that Kanazawa is quite a wack job himself; try this one, which quotes Kanazawa to the effect that "we" (the US) "are losing this war" (World War III, as Thomas Friedman calls it) because "we don’t hate our enemies nearly as much as they hate us ... And [in past wars] we didn’t think twice about dropping bombs on them, to kill them and their wives and children. (As many commentators have pointed out, the distinction between combatants and civilians does not make sense in World War III, and the Geneva Convention -- an agreement among nations -- is no longer applicable, because our enemies are not nation states.)" And that's why Kanazawa thinks that "this war" is taking longer to win than World Wars One and Two (though he evidently forgets that the War in Vietnam lasted longer than those two put together, and we didn't win it either).

Oh, not to worry, Tosh, "we" are still killing "their wives and children" with gay abandon. The strangest thing in this very strange rant is Kanazawa's claim that "terrorist" is a "euphemism," though it's not far ahead of his notion that Americans don't refer to their current enemies with racial slurs, as they did in the good wars of the past.

No, no, the CNN article must be satirical, like The Onion but not as far over the top, and the study should have appeared in the April Fools issue of Social Psychology Quarterly.