Skip to content

IQ and you-know-who: the great white hero faces reality

Black and white bell curves according to Herrnstein and Murray
Black and white bell curves according to La Griffe du Lion

Bell Curves, black and white — Slightly different? Top from The Bell Curve, by Richard J. Herrnstein and Charles Murray (viasee also). Bottom from La Griffe du Lion, a pseudonymous American academic.

Back at the end of November, Andrew Sullivan wrote a short post claiming that the study of intelligence — not just its racial aspects but the general study — has “been strangled by p.c. egalitarianism.” He admitted later that his conclusion was “an over-reach” (“Sometimes that happens when you respond as a blogger to a story”). But it opened up the whole race-and-IQ can of worms. And, well, there they were, old and tired but still squirming.

Sullivan has a history with the issue. In the 90s, when he was the editor of The New Republic, he devoted an issue to trying (and apparently failing) to “foment a sane discussion of The Bell Curve.” Ta-Nehisi Coates was one of the “young intellectuals who’d gather under the flag-pole” at Howard University at the time. He remembers it as a galvanizing moment: “That the in flight magazine of Air Force One would argue that all the world I’d known was brain addled set me afire.” He brought some of the fire to the latest round of the old debate, talking back to his friend Andrew. [⇓ the whole thread ⇓]

It’s an honorable role Sullivan is trying to play, though, and not an easy one. He’s read The Bell Curve and it’s not some flimsy racist tract, it’s a serious statistical analysis of a longitudinal study that followed nearly 12,000 Americans for 12 years, starting at age 17. He’s looked into some other similar studies, too, and he’s “gob-smacked by the resilience of IQ differences between broad racial groups.” But as Coates says, the main conclusion — that Africans, on average, are not as intelligent as Europeans — is “a long-held tenet of slave-holders and white supremacists.” And there’s no clean break between the impressive late-20th-century studies and the old racist pseudoscience. They share the same preoccupation with the problem of black people and the same conviction that it will always be with us. But that doesn’t mean the science is wrong, it just means that it takes balls to stand up for it.

I doubt that anyone has stood up for it as heroically as Will Saletan. In 2007, after DNA-discoverer James Watson made some foolish remarks about the intelligence of Africans, Saletan wrote a series of articles for Slate. He “soak[ed] [his] head in each side’s computations and arguments” and concluded that African genes probably carry a mental deficit. Since it’s an idea that makes a lot of people angry and uncomfortable, he looked over the precipice for us and found that “the truth isn’t as bad as our ignorant, half-formed fears and suspicions about it.” The truth set him free, anyway, and his voice rang out across the land.

Don’t tell me it isn’t genetic. Don’t tell me it’s God’s will. And in the age of genetic modification, don’t tell me we can’t do anything about it.

No, we are not created equal. But we are endowed by our Creator with the ideal of equality, and the intelligence to finish the job.

In other words, it’s a chance for all the p.c. egalitarians to get off their butts and pick up the White Man’s Burden. White Person’s Burden, I should say. Naturally, Saletan aimed his pep talk at people like himself — the educated non-blacks who have to somehow live with themselves after they give in to this politically-incorrect truth. How the folks running around with a bunch of African genes are supposed to deal with it, he didn’t say. They’re the problem, after all, not the solution.

Sullivan is 100% behind the hard truth, too, but he doesn’t go overboard with the heroics. His bottom line is that we need “unfettered research into reality.” In the mean time he seems to be unsure what the gob-smacking data means or even whether it means much of anything. But “conservatism is about facing reality.” And “research is not about helping people; it’s about finding out stuff.” So he’s ready and waiting to face the facts, if only the researchers could get to work finding out the stuff he needs to know.

We shouldn’t be too worried, though. The data on IQ differences may be way too persuasive to ignore, but when he answers Coates’s initial criticism, Sullivan stresses that the differences are subtle.

No one is arguing that “that black people are dumber than white,” just that the distribution of IQ is slightly different among different racial populations, and these differences also hold true for all broad racial groups.

The last point is good. “The differential between Caucasians and Asians - or between Ashkenazi and Sephardim Jews - is also striking in the data.” So Coates is wrong to say that the project is all about showing “a genetic relationship between the darkness of skin and the potency of neurons. (Only for ‘Africans,’ mind you.)” The research has deepened to the point that it’s confirmed all our stereotypes. It’s not just some narrow-minded racist thing about white superiority anymore.

Still, at least in America, the black-white difference is the elephant in the living room. That might be true even if it really was just a matter of “slightly different” distributions. But the distributions are the same old bell curves, and the difference that matters is the distance between the peaks — the gap in average IQ. The number that’s tossed around most of the time is roughly 1 standard deviation. That’s what you’ll find in The Bell Curve, for instance, and in “Thirty Years of Research on Race Differences in Cognitive Ability,” the 2005 article by J. Philippe Rushton and Arthur R. Jensen that Saletan soaked his head in (Saletan wrote that “everyone agrees that the black-white IQ gap closed significantly during the 20th century,” but that’s not what I’m finding).

There’s nothing slight about a 1 standard deviation difference. It means that going from a typical white population to a typical African American one should be kind of like going from the regular class to the remedial class. Daniel Seligman went over the implications in his 1992 book A Question of Intelligence: The IQ Debate in America. (I should say that I’m not quoting directly from the book, I’m quoting from the website No doubt their quoting is flawless, too.)

If you tell yourself that the top professional and managerial jobs in this country require an IQ of at least 115 or thereabouts, then you also have to tell yourself that only about 2.5 percent of blacks appear able to compete for those jobs. The comparable figure for whites would be about 16 percent. […]

The data are even more depressing on the downside. An IQ in the 70-75 range, which many psychologists would label “borderline retarded,” implies a life that is guaranteed to be short of opportunities. Very few students in that range will absorb much of what elementary schools teach, and virtually none will graduate from high school; few will succeed in finding and keeping good jobs. None will be admitted into the armed forces (required by law to screen out the lowest ten percent of the distribution). The bad news is that a substantial minority — apparently more than one in five — of American blacks have IQs below 75. Around one in twenty whites are below 75.

One of the proud racists at the Vanguard News Network goes through the same reasoning to show the folks “Rattling the Monkey Cage” at that “the differences in those distributions explain a lot of what you blacks blame racism for.” He even shows his work.

[S]uppose that a certain kind of work requires that the person doing it have an IQ of 130 or more. There are jobs of this kind, which require more brains than muscle. Job applicants of every race approach the employer seeking jobs of that kind, and the employer, to stay competitive in his industry, must select only those applicants who meet the minimum IQ qualification

And the answer is that “about 1 out of every 20 whites qualify,” but “only 1 out of every 7030 blacks do.” Assuming that an employer is drawing from a population with 1 black for every 5 1/2 whites, then…

If among his hirelings there is a lower ratio of whites to blacks than 1900, then the employer is being racist in favor of blacks, though perhaps he is being forced into it because of Affirmative Action laws.

[⇓ about those statistics ⇓]

(According to Ron Paul, you could look forever and you’d never find a black person with the brain power to make it in the quantitative realms of investment banking. Back in the 1990s, Paul told his followers that “complex embezzling” was “100 percent white and Asian.” He must have been thinking generically about the different races, since he couldn’t possibly have know every complex embezzler out there. So using your higher math skills to monetize risk in such a way that you soak both your investors and the taxpayers, for instance, must be a job for super geniuses, people so high up in the IQ stratosphere that there’s not a black person in sight. The darker criminals aren’t the brainy ones, but as Paul [⇓ or whoever] said, they’re “unbelievably fleet-footed.”)

There’s a good reason the book is called The Bell Curve and not, say, The Big Bad IQ Gap. An effect of the two offset distributions is that the discrepancies grow exponentially as you move away from the average. Even the relatively small IQ difference between whites and Asians (3 points, according to Rushton and Jensen) means that as a proportion of each population there should be several times more Asian than white geniuses and vice-versa for simpletons. The disparities between the white and black curves are much larger.

Sullivan owns that he wouldn’t mind if affirmative action was a casualty in this debate but he doesn’t go into any detail. According to the race-and-IQ hypothesis, the proportion of African Americans who are suited even to be middle managers, never mind engineers or lawyers, is much smaller than the proportion of European Americans. If anything, to the extent that this theory circulates as fact, it strengthens the case for affirmative action. It makes a compelling and seemingly rational case that it’s a huge waste of time to look for qualified black candidates for any “high IQ” job. But then it’s the job of a reality-facing pundit to take those big risks on behalf of the people who stand to lose. There’s no way they’re gonna do it, right?

Whether the underlying cause is genetic is an endlessly fascinating question — will the problem of black people always be with us? Researchers like Rushton and Jenson are sure that it will be. But those two displaced bell curves supposedly describe the world as it is now. There are all sorts of assumptions buried in the data and in the concept of IQ, and there’s no way I’ll have the time and interest to work through the primary literature and sort it all out. But what comes out in the end is claims like this: you’d have to go through 5 or 10 times as many blacks as whites to find someone who’s mentally equipped to be a teacher or an accountant. I find that very hard to believe. I might get into the reasons in another post. Right now my point is that this is where the theory touches down in the real world. It makes some very strong claims about the here and now, and there’s nothing subtle about them.

I’m sure Sullivan knows what the bell curves mean. I guess he didn’t feel like facing that much reality.

~   Notes & Stuff   ~

  • the thread Sullivan’s first post was on Nov. 21. A couple of days later he responded to a reader. It was about a week before Coates fired back. After that, it was Sullivan 3, Coates 2, Sullivan 4, Sullivan 5, Coates 3, Sullivan 6, Coates 4. Rod Dreher has a highlights reel.

  • about those statistics The bell curve is a normal distribution. Once you have a mean and a standard deviation you can get the percentage of people who have an IQ above X or below Y from a table or a web form. Different writers use somewhat different mean and standard deviation values, though, and it’s often not clear what the original source was. Actually the means are pretty consistent — 100 or 103 for whites, 85 for blacks. The standard deviations are not so consistent. According to La Griffe du Lion, the means are 100 and 85, the standard deviations are 15 and 13.5 (black and white) — see the red and blue graph at the top of the post. The VNN racist uses 103 and 85 as the means, 16.4 and 12.4 as the standard deviations. That means the discrepancies he calculates are extra large, and you’d pretty much expect a person like that to play them up.

    I’m not endorsing any of those numbers. I don’t have much faith in the research behind them. And the main point doesn’t depend on an exact value of the mean and standard deviation. Any two bell curves centered about 15 points apart with standard deviations of about 15, plus or minus a few, will produce the same general effect — large discrepancies in frequency of IQs that are significantly above or below average.

  • or whoever Yeah, this stuff is from the Ron Paul Newsletters, and as everyone knows, just because his name is on the top and the whole thing purports to be financial and survival advice from Ron Paul, that doesn’t mean that Ron Paul actually wrote it.

{ 3 } Trackbacks

  1. […] good reason to question the conventional wisdom is not something that liberal media will discuss. Nor will they willingly allow discussion that might lead to a questioning of the view that there are… in areas like criminal activity, violence, drug use and other self-destructive behavior that might […]

  2. […] (11), Derbyshire says, “The mean intelligence of blacks is much lower than for whites. The least intelligent ten percent of whites have IQs below 81; forty percent of blacks have IQs that low. Only one black […]

  3. […] If you’re following the link from John Derbyshire’s “Talk,” this is the graph you’re looking for. Drink it in, because according to ThinkProgress, it’s the “only ‘fact’ included in [his] entire piece” (source info). […]