August 27, 2005

Intellectual Vices, and Other Kinds

Yglesias observes that the parts of The Bell Curve that deal with race don't hold up well. Which is to say they're incompetent or malicious nonsense, as is the rest of the book; more from DeLong, Thomas Sowell, Stephen Jay Gould, Nicholas Lemann. Yglesias then goes on to make what I find a disturbing argument:

To answer Atrios's question, I'm reasonably confident that this is a "fools" problem rather than a "bigots" one. The errors in TBC's discussion of this problem are exactly the sort of subtle -- but utterly devasting -- technical problems that have rendered most contemporary conservatives... incapable of grappling with complicated policy questions.
[huge snip]
's tempting to conclude that, given the shoddiness of the analysis and the inflammatory nature of the subject matter, that all this can only be evidence of bigotry. I disagree, however. This is research coming from the same movement that sought to revamp Social Security without having done any of the math, that advocating invading Iraq in part on the basis of the claim that there was no history of ethnic strife in that country, and which appears to believe that the theory of evolution is a massive conspiracy. The human capacity for foolishness is large, and that of the contemporary American right is enormous.

But what's going on here can't be anything as innocuous as pure foolishness. It's an intellectual shortcoming to believe things on shoddy or no evidence--it's stupidity. But it's something more--an intellectual vice and a moral one as well--to believe things on shoddy or no evidence when they support your other views, and only then. And it's really a moral vice to believe things on shoddy or no evidence when they support views that are in themselves vicious, or would be vicious without the factual claims that you've conned yourself into believing.

It's no coincidence that Charles Murray, who argued that AFDC was a major contributor to the problems of African-Americans, went on to argue that African-Americans are genetically inferior. Because the crushingly obvious reason that African-Americans tended to be poor before AFDC--racism and its effects--continues to hold after AFDC. Better come up with an alternate explanation! (See Scott Lemieux on Herrnstein and Murray's idea of America's meritocratic past.)

You just couldn't do work that shoddy on race and IQ if you weren't predisposed to those conclusions. If Murray and Herrnstein hadn't wanted to discover that African-Americans were genetically inferior, they would have checked their numbers and sources. And if they'd done that, they wouldn't have reached those conclusions. Or, as Lemann says, "Unsurprisingly, all the mistakes are in the direction of supporting the authors' thesis."

(Similarly, the people who haven't done the math on Social Security are people who want to get rid of it anyway; the people who believed that there wasn't any ethnic strife in Iraq tended not to be peaceniks; and the people who think science doesn't support evolution tend to be committed to some religious creation stories, or politically allied with those who are.)

And honestly, I think this point goes past Herrnstein and Murray to the people who are still pushing The Bell Curve. When the book first came out, it was possible to think that it was telling disturbing truths, because--as Lemann points out--its release was handled so as to delay informed criticism. But by now it oughtn't to be too hard to find out about the debate surrounding The Bell Curve, if you look. But if you're inclined to accept the books's thesis, you may not be inclined to look for the criticism. And that doesn't speak well of you.

So on the bigots or fools questions, fools; or maybe foolish like a fox. But foolish in a way that takes a certain predisposition to bigotry.

Posted by Matt Weiner at August 27, 2005 12:39 PM

The late 20th/ early 21th century love affair with high IQs is driven by scarcity. If you have a high IQ, you can make a lot of money.

This probably won't last forever. Once you can get AI as smart as humans, you can get AI smarter than humans. Lots of AI smarter than humans. And, some other bottleneck trait will be the one that will be the new ideal. Lord knows noone will put up with those mesa guys any longer than they have too.

Posted by: joe o at August 30, 2005 03:50 PM

I used to argue (playfully) that AI was soon going to be smarter than people because people were getting dumber.

High IQ is the ideal only among the limited set of people who read long, wonky books, or some of them. Abs of steel, lots of money, smartmouthedness, obedience, seem to be the ideals of much larger sets of people.

Posted by: Matt's mom at August 31, 2005 09:25 AM