Dear Jim: [Email Jim Manzi]
I've thought some more about why your National Review cover story "Escaping the Tyranny of Genes," [June 2, 2008], into which you clearly put a lot of effort, is getting such a skeptical reaction from the small number of people whose respect you should worry about. (For example, I'm told that Richard Lynn, when he came to the part about the Minnesota Transracial Adoption Study, simply stopped reading your article.)
I think I've figured out how you went off track.
You started with the reasonable goal, one that I've pursued myself several times, of trying to criticize the pop journalism about genetics that has been common for the last 15 years. There have been repeated sloppy headlines about the discovery of "A Gene for ... Homosexuality (or Happiness or Infidelity or Whatever) ". Some of those "Gene for" headlines have turned out to be wrong.
For example, gay geneticist Dean Hamer got enormous publicity in 1993 when he declared he had found "The Gay Gene" (at least for men). This was hugely popular in the media for a while because a genetic cause for homosexuality is politically correct—it's assumed to be a rebuke to Christians. But 15 years later, you never hear much anymore about Hamer's "discovery".
It's probably not very true. As physicist turned evolutionary theorist Greg Cochran has argued since the 1990s, it's unlikely that a gene for gayness could evolve, because gay men have so fewer children.
Your NR article didn't spell out what bad effects you expect to be caused by credulous science journalism. When you were pushed to clarify your fears in the comments section of The American Scene blog, you wrote:
"I suspect that the analogous policies that might be established if an (incorrect) view of the linkage between gene patterns and mental characteristics and capabilities became more widely and deeply entrenched would be unpredictable, but more likely to be related to the relaxation of the notion of personal responsibility—replacing justice with therapy, greater paternalism in constraining economic, political and lifestyle decisions for those who are 'unable' to exercise 'true' choice, targeting government services based on genetic content and so on."
That's pretty vague. But perhaps you fear a "liberal therapeutic regime" rather like the one Anthony Burgess described in A Clockwork Orange, where the young thug Alex, rather than being locked up, is conditioned into not liking violence anymore.
Unfortunately, you didn't spend much time at all on these valid examples of weak pop journalism that might support your thesis that the press is overemphasizing genetic explanations. Instead, you chose to devote a huge amount of space to a single example—race and IQ—so incredibly ill-chosen as a case study for your argument that it has proven disastrous to the reception of your article.
As we all know, but you ignored to your credibility's severe detriment, much as the mainstream media want to hear about the Gay Gene and such, they do NOT want to hear about racial differences in IQ. And, the MSM especially do not want to hear about evidence for genetic causes for racial differences in IQ. How many voices in the press stood up to defend America's most eminent living scientist, James Watson, when he got fired last year? (Answer: none—and notably not National Review).
Moreover, the small number of race-and-IQ researchers, the Arthur Jensens and Charles Murrays, are not slapdash Dean Hamers going with the flow of popular opinion. They tend to be cautious and careful scientists aware that they are infringing elite taboos by carrying out unpopular studies certain to be picked at by legions of hostile critics.
Real IQ scientists, like Cochran and Henry Harpending, authors of the 2005 theory [PDF] attempting to explain the evolution of high average IQs among Ashkenazi Jews, are generally close students of the theory of natural selection. So they are less likely to fall for evolutionarily dubious ideas like the Gay Gene.
The evidence for a genetic link between IQ and race is broad but not conclusive. For example, Jensen and Rushton's 2005 summary paper [Thirty Years Of Research On Race Differences In Cognitive Ability (PDF)] listed, I believe, ten different lines of non-genetic evidence for a genetic link.
Occam's Razor, which tells us that the simplest explanation is most likely right, suggests that Jensen and Rushton are probably correct, especially because there is so little evidence for the more socially acceptable opposite view.
You mention Sandra Scarr's Minnesota Transracial Adoption Study, but what you don't mention is that it was originally trumpeted in the 1980s as proof of closing of the racial gap through improved home environments for black children. (The adoptive fathers averaged a year of grad school each.) When the black adoptees were tested as 7-years-olds, they averaged around 100. This was a very popular study at the time.
Then when Scarr went back and retested the kids when they were teenagers, their average IQs only came out to 89. This was horrible news and so she buried it in her subsequent paper. Nobody noticed what had actually happened except a CCNY philosopher named Michael Levin, who publicized the actual results.[Comment on the Minnesota transracial adoption study. Intelligence , 19 , 13-20, 1994]This led poor Dr. Scarr to do a lot of soul searching. [PDF]
There is the Flynn Effect—the tendency for average IQs to rise over time—which shows we don't fully understand IQ. But otherwise, even though any social scientist who could publish a valid study showing the race gap in IQ could be eliminated would become an academic superstar, there is remarkably little evidence supporting the conventional wisdom. Thus, when James Flynn debates Murray, he ends up harping on Eyferth's unreplicated 1959 study of the children of black American soldiers and German women for lack of anything better to cite in the way of positive evidence.
But the Jensens and Murrays do NOT claim they've proven their case. They hope to live long enough to see the genome analyses dramatically lower the uncertainty level.
Murray said in 2003 that we'll know from the genome studies one way or another within a few decades. James Watson guesstimated in 2006 that it would take 15 years, but on second thought decided it might be as little as ten.
In the long run, the number of years or decades doesn't much matter. We'll find out, one way or another.
Hence, your race-IQ example is precisely backwards and undermines the point of your article.
Jim, I imagine you are upset at present that your article has elicited so much scoffing. I hope this helps you understand where your chain of argument derailed itself—so you can get back on track in the future.