Andrew Sullivan’s long obsession with race and IQ

For reasons that are obscure to me, Andrew Sullivan and his blog The Dish are highly popular. He is often cited as someone whose opinion is worth considering and is a frequent guest on talk shows. But he has always struck me as someone who has no internal compass to guide him but worships power and those who possess it. The only purpose he serves to me is as a reliable indicator of where the boundaries of conventional wisdom lie, because he cruises close enough to give himself the air of a daring thinker while not threatening the current social order.

He was an enthusiastic warmonger who cheered lustily as George W. Bush went on his war rampages and he viciously castigated as terror-loving fifth columnists anyone who doubted the wisdom of these actions. He displayed all the excessive jingoism of an immigrant who wants to outdo Americans in their patriotism, a quality he shared with his compatriot the late Christopher Hitchens. Later, when the war turned sour, he turned against it and now has become an Obama worshipper who attacks anyone who dares question his Dear Leader’s drone war on Muslims. His latest foray was his shameful and outright dishonest distortions of Glenn Greenwald’s questioning of the selective the use of the word ‘terrorist’, which reached a new low even for him.

But Sullivan’s crimes go back to the days when he was the editor of The New Republic and heavily promoted the release of the book The Bell Curve in 1994. As a result of the beating that he and the book’s authors took on their entire methodology, he has backtracked from some of those ideas but, in the context of the recent Jason Richwine debacle, he is still pushing the idea that the differences in mean IQ between groups mean something significant enough to be the basis of policy decisions. Most recently, he has been engaging with Ta-Nehisi Coats on the meaning of the Richwine analysis in which he does his usual trick of imputing to his opponents broad assertions that they do not make.

If we are discussing “subgroups with genetic differences,” we avoid the pitfalls of race as an overly-broad category. But we do not deny biological genetic differences in these subgroups, which can correlate with various degrees of accuracy with our crude racial terminology.

It’s really futile and I would argue self-defeating for liberals to deny this reality. These days, you can actually find out the exact subgroups with genetic differences that your DNA most closely resembles.

I don’t know what Sullivan’s expertise is in quantitative analysis but he strikes me as one of those people who are easily impressed when people toss out statistical and other mathematical terms. He seems to think that the fact that certain groups have a greater chance of having certain genetic traits (for example people with African ancestry with sickle cell and Ashkenazi Jews with Tay-Sachs) is somehow deeply significant and lends credence to Richwine’s argument that Hispanics are somehow doomed to be lower in intelligence than white people.

The big problem is of course that ‘intelligence’ is vaguely defined and it is not clear how it relates to IQ tests, and scores on IQ tests have been rising too rapidly to be caused by changes in genes. So the differential distributions of IQ scores between identifiable groups may be measuring something but what that thing is and what consequences follow from it are not at all obvious.

The fact is that any sub-population that tends to significantly interbreed within itself will experience a certain amount of genetic variation that are statistically from that of another population. So take the population of an island like Sri Lanka and measure the gene frequencies in its population and it would not be at all surprising to find a distribution for some genes that differs from that of the population of another island like Madagascar. So what? What follows from this other than providing evidence for how evolution occurs?

The problem is that what we find are just correlations and not strictly deterministic. By measuring the genes of any given individual, you would not be able to tell whether they are were Sri Lankan or Madagascan but could only give very crude odds. It is this lack of predictivity that makes gene frequency knowledge, like IQ test results, not lend themselves to policy prescriptions at the individual level, because of the vast number of false positives that would inevitably ensure.

If you are an Ashkenazi Jew it may make sense to test for that gene for health reasons but it would be outrageous for the US to forbid immigration of this entire group if it ever decided that Tay-Sachs were a serious health hazard it did not want to deal with. The same is true with any other trait that is statistically distributed like sickle cell.

And IQ is just another trait that has even less consequence than medical traits.


  1. unbound says

    The whole problem with defining what exactly is intelligence and what an IQ test measures should be the primary focus. By not being able to define intelligence very well, the whole concept of an IQ test should have been laughed out of existence long ago (even Binet himself, essentially the creator of the test, didn’t think it could be used to determine an individual’s intelligence…the IQ test was developed to help identify children that may struggle in school). As my psychology professor put it, “IQ tests are very good at what they measure, but we really don’t know exactly what it measures”.

    But once the laughable IQ test was tested, it leads to further nonsense like “The Bell Curve” where many people debate the merits of the book without pointing out that it really wasn’t much better than a study of how people respond to tarot card readings.

  2. kestra says

    I read Sullivan because he is always engaging with other voices that I may not be familiar with, and he’s good at aggregating the kind of news (international and political commentary) that I’m looking for, but he does have his obsessions, and seems incapable of being objective about them. Catholicism and his willful downplaying of its abusive aspects (or even divorcing “Catholicism” from Popes he doesn’t like!) while gleefully demonizing Islam in the strongest possible terms is one. IQ tests and testing is another.

    Sullivan took an IQ test early in his academic career, and Britain’s education system was set up at the time so that he was automatically placed in a more rigorous school which he views as the basis for the career in commentary he subsequently built. Whether he could have been as successful if he never took such a test and was never placed in a higher academic track, of course can’t be known. But his vehement defense of IQ testing as a result is really kind of inexcusable given how fungible and unreliable the data is known to be. He certainly hasn’t come out saying that low-IQ-testing US citizens should be penalized for it, so why would it follow that the IQ levels of immigrants should be considered in the debate about immigration policy and amnesty? It doesn’t.

    Sullivan just likes to flatter himself that since an IQ test proved his ability and gave him the resources he needed (deserved), and so they can’t possibly be biased or slanted against certain populations, because that would somehow make his success un-earned. Being as he is a HIV+ gay man who faced discrimination regarding his own immigration status in the US, his embrace of privilege without calling it privilege is pretty embarrassing to watch.

  3. sailor1031 says

    I think when many people refer to IQ (which is not supposed to be influenced by education and culture -- ha ha) they really mean the even-less-well-defined “smart” -- which certainly is influenced by education and culture. At home here we have four cats, all domestic short hair variety with little DNA variation other than hair colour. But they sure do vary regarding their smart. Does this mean that IQ varies among cats? If so that rather disposes of the theory that, whatever it is, IQ is determined by DNA

  4. MNb says

    Wow, you sound sincerily pissed off -- the first time as far as I know. I guess Sullivan deserves it. I’m know the name, but am not terribly familiar with him. The ideas you describe are pretty popular in The Netherlands too and I indeed had noticed it’s pseudoscience -- one of the more vicious variations.

  5. jpmeyer says

    “I don’t know what Sullivan’s expertise is in quantitative analysis but he strikes me as one of those people who are easily impressed when people toss out statistical and other mathematical terms.”

    That’s him to a T, actually. He gushed all over the Paul Ryan budget for example until people beat him over the head repeatedly with real math that showed that it was a load of hot air.

    Also, more or less the reason that Sullivan gets widely read is that he is more or less the least bad conservative pundit.

  6. glendenb says

    Andrew Sullivan is a good example of everything that is wrong with the punditocracy. they swooned over George Bush because 9/11 scared them senseless and they wanted a strong father figure to keep them safe. The minute the wheels came off the Bush administration they turned on him. So many of the pundits are basic authoritarians who yearn for strong leaders. Sullivan yearns for strong leaders who can stand up to the menace. That’s connected to his nonsense on race and IQ because it supports the notion that some people are meant to govern and some to be governed.

  7. slc1 says

    Prof. Singham, in his discussions of IQ fails to consider the Achilles heal of the entire subject, namely to what extent is whatever IQ measures inheritable. That’s the biggest problem with the Bell Curve,. Much of the work on this issue earlier then the 1970s was based on the twin “studies” of Sir Cyril Burt Unfortunately, when Burt’s studies were shown to involve twins who didn’t exist, these works were also discredited. For example, Berkeley Professor Arthur Jensen, whose controversial paper in the Harvard Educational Review was heavily based on Burt’s “results”, was discredited when Burt’s work was found to be fraudulent.

  8. slc1 says

    My biggest problem with Sullivan, who I consider worth reading even when I disagree with him is his instance in remaining a communicant of the Raping Children Church, despite the pedophilia scandals and it’s virulent anti-gay stance. For a gay man, who is legally married to another man, to remain in an organization that opposes same sex marriage and discriminates and encourages hatred and bigotry against gay men is unfathomable. Somewhere, Sullivan has a screw loose.

  9. Nathan & the Cynic says

    >>it would be outrageous for the US to forbid immigration of this entire group if it ever decided
    >>that Tay-Sachs were a serious health hazard it did not want to deal with.

    The U.S. did this for years and years. A cousin of mine has a story about her father almost being sent back to Europe because they thought he had a heart murmur; he only got in because other relatives were already here and took an oath that they’d take care of him if he got sick. Ellis Island checked for TB, glaucoma and all sorts of other things. The steamship lines were on the hook to take the rejected people back to Europe.

    So the question is -- if we accept that the U.S. (or any country) has a right to control who enters that country, why would this be more less outrageous than any other criteria?

  10. Mano Singham says

    It is one thing to test individuals for some trait and if they have it to exclude them, another to exclude broad groups based on statistical group differences. It is the latter that is outrageous on its face.

    In the case of individual testing and exclusion, then the issue is a different one of the predictivity and morality of the bases for exclusion. We are never going to get ‘perfect’ people applying for immigration. You then have to ask which traits are the ones you are are going to test for and exclude. If you have one person with high IQ but with the Tay-Sachs gene and another with lower IQ but no Tay-Sachs, which one would you admit and which one would you exclude? It would be a moral nightmare.

Leave a Reply

Your email address will not be published. Required fields are marked *