Bering’s post focuses on a series of studies by the evolutionary psychologist Gordon Gallup. Gallup was interested in the question of whether there might be an adaptive explanation for homophobia–which, given the fact that many (although far from all) human cultures treat homosexuality as a taboo–is a fair question for research. He hypothesized that treating homosexuality as taboo helped to prevent homosexual adults from contacting a homophobic parent’s children, which would reduce, however slightly, the prospects of those children growing up to be homosexual, and ensure more grandchildren for the homophobe.
Gallup supported this adaptive hypothesis with … evidence that straight people were uncomfortable about homosexuals coming into contact with children [$a]. Here’s the opening sentence of that paper’s abstract:
In a series of four surveys administered either to college students or adults, reactions toward homosexuals were found to vary as a function of (1) the homosexual’s likelihood of having contact with children and (2) the reproductive status (either real or imagined) of the respondent.
If you’ve noticed that this doesn’t mention evidence of heritability or a fitness benefit to homophobia, that’s not because I left it out–that’s because Gallup’s work contains no data to support either.
I’m starting to see a pattern here. You might remember that Bering wrote a piece for Slate, in which he argued that rape avoidance was adaptive.
There too, Bering cited an article with weak evidence:
Claim #3 is really problematic. The article claims:
German investigators Arndt Bröder and Natalia Hohmann established, ovulating women are not less active in general–they’re still busy shopping, going to church, visiting friends, and so on–but they avoid doing those things that make them sexually vulnerable. (link is to a pdf)
I’m not sure they’ve established much of anything. Risky versus non-risky activities were scored by 23 women on a scale of 1-5 (5 being the most risky), and the average score was used. I would have used the median score instead: a few ‘nervous Nellies’ could easily raise a score. This sounds technical, but it could make a huge difference in scoring. I’ll also note that college students were used to assess risk, and college students are fantastic risk assessors (Got Four Loko?) But what really bothers me is this figure:
The largest change in behavior is among women not on the pill (the pill simulates some of the effects of pregnancy, most notably preventing ovulation): they increase non-risky activities, far more than they decrease risky activities (which is still statistically significant). Ovulating women seem to be more active overall. Maybe ovulation makes women want to do more stuff, and that causes a decrease in risky activity–there are only so many hours in a day. Finally, by the authors’ own admission, this is a relatively weak effect (related to this, keep in mind the scoring caveat noted above). One study that demonstrates a weak effect needs a lot more follow up (didn’t the sciencebloggysphere just have a humongous conniption fit over the Decline Effect?).
What bothers me about Bering is that he treats barstool speculation based on weak or non-existent evidence as settled consensus. Then it follows the scientific equivalent of Cokie’s Law: once it’s out there, we have to discuss and refute it, even if the premise is poor.
That’s makes Bering’s statement of intellectual boldness all the more ridiculous, as Yoder notes:
Why on Earth would Bering dredge up Gallup’s adaptive fairytale a decade and a half after it was published, if it was baseless to begin with and no new evidence supports it? Well, according to Bering, because he’s a hard-nosed scientist who isn’t afraid to consider uncomfortable possibilities.
Sometimes, science can be exceedingly rude–unpalatable, even. The rare batch of data, especially from the psychological sciences, can abruptly expose a society’s hypocrisies and capital delusions, all the ugly little seams in a culturally valued fable. I have always had a special affection for those scientists like Gallup who, in investigating highly charged subject matter, operate without curtseying to the court of public opinion.
Of course, says Bering, Gallup’s work isn’t conclusive, but it sure would be interesting if someone tested it.
Except, when Gallup was forming his hypotheses about the evolutionary benefits of gay-hating–he first proposed the idea in a 1983 article–he was hardly thumbing his nose at public opinion. He was, in fact, giving natural selection’s approval to the prevailing ugly stereotypes about gay men.
A responsible scientist who cares about his discipline doesn’t mainstream speculative claims. This is not helping the field of evolutionary psychology–and that’s too bad, because if done rigorously, we could learn a lot about human behavior.