I’m not a big fan of “feminist science”—the idea (promulgated by people like Evelyn Fox Keller and Sandra Harding) that women’s psychology gives them unique insights into nature and unique ways to study it. But I am a fan of “feminist science criticism”: the idea that women can sometimes point out male biases in research strategies and in the interpretation of scientific results. And a prime example of feminist criticism is on offer this week on Slate. Three of its columnists have taken on evolutionary psychology, using as a springboard the Slate article by Jesse Bering that I wrote about a few days ago.
As you recall, Bering highlighted a number of “scientific” studies purporting to show that women have a genetic “rape kit”: an evolved set of behavioral modules that act only during ovulation to reduce the possibility of rape. These behaviors include increased grip strength, avoidance of places where they’re likely to be assaulted, and increased racism (the latter is supposed to be an adaptation to avoid being raped by a member of an “outgroup”). I questioned these results because they were based on single studies using using a small sample size of undergraduates at single American colleges. Further, the authors neglected potential weaknesses of the work—or alternative, non-evolutionary interpretations. (P. Z. also leveled these criticisms at Pharyngula.)
A couple of evolutionary psychologists went after me in the comments, claiming that I was tarring the field by criticizing some articles that were, after all, in the popular press. What these critics don’t seem to realize is that many evolutionary-psychology papers themselves—papers from the primary scientific literature—are also lame, dubious, or even laughable.
And that’s what two of the Slate columnists found. They simply read the original papers.
In a post at Slate’s XX Factor, “Ovulating woman seeks better science,” Amanda Schaffer dissected the paper by Navarrete et al. (reference and link below) that purports to show that women become more racist during ovulation. I’ve now read it as well. Schaffer points out that the authors (and Bering, who reports their results) neglect two results that go against the authors’ own hypothesis:
In the single, unreplicated study he [Bering] cites, researchers gave women several tests of racial prejudice and tried to correlate the results with their menstrual cycles. On a few measures of implicit or unconscious racial bias, they found that women who were more likely to be ovulating tended to score higher. On a measure of explicit or conscious racial bias, however, they found no link between ovulation and prejudice. Now, we could delve into the minutiae of those tests and argue about whether implicit or explicit attitudes about race matter most. But Bering doesn’t do that. He simply ignores the negative finding.
Yep, he sure does, and so do the authors of the study. They—and Bering—also neglect the observation that there is no correlation between ovulation and fear of rape, something that should certainly be seen if racism is a result of that fear:
Meanwhile, that same study complicates his argument in another, unacknowledged way. The researchers measured women’s feelings of “vulnerability to sexual coercion” using something called the “fear of rape scale.” And they reported that women who are both more likely to be ovulating and more afraid of sexual coercion were more likely to show racial bias. But here’s the confusing part: they also looked at the more basic question of whether ovulating women were, in fact, more likely to fear rape or take measures to avoid it in the first place – and they found no connection. That would seem to be a problem for Bering. Higher up in his piece, after all, he argues that “ovulating women overestimate strange males’ probability of being rapists.” He cites a study that claims: “women perceive men as more sexually coercive at fertile points of their cycle than at non-fertile points.” This is supposed to help prove that rape-avoidance around ovulation is an evolved adaptation. Yet when another study in his grab bag turns up negative evidence, as in no link between fertility and fear of rape, he never mentions it. The study authors don’t highlight it either—it’s tucked in a data table. And they all go blithely on their story-telling way.
Do we need a layperson to tell the authors this? Why didn’t the reviewers catch it?
It’s typical of many evo-psychology studies that they simply ignore, or downplay, results that don’t support the authors’ a priori Darwinian hypothesis. Indeed, in many ways evolutionary psychology resembles religious belief—at least in the fervor of many of its advocates and their tendency to completely ignore data that don’t support their hypothesis.
That tendency is highlighted by another Slate piece, one by the ever-thoughtful (and entertaining) Emily Yoffe, “Not that time of the month again.” Yoffee goes after a different paper: one by Meghan Provost et al. (citation and link below) that studied changes in women’s walking gait during their menstrual cycle. As author Provost reported in an interview with BBC news, the authors expected to find that women would walk in a “sexier” manner during ovulation, adopting the salacious hip-swinging gait made famous by Marilyn Monroe. This prediction stemmed from previous evo-psychology studies showing that women become more attractive during ovulation, presumably to attract a mate when they’re most fertile. (These earlier studies claimed to show that men found women’s faces more attactive during ovulation than at other times; the same held for the smell of vaginal secretions.)
But what Provost et al. found was the exact opposite of their prediction. Women walked in an “unsexier” way during ovulation, swinging their hips less and keeping their knees closer together. And when men were asked to judge the attractiveness of women’s walks at different stages of their cycle, they found them sexier when women weren’t ovulating! So the author’s thesis was disproved. Did that show that variations in walk during the menstrual cycle had nothing to do with ovulation? Nope, the authors simply told another story. As Yoffe notes:
Having their thesis disproved was no obstacle for the researchers. They just came up with an equally spurious thesis: Evolution made women want to look “unsexy” during ovulation to discourage nonpartners from raping them. Actually what the study proves is that there is no finding that won’t prove something about women and ovulation. However, I have the feeling when, say, Mongol horsemen galloped into a village they weren’t calculating, “Hmmm, that one, not so much hip swing, not going to bother to rape her!”
It’s actually worse than this. The authors have to reconcile their finding that women’s walks are less attractive when they’re ovulating with the earlier findings that female faces and vaginal secretions are more attractive when they’re ovulating. Provost et al.’s theory—wait for it—is that women give different signals to men at different distances. Here’s what the authors say:
It is possible that faces and gait present different information because of the intimacy with which the stimulus is viewed. For example, faces can only be seen in a fairly close encounter, whereas gait patterns can be seen from a large distance. If women are trying to protect themselves from sexual assault at times of peak fertility, it would make sense for them to advertise attractiveness on a broad scale when they are not fertile, yet still being attractive to people they choose to be with (i.e., during face-to-face interactions).
Like the stories of the Bible, there’s no evolutionary psychology hypothesis that can be disconfirmed by data. If your story doesn’t hold up, simply concoct another story. Of course, there’s no evidence for the alternative stories, either.
In a Slate post called “Rape likely not genetic adapation,” Amanda Marcotte is less successful in her criticisms. Although she mentions the posts by both P.Z. and me, she also goes off the mark a bit. For example, she says this about Bering’s article:
There’s also the weird side assumption that features prominently in many half-baked evolutionary theories, which is that sex is strictly about reproduction in a species that has homosexuality, contraception, and old people who get it on.
This is based on a misunderstanding of evolutionary psychology. Evolution has given us orgasms as a proximate cue to reproduce: orgasms almost certainly evolved to make us want to copulate, since they provide profound pleasure associated with the act of copulation. But using our evolved brain, we can experience that pleasure without reproducing (other animals can do this too, as you may know from visiting the zoo!) “Getting it on” as a gay, a lone masturbater, or an old person, simply takes advantage of the orgasm cue in a way that dissociates it from its evolved “purpose.” The same thing happens when we eat too much fat or sugar—we’re using cues that were adaptive in our ancestors (our taste for rare but nutritious substances) in a way that doesn’t further our reproductive output.
Marcotte also accuses Bering of being soft on rape:
Even with his disavowals, the fact remains that Bering’s article downplays the severity of rape. It suggests that there’s not much to be done about rape and that men are just programmed to do it, and it distracts from the fact that it’s a violent act, experienced by both victim and assailant as assault.
I think this accusation is unfair, since Bering does go to great pains to say that he recognizes the severity of rape, and certainly does not say there’s nothing to be done about it. And yes, rape is a violent act, but it’s also a sexual act. Granted, some rapists use condoms, or engage in non-vaginal penetration, but there is still a sexual side to the act: the attainment of orgasm. And orgasms were our evolved cue to reproduce. My own theory is that rapists tend to be men with violent tendencies who use violence to both dominate women and have sex. But that’s just a hypothesis, and I certainly do not think that rape is an evolved behavioral module in males.
At any rate, the fact that two laypeople can read papers in evolutionary psychology, and find obvious problems with them, shows that it’s not just the popular press that distorts the scientific findings. It’s the authors of the papers themselves who are lax, as behavioral biologist Marlene Zuk has pointed out in her comments on this site.
I maintain my claim that much of evolutionary psychology is scientifically weak: little more than exercises in story-telling with a thin veneer of science. I emphasize again that not every study in the field is weak or flawed: there is some good work in evolutionary psychology. But the field suffers in general from not only scientific lassitude, but a failure of its practitioners to police the discipline. Many of them have an interest in selling the field (which, of course, enhances their careers), and you can’t do that if you spend your time criticizing shoddy work by your colleagues. It’s this failure of policing that leads me—and Yoffe and Schaffer—to put on our badges and nightsticks.
Navarrete, C. D., D. M. Fessler, et al. (2009). “Race bias tracks conception risk across the menstrual cycle.” Psychol. Sci. 20(6): 661-5.
Provost, M. P., V. L. Quinsey, et al. (2008). “Differences in gait across the menstrual cycle and their attractiveness to men.” Arch. Sex. Behav. 37(4): 598-604.