UPDATE: Over at Pharnygula, P.Z. has his own take (negative) on Bering’s paper and the lax standards of evolutionary psychology. P.Z. notes that the “adaptive” results of the handgrip study cited by Bering have not in fact been replicated by other investigators. That calls the results into question, something that Bering conveniently omits from his piece.
In the past week I’ve read two “pop” articles on evolutionary psychology. One, “Social Animal,” in the latest New Yorker, is by New York Times op-ed columnist David Brooks. It’s apparently an extract from an eponymous upcoming book. The piece is cleverly framed: it’s the account of a first date by two fictional people, “Harold” and “Erica,” in which all of their conversation, their gestures, and their subsequent courtship and marriage are couched in terms of evolutionary psychology, with Brooks describing the research backing up the story. Here’s an example:
As Erica and Harold semi-embraced, they took in each other’s pheromones. Smell is a surprisingly powerful sense in these situations. People who lose their sense of smell eventually suffer greater emotional deterioration than people who lose their vision. In one experiment conducted at the Monell Center, in Philadelphia, researchers asked men and women to tape gauze pads under their arms and then watch either a horror movie or a comedy. Research subjects, presumably well compensated, then sniffed the pads. They could somehow tell, at rates higher than chance, which pads had the smell of laughter and which pads had the smell of fear, and women were much better at this test than men.
A bit more distressing is a piece in the January 13 Slate, “Darwin’s Rape Whistle“, by Jesse Bering, who should know better. Bering is a psychologist at Queen’s University Belfast, a writer for Scientific American, and author of the forthcoming book, The Belief Instinct: The Psychology of Souls, Destiny, and the Meaning of Life. (He’s also funded by the Templeton Foundation). The subtitle of Bering’s piece is “Have Women Evolved to Protect Themselves from Sexual Assault?”, and you know what the answer is going to be without reading the piece. If it were “no,” the piece wouldn’t have appeared anywhere—so popular are Darwinian explanations of modern human behavior.
Bering’s piece begins by highlighting Randy Thornhill and Craig Palmer’s idea (described in their book A Natural History of Rape: Biological Bases of Sexual Coercion) that the human brain contains an evolved “rape module”: a neuronal circuit that impels men to subdue and copulate with women when they can get away with it. The idea is that this would have been adaptive in our ancestors, giving disenfranchised males a way to spread their genes when they couldn’t do so through the usual route of pair-bonding.
Thornhill and Palmer’s book was controversial, with many critics claiming that the authors were trying to excuse or justify rape. Bering takes after these critics, properly noting that “‘adaptive'”does not mean ‘justifiable’,” but rather only mechanistically viable.” But what he doesn’t mention is that there were strong scientific critiques of the “rape module” idea as well. I produced two of them myself, a long one in The New Republic and a short one with Andrew Berry in Nature, pointing out not only scientific weaknesses in the evolutionary scenario but Thornhill and Palmer’s unsavory fiddling with statistics, distorting what the primary data on sexual assault really said. Bering doesn’t mention the scientific controversy, noting only that “it’s debatable that a rape module lurks in the male brain.”
Bering also asserts without question that even the shape of the human penis testifies to pervasive sexual coercion in our ancestors:
In fact, the distinctive, mushroom-capped shape of the human penis is designed to perform the specialized function of removing competitors’ sperm, which indicates an ancestral history of females having sex with multiple males within a 24-hr period.
Note that there is no reservation here, no claim that “there are data supporting this theory but it may be wrong.” Just a simple factual claim about what science has found. But, as Bering described in Scientific American, the original study used dildos inserted into “realistic latex vagina[s] sold as a masturbation pal for lonely straight men” that had been filled with an artificial substance cooked up to mimic sperm. This bizarre experiment leaves considerable doubt about why the human penis is shaped like it is! Yet Bering has no such doubts. His journalistic certainty in the face of experimental doubt is a hallmark of much reportage on evolutionary psychology. It certainly characterizes David Brooks’s piece as well.
Bering goes on to the centerpiece of his article, the notion of a “rape arms race” that has given human females genetic tools to withstand rape. And by “genetic tools,” I don’t just mean the willingness and desire of women to fight off sexual assault. No, women have specific psychological and physiological modules that supposedly evolved in our ancestors as “anti-rape” tools. Bering sees four of these:
- When threatened by sexual assault, ovulating women display a measurable increase in physical strength.
- Ovulating women overestimate strange males’ probability of being rapists.
- Ovulating women play it safe by avoiding situations that place them at increased risk of being raped.
- Women become more racist when they’re ovulating.
The “increased racism” module is supposed to have evolved to prevent mixing with outgroups; as Bering says:
In this case, skin color serves as a convenient marker of group identity. (The authors concede that people of different skin colors came into contact with one another only in recent times, evolutionarily speaking, but propose that any physical trait that serves to demarcate an out-group member would be processed by ovulating females as a sort of “hazard heuristic.”)
There is, as far as I can see, exactly one study supporting each of these four points, with at least two of them based on surveys of undergraduates at single American colleges. Here’s Bering’s summary of the data supporting handgrip strength:
In 2002, SUNY-Albany psychologists Sandra Petralia and Gordon Gallup had 192 female undergraduate students read a story about either a female character being stalked by a suspicious male stranger in a parking lot (ending with: “As she inserts the key into her car door she feels his cold hand on her shoulder …”) or a similar story in which the female character is surrounded by happy people on a warm summer’s day (ending with: “She starts her car, adjusts the stereo, and as she pulls out of the parking lot those nearby can hear her music blasting”). The researchers measured the handgrip strength of each participant before and after she read the story, and compared the scores. Petralia and Gallup also knew from the results of a urine-based ovulation test kit where in their reproductive cycles each participant was, so the researchers could differentiate among women in the menstrual, follicular, ovulatory, and luteal phases. A fifth group consisted of those women who were on contraceptives at the time of the study. The results were unambiguous: Only the ovulating women who read the sexual assault scenario exhibited an increase in handgrip strength. Ovulating women who read the control passage and nonovulatory women who read the sexual assault material grasped with the same intensity as before.
Well, one can debate whether reading a story about rape is the same thing as being sexually assaulted, or whether a marginal increase in handgrip strength would have been sufficient in our ancestors to fight off a rapist. But the important part of these studies is that they were apparently one-offs—they have not, as far as I know, been replicated by other researchers. Do we accept single results, based on surveys of American undergraduates at a single university, as characterizing all modern women?
As we know, many studies in science, when repeated, fail to replicate the initial results. Think of all the reports of single genes for homosexuality, depression, and other behavioral traits that fell apart when researchers tested those results on other groups of people! And if an author did an initial study (not a replication) of handgrip strength that didn’t show the relationship with ovulation, would that even be publishable? I think not.
I suggest, then, that the results of evolutionary psychology often reflect ascertainment bias. If you find a result that comports with the idea that a trait is “adaptive,” it gets published. If you don’t, it doesn’t. That leads to the literature being filled with positive results, and gives the public a false idea of the strength of scientific data supporting the evolutionary roots of human behavior.
Nevertheless, Bering, who apparently has never seen a Darwinian explanation he doesn’t like, finds all this very convincing:
I don’t know about you, but I’m riveted, and convinced, by much of the logic in this anti-rape area. And researchers are just getting started.
Well, pardon me if I’m not quite so convinced. It takes more than a small study on American college women at a single school to convince me that a behavior is an evolved adaptation to prevent rape.
Now I don’t oppose evolutionary psychology on principle. The evolutionary source of our behavior is a fascinating topic, and I’m convinced that the genetic influences are far stronger than, say, posited by anti-determinists like Dick Lewontin, Steve Rose, and Steve Gould. Evolved adaptations are particularly likely to be found in sexual behavior, which is intimately connected with the real object of selection: the currency of reproduction. I’m far closer in my views on this topic to Steve Pinker than to Steve Gould. And there are many good studies in the field, so I don’t mean to tar the whole endeavor.
But, for crying out loud, let’s have the journalists and scientists show a little more responsibility when reporting on evolutionary psychology. If there are problems with a study, describe them. If an idea is pure speculation, say it. If there are other explanations for a phenomenon, give them. Let’s not gull the public into claiming that we understand something with near certainty when we don’t. These lax reportorial standards, pervasive in evolutionary psychology, seem to be much tighter in other areas of science, like physics or molecular biology. And this despite the enormous difficulty of demonstrating that any human behavior is an evolved adaptation.
Every time I write a piece like this, one that’s critical of evolutionary psychology, I get emails from its practitioners, chewing me out for being so hard on their field. And my response is always the same: I’ll stop being so hard on your field when you guys start being more critical yourselves. If you policed your own discipline better, I wouldn’t have to.
But since humans are so fascinated by scientific explanations of their own behavior, and so impatient with uncertainty and doubt, there’s not much incentive for the field to clean itself up.