Preposterous Universe

 
Sex and science!

Hey, have you heard? Apparently Harvard President Lawrence Summers has made some sort of remarks about underrepresentation of women in science being attributable in part to innate cognitive differences, rather than some sort of discrimination. And it started a bit of a kerfuffle! Who would have guessed?

I mentioned the incident briefly, and others in the blogosphere have not been shy about offering their opinions: see Mark Graber, Bitch, Ph.D., Kriston, Unrequited Narcissism, Lubos Motl (don't miss Ann Nelson's comment), Michael Bérubé, Matthew Yglesias, Barb Mattson, Becky Stanek, Juan Non-Volokh, PZ Myers, Universal Acid, Kristy Elesko. A nice review of the actual research on cognitive differences is given by Mixing Memory. Summers has been scolded by the Harvard faculty, defended by Steven Pinker, corrected by the sociologists he quoted, and has issued a slightly more contrite apology than was originally squeezed out of him.

Perhaps more commentary isn't really necessary. But it's hard to miss the fact that Summers' defenders and critics are mostly talking past each other. As one of the critics, I'm especially baffled/annoyed at the tack taken by most of the defenders. The basic line seems to be that Summers was simply offering a scientific hypothesis, one that is worth investigating, and if you are in any way offended you must be letting political correctness compromise your interest in finding the truth. This seems to miss the point entirely, so perhaps it's worth just a little more blather on the subject.

Men and women are indeed different. I've noticed differences, anyway, and thank goodness. Biological differences are obvious, and I'm someone who believes that the mind is completely tied to the brain, which is part of the body, so it's certainly possible in principle to imagine that there are innate cognitive differences. I know that some people disagree, and deny even the possibility of cognitive differences, but I think they are unreasonably extreme.

So, except for the fact that "scientific ability" is something hopelessly hard to quantify, I'm happy to contemplate the possibility that men have some tiny innate superiority to women when it comes to science. I am equally happy to contemplate the possibility that women have some tiny innate superiority to men when it comes to science. The point is that we have no strong evidence one way or another. It's impossible, given the current state of the art, to reliably measure "innate ability" in a way that isn't hopelessly noisy and compromised by cultural factors. It's perfectly clear that the differences between individual people are typically much larger than the difference between some hypothetical average man and average woman, just as it is perfectly obvious that the expression of innate ability is tremendously affected by social and cultural factors.

Don't these people read any history at all? Whenever some group is discriminated against by some other group, people inevitably suggest that the differences in situation can be traced to innate features distinguishing between the groups, and they are never right! If you would like to suggest that innate differences are responsible for some current discrepancies in people's fortunes, the minimal burden you face is to acknowledge that such explanations have been spectacular failures in similar circumstances throughout history, and explain why we have compelling reasons to think the situation is different this time. Maybe it is, but the presumption is strongly against you.

Systematic biases against women in science are real. I've talked about this before, so didn't think it was worth rehearsing, but apparently there are a lot of folks who don't quite see it. They must not be looking. It might be better to refer to "systematic biases" rather than "discrimination," because many of the pressures that work against women are brought to bear by men who have no idea that they are discriminating, and could even be said to have the best of intentions. The truth is that girls are dissuaded from pursuing science from almost the moment they are born, and the pressures are equally real at the university level. If the current differences in representation between men and women were due to innate differences, the U.S. wouldn't have such low numbers compared to other countries, and the numbers wouldn't actually be gradually but steadily increasing (as they are). There are very few role models for talented women students. There is a culture within science that stretches from offputting to downright misogynist. There are teachers in elementary and secondary schools that steer girls away from math and science. There are societal stereotypes that discourage women from pursuing scientific careers. There are many male professors who deep down just don't think that women are cut out for the job. The list of biases goes on and on, and only someone willfully blind or extraordinarily simplistic could miss them.

Summers, of course, casually dismisses the idea that differences between the representation of men and women in science can be traced to systematic biases. His argument is based on rational markets: if there were a lot of talented women out there who were being discriminated against, a clever university could dominate the competition by hiring them all up, but this doesn't happen. This is the kind of idea so dumb that it could only be entertained by a professional economist. By similar logic, shouldn't smart baseball executives in the first half of the twentieth century been able to win multiple World Series by simply scooping up all of the African-American players that their racist colleagues were reluctant to hire? Somehow that didn't happen. When the biases are widespread, subtle, and diffuse, they affect all institutions. This is especially true (and obvious) in the case of women scientists, where the pressures working against them stretch from preschool to university departments (and administrations!). No one person or institution can undo the damage, which is one of the reasons why it's so important to acknowledge that the damage is real.

Entertaining hypotheses can, in context, be offensive. Summers' own defense was that he was simply offering an hypothesis, and even hoped that he would be proven wrong. How can innocent scientific inquiry upset people so much? We should be devoted to the truth, right?

Okay, imagine you like to play chess, but the only person you know with a chess set is your friend (let's call him "Larry"), so you have to play with him over and over. You believe that the two of you are evenly matched, so the games should be competitive. Except that, while you are an extremely polite and considerate player, Larry is consistently obnoxious. When it is your turn to move, Larry likes to take out his trumpet and practice scales (he's a terrible trumpet player). Also, he tends to flick the light switch on and off while you are thinking. And he is consistently jiggling the chessboard slightly, so that the pieces are vibrating around. Occasionally, at crucial points during the game, he will poke you in the side with a sharp stick. And more than once, when it looked like you were about to win the game, he would "accidentally" spill his coffee on the board, knocking over the pieces, and declare the game a draw by forfeit.

You put up with this behavior (he does, after all, own the chess set), but you are only able to win about ten percent of the games. Eventually, in frustration, you complain that his behavior is unfair and he should cut it out. "Well," says Larry, "let's entertain the hypothesis that you usually lose because you just aren't as good a chess player as I am. I suggest that you are just a sore loser with inferior cognitive capacity, although I'd love to be wrong about this."

Perhaps he is correct -- but in context, you have every right to slap him. Nobody should be against seeking the truth and exploring different hypotheses. But when systematic biases are widespread and perfectly obvious, and these biases are strongly affecting the representation of a group such as women, people have every right to be offended when the president of the most famous university in the world suggests that discrimination is imaginary, and it's women's own fault that there aren't more female scientists. Of course psychologists and sociologists should continue to do research on all sorts of hypotheses, and perhaps some day we will have a playing field that is sufficiently level that any remaining differences in the numbers of working scientists can be plausibly attributed to innate capacities. But in the meantime, we should be focused on overcoming the ridiculous biases that plague our field, not in pretending that they don't exist.

~ posted by Sean Carroll at 10:15 AM ~ link ~ ~ ~