February 22, 2005

The Trouble With Larry

Most of what I have to say about Larry Summers has been said already by others. He is not a martyr to political correctness. Many of his critics were exaggerated or extreme in their reaction, but the speech he gave was really quite weak.

It’s perfectly ok to get up and say something like, “We have to remain open to a variety of explanations for the relative lack of women in the sciences, including genetic or innate differences between men and women”. But Summers didn’t say that: he went on to speculate that this was the correct hypothesis. The current state of knowledge on this subject suggests fairly strongly that this is not a good hypothesis. If you’re the president of Harvard, you ought to know that if you’re going to shoot your mouth off on the topic.

One thing that I think many observers have overlooked, however, is that the most inexcusable thing about Summers’ provocation is that even if he’s completely right in his hypothesis, it has nothing to do with the representation of women on the Harvard faculty.

Let’s ignore the large body of research that casts doubt on or hugely complicates the working hypothesis that men are somehow adaptively better at science and mathematics. Let’s assume that Summers’ hypothesis is valid. Even in the best case scenario for this kind of conjecture, we’re only talking about tendencies, not gender-based absolutes. Meaning that even if Summers’ hypothesis actually is the best explanation for the imbalance in the sciences, this imbalance should pose no difficulty for Harvard should Harvard judge it desirable to have more women on its science faculty.

Harvard is wealthy and powerful enough that should its president deem it to be a priority to staff its faculty with the most brilliant left-handed sociologists who cook a mean risotto and have surnames that start with the letter “M”, they could do so. Even if you wanted to be generous to the argument that affirmative action goals result in declining standards, it only applies to the average institution, to institutions which are presumed to lack the clout or financial power to compete for scarce goods and which therefore are presumed to have to lower their standards in order to achieve diversity.

None of this applies to Harvard, Even if genetic or innate differences mean that no more than 15% of the top scientists and mathematicians are women, Harvard could pay whatever was necessary to recruit from that 15% and achieve a faculty which had a 50-50 balance of men and women.

So not only is Summers’ hypothesis a poor one in light of available research, it isn’t an alibi or explanation for gender imbalances at Harvard. The only way Summers could account for that imbalance would be to say that in his opinion achieving gender balance is an unimportant objective, or at least not worth the trouble involved. Now that, if he said it, would be a much bolder and more provocative statement, and curiously enough, a more defensible one than what he actually had to say. What Summers actually said was dubious factually and intellectually, and it was a lousy explanation for the specific institutional problem he was attempting to grapple with. If Summers wanted to get up and say, “Look, I don’t actually care what the genesis of the imbalance between men and women in the sciences is: it is in my judgment too expensive and labor-intensive for one institution like Harvard to heroically compensate for it”, then at least he would have started a conversation that could shed more light than heat. If you want to make a critical reply to that statement, you actually have to either demonstrate that it’s not that expensive or difficult to do, or that for some reason achieving gender balance is such a pressingly important objective that it outweighs many other priorities that might exist in the process of hiring and tenuring. These are both useful claims to constantly revisit, revise and challenge even when you agree with them.