A dilemma for defenders of sex differences

One of my favorite philosophers, Allen Buchanan, makes a great point about the limitations of “human nature” as an explanatory concept for behavior in his writing. Let’s characterize “human nature” as the set of traits and capacities that biological human beings normally acquire in their development, and which set is stable across time and space.

Suppose there is such a thing as “human nature.” Most likely, the capacities and traits that comprise human nature would be highly abstract, something like “a capacity for culture,” or obvious, like “bipedal.” They’d have to be this abstract or obvious to meet the generality requirement that the concept imposes. The problem, however, is that human nature looks pretty uninteresting–specifically, it can’t explain any of the specific behaviors or patterns that excite our interest.

For example, it won’t do to say that part of human nature is a willingness to cooperate with complete strangers, since this will only delay the question of why such willingness is part of our nature at all. It also risks incoherence in the concept, since human beings are not fully altruistic and cooperation frequently breaks down. Any specific, interesting form of human behavior will be so confounded with environmental and cultural factors that an appeal to “human nature” won’t cut much ice.

Buchanan’s basic point is this: “human nature,” as a concept for explaining behavior, is either 1) uninteresting or 2) thoroughly inadequate. If there is such a thing as human nature, it’s not very interesting. And if we try to make human nature interesting, we risk making it incoherent by introducing confounding cultural and environmental factors.

I think a similar point applies to the debate about sex differences. Suppose their are sex differences between men and women. If there are, they are so abstract or obvious that they can’t explain much anything interesting. And if we try to explain specific behavior by appealing to sex differences, we will confound our analysis with other factors.

Take the recent example of the Google diversity memo that’s been floating around. Even if there are sex-based personality differences between men and women, why think at all that those differences are at all the most important in explaining the gender gap at Google? As the author of the memo himself concedes, we shouldn’t commit the “statistical fallacy” of explaining individual-level differences through population-level differences. But the same point precisely undermines any argument that sex-based differences explain the gender gap at one particular company at one particular time.