Are Generalizations Useful?

William Blake noted that "To generalize is to be an idiot. To particularize alone is a distinction of merit." Journalism is primarily about anecdotes, witness the Malcom Gladwell phenomenon.

Robin Hanson recently posted an interesting blog post on women using sex as a bargaining chip in marriages. Midge Decter (wife of Norman Podhoretz) noted this in her autobiography, that a woman's main power over men was that they needed sex less than men did, and this actually countered their disadvantage in earnings and strength quite effectively.

Yet, there seemed many feminists unhappy with Hanson's cynical evaluation of male-female relationships. One commenter (a 'sex activist') noted that

it would be nice to see some recognition that women sometimes don’t get as much sex as (or the kind of sex that) they want out of long-term relationships. And that there are men who have considerably more complex experiences of sexuality than you seem to acknowledge here.

So, Robin noted a generalization (women have a propensity to do X more than men), and some women got hysterical (heh) about the insight. Let's be clear. A stereotype is that all A are A. These are generally untrue, because there are exceptions to most assertions (eg, 'men are taller than women'). But generalizations are true and useful. It is useful to know that walking in a dark alley containing a bunch of young men is dangerous for a young woman. Not that all such situations mean bad things will happen, but the odds imply it is not a good idea.

Most things come down to probabilities, and learning to slant them in your favor is part of an intelligent life. For those who think only deterministic relations are interesting, or known facts about individuals, I just can't empathize. I find it a childish view of life. Not that anecdotes and biographies aren't interesting, but they are primarily interesting as how they are relevant to everyone's story, or a metaphor for something larger.