To answer the question of how many homosexuals there are, you have to answer a prior question: What is a homosexual?
Gary Gates at the Williams Institute is doing his best to answer the first question in a new report. And while the results are a bit more clarifying than what has come before, they’re still no better than the imperfect answer to the more fundamental question.
Gates goes with the most minimal approach to the question of whom to count: people who identify as gay, lesbian or bisexual. That’s fair, but is obviously underinclusive. There are still probably millions of people in the country who are homosexual but in the closet in some measure. It’s also a bit overinclusive, since the fluidity of bisexuality can sometimes give them a hall pass out of the laws (at least) that disadvantage those whose sexual orientation is more fixed. Gates separately counts the number of transgender people, whose sexual orientation is independent of their gender or gender presentation.
Timothy Kincaid has some thoughtful criticisms of Gates’ methodology over at Box Turtle Bulletin, and Gates defends himself (but not his methodology) at the Washington Post.
The twin problems of whom to count and how to count them seem insoluble to me. As with race, the number of confounding personal and subjective factors means that the very best we can do is approximate an approximation. And with sexual orientation the subjectivity almost eclipses any objective criteria – except, of course, identification. That is the one part of sexual orientation that is most clearly visible and public, and as close to objective (though still not truly objective) as science can accept.
What this new chapter reveals has less to do with sexual orientation than with our current cultural preoccupation with the biases of social science. Gates says that it is important to have an accurate count of homosexuals “because legislatures, courts and voters across the country are debating how LGBT people should live their lives.” But how do, or would, those discussions change if Kinsey’s old guesstimate that 10% of the population is homosexual is instead 3.5% or 1.7%?
The anti-gay folks are already subdividing the numbers Gates arrived at, and are trying their best to use his scientific candor to their advantage. But their game exists in the same parallel universe that Gates inhabits. The debate in those legislatures and courts is not a demographic one, nor is it a scientific one. It is a moral debate about equality, and a legal one about the meaning of some pretty specific state and federal constitutional guarantees. Numbers are, in fact, a distraction from that discussion, a detour some politicians are all too happy to take.
Nothing in the legal notion of equality requires a statistical threshold. Journalists, in particular, are eager to entertain the social scientists because numbers always sound like they are meaningful and objective. In this case, though, they are deceptive at worst, and flawed at their very best. The only numbers we need are among voters and politicians. Let’s devote our efforts to counting what counts.