Alternative formulation of sexual orientation

Sexual orientation is often expressed as one of heterosexual (attracted to opposite sex), homosexual (attracted to same sex), or one of several extrapolated terms.

We may alternatively express it as one of attracted-to-males, attracted-to-females, or one of several extrapolated terms. From Greek-derived morphemes, we may construct the words "androphilic" and "gynephilic", respectively. (It's obviously a coincidence that these words I derived happen to already be in use! ... no, I'm just here to popularise them and make a couple original remarks.)

Then, instead of orientation being a majority/occasional-alternative distinction, it is a correlative component of gender, correlated more closely than hair length and less than reproductive anatomy.

Some consequences of this formulation:

• It hugs the query. The main reason to consider a person's orientation is to determine to which people they'd be attracted. You could easily infer that from hom/het, but AP/GP is more direct on that matter.
• The pair (gender, hom/het) has the same information as (gender, AP/GP), but AP/GP on its own is more informative than hom/het, in a very precise sense: H(AP/GP) = -P(AP) log(P(AP)) - P(GP) log(P(GP)) ≈ 0.5·1 + 0.5·1 = 1 bit, but H(hom/het) = P(hom) log(P(hom)) + P(het) log(P(het)) ≈ 0.02·5.64 + 0.98·0.03 = 0.14 bits.
• It generalises to transgender people more easily. No need to keep track of whether hom/het are defined wrt sex or gender, or whether you accept their self-identification as their gender (which Obviously You Should, but some people don't want to).
• It generalises to non-binary people more easily. Hom/het, applied to non-binary people, either introduce (probably undesirable) consideration of their sex, or are meaningless.
• "Bisexual" still makes sense, with the same meaning as before.

I don't intend to be one-sided about this, so here are some downsides:

• AP/GP can't describe couples well. "AP/GP couple" is either meaningless or equivalent to "male/female hom couple", and there is no sane way, in this formulation, to describe het couples.
• Some of the consequences I presented as positive may be interpreted as negative.