While that's certainly a conceivable explanation (the "gay uncle" theory, that the gay members of a family help raise their family's offspring), I've always fou

While that's certainly a conceivable explanation (the "gay uncle" theory, that the gay members of a family help raise their family's offspring), I've always found it really tough to swallow the numbers there. More or less, it would require that each gay uncle increases the fitness of the family [see note 1 below, "family" is very important here, as opposed to "group"!] by enough to make sure that on average, more than 1 additional child is produced per gay uncle. Put another way, with a 10% rate of homosexuality in the population at large, that would mean that at least 10% of our total reproductive success should be due to the helping hand that gays provide for family members' children. That's a lot, probably more than I'd realistically believe. More likely, the gene that predisposes us to homosexuality (it doesn't cause it, as MZ vs DZ twin studies have shown that homosexuality has well below 50% genetic cause, with a much higher component based on individual environment, both pre and post-natal) is either beneficial in some other way (for instance, a recent study showed that female relatives of homosexuals tend to have more children, which could mean that the heritable bits behind homosexuality actually increase female reproductive rates enough to overwhelm the negatives on the male end), or live close on the chromosome to another gene that's more important, so the gene tends to stick around because of linkage even if it's harmful in some cases. [1] The "gay uncle" theory relies on the fact that it's family members, i.e. people that are causally likely to share the gene in question, that are helped by the behavior. Google "kin selection" and "group selection" to understand the difference - briefly, kin selection is a totally legitimate effect because a gene helps itself through its prevalence in a family (and specifically, due to the fact that the gene will appear in the offspring that the gene helps make possible, even if it comes from other individuals), but group selection is completely bogus, because Prisoner's Dilemma-style defection is always the right strategy for a gene there.