For decades there has been a debate about the root causes of homosexuality. One camp favors nurture, one camp favors nature. For the nature camp it has always been problematic explaining how homosexuality is adaptive and sends it genetic legacy forward. This is also a problem for the nurture camp, because if homosexuality is maladaptive then strong defenses against it should occur even if nurture plays a factor in its expression.
In centuries past there was less debate; homosexuality was outlawed; tolerantly accepted; or in some societies like ancient Greece considered just one variation of human sexuality to be pursued or not as long as it didn't interfere with siring an heir.
About 10% of the male population in America identify themselves as being homosexual with some studies citing as many as 1 in 3 males having had at least one homosexual encounter. With this high an incidence there must be some evolutionary explanation.
Many men who don't consider themselves homosexuals engage in male on male sex under certain circumstances; prison inmates for instance. In a prison setting dominant males force or coerce less dominant males into providing sexual release. Much the same has occurred in other all-male settings throughout history. I doubt the modern Navy has a much higher occurrence of homosexuality than any other sector of modern society, but it was definitely a feature of ancient seafaring life.
Most of the evolution that led to modern man occurred before the written word, teasing out the sexual practices of our Neolithic ancestors might be less than straightforward. Polygamy has been common all through recorded history. It would not be hard to imagine alpha males hoarding all the reproductive age females during most of the ascendancy of Homo sapiens. Where there are alpha males there are also beta males and gamma males and a system arises to determine dominance. Submitting sexually to more aggressive males probably allowed males lower in the caste system to survive and possibly procreate at a later date. Even future alpha males would have to had make it through the challenging time of adolescence before developing a physicality allowing them to survive not being submissive. Given millennia of evolution it is a possibility the anus adapted to accommodate the male penis and even derive pleasure from the experience so as to make submission more likely, lessening the risk of serious physical harm from be forced to submit.
All this is prelude to speculating that the evolutionary installed tripwire for homosexuality is male submissiveness. This would explain why homosexuality seems to have genetic and developmental environment components. Males raised in households with dominant mothers and passive fathers have a higher expression rate for male homosexuality. I suspect there is a tendency to learn submissiveness in this environment, which in turns then triggers the desire to please and submit to other males. Neither genes nor environment would be absolute in determining the final sexual orientation, but would work in concert. There would be males genetically predisposed towards submissiveness that nurture in such a way that they do not become homosexual. Some males would be dominant by genetic heritage, but nurture in such a way that they become submissive and more likely to become homosexual.
Not all homosexuals are submissive, homosexual couples often (mostly?) have a dominant male and a submissive male. It seems unlikely passivity is the sole determining factor in determining homosexual tendencies. We have already given examples where dominate males in all male situations will demand sexual gratification (the prison settings or ancient mariner settings we cited earlier). Beta males of our very distant ancestors may have used forcing sexually submission of lower caste males to maintain their right to succession when the current alpha male dies, ages, or becomes injured. It probably would also make them more attractive to lower caste females that would engage in sex with them when given unobserved mating opportunities. I have no figures, but this would be predictive for the dominant member of a homosexual pair to being more likely to engage in bisexual behavior.
Our sexual preferences are probably not totally hardwired at birth. This would be adaptive in that taking on whatever sexual role society offers (whether officially approved or not) and probably leads to the best overall compromise between opportunities to mate versus survival.
This still does not address what adaptive value there would be to totally submissive males that would never engage in heterosexuality voluntarily. And yet ironically the most vocal self-avowed homosexual males often seem the most obsessed with female culture and expression. Many females, especially young female are attracted to submissive, androgynous, non-threatening males. In this case the female may more often initiate the sexual liaison even though the submissive male is less naturally attracted. It is easy to imagine male homosexuals trying to perform an act of reciprocation for their close female friends even though it is not a natural act for them. This provides and evolutionary path forward even for the most submissive males. Given this would be a direct lineage, even if it were a rare occurrence, would probably factor in with equal or greater force than other kin-support theories I have seen use to explain the persistence of homosexuality in Homo sapiens.
I am not a psychiatrist, evolutionary biologist, archeologist, or sociologist so perhaps these speculations are naive and flawed, but I haven't heard these particular speculations put forward as and explanation for the expression and evolutionary persistence of male homosexuality in society. If this theory were true (and I will leave that for others to determine) it would answer whether homosexuality is a sickness or maladaptive -- the answer to both would be a firm no. It also puts nurture and nature on near equal footings.
One prediction of this theory is that modifying behavior to induce submissiveness should be accompanied with an increase in homosexual expression, though this effect would be less pronounced once the subjects have passed puberty and sexual preferences have solidified. It would be hard to perform this experiment directly in an ethical fashion, but it should be possible to find statistical evidence for it. The dominant female mother household having more homosexual male children would already be one example cited in the literature on the subject supporting this view.
I don't know with what accuracy rate people who claim to be able to identify homosexuals in a crowd are actually able to do so. It is not impossible though that there are body-language clues or even voice-pitch clues that point to submissive tendencies which then might be used as rough predictors.
I have not speculated on factors that would lead to female homosexuality, but they probably have some of the same components as speculated for male homosexuality. Except for all but the most dominate of female lesbians the tendency towards lesbianism seems less set in stone. It could be that female sexual preferences are less hard wired than in males, in which case the old saying: "Homosexuals are born. Lesbians are made" may be more than a little bit true. Some adaptations that benefit males would probably express in females as well as long as they simplified genetic expression and didn't decrease reproductive opportunities too unduly."