The evolutionary mystery of homosexuality

The theory of evolution by natural selection says that changes come about incrementally, as a result of the long-term consequences of small selection advantages for favorable traits. The selection advantage is measured by differential rates in the production of offspring. If organisms with a new and favorable trait produce 101 offspring for every 100 produced by the older forms, the selection advantage s is said to be 0.01. The changes produced by even such a small reproductive advantage can be quite dramatic. If we start with a trait that is present in just 0.1% of the population and if this has a small selection advantage of size s=0.01, this variety will grow to become 99.9% of the population in just under 1,400 generations (in the codominant mode of selection) which is a very short time on the evolutionary scale. (Molecular Evolution, Wen-Hsiung Li, 1997, p. 39)

This presents a bit of a puzzle when it comes to explaining the existence of homosexuality if it is a biological trait, since on the surface it seems to have a selective disadvantage. David Barash has an article that explores what we have learned about the six main hypotheses that have been proposed to get around this problem: kin selection, social prestige, group selection, balanced polymorphisms, sexually antagonistic selection, and nonadaptive byproduct.

He concludes that none of them is the clear winner. But there are many as yet unanswered questions in evolutionary theory that researchers are investigating and this is one of them. He concludes that, like the others, it will be solved eventually.

There are lots of other hypotheses for the evolution of homosexuality, although they are not the “infinite cornucopia” that Leszek Kolakowski postulated could be argued for any given position. At this point, we know enough to know that we have a real mystery: Homosexuality does have biological roots, and the question is how the biological mechanism developed over evolutionary time.

We can be confident that eventually, nature will tell.

In terms of social policy, the answer to this question does not really matter. Treating people equitably irrespective of their sexual orientation means that we do not discriminate based on theories of the origins of the behavior either.


  1. slc1 says

    This may turn out to be something like the gene for sickle cell anemia, which is recessive. Having the gene on one chromosome provides resistance to malaria and so has persisted in African populations that live in areas with high incidences of malaria bearing mosquitoes.

  2. says

    It’s definitely an interesting puzzle.

    But you’re right. Even if it were evolutionarily disadvantageous, using that fact as an excuse to discriminate against homosexuality makes as much sense as using my asthma to justify denying me civil rights. Our personal freedoms are not contingent on our ability to propagate.

  3. skaduskitai says

    Another thing to consider is that it’s largely dependant on the society wether homosexuality is a reproductive disadvantage or not. Lots of homosexuals have reproduced, especially in societies where you pretty much don’t have a choice.

  4. F [disappearing] says

    This presents a bit of a puzzle when it comes to explaining the existence of homosexuality if it is a completely genetically determined biological trait.

    For the bigots who like to use pseudo-scientific reasoning against teh gay, I’m sure they are equally astounded that there is any material universe left over after the era of matter-antimatter annihilation.

  5. says

    What you are basically asking is how does a trait that seems at a reproductive disadvantage remain in the population? First, one of my gay friends has a daughter, his own; he was married to a woman when young, and figured himself out later. So some homosexual men and women do have biological children. But as mentioned recessive traits remain hidden from selection, and homosexuality seems to be around 2% of the population. Homosexuals have as many siblings on average as anyone else, so in the USA, about 1, and they will each have a 50% chance of being a carrier. The rest of the Hardy-Weinburg math is beyond me right now, but this seems quite adequate to maintain a homosexual gene in the population even when the homozygotes don’t reproduce.

  6. baal says

    Phrasing like “the mystery of homosexuality” is misleading at best and I question the sexual selection disadvantage.

    If you survey same sex sexual relations in animals, you find homosexual behaviour is widespread. This means that a population can have a certain level of gayity and still be perfectly able to utilize reproductive opportunities or that there is a fitness cost to the population for getting rid of homosexuality fully.

    Evenmoreso, the usual bigger problem is getting offspring to reproductive age -- i.e. there is rarely a shortage of offspring producing sex. In species which do seem to be sex limited -- pandas and platypus for example, the problem isn’t usually gay sex but rather getting a putative hetero male and a putative hetero female to have sex instead of fighting or running away.

    Opposite sex identification and attraction systems are both complex and need some plasticity. Given how I understand the rest of biology (used to be a dev. bio. phd), those two requirements mean you’ll get a range of phenotypes. A too stringent a system and you’d have perfectly otherwise fit males and females who have some minor deviation not reproducing. You can think of a number of scenarios with too much or too little stringency and survey the current species for examples who match those niches. When I run that exercise (bench in my head science, not the best but enough to refute the idea that homosexuality is unnatural or needs an special explanation beyond the general explanation for the system generally), there isn’t a reason (selective pressure) to stop gay sex.

    I should probably stop now and go read the paper.

  7. cp3o says


    Professor Andrea Camperio Ciani at the University of Padova in Italy has tested various hypotheses by studying 100 families of gay men. Not only did he replicate Blanchard’s birth order effect, he also detected inheritance of homosexuality on the mother’s side, supporting Hamer’s idea of a gay gene on chromosome X. The maternal inheritance effect seems most important too.
    “Genetics explains 20-25 per cent for the moment,” says Camperio Ciani. “The rest is unknown. A part is environment; a part can be other genetic elements that we cannot perceive with our study.” In principle, the genetic component might even be the Xq28 region.
    Regardless of which regions of DNA are linked to homosexuality, the very existence of ‘gay genes’ creates a Darwinian paradox. How would genes that cause homosexuality pass from one generation to the next, given that gay people reproduce less than heterosexuals? Natural selection opposes anything that might cause even a small reduction in the number of offspring you produce, so a gay trait would soon disappear from the gene pool. “If you carry a trait that reduces your fecundity [the number of offspring you produce] by 10 per cent, in seven to eight generations your trait and all your descendents disappear,” says Camperio Ciani.
    The paradox was finally resolved by his 15-year-old daughter. After Camperio Ciani described the observed patterns in pedigrees of homosexuality – the effects of maternal inheritance and birth order – his daughter suggested that he re-check his data to see if the female relatives of gay men had more children on the mother’s side. When Camperio Ciani went back to the lab, that’s exactly what he found. “Mothers and aunts on the maternal line of homosexuals had around one-fifth to one-fourth more kids than the heterosexual comparison, and also than the paternal line.”
    He thinks that the evolution of homosexuality is driven by a process called sexually antagonistic selection. It’s where a genetic factor confers an advantage when expressed in one sex, but incurs an evolutionary cost in the other. In this instance, the ‘gay genes’ don’t exist to make men homosexual, instead they’re a consequence of ‘fertility factors’ that help women reproduce.

  8. Ray de Silva says

    There’s a lot of evidence that nearly all teen girls go through a transient homosexual phase, whose biological purpose seems to be group bonding. This phase, though, is quite short -- a few years -- and is usually completely replaced by “normal” heterosexual attraction by around 17. The pity of it is that most researchers have assumed that all homosexuality is abnormal, and their entire theses are biased from that viewpoint.

  9. theAtheistAxolotl says

    The problem when you assume that a trait is under “Hardy-Weinberg” math is that these calculations assume genes controlled by a single loci. This doesn’t work for traits controlled by multiple loci, which usually exhibit a range of possible outcomes rather than two or three possibilities. Height, for instance, can’t be (easily) represented by these calculations because it relies on several genetic factors all working together (as well as environmental factors). Since sexuality exists in a range rather than set states, it seems unlikely, in my mind, that it could possibly be controlled by a single loci. Also, most of the current hypothesis for the evolution of homosexuality deal with it being a selective advantage for a small portion of the population (through kin selection).

  10. invivoMark says

    Eh, I am not impressed with the article. For one thing, it fails to mention the maternal hormone effect, which is currently the hypothesis overwhelmingly supported by biologists as the most important factor determining sexual orientation. The maternal hormone effect has been recapitulated in the ram, which is the current most popular animal model for homosexuality studies (PubMed it, there’s lots of interesting stuff out there).

    So any discussion of the genetics of homosexuality will be at best incomplete, and at worst entirely misleading, if maternal hormones are not mentioned.

    Given that it is mostly a maternal hormone effect, this opens up several other explanations (health of the fetus, health of the mother, maternal age factors, maternal diet…) that are each far more successful explanations for the genetic control of homosexuality than any of the options mentioned in David Barash’s article, with the exception of “nonadaptive byproduct”.

    Furthermore, it’s important to recognize that homosexuality is not an either-or thing. Sexual flexibility/fluidity may be the trait under selection, while strict homosexuality is a nonadaptive byproduct of this tendency.

  11. Kilian Hekhuis says

    “if it is a completely genetically determined biological trait” -- but it very likely isn’t, the strongest cases made by identical twins with non-identical sexual preferences.

  12. raymoscow says

    Sexual reproduction itself is hard to explain by evolutionary theory, since each parent loses half its genes in each generation — as compared to asexual reproduction, in which all the genes are passed on — and it’s hard to find benefits that overcome this obvious detriment (from the gene’s point of view). But of course there has to be an explanation, and some tentative ones have been advanced.

    There’s no doubt that prevalence of homosexuality in many species, including ours, has a natural explanation as well, although it might be quite complex or difficult to demonstrate.

  13. Kilian Hekhuis says

    Sexual reproduction itself is hard to explain by evolutionary theory, since each parent loses half its genes in each generation

    Perhaps the explanation is the same as with horizontal gene transfer of asexually reproducing species, viz. incorporating other’s succesfull survival strategies within your own.

    There’s no doubt that prevalence of homosexuality in many species, including ours, has a natural explanation as well

    Homosexuality may very well be a byproduct of the way the foetus develops, and even if it has a genetical basis, may not be subject to evolutionary preasures.

  14. says

    As for genetic predisposition, I’d like to say that “it’s a choice” vs. “it’s genetically determined” is a false dichotomy: there are plenty of individual human attributes that are neither chosen nor genetic, most of them (IMHO at least) determined by environment and conditioning. Take musical taste for example: I can’t choose which music I find pleasant and which makes me want to gouge out my ears — does that mean I’m “genetically programmed” to like Arabic music and hate Toby Keith? I find that idea very implausible — especially when I remember that my musical tastes have changed over the years, and I can’t choose how they change either.

    It may be much the same with sexual orientation: there may be a genetic component, but there may also be environmental factors that do most of the actual “predetermination;” which would then mean that there isn’t as much of an “evolutionary mystery” to homosexuality as some seem to think.

  15. says

    Furthermore, it’s important to recognize that homosexuality is not an either-or thing. Sexual flexibility/fluidity may be the trait under selection…

    This is a very good point. It should also be noted that reproduction is not the only behavior that is beneficial to a tribe (especially for more complex social creatures like humans): if a certain number of people don’t have any kids, but spend their time doing something else beneficial to the next generation, then they could well enhance the tribe’s viability more by that action than they would if they had more kids and left the other activity undone. (Menopause has been mentioned in this context: when older women stop having kids, they have more time and energy to spend on such necessities as teaching, socializing, child-rearing, and passing on accumulated wisdom.)

  16. richardrobinson says

    I’ve often wondered if we aren’t a little blinded by our culture in this research.

    Ostensibly, we’re a monogamous species that abstains from reproducing until marriage.

    In practice, we have loads of sex with lots of people. Given that something like only 1 in 1000-10000 acts of coitus results in reproduction, sex obviously serves an important function outside reproduction.

    Therefore I question the whole premise of this exercise. Does sexual orientation have as big an impact on reproduction as is assumed here?

Leave a Reply

Your email address will not be published. Required fields are marked *