Thursday, February 12, 2009

Origins of male homosexuality

Male homosexuality has always puzzled evolutionary biologists. It has 30-45% heritability, so there must be some genetic predisposition (Bailey et al., 2000). But how could such a predisposition get passed down from generation to generation? It ought to die out, by its very nature.

Well, not all predispositions are fully expressed. This observation led the economist and evolutionary psychologist Ed Miller to propose that male homosexuality has survived by means of a balanced polymorphism (Miller, 2000). In the human species, the male has to invest more in his offspring, particularly if the female is less able to feed herself and her offspring through food gathering, as in non-tropical environments. To change male behavior in this direction, the fastest way, with the least genetic change, is to partially feminize the male brain. This could be done through heterozygote advantage at some gene loci: if only one allele is changed, a man would become more child-oriented while still being heterosexual; if both alleles are changed, he would also become feminized in his sexual orientation.

But heterosis effects are absent at most loci. There is usually a simple dominant/recessive mode of action. Nonetheless, a balanced polymorphism is still possible through additive effects over many loci. If a certain proportion have alleles for mental feminization, a man will take more interest in his children without becoming feminized in his object of sexual interest. There may, however, be more than one ‘right’ mix of feminization/non-feminization alleles (the number of possible permutations being a function of the number of relevant genes). Even if all men in one generation have the right mix, reshuffling of genes from their generation to the next would produce some male offspring who are too feminized and others not enough.

Of course, this situation would be gradually corrected by natural selection. The ‘right’ mixes would not be exactly equal in their adaptive value. Natural selection would tend to favor some over others and eventually there would be fixation of one mix of alleles. But this would take a long time and paternal investment varies in importance even among the relatively young populations of Homo sapiens—itself a young species.

So this is Miller’s theory: male homosexuality is the tail end of a balanced polymorphism due to selection for men who invest more in rearing their children and provisioning their families. Santtila et al. (2009) tried to test this theory, which they describe as follows:

Miller (2000) speculates that if a typical man inherits only a few of the alleles partially preventing androgenization, he would express more kindness, sensitivity, tendermindedness, and empathy. Only occasionally would a sufficient number of these alleles come together to produce homosexuality. Heterosexual carriers of these alleles would be better fathers and more attractive mates as men with such traits are less likely to harm their partner or children (Miller, 2000). These traits in men would help them to attract women and, thus, lead to greater reproductive success, ensuring the survival of the alleles in the gene pool. A heterosexual man without any of these alleles would exhibit traits such as ruthlessness, selfishness, insensitivity, and cruelty, making them unattractive to women.

This isn’t quite what Miller had in mind. Yes, a heterosexual carrier would be less likely to harm his partner and children, but a more critical factor would be his willingness to provide for them. Nor is it necessary to assume that such men are preferred as mates. Their greater reproductive success would come from the higher survival rate of their children. We’re talking natural selection here, not sexual selection.

This misunderstanding affected Santilla et al.’s experimental design. Heterosexual men with homosexual brothers were questioned on a wide range of behaviors: psychopathy, sexual aggression, sexual coercion, sexual activity, and number of children. No questions were asked about interest in children or willingness to care for them. Number of children might inversely correlate with paternal investment (some could die through neglect or insufficient care), but only if there are enough of them. Nowadays, family size is too small for this factor to be significant. Indeed, the heterosexual subjects had each fathered less than one child on average.

No significant differences were found between the heterosexual ‘carriers’ and a control heterosexual group. But it’s not clear to me that a difference should be expected, given the experimental design.


Bailey, J.M., M.P. Dunne, & N.G. Martin. (2000). Genetic and environmental influences on sexual orientation and its correlates in an Australian twin sample. Journal of Personality and Social Psychology, 78, 524-536.

Miller, E.M. (2000). Homosexuality, birth order and evolution: Toward an equilibrium reproductive economics of homosexuality. Archives of Sexual Behavior, 29, 1-34.

Santtila, P., A-L. Högbacka, P. Jern, A. Johansson, M. Varjonen, K. Witting, B. von der Pahlen, N. K. Sandnabba. (2009). Testing Miller’s theory of alleles preventing androgenization as an evolutionary explanation for the genetic predisposition for male homosexuality. Evolution and Human Behavior, 30, 58-65.


Anonymous said...

I told him that the prevalence of homosexuals is at least as high among blacks and Hispanics. According to the General Social Survey, it's 3.0% of white men, 5.4% of Hispanics, and 3.8% of blacks. I suggested that non-white homosexuals are more invisible since their ethnic groups are small, and there is less acceptance of the orientation among these groups.

According to The National Longitudinal Study of Adolescent Health, 7.0% of white adolescent boys say that they have been romantically attracted to another boy. The corresponding number for blacks is 8.2% and 9.0% for Hispanics (N = 3,101). Interesting how the ranking is the same for the two surveys.

Anonymous said...

It is interesting that among sheep around 10% of rams are exclusively interested in and only attempt to mate with other rams.

There would appear to be similarities between the reduced pre-optic area in homosexual rams and human male homosexuals.

While this theory is very interesting I still like Cochran's virus theory better.

Peter Frost said...

I have problems with Greg Cochran's theory. If male homosexuality is caused by a viral infection, it should be less common in small hunter-gatherer societies than in larger agricultural societies (where the possibilities of transmission are much greater). Yet male homosexuality is widely attested in Amerindian groups of all categories: hunter-gatherers, simple agriculturalists, and complex agricultural societies with state formation.

In any case, there is probably more than one cause for male homosexuality. I suspect that the rising level of 'environmental estrogens' is another factor.

Most social surveys make no distinction between exclusive homosexuals (who avoid heterosexual sex) and heterosexuals who engage in homosexuality as the 'active' partner. Men in the latter category usually deny they are 'gay'. I don't know about blacks and Hispanic Americans, but I know there is a difference between northern and southern Europe in this respect. In the South, male homosexuality is less common in the exclusive form and tends to be seen as an expression of male heterosexuality. In the North, exclusive male homosexuality is the most common form.

Anonymous said...

Could it be a a side effect of a sexual competition that is too intense for males, an expression of the "survivor" syndrome?
"Survivor" is this TV show where participants have to eliminate the bests players because they are a greater threat, so that at one point teams are composed of one strong player surrounded by a courtship of weaker players.
Under intense male competition, this is a selective adaptation for the dominant male to select the less agressive males instead of more agressive competitors.

Is it possible that under less intense competition between males, there would be less "dominance-associated" features and accordingly less "submissive-associated" features ?


Anonymous said...

If this explanation is correct then a prediction would be that different groups would have different rates of male homosexuality and it would depend on the extent to which females need help in rearing children in the environment in which they evolved.

Do we have stats on homosexuality rates for sub-Saharan Africans, East Asians and Caucasians?

Anonymous said...

Last anonymous, yes we do. See the first post.

Tod said...

"I have problems with Greg Cochran's theory. If male homosexuality is caused by a viral infection, it should be less common in small hunter-gatherer societies than in larger agricultural societies (where the possibilities of transmission are much greater). Yet male homosexuality is widely attested in Amerindian groups of all categories: hunter-gatherers, simple agriculturalists, and complex agricultural societies with state formation".

"among the Plains Indians there were males, called berdache, who did not participate in the “normal” male world but instead were of ambiguous sexuality"

From what I can gather on the net the Inuit never had this tradition. Moreover they are seemingly a lot less accepting of homosexuality than other Amerindians, and formerly lived in smaller groups. To my way of thinking Cochran's theory is consistant with these facts. Maybe you are rethinking things but in Testosterone and Greenland Inuit "Inuit Greenlanders have very low levels of prostate cancer because they have fewer alleles of the sort that increase androgen receptor activity or facilitate testosterone to DHT conversion.

The difference was striking, even in comparison with Swedish subjects—whose incidence of prostrate cancer falls within the world average. The authors concluded:

Our results suggest that Greenlanders are genetically predisposed to a lower activity in testosterone to 5α-dihydrotestosterone turnover and to lower AR activity, which, at least partly, could explain their low incidence of prostate cancer"
Wouldn't Miller's theory lead us to expect a much higher level of homosexuality among the Inuit who are lacking in androgenization?

"Boys who grow up in urban
areas are something like three times more likely to be homosexual than those who grow up in rural areas. The stats for schizophrenia are quite similar. Your idea that identical twins would end up with identical fates when exposed to an
infectious diseases is often not the case. Concordance for polio among identical twins is 36%. You know they both had to be exposed. Exactly how this works is unknown. We do know that the vast majority, at least 99%, of people who contract polio do not get paralyzed. The polio virus is really an intestinal bug, but sometimes it wanders into
motor nerves and causes trouble. MZ twins are more likely to be concordant than DZ twins
( 6% concordance for DZ) ( also true of homosexuality) but most are still not concordant.
MZ twins are concordant about half the time for TB"

What Cochran says about urban areas sounds very telling to me, the greater acceptance in urban areas might lead to them having fewer people lying about being homosexual of course but the data on schizophrenia doesn't suffer from the the same drawback and hence it supports Cochran. His main argument is it is too common "Homosexual men have low fitness, around 0.20 in recent
US surveys. And they aren't what you'd call rare: probably over 1%, possibly as high as
4%. That is an anomaly. You don't expect any substantial fraction - and for these purposes
1% is pretty substantial - of the population in _any_ species to pursue a strategy that is
never going to work well. Natural selection is continually pruning any gene that would
cause something like that, and for that matter genes that tended to prevent exterior
influences from inducing that strategy would increase."

He's mistaken on some things like the importance of vitamin D, which is basically the natural selection for climate idea of William Charles Wells, but the virus theory has more going for it.

Miller's theory is relevant to northern as compared with southern Europe. In this post it's being applied to a specific kind of homosexuality: the 'passive' or unmasculine kind which may be most common in Northern Europe as the byproduct of an environment where selection for domestic non-violence and provisioning was most intense. Judged by how accepted homosexuality is in northern Europe this may well be true. Denmark is more accepting of homosexuality than any other state, even in liberal Scandinavia.

In Denmark homosexuality was legalised in 1933, the age of consent is 15, and it was the first country to legalize same sex unions

What is the average level of masculinity in Danish men?
Danish men may be feminized by comparison with other countries a study found them to have a higher 2D:4D finger ratio than men from other countries,(High digit ratio in danish men, M. Vorasek & M.Dressler. University of Vienna) Although the evidence so far is not conclusive this is suported by indications of reduced fertility in Danish men.

Men tend to be less accepting of gays than women and thus the most masculine men are generally the least approving of all. A community where more masculine men are relatively common may, it can be predicted, be unlikely to approve of or, even accept gays than one where men are less masculine on average. Jamaica is a society where, until foreign white people forced them to accept white standards, you could be a very successful artist with songs about murderous violence towards homosexuals as Beany Man and half dozen others did in Jamaica (one of many examples goes "I'm dreaming of a new Jamaica, come to execute all the Gays")

Assuming that your ideas on sexual selection are correct then I think 'passive' homosexuality, to the extent that it occurs among black Africans and Hispanics, must be due to somewhat different causes than it is when present in Europeans.

Peter Frost said...


Most of the studies I've seen on ethnic/racial differences do not distinguish between different kinds of homosexual behavior.


Homosexuality is not uncommon among the Inuit, but tends to be of the facultative sort, e.g., two men go out on a hunting trip and engage in sex because no women are available. But you're right. The 'berdache' phenomenon is not attested among the Inuit whereas it was common in Amerindian societies.

But how would Cochran explain this discrepancy? Berdaches were common among the Ojibwa of Manitoba, who were small bands of individuals scattered over vast territory. The following is an 18th century account of an encounter with a berdache in what is now Manitoba:

"Some time in the course of this winter, there came to our lodge one of the sons of the celebrated Ojibbeway chief, called Wesh-ko-bug, (the sweet)... This man was one of those who make themselves women, and are called women by the Indians. There are several of this sort among most, if not all the Indian tribes. They are commonly called A-go-kwa, a word which is expressive of their condition. This creature, called Ozaw-wen-dib, (the yellow head), was now near fifty years old, and had lived with many husbands. I do not know whether she had seen me, or only heard of me, but she soon let me know she had come a long distance to see me, and with the hope of living with me. She often offered herself to me, but not being discouraged with one refusal, she repeated her disgusting advances until I was almost driven from the lodge."

I contacted Ed Miller for his thoughts on the matter, but he has not answered so far.

Peter Frost said...

The link should be:

Anonymous said...

The main problem I have here is that this notion that feminization of attitudes towards children and an increased willingness to provision females leads also to errors in mate choice, as if these three items were somehow inextricably linked from a genetic perspective.

However, it is not at all clear that all homosexuals show all three traits that are being advanced as linked. Based on my reading, a great many would seem to have a heightened sex drive (an essentially male trait) and an attraction to the same sex but with no interest at all in Children.

Furthermore, since there seems to be considerable evidence from hunter gathers that males who are successful at hunting tend to get more sex and thus more offspring, selection could work on genes controlling such behavior to shift it towards provisioning the mother(s) of successful males offspring more than others within the group.

Indeed, it would seem that any variance towards hard work together in an agricultural setting would tend to be selected for.

So, I am not convinced that feminization is the answer.

Tod said...

Of course, this situation would be gradually corrected by natural selection. The ‘right’ mixes would not be exactly equal in their adaptive value. Natural selection would tend to favor some over others and eventually there would be fixation of one mix of alleles. But this would take a long time and paternal investment varies in importance

Might this process of correction have been greatly speeded up for Inuit men if any excessively reduced androgenisation (conferring a tendency to non-faculative homosexuality in addition to provisioning) also tended to make them poor hunters.

In a time of scarcity ineffective hunters in Greenland would have been subjected to natural selection an order of magnitude greater than that in Manitoba.

Bandar Bola said...

I'm now not sure the place you are getting your information, but good topic. I must spend a while finding out more or understanding more. Thank you for fantastic information I used to be searching for this info for my mission. Taruhan Bola

Matthew Wettlaufer said...

As a psychologist I am alarmed at some of the comments on this blog passing for scientific discussion--you are speaking of people as if they have a disease. The way some of the commentators have talked about homosexuality reminds me of 19th century discussions of eugenics regarding Black people--where the speaker is completely unaware of their bias and power in the discussion. Be careful about that--if you speak about a group of people as if they were statistics or lab experiments, it will discredit anything you are trying to say. It's fascinating for example how many assumptions have been included in the discussions--that homosexuals are weaker than heterosexuals (regarding indigenous cultures) or that homosexuality implies a feminization of men (really? is that always the case?) or the fact that homosexual women have been conveniently left out of the conversation entirely. Constructing a thread along these lines says more about the contributors than it does about the topic! For my part I would suggest that 1) sexual identity is not reducible to sexual activity, it is not the same thing, and what constitutes sexual identity is far more complex than is being understood in this thread. 2) evolution and natural selection could be twisted into explaining homosexual ACTIVITY, perhaps, as has been shown, by additions that are far fetched at best (viruses for example) but it does not explain sexual identity (love idealization, spirituality, worldview). 3) Because of this, a biologistic explanation of sexual identity will inevitably fail under the weight of its own presumptions and lack of complexity that must factor in non-biological variables. Acknowledge the limits of your field--I would never suggest psychology can explain something this difficult by itself. Also ask yourself the question as to why you are trying to "explain" it in the first place? Would you be trying to explain why Black people are black? The act of trying to "explain" the existence of a body of people is highly suspicious--it suggests by default an assumption that there is something out of keeping with the existence of that body of people, besides being highly disrespectful and inhumane (eugenics again). Don't hide prejudice behind a facade of objectivity--it will poison your work. Remember that you are talking about people.