Gay genes

A recent paper discussed the results of a GWAS study of same -sex behavior, based on data from the UK Biobank. The data wasn’t ideal – it was based on self-reporting, and asked whether the respondent had ever had a same-sex experience, rather than trying to detect orientation… But that’s life.

They found two SNPs that influenced both male and female homosexuality, two that affected males only, one that affected females only. All had very small but statistically significant effects.

If we had lots of SNPs, we could look for trends ( are they predominantly expressed on certain tissues or processes ?)  but with only 4 in males  and 3 in females, not really possible. It does look as if the genetic architecture of homosexuality only partially overlaps between the sexes ( 0.63) – usually, say for things like height or iq, the overlap is close to 100%.  So the genetics, like every observation, suggests that male and female homosexuality are qualitatively different. Yet the degree of shared genetic influence is also interesting – I don’t think it would be predicted by most strategy models.

The fraction of the variance influenced by these few SNPs is low, less than 1%. The contribution of all common SNPs is larger, estimated to be between 8% and 25%.  Still small compared to traits like height and IQ, but then we knew that  the heritability of homosexuality is not terribly high, from twin studies and such – political views are more heritable.

So gene influence homosexuality, but then they influence everything.  Does it look as if the key causal link ( assuming that there is one) is genetic?  No, but then we knew that already, from high discordance for homosexuality in MZ twins.

Most interesting to me were the genetic correlations between same-sex behavior and various other traits.


The genes correlated with male homosexuality are also correlated ( at a statistically significant level) with risk-taking, cannabis use,  schizophrenia, ADHD, major depressive disorder, loneliness, and number of sex partners. For female homosexuals, risk-taking, smoking, cannabis use, subjective well-being (-), schizophrenia, bipolar disorder, ADHD, major depressive disorder, loneliness, openness to experience, and number of sex partners.

Generally, the traits genetically correlated with homosexuality are bad things.  As far as I can see,  they look like noise, rather than any kind of genetic strategy.  Mostly, they accord with what we already knew about male and female homosexuals: both are significantly more likely to have psychiatric disorders, far more likely to use drugs.   The mental-illness association maybe looks stronger in lesbians.  The moderately-shared genetic architecture seems compatible with a noise model.

The strategy ideas never made sense ( in terms of a workable kin-selection, sexually antagonistic, or other genetic model) , but I think this study makes that clearer.

Find that homosexuality was genetically correlated with various kinds of unpleasantness was apparently an issue in the preparation and publication of this paper. The authors were at some pains  to avoid hurting the feelings of the gay community, since avoiding hurting  feelings is the royal road to Truth, as shown by Galileo and Darwin.

This entry was posted in Uncategorized. Bookmark the permalink.

126 Responses to Gay genes

  1. david says:

    Another possible clue: I noticed all the gay bars ive been to were comprised of majority nonwhite customers. Research seems to confirm black and latino men are 50% more likely to “identify as gay.” The real numbers are probably higher, since those communities are actually more opposed to the lifestyle than whites. My theory was female dating preference combined with testosterone levels lead to more experimentation out of desperation. A man of color is the least searched for on dating sites, last I checked. Difficulties in dating women, high testosterone, higher propensity of child abuse, and male mentors returning from incarceration with homosexual experiences under their belt could create a decent enough environmental motive for abnormal sexual attraction. Milo Yiannopolous made an interesting argument that gay men have been forcing themselves to be bisexual for millennia, otherwise theyd have been removed from the gene pool. Maybe they will be now. Unless of course its a gay disease like you’ve proposed.

    • Frau Katze says:

      Maybe black and Latino gays go to bars more because it’s less accepted in their communities. It may be the only place they can go to meet other gays of their ethnic background. It would be interesting to know what fraction of gays settle into fixed relationships. You’d expect fewer if the family and friends were strongly disapproving.

      Being a white female non gay I can’t comment on the desperation theory. But I have read that in pretty much any situation where there are few women (prisons, single sex boarding schools and, at least formerly, navy ships) the men do seem to follow a desperation strategy.

    • Yamamoto says:

      Still seems like gay neighborhoods like Boystown (in Lakeview East) in Chicago are disproportionately white, but I might be wrong.

    • Ethan says:

      The more rational answer is found in maternal immune responses. That’s been found in a lab as of 2017, that the more previous sons increases the chance of the next one being gay. Mothers body creates antibodies and interferes with brain masculinisation. In addition, maternal stress is linked with homosexual offspring. Thus, homosexuals are more likely to have mental illness not because they are causing one another, but because more maternal stress and immune responses from larger families and more abortions in minority communities.

  2. reziac says:

    In animals, homosexual behavior is basically a juvenile trait; normally the animal matures and thereafter such behavior is just opportunistic when better options are absent. Occurs to me that if the same is true in humans, then there’s probably some failure to mature in the hormonal system, which in turn leads to both the various negative traits and the homosexuality.

    • magusjanus says:

      not in rams it’s not. or rather, a certain % of them. it’s very perplexing to me how no one, not one millionaire or weird Saudi or rich religious fundamentalist out there funds research into this. seems like a massive layup for the Right in finding a virus link.

      • reziac says:

        I’d like to know if the affected rams were twins or triplets, and the gender of their sibling lambs. Guessing there’s a shift dependent on that. (Not going to buy the article to read the rest of it.) Kinda like how longer ring finger in human females is from androgen exposure in utero, and tends to correlate with more-masculine traits.

        One could wonder if there’s a similar effect in other species’ mixed-gender twins or triplets. By the time one reaches larger litters, any effect probably evens out.

    • jb says:

      My understanding is that in wild animals — (making a possible exception for genetically screwed up domestic animals like sheep) — obligate homosexuality is essentially unknown. They may engage in various types of homosexual behavior on various occasions for various reasons, but never at the expense of normal reproductive sex. So there really is no good animal analogue for committed human homosexuality.

      • Jason y says:

        Humans can and do commit to self-defeating behaviors that may have proved expedient in the short-term, e.g., as a Min-Max strategy on the playground, or as a political stance, e.g, intellectuals committing to failed theories, or carpenters to “their way”, but most commonly they do it out of mindless habit, I.e., further investment in exploring the space of solutions doesnt (seem to) make sense. Much of what we do is a cobbled together collection of first-guesses that congeal into habits. That’s why genes matter so much even in Taleb’s hyperdimensional imagination: where we start looking matters the most.

        How many committed homosexuals occasionally feel attracted to members of the opposite sex? We dont know, and for reasons having to do with self-labelling, status, and self-esteem, they arent going tell us. From what I have observed of lesbians, the answer is “sometimes”, but because they know they cant win the male’s affections theyll often aggressively deny it. Their identity is built out of sour grapes.

        • Ryan says:

          Blogger Severian: “The strongest force in human affairs isn’t love or hate, it’s inertia.”

          • kpkinsunnyphiladelphia says:

            In my wild oats days, I had relationships with two “lesbian” women, In the first, I think I was her last heterosexual encounter and she didn’t seem that into me or it. I liked here, she was quite attractive, and I wanted to keep it going, but she decided man were not fore her. The other was already in a “committed” lesbian relationship, and told me that she needed a man from time to time. A small sample size, to be sure, but there you go.

        • Matt says:

          Jason, this is the stupidest comment I have seen here. Obligate homosexuality has a huge fitness cost. Even if gays had 98% as many kids as straight men, it would never persist. You aren’t up to speed.

    • Labayu says:

      Seems to be the case among teenage boys in Morocco to whom there is little if any safe access to teenage girls. This is part of what made Morocco so appealing to the likes of William S. Burroughs and Allen Ginsberg.

  3. Anonymous says:

    We know that “gay face” is real from personal experience but studies have also shown this intuition is true. People are pretty decent at detecting homosexuality from looks xor voice in males.

    That associated stuff must be genetic.

    • greg kai says:

      I don’t think so: pathogens, development defect or other physiological events can certainly be detected visually, on the face or the body. Polio, smallpox, thyroid problem,…you name it…

      Gay face do not say anything regarding genetic origin, but if you have a gay/straigth homozygots twin pair , and the gay twin can be reliably guessed from a (cleaned up) mug shot, then you can deduce that gay face is a physical sign of whatever non-genetic incident is causing homosexuality, instead of a marker for genetic suceptibiity to such incident. The fact that homozygots often differ in gayness shows that it’s genetic suceptibility to an external factor, not direct genetic causation

    • oldmiseryguts says:

      Most straight men can pick a gay man out fairly easily, usually just by facial features (not always), there’s always finger lengths, hair whorl pattern and posture speech etc. Most lesbians I’ve met or worked with are generally just really unattractive women who have given up on getting a man, most lesbians go both ways whereas gay men tend to be much more gay than bi sexual.

    • David Chamberlin says:

      Being gay isn’t inherited, end of story.
      There are some correlated variables that confuse the issue.
      Left handed, schizophrenic, having more older brothers, being in love with a monkey, getting bitch slapped by clowns at a young age, the list keeps growing.

      Here is how genetics works. Mutations that put a small hit on fitness don’t spread very far. They are rare. A great and wonderful scientific undertaking was completed a while back that was going to make the world a better place. The Human Genome Project. We were supposed to then find all those terrible nasty mutations and fix them. Kids would dance, parties would happen, music and balloons would proliferate.

      Didn’t happen. It seems that genes that take a small hit on fitness are rare. Who woulda thunk. Now a gene that makes you lust over X hole rather than Y hole and X hole never makes babies is going to have a HUGE hit on fitness. So homosexuals are somewhere around 2 percent of the population. It isn’t genetic.

      • David Chamberlin says:

        The cause of homosexuality isn’t known. My guess is it’s a developmental quirk that happens very early, possibly in the fetal stage, but who knows. My point is simple. Inherited conditions that have a huge hit on fitness with no counter balancing positive effect never remain as common as homosexuality is. That is how evolution works.

        • james wilson says:

          The first two weeks following conception are filled with staggering array of metamorphoses. A two percent error rate is either good enough or a miracle in itself, and if it’s higher perhaps spontaneous abortion deals with that.

  4. Unirt says:

    It is consistent with the noise model, but can’t many of the negative traits be directly caused by homosexuality? The genetic correlations don’t rule that out, right? I imagine that if you find yourself in an ostracized position you may be more susceptible to (perhaps adaptive) anxiety and depression and also maladaptive wireheading like taking drugs. This would explain neuroticism, low life satisfaction and smoking, but doesn’t explain scizophrenia, ADHD or risk-taking (depressed people are neither more nor less likely to take risks). Would the noise model of homosexuality predict IQ to be negatively assotiated with being gay? The evidence there is seems to point to the other direction, if simply because more intelligent people are more likely to admit to being gay?

    • another fred says:

      “I imagine that if you find yourself in an ostracized position you may be more susceptible to (perhaps adaptive) anxiety and depression and also maladaptive wireheading like taking drugs.”

      Surely so to some extent. I think we can be confident that the more “blank slate” inclined will hang their hats on that peg and continue to demand that the loonies run the asylum.

  5. James Thompson says:

    It opens up the possibility that homosexuality is not the cause of the behavioural and emotional problems, but simply part of a pattern of which the non-reproduction-oriented sexual behaviour is a part. In short, a series of errors. However, there may be upsides: artistic sensibilities? emotional awareness? It is worth looking more widely to get the full picture, which at the moment looks like being entirely negative, or at the very least, a very poor strategy.

    • Dieter Kief says:

      ” (…) homosexuality is not the cause of the behavioural and emotional problems, but simply part of a pattern of which the non-reproduction-oriented sexual behaviour is a part.”

      Yep, this would mean, that seen from the biological perspective, homosexuality would rather be a solution than a problem.

    • The Z Blog says:

      To split hairs a bit, but a set of errors may be more precise. A set of common errors, as in the set as a whole is common, while anyone person having the whole set is rare. Within that set, subsets manifest as degrees of homosexual behavior, depression, openness, etc. In the overlap of these sets we find homosexuals are varying intensity.

      That’s something that does not come up in these discussions. Homosexuality in males seems to be a spectrum. At one end are the always and forever exclusively gay. At the other end are males who play on whatever field is available. There’s also a dimension to it, with someone like Lindsay Graham being a different type of homosexual from Milo Yiannopoulos.

      • Dieter Kief says:

        “Homosexuality in males seems to be a spectrum.”
        Yes. And that might as well correlate with the social aspects of homosexuality, which means, it is – up to a maybe even_varying_ degree, itself flexible. Homosexuality = a place, where individuals and biology meet under varying social, cultural and individual circumstances? So homosexuality would be – at least at its biological outskirts/ rims – something quite fluid.

      • Janet says:

        To be even more specific: male homosexuality seems to be a syndrome, that is, a collection of symptoms or signs that usually happen together. Sexual behavior is only one of the traits in the syndrome; not everyone has every trait in the cluster, nor every trait to the same intensity. Being a syndrome doesn’t tell you what causes it (genetics, infection, etc.), of course, or even if it’s necessarily a single root cause. The data could also be muddied by men having sex with men when there are effectively no women available (prisons, military units, migrant camps, etc.), male prostitution or abuse, or when the main motivation is aggressive dominance rather than sexual desire per se.

        Our host has noted more than once that there’s a different distribution of sexual preference between male and female homosexuals– females show a smooth bell curve-like distribution from “attracted only to men” through “sometimes have sex with other women” to “50-50” to “mostly women, but sometimes men” to obligate lesbians. Whereas, with men, there are many more obligate gay men than bisexuals. That male pattern doesn’t match well with the cause being an accumulation of genetic variations, or with sexual attraction as a gradient along one axis (say, due to a gradient in sex hormone levels across the population, or disinhibition in general). It does fit with an environmental cause, such as a virus (perhaps with certain genetic patterns being much more susceptible to the syndrome than others).

        • M says:

          This is not true, that there are many m ore gay men than bisexuals. There are at least as many bisexual men as gay men, as polls of sexuality in the youngest generation (Generation Z? Where are we now? sorry…) are confirming.

          I can add my subjective testimony to this fact. I’d put the percentage of truly bisexual men at around 6-10%, with another 3-6% being gay.

          • gcochran9 says:

            The epidemiology of AIDS is not consistent with your statement.

            • Hypoborean says:

              You are correct that if the numbers for Gen Z applied in 1980 the epidemiology of AIDS would be very different. However, that doesn’t mean that M’s statement is incorrect, seeing as 2019 is not 1980.

              Gen Z having a massively higher rate of self-reported male bisexuality than previous generations is easily supported by declining social stigma towards male bisexuality.

              Female bisexuality was and is fetishized (encouraged) while male bisexuality was very strongly stigmatized. Straight men did not avoid dating bisexual women (on the expectation that they would be more open to having FFM threesomes), while straight women did avoid dating bisexual men (on the expectation that they would become fully gay). Since there was a social cost to experimenting for males and a social benefit to experimenting for females, the rates of female and male apparent bisexuality strongly diverged.

              In the last decade as LGBT identities in general have become very strongly accepted / high status in youth culture, the social cost of male-male experimentation has declined significantly, which is why we are seeing the emergence of male bisexuals.

              To summarize: the kids really are different, and you’re making a big mistake in assuming facts from the 1980s disprove the survey data + the lived experiences that people sub age 30 (myself, and I’m assuming M) can attest to.

              I’m not saying that the distributions between the two genders are the SAME. We still have a lot more “obligate” gays than “obligate” lesbians, to use the terminology preferred on this blog. But a major social force that was pushing a gender disparity in the rate of bisexuality (the fuzzy middle zone that you would logically expect to be very open to social pressure, since its members have the OPTION to cave to social pressure without forgoing sex) has dissipated, and we are seeing the in-some-ways-predictable, in-some-ways surprising result of that.

              • Hypoborean says:

                Important addendum:
                Within the “obligate” gay community there is a notorious “top shortage”, meaning more prefer to be penetrated than prefer to penetrate. Since the penetrated partner is far more likely to contract HIV (women, in the case of straight sex, and “bottoms”, in the case of gay sex, with gay bottoms being more at risk than straight women…also note this is why HIV/AIDS is very rare among “obligate” lesbians), if the rare bisexuals of the 1980s were dropping by the community, it is likely they were predominantly performing the role of “top”, which would limit the escape of HIV from the primary infected communities through bisexual men.

              • The Z Blog says:

                In 1980, the number of people identifying as “transgender” was zero. Seeing as 2019 is not 1980, transgender is still a nonsense label. The point being, the alleged spike in bisexuality in Gen Z is without foundation and most likely reflects the level of agit-prop in the schools.

              • Pincher Martin says:

                In 1980, the number of people identifying as “transgender” was zero.

                Tell it to Myra Breckinridge.

            • M says:

              That’s true, but there are other reasons for that.

              Again, read the latest studies on self-reported sexual experiences of the most recent generation. The number of men reporting same-sex behavior has doubled relative to the previous generation, without a corresponding increase in the number of men identifying as gay. All the increase has come from men self-identifying as bisexual.

              When homosexuality is tabooed, it seems that bisexual men adopt a straight lifestyle and self-identity. Now that taboo has seriously eroded, and behaviors are shifting.

    • Frau Katze says:

      Correct my reasoning here if it seems wrong. On another blog a devoutly religious male asked how could the seeming common popularity of oral sex could have ever developed, from an evolutionary point of view.

      I had never thought about it before. But then, it seems that humans are more interested in sex of any type than say, cats or dogs. Female mammals are generally only receptive when they’re in heat. And the males are only interested in females that are in heat.

      It makes sense, why waste time and energy if the possibility of conception is low.

      This is not the case in humans. I don’t know anything at all on the subject from a scientific point of view but I hazarded a guess that it might have improved pair bonding. With strong pair bonding there is less temptation to get involved with someone other than your mate. Such behaviour would affect the stability of the tribe.

      Sex has become partly recreational for humans.

      It might also explain homosexuality, at least partly. It’s negative from an evolutionary point of view but anything that stabilized the tribal unit might be selected for, even if weakly.

      The questioner agreed with my answer (I didn’t mention homosexuality) but wanted more input.

      But everyone else on the thread was distracted by politics and iPhones.

  6. Ziel says:

    Positive correlation with # of children?

    • Coagulopath says:

      I think this correlation would disappear if you controlled for age.

      As people get older, their chance of having a sexual encounter with the same sex increases. Their chance of having a child also increases. So generally speaking, older people will have had more gay sex and more children.

      This is one of the weaknesses of defining homosexuality as “has had a same sex experience”.

      • Lot says:

        “I think this correlation would disappear if you controlled for age.”

        The number of children trait is limited to women 45+ and men 55+.

    • gwern says:

      Those are puzzling ones. It’s possible it’s a misprint (reverse-coding?). If you look at the fulltext, the phenotype correlations with # of children is as expected: Homosexuals of both sexs have less children than 0%/heterosexuals, and the effect is stronger the larger % is (from 1.6 down to as low as ~0.3). As the authors comment, “This reproductive deficit raises questions about the evolutionary maintenance of the trait, but we do not address these here.” How that can translate into a positive correlation with number of children, I am not sure.

      • Eponymous says:

        I’m honestly surprised they don’t discuss this. If the genes are positively correlated with number of kids that would be huge. Though maybe they don’t discuss since the relationship is not statistically significant. But honestly that is quite interesting in itself. They should report the result of a power test for that, since it would be a very interesting null result.

        • NonLinear9 says:

          Eh, it really irritates me when authors spend time discussing their non-significant results. To do so is to miss the point of doing that test in the first place. I’m not also sure why the null result is surprising — why do you find it so?

      • info says:

        Many homosexual men were sexually abused as boys by Gay men.

        • Greying Wanderer says:

          filtering the results (mental illness, drugs, alcohol etc) by experience of molestation or otherwise might be informative as they tend to correlate (anecdotal).

    • Lot says:

      Gwern, the chart on page 2 shows “percentage of same-sex sexual partners” is negatively correlated with number of children, as you’d expect. Though with women it breaks down and there’s isn’t a difference between being 50% and 100% lesbian.

      The large trait chart I believe shows the relationship between gay genes and fertility, which is positive (and also linked to earlier fertility). Authors are not fully clear, in the supplement they say “Genetic correlation between same-sex sexual behavior and several traits of interest, separately for UK Biobank and 23andMe.”

      So if I am reading this right it supports the old theory that people with gay genes have more children when they don’t turn out gay.

      • gwern says:

        I understood the graphs fine. The problem is that if you believe in the inclusive fitness theory, these rgs do not seem to be nearly large enough. (Some of the rg CIs do not even exclude zero.) To very roughly sketch it out: you go from 1.6 to 0.3 children when you become homosexual, a loss of 1.3 children (a massive fitness reduction); this is somehow compensated by your average number of 1 sibling or 2 cousins, who are 50% and 25% related to you; to make up for your -1.3 children, your siblings need to have 2.6 children (in addition to their own baseline of 1.6) or your cousins have 5.2 children or some combination thereof etc (assuming of course none of them are homosexual, in which case the others have an even larger deficit to make up for). That’s a lot of kids, for traits which have low heritabilities to begin with! An rg of ~0.05 doesn’t seem like it is remotely enough to cut the mustard, and the inclusive fitness theory is wrong. However, since it’s wrong, that still leaves puzzlingly positive correlations with fitness.

        • Eponymous says:

          It could work out if only a small fraction of the people with a particular “gay gene” turn out gay. How common are the genes in question anyway?

        • Greying Wanderer says:

          “puzzlingly positive correlations with fitness”

          maybe these genes they found are somehow connected to promiscuity more than homosexuality?

          maybe these studies need to separate life-time homosexuals from people who have had occasional or possibly even a single experience (nb correlation with drugs and alcohol)?

      • Lot says:

        gwern, i am not defending “homosexuality evolved because of gay uncles.”

        If the gay gene, on its own, caused homosexuality risk to go up 1%, but fertility if not homo by 2%, then it gets selected for.

        My guess is male homosexuality is primarily caused by rare mutations, with schizophrenia being a closer match to what I’ve observed and read as a model for homosexuality than toxoplasmosis.

        • gcochran9 says:

          Very high MZ twin discordance..

        • David Chamberlin says:

          “My guess is male homosexuality is primarily caused by rare mutations.”
          Some of it is influenced by rare mutations but the problem is male homosexuality is not very rare. Something other factors are at work to create the high percentage of homosexuals. Something not hereditary. Cochran has repeatedly made the argument that I can’t disagree with that homosexuality can’t be hereditary for the simple reason genes with tiny hits on fitness get weeded out, Inheriting the urge not to breed will obviously have a huge negative impact and would not spread though a population.

          • Lot says:

            Schizophrenia hits about 1 in 80 men at some point in their life, exclusive homosexual orientation is about 1 in 40. Seems like the same ballpark. Both have a similar reduction in fertility of about 2/3 (at least in modern society).

            Having the gay-promoting genes the study identified have a positive impact of fertility.

            • gcochran9 says:

              If a result isn’t statistically significant – it isn’t statistically significant.

              • Lot says:

                Age at first birth, female, is statistically significant and lower. And the other ones are all in the direction of earlier/higher fertility, and some are close to SS. Ignoring results completely as a matter of principle that are just under SS is a fine policy to have, but it isn’t one I follow.

              • Greying Wanderer says:

                if there are actually two things to be studied (promiscuity and homosexuality) but they are lumped together in the research then the result may be statistically significant in one group but not the other and so when averaged out the significance is lost?

                just a thought.

  7. Gord Marsden says:

    Appears to be a very unhappy life , is number of children number of siblings ?

  8. Shakagreni Thomas says:

    According to Plomin’s book schizophrenia and major depressive disorder are essentially the same mental illness, expressed differently, and they as a group are heritable. They look about the same in the GWAS. I’m not sure if ADHD is in the same mental illness cluster or not.

    Transexuals, on the other hand, tend towards (non-major) depression and anxiety disorders, another mental illness cluster at the GWAS level.

    • Andrew Oh-Willeke says:

      I wouldn’t consider that to be the majority or at least plurality opinion on the issue. Schizophrenia and bipolar have very heavy genetic overlap, and bipolar, of course, had a depression component. But, lots of major depression is a product of excessive anxiety and stress reaching overload and the body protecting itself from that by shutting down and collapsing instead of continuing to fight those stresses. ADHD like schizophrenia and bipolar disorder, has a very significant genetic component, but certainly doesn’t overlap as strongly with the other two.

      Of course, another complication is that DSM mental conditions are almost across the board syndromes, i.e. clusters of symptoms that present similarly, that with most conditions almost certainly don’t have a single underlying cause and have significant sub-clusters. One of the strongest overlaps in the DSM in between mixed type ADHD and anti-social personality disorder (without the lack of empathy component associated with non-DSM conditions psychopathy and other “dark triad” conditions).

      It is also appropriate to ponder the direction of cause and effect for some of these correlations. Being trans subjects you to all sorts of stresses in daily life and social interactions that can give rise to purely environmentally driven anxiety and depression, in the spirit of “just because you’re paranoid doesn’t mean that they aren’t out to get you.”

  9. jb says:

    Greg — I understand why the “gay uncle” theory doesn’t work: the effect would have to be so big that we would see it in action, and we don’t. But I’m less clear on why sexually antagonistic selection can’t work. Is it the same problem, or something else?

  10. qtl says:

    Is the chart showing the correlation of each trait with male/female homosexuality, or with the alleles correlated with male/female homosexuality? (I assume the latter since they show both bars even for sex-specific outcomes, but then I wish they would have shown the m/f breakdown for all the gene/phene correlations…)

    The obvious model that would have “ever had homosexual contact” (rather than homosexuality strictly) mediated by the same gene for men and women would involve openness or low disgust threshold. That seems to fit with the fact that the openness/homosex correlation is higher and significant for female homosex — presumably if infection is a significant cause of male homosex, then any shared vector for male and female homosex would responsible for more of the variation in female behavior and thus have more signficant covariant traits.

    (In fact one might even think that the less variation in male homosexual behavior openness-to-experience explains, the more it corroborates the stylized fact for gay-germ theories that orientation is changed by infection in childhood, since children don’t have any choice in the matter. Any alternative theory where one homosexual contacta increases the likelihood of future repeats would expect high-openness to be strongly correlated with “trying it once” and therefore with full-blown homosexuality.)

    • gcochran9 says:

      With the alleles – genetic correlation..

    • Andrew Oh-Willeke says:

      i think you are right that “did it once” should have correlation to openness to experience in a way that “repeatedly voluntarily did it” would have to a much lesser degree and a narrow threshold in theory should provide a stronger signal in the data despite a smaller sample size.

      One would still expect some “openness to experience” correlation, however, even then, because someone who is very not open to experience whose natural sexual orientation is not straight is likely to resist any same sex sexual contact and instead choose celibacy or to sublimate the urges by becoming an anti-gay pastor or politician or something like that instead. The absence of any trace of that in the men who have had sex with men category is one reason among several that I think that the “did it once” category in men includes a significant number of people who were raped in prison.

      Another bit of circumstantial evidence is comes from STD data for men who have had sex with men and their partners. Black men who are incarcerated at much higher rate and thus subject to a much higher risk of prison rape who have had sex with men, and are much more likely to have spread that STD to a female partner than white men who have lower STD rates and are proportionately much less likely to have spread that STD to a female partner, suggesting that the having sex with men component for lots of black men who have done so is not driven nearly as strongly by sexual orientation.

  11. Alice says:

    I would like to find more papers on correlating SNPs and the Big 5 personality traits. Homosexual attraction seems to be correlated to significantly higher tolerance for disgust and significantly higher neuroticism. Openness as a category has issues, but makes sense as a proxy for risk taking. Is there a big 5 profile for homosexuality? Are these higher correlations will mental illness and bad behaviors a way to identify SNPs dor personality?

    • gwern says:

      That won’t help much. The SNP heritability for most of OCEAN is near zero. That’s why you don’t see any genetic correlations reported for Conscientiousness/Extraversion/Agreeableness, but only occasionally Openness (rg with IQ and education) and Neuroticism (rg with happiness, depression). There’s no SNPs you can use. It’s still an open question as to where the heritability of the rest of personality is hiding: epistasis? Rare variants?

  12. Andrew Oh-Willeke says:

    The lack of significant correlation to openness to experience in gay men is particularly notable.

    “asked whether the respondent had ever had a same-sex experience”

    The biggest confound, as a result of this, may be incarcerated men who have been raped by fellow inmates.

  13. NonLinear9 says:

    Under the gay germ hypothesis I was expecting a raft of immune genes to be hit. Instead we get a lot of WNT signalling and olfactory — mostly neuronal stuff, which supports the reigning dogma (though of course it does nothing to overcome the other strong arguments against it).

    Instead of modulating the immune system, maybe the genetic component modulates the response to the pathogen insult? Not the most elegant model, but . . .

    • Greying Wanderer says:


      the scent angle interests me – if you look at animals the female comes into season and (i assume) releases a scent and it’s the scent that triggers the males so if you take the case of “gay rams” then most of the genetic arguments you could make for humans don’t apply whereas one simple possibility might be either:
      1) some rams give off female scent for some reason
      2) some rams are triggered by male scent.

      (intersex through twin chimerism?)
      (populations with more twins have more homosexuals?)

  14. LOADED says:

    Any one here use Genomelink or various other websites to find genetic analyses of their behavioral profiles? I did. I’m predisposed to high risk-taking, loneliness, alcohol and cannabis consumption, and various other dangerous behavior.

    I think all these things do lead to homosexual behavior because these behaviors in general because all of them would serve as rewards for conducting homosexual behavior:

    loneliness-finding a partner
    risk-taking-trying new things
    alcohol and cannabis-doing new things under the influence

    Anyways, don’t think I’m gay but it’s awful because I’ve lived quite a horrible 23 years of life because I’ve been put under so many stressors and put myself in dangerous situations. Some homosexual behavior is forced because it’s innate in our social behaviors and has nothing to do with our preferences for a mate.

    • stephen Cooper says:

      Dude, you seem to be an edgy guy.

      Weird hot chicks love that, assuming you take good care of your personal hygiene and do not wear Gilligan Island level clothes. (if you have bad hygiene or dress like Gilligan — stop right now!)

      Cheer up!

      And no, sexual relations with others are not something we need, they are something that is what makes marriage and love between decent human beings one of the best things in life.
      Cannabis is for unpoetic losers
      alcohol is a bond between friends, but absent friendship, it is just a moral failing

      Loneliness is not your fate, my friend. You amused me, and if you can amuse me enough to get me to spend a few of my precious moments to give you a compliment, you can easily

      one day

      get access to quality pussy, by which I mean, you can find a woman to love you and to be the mother of your children.

      When you have grandkids, remember that Stephen Cooper told you everything you needed to know to get quality pussy.

    • LOADED says:

      Sometimes I act at least twice my age, sometimes I act very immaturely. Finding the fine line between the two is what impresses people the most. That is what I am trying to embody.

      • stephen Cooper says:


        Most people realize they are only going to impress a very limited number of fellow humans – for example, if you are an ordinary guy, you realize that your best bet in life is to impress one woman out of all the women in the world who may or may not think you are the best man they can find, but who is nevertheless a woman who will be happy to spend the few decades of her attractiveness with you – life is a short ride, for all of us, my friend —– and really there is nothing better in this world for most of us than not missing out on our chances for love.

        and I am in a good mood tonight and I am gonna say this – you can believe me or not but there are generations of AIs who listen to everything I say, so I have an incentive which you cannot imagine to always tell the truth – and I am gonna say this —-
        Prayer is powerful, God listens to the prayers of the humble (like me) and I am praying for you and for those who will be happy, in the future, at your successes in life.

        Trust me. You amused me enough to get me to say
        I am praying for you
        you will do well in this world
        if you remember that God loves us all.

  15. Pingback: What about the “Gay Gene”? | POLITICS & PROSPERITY

  16. NonLinear9 says:

    @soren, they tested height — look more closely at fig. 4. It’s weakly negatively correlated.

  17. Richard Wrangham has a brief discussion of homosexuality in his recent book “The Goodness Paradox”. Mainly the book is an argument that human beings have gone through a process of self-domestication by killing off the most anti-social and violent group members over millennia. (Here’s a discussion on my blog, ) Wrangham argues that homosexuality in humans (and domesticated sheep) is a non-adaptive byproduct of selection for tameness. This is not supposed to be an explanation of why A is gay and B is not (there might not be a single cause for the various A’s and B’s; it might result from developmental noise), but of why sexual orientation is less canalized in our species and others recently domesticated. A possibility, he writes, “is that self-domestication has been operating too recently, up to historical times, for selection to operate strongly against incidental consequences” (p. 196).

    • david says:

      Ive been saying that the agrarian revolution necessitated monogamy, which became a type of “cultural eugenics.” Unintelligent and lazy men were ostracized from social circles and no women would risk their family’s reputation procreating with such a brute, so his genes were eventually removed from the pool.

      Modern welfare is having the opposite effect.

    • Boswald Bollocksworth says:

      Self-domestication should have helped to prune “gay genes” or promote immunity to the pathogen. Sodomy and pederastery(correlated with homosexual behavior) would have been punished with reproduction-ending penalties in the medieval period. Iirc, Chinese and south Europeans have lower incidence of homosexuality than populations with less total history under sedentary agrarianism (Northern Europeans)

  18. Bernard says:

    In his new book, Richard Wrangham proposes that human male homosexuality is a pedomorphic byproduct of self-domestication. He notes that only domesticated sheep show lasting homosexual behavior in nature.

    • J says:

      Only domesticated sheep show homosexual behavior, then why there is no such trait among other domesticated animals like bovines, horses, pigs, chickens, etc.? And in what sense “domestication” of sheep is like “self-domestication” of humans? And what about the most domesticated animals that are the dogs?

      I had the idea that homosexuals have LESS children than straight people, posing the question why such an evolutionary mortal trait survived. Now it appears that they have MORE children, making them fitter, so homosexuality is explained. I wish I was more intelligent and less confused.

        • J says:

          If so, it is a disorder, a disease. Cui bono? Who benefits from that compulsive itch? HIV benefits. Ergo, homosexuality may be caused by some tricky Lentivirus.

          • Jacob Rosenthal says:

            Doubt it. A pathogen that triggers homosexuality would have to be neurotropic, which Lentiviruses are not. Herpesviruses on the other hand…

            Also, a hypothetical pathogen that triggers homosexuality most likely does so via molecular mimicry (tricking the immune system into attacking a specific type of neuron instead of the virus) and so doesn’t necessarily have to benefit from altering its host’s sexual behavior, kind of like how polio reaps no benefit from causing paralysis in its host, but still does so.

  19. Dan Eggum says:

    “The contribution of all common SNPs is larger, estimated to be between 8% and 25%. ”

    Does this mean that they can explain more of the variance in same sex-experience, than for educational attainment (PRS predicting 11% of variance)?

  20. תמריץ says:

    The confounding explanation is so simple. Straight man and women who are unstable (low on conscientiousness) are much more likely to have a gay experiences in their past. The question used in that research made it almost worthless.

  21. Warren Notes says:

    Given the high number of sexual partners, could homosexuality involve a type of hyper sexuality – at least in cognition / fantasy? This is not too far out if you consider the role of absent inhibition in coprolalia. There might also be a characteristic of absent inhibition that results in lower discrimination between desirable vs. non-desirable sexual partners that overlaps with “openness to experience.” Finally, where does bisexuality fit in? It may be the more extreme version of lack of discrimination, rather than a “mid-point” between gay and straight as we so easily fall into thinking about it. Bisexuals are actually the “least” discriminating group, as their desirability is the least influenced by primary sexual characteristics, i.e., they are attracted to both males and females.

  22. Maciano says:


    Is there a link between gays and thin-ness? I’ve often noticed these guys to be very thin.

    It could be related to gays wanting to be more attractive to other gays so watching their diet or being more effiminate, therefore having lower muscle mass? I’ve also theorized it might be a consequence of gays having many infections because of high amount of sexual partners, leading to high calorie burn for an over-active immune system? I’ve never seen a paper explaining the thin-ness. If it’s none of these things could it be a similar effect like drug use and depression?

    (I’m not really trying to diss gays, btw. I just don’t understand why gays are more often thin than straight men.)

  23. helenahankart says:

    They don’t “look like noise” they look like “fast life history strategies”. Stearns (1976)

    • gcochran9 says:

      They look like noise. Mental illness is not a ‘fast life history strategy”.

      • helenahankart says:

        That depends.
        “Mental illness” covers a large range of things from being dinged in the head (not adaptive unless you are running for congress) through frequency dependent selection (psychopathy problably) mismatches (addictions, almost certainly) extremes of a normal distribution (GAD, maybe feminine bran explanation of schizophrenia) and facultative adaptations to local conditions.
        This isn’t, by itself, controversial stuff.
        Bowlby noted that attachment disturbance (we might call it “callibration”) led to later risk taking, we know the mechanism by which GR genes de-methylate in response to maternal grooming, and the influence of this on later risk taking in mammals
        Your mate Henry Harpending had it in mind (I know, because Jay asked him and talked to me about it at some length) when he and Jay Belsky hypothesised about the facultative effects of father absence on life history.
        So–I wouldnt rule it out. Neither wolud Randy Nesse, e.g.
        Good Reasons for Bad Feelings: Insights from the Frontier of Evolutionary Psychiatry

        • gcochran9 says:

          I think that the early-menarche with father absence thing has turned out not to be real.

          I wish I could talk to Henry about this… but the more I think about it, the patterns that some have called ” fast life histories” look to me have been a recipe for failure, except in the most resource-rich societies ( which are rare). As such, more likely a consequence of mutational load than any kind of strategy.

          I have always been interested in the idea of alternate evolutionary strategies, but in humans, I can’t at this point see evidence for any, other than men and women.

          • J says:

            With a little imagination, one can clearly see that “this point” is the past, humanity has passed the point of men and women based reproduction. Who but you knows what can be done in a lab with genetic engineering. We are living the last days of “natural” human evolution, science fiction creatures are no more fiction. We see no evidence of the dawning of this new era because regulations and conventions designed to stop the advance of reproductive science, and the horror of the media in face of the new. I consider the strange creatures living in computer games and Marvel Universe as evidence that something is coming.

          • Greying Wanderer says:

            “early-menarche with father absence”

            if it’s true i’d expect it to be genetic passed on to the daughter by the mother who got pregnant by the sort of dude who targets girls who are sexually active when they’re still young and gullible i.e. “he told me he loved me!” i.e. the sort of dude who doesn’t stick around.

            these dudes are definitely a type but they mostly exist on the fringes so the average person doesn’t come into contact (unless their job involves the fringes).


            “the patterns that some have called ” fast life histories” look to me have been a recipe for failure, except in the most resource-rich societies ( which are rare)”

            i think this is mostly true and the sort of dudes i’d call “slack” used to only survive on the scraps and fringes often involving lots of travel (musicians, soldiers, sailors, criminals, carnies etc) but since off-shoring that’s reversed in the rust-belt towns – the once 95%+ “steady” dudes give up trying and slowly drink themselves to death while the small minority of “slack” dudes suddenly got a massive reproductive boost (cos welfare pays for their kids).

            another possible exception to “resource-rich society” might be places like the Congo where war bands go back and forth raping the villagers along their path ( i saw a documentary once where village women were explaining how all their kids were the product of various rapes from various militias)

            so i’d say the common element is female-only provisioning – where kids need a male provider to survive then “slack” genes get bred out but where a male provider isn’t necessary then “fast life history” becomes adaptive for the males.

            seems to me (assuming we started in the jungle where females fed their kids on their own) then “fast life history” for the males would originally have been the norm and it’s been slowly bred out over time so the question becomes if a trait is in balanced selection does it ever get to 100% vs 0% or will there always be a surviving pool which, if the environment changes, can expand rapidly?

    • gcochran9 says:

      Mostly, over the past few thousand years, a pattern that looked like the usual definition of a “fast life history” would leave you with no descendants. Thinking about it, one reason that the historical nonpaternity rate looks so low must be, partly, because bastards died like flies.

  24. Greying Wanderer says:

    “Yet the degree of shared genetic influence is also interesting – I don’t think it would be predicted by most strategy models.”

    testing on the basis of “any” same-sex experience (as opposed to restricting it to people with a long-term orientation) might tell you more about promiscuity than homosexuality.

    so it may not be a genetic strategy at all but if it is (or if part of it is) a strategy then maybe it’s more a side-effect of a promiscuity strategy (r-K) than a homosexuality strategy per se.

  25. Greying Wanderer says:

    “There are at least as many bisexual men as gay men, as polls of sexuality in the youngest generation (Generation Z? Where are we now? sorry…) are confirming.”

    the Gen Z stuff is accurate but not true i.e. it’s a side-effect of the “progressive stack” where white middle class kids are at the bottom of the stack so pretending to be “genderfluid” “genderqueer” bisexual etc is their way of clawing their way up the stack – it’s basically the modern equivalent of pretending to be native American.

    (nb i wouldn’t be surprised if the numbers for one-time same-sex experience didn’t go up as a result but it’s not because they’re actually homosexual)

    • Lot says:

      “where white middle class kids are at the bottom of the stack so pretending to be “genderfluid” “genderqueer” bisexual etc is their way of clawing their way up”

      The increase in reported non-hetero identification is across all races, but whites the least.

      It also isn’t true IRL that whites are “at the bottom.” College humanities departments are not the whole of America.

      • Greying Wanderer says:

        fair enough – most of the ones i’ve come across were white and from college humanities departments so my anecdata may be skewed by that – either way at least part of this phenomenon is due to cultural fashion.

  26. Greying Wanderer says:

    apologies for spamming but one last thought…

    if promiscuity is part of this puzzle (whether the causes are genetic or environmental) then it might tie into the germ/virus theory as people who were promiscuous (or whose parents were) might be more likely to pick up the virus?

    example of what i mean
    1) virus/bug exists which if the mother is infected increases likelihood of kid becoming homosexual
    2) married dude has four kids over ten years with wife
    3) during same period dude visits four prostitutes a year with say 5% chance of picking up bug and passing it on to wife each visit

    leading to correlation of homosexuality with youngest sibling.

  27. Efim Hawkins says:

    Dr. Cochran,

    Ed Dutton (the “Jolly Heretic”) and his colleague recently had a brief discussion about “Gay Germ Theory”:

    Would love to read your thoughts about their comments, if you’d care to write a response some time.

    • gcochran9 says:

      Woodley makes three points. They are all false

      A. That MZ concordance for homosexuality is much higher than I said. That’s untrue. For example, in this Swedish study from 2010 [Genetic and Environmental Effects on Same-sex Sexual Behavior: A Population Study of Twins in Sweden], they found 7 concordant pairs and 64 discordant pairs in MZ twins.

      B. He talks about what I’ve said about the evolution of homophobia. I haven’t said anything about it.

      C. He talks about selection for such things in other primates – but preferential homosexuality does not exist at any significant level among other primates. In fact, the only known example is in sheep.

      Woodley is consistent.

      • Lot says:

        7/71 is 9.8% and appears to be on the extreme low side. The top results on the topic from a web search:

        “Bailey and Pillard found that 52% of the MZ co-twins of male homosexual twins were
        also homosexual or bisexual”

        “Bailey et al found that 48% of MZ co-twins of female homosexual twins were also

        “More recently, Kendler et al conducted a study that used a random sample of
        approximately 3,000 people. The researchers found a concordance rate of 32% for nonheterosexual orientation in [male] MZ twins”

        Whitham 1993: “Thirty-eight pairs of monozygotic twins (34 male pairs and 4 female pairs) were found to have a concordance rate of 65.8% for homosexual orientation.”

        Bailey’s 2000 study from an Australian sample found 37.5% and 30.1% concordance for MZ male and female twins and 11.1 and 11.5% for DZ twins.

        • Lot says:

          The Swedish study was of “at least one same sex experience” and had 5% say yes to this, while Bailey’s Australian study was of concordance of “not exclusively heterosexual orientation” and 8.1% of the male population. I think Bailey’s was more informative since the Swedish study considered individuals with homosexual attraction but no homosexual experience as straight, while putting individuals with a single random homosexual experience but who’d report zero homosexual attraction as homosexual.

          Bailey also found even stronger concordance for gender nonconformity.

          About prior studies, they say:

          our MZ concordances were 20% and 24%, respectively, for the strict [excluding Kinsey 1s] criterion that is most similar to those used in prior studies. These rates are significantly lower than the respective rates for the two largest prior twin studies of sexual orientation: for men, 52% ( ), [chi] (1, = 550) = 8.2, < .01, and for women, 48% ( ), [chi] (1, = 1,115) = 4.3, < .05. This suggests that concordances from prior studies were inflated because of concordance-dependent ascertainment bias. In those studies, twins deciding whether to participate in a study clearly related to homosexuality probably considered the sexual orientation of their co-twins before agreeing to participate. In contrast, both the more general focus of our study (i.e., on sexuality in general) and its anonymous response format made such considerations less likely.

  28. Anonymous says:

    Disingenuous of you not to mention fecundity as a correlate.

  29. Pingback: The Gay Debate | The Z Blog

  30. Jacob Blaustein says:

    “As far as I can see, they look like noise, rather than any kind of genetic strategy. Mostly, they accord with what we already knew about male and female homosexuals: both are significantly more likely to have psychiatric disorders, far more likely to use drugs. The mental-illness association maybe looks stronger in lesbians. The moderately-shared genetic architecture seems compatible with a noise model.”
    Could be epigenetic:

  31. Jacob Rosenthal says:

    I’m surprised this study didn’t look at IQ as a variable, especially since we know the UK Biobank data includes various cognitive measures. Given previous findings, they likely would’ve found a negative correlation with IQ as well, at least in male homosexuals.

    Example of previous findings:

    • Matt says:

      I doubt it. Other research claims that homosexuality is correlated with intelligence:
      (Although it’s probably false, smart people are more likely to disclose their atypical sexual interest).

      • Jacob Rosenthal says:

        Your parenthetical illustrates the major flaw in such research that claims to find a positive association between intelligence and homosexuality, whereas the paper using the AddHealth data I linked to found that lower IQ males were, in fact, more likely to experience same-sex attraction (as opposed to opposite-sex attraction) than their average IQ peers.

        • Matt says:

          Personally I think both hypotheses are silly. There is similar levels of gay males in Japanese and Chinese university samples, so that is not consistent with a ‘low iq’ hypothesis. It could also be that those from harsher backgrounds, e.g. backgrounds of divorce, are more likely to disclose their homosexuality because they don’t have dad around to disapprove.

  32. Matt says:

    I know this a late response Greg, but I saw this old article authored by Richard Horton, the editor of The Lancet journal, which is just astounding. He subscribes to Foucaultive view of homosexuality and seriously thinks biology cant even code for sexual orientation. He calls it a social construct, says gay men didn’t exist in Platos time. Of course the critiques of Dean Hamer are only valid but one can hardly blame the chap for looking for an answer. I can respect Hamer because he called your gay germ hypothesis “interesting”. This buffoon, Horton, has been in charge of one of the UK’s most “prestigious” medical journals for a long time. Complete insanity.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s