Determinants and evolutionary mechanisms of homosexuality

Debates over the rights of homosexuals in the United States, particularly the right to marry, often get hung up on a thoroughly inane point: whether homosexuality is "chosen" or "innate". While this may seem to be a question of moral import, it is not, and moreover it presents a false dichotomy. Like nearly all human behaviors, sexuality is too complex to be reduced to a choice or a destiny; it is neither, or both, depending on your view. However, the degree to which different factors contribute to sexuality, and the mechanisms by which they do this, are fit subjects for scientific inquiry. Two articles this week present interesting findings, sure to be distorted by all sides of the argument, that may prove enlightening in this regard. I will endeavor, along with others, to be a resource providing an unbiased view.

First, however, a plea for sanity. If science finds, by some transcension of nature, that sexual orientation is entirely chosen, or entirely innate, it does not matter to any debate over the rights of homosexuals. Men may have an innate tendency to try to spread their genes as widely as possible, but we would not forgive adultery on this basis. Toddlers have an innate tendency to become frustrated and throw tantrums, but we still make them sit in the corner. That a behavior is innate is not a basis for withholding moral judgment. And if sexual orientation is a choice? Well, we frequently forbid discrimination on the basis of chosen behaviors—religion, for instance, or political affiliation. What is truly at issue is not choice, but whether it is just to deny rights and protections to one group of citizens for no reason beyond the moral opprobrium of another group.

Thus, the question of rights for homosexuals does not depend, one way or the other, on whether people choose to be gay. I firmly believe one side of this question to be in the right, but this opinion is not informed by my scientific knowledge, because it cannot be. I would urge my readers (all three of you) to view these results strictly as what they are: interesting scientific findings related to a political question that do not support one side of the argument or the other. I would ask advocates for both sides to refrain (for once) from distorting the conclusions of these reports, not only because of the raw immorality of lying, but because by misrepresenting these findings they will have sacrificed their integrity for no gain in the debate.

There, I feel better now. On to the science!

In the first study, a team analyzed the results of a survey of Swedish twins in hopes of parsing out the relative contributions of heredity, shared environment, and unshared environment in shaping sexuality (1). Although the survey was answered by a fairly large number of twins, the authors draw their conclusions using two questions that do not directly ask for sexual orientation. The survey only requested information about actual sexual partners, and did not address homosexual feelings that might not have been acted upon. After excluding twin pairs that were opposite-sex or unclear with respect to zygosity, they had 3826 pairs to work with, of which 5% of men and 8% of women reported at least one same-sex sexual encounter. Because it is suspected that the factors influencing homosexuality may differ between the sexes, males and females were treated separately. By comparing the concordance and discordance of sexual behaviors between monozygotic and dizygotic twins it should be possible to parse out the degree to which genetics and the environment contribute to sexuality.

Despite the limited materials, the authors were able to reach some conclusions, with the caveat that the 95% confidence intervals were quite wide. For instance, for males they found that genetic factors explained 39% of the observations with respect to whether a twin had any same-sex partner in his life. However, the 95% CI on this prediction was 0% to 59%. For men, shared environmental factors appeared to contribute nothing, while unique environmental factors explained 61%. For women, it was determined that genetic factors contributed 19%, shared environment 17%, and unique environment 64%. Similar distributions were seen for comparisons of total numbers of same-sex partners. While the confidence intervals for all factors are quite large, the numbers largely agree with a previous study on Australian twins (less so with a study on American twins).

Obviously, the small sample size and broad confidence intervals on these results suggest that they should be interpreted cautiously. It should also be noted that "unique environmental factors" may run the gamut from hormone exposure in utero to childhood illness to personal experiences. Many unique environmental factors, even for twins, are just as involuntary as genetics, but some result from conscious choices of the individual (which is different from choosing to be gay). Despite their limitations, these results generally support the idea that sexual orientation results from a confluence of genetic and environmental factors.

That genetics play a role in homosexuality may seem curious, because in terms of the classic expression, "survival of the fittest", homosexuality would appear to be a non-starter. After all, a reluctance or outright inability to mate with the opposite sex would seem to result in a substantial reduction in reproductive fitness. However, contrary to what a certain ignoramus would have you believe, the Theory of Evolution has advanced substantially since the days of Darwin, and we are aware of numerous additional evolutionary mechanisms that operate alongside the law of natural selection. In the case of male homosexuality, a new paper by Camperio Ciani et al. argues that sexually antagonistic selection may be at work (2). PLoS ONE is open access, so feel free to open up the article in another window and skim it yourself.

Camperio Ciani et al. begin with the observations that male homosexuality has a matrilineal association, and that the mothers (and maternal aunts) of homosexuals are somewhat more fecund than the population at large. From these pieces of data, and from the fact that homosexuality appears to have been present at low levels in every society that has left written records, the researchers created a set of requirements for some evolutionary simulations, based on different supposed properties of the genetic factors influencing male homosexuality (GFMH). Most of the simulations failed to satisfy the parameters. In many cases (especially with single-locus traits) the GFMH either became extinct or gained too high of a frequency; in others the matrilineal association was not preserved.

Ultimately, the researchers found that the model that best fit the parameters featured two alleles (one of them X-linked), and was sexually antagonistic. What this means is that the trait increases the reproductive fitness of one sex while decreasing that of the other. For instance, a heightened sexual response to men could make women more likely to pass on their genes, while making men possessing the trait less likely to do so. Provided that the effects of this trait are balanced with respect to the population proportion of each gender, it should be possible for it to survive in a population at a relatively constant level.

This result is interesting, and provides some hypotheses that can be tested with genetics. However, it does not prove that homosexuality is genetic, or even that it has a genetic component. Like all simulations, these results merely inform us that a particular possibility is consistent with what we already know. In this case, we now know that the observed aspects of homosexuality are consistent with a 2-locus trait that is sexually antagonistic. However, this model was arrived at simply through process of elimination, and there may be some superior model or more-accurate mechanism that simply hasn't yet been tested. There is always a model we haven't thought of; sometimes that model is the right one. Moreover, as Långström et al. note, some of the data used to determine criteria for successful simulations remain controversial. Camperio Ciani et al. convincingly show how the preservation of homosexuality through evolution could happen, but that is not the same as demonstrating how it did happen. That will require a positive identification of the actual GFMH.

The results of Långström et al. indicate that any GFMH eventually identified, whether or not they materially resemble the predictions of Camperio Ciani et al., will only give rise to a heightened propensity for homosexuality. Environmental factors play a significant, perhaps even dominant, role in determining sexual orientation. Whether genetic or environmental, most factors contributing to homosexuality are involuntary, but some are chosen. If that answer doesn't satisfy you, perhaps you were asking the wrong question.

1. Långström, N., Rahman, Q., Carlström, E., Lichtenstein, P. (2008). Genetic and Environmental Effects on Same-sex Sexual Behavior: A Population Study of Twins in Sweden. Archives of Sexual Behavior DOI: 10.1007/s10508-008-9386-1

2. Camperio Ciani, A., Cermelli, P., Zanzotto, G., Brooks, R. (2008). Sexually Antagonistic Selection in Human Male Homosexuality. PLoS ONE, 3(6), e2282. DOI: 10.1371/journal.pone.0002282 OPEN ACCESS