AWA: Academic Writing at Auckland

Sexual orientation is extremely socially relevant due to the current media focus on the United States of America’s impending review of the Defense of Marriage Act. Public debate about sexual orientation frequently centres on questions of aetiology, with gay-rights activists asserting that being gay is innate and their opponents declaring it a lifestyle choice. In this essay, I will examine scientific findings from three broad research areas (family studies, neuroendocrinology, and anatomical comparisons) in order to critically evaluate the claim that being gay is innate.

The concept of ‘being gay’ is not as straightforward as it may seem; indeed definitions in popular discourse and in scientific publications vary widely (Sell, 1997). The Kinsey Scale (Kinsey, Pomeroy, & Martin, 1948) ranks sexual orientation on a six-point continuum, taking into account psychosexual and behavioural aspects of sexual orientation, and the Klein Sexual Orientation Grid (Klein, F., Sepekoff, B.,  & Wolf, T. 1985) utilises seven variables. However, much of the research into the innateness of sexual orientation uses a dichotomous ‘gay’ or ‘straight’ definition. For the purposes of this essay I will define ‘being gay’ as being primarily sexually attracted to and physically sexually involved with members of the same sex. ‘Gay’ commonly refers specifically to male homosexuals, but I will include female homosexuals (lesbians) in this evaluation.

Considerable research has been devoted to the question of whether or not being gay is innate. But what exactly does ‘innate’ mean? A standard dictionary definition is ‘inborn, natural’  (Deverson, 1997). ‘Natural’ is perhaps what media commentators and human rights activists mean when they use the term ‘innate’ in relation to sexual orientation, but this is not specific enough for use in the sciences. Mameli and Bateson (2006) identify no fewer than twenty-six possible scientific definitions for ‘innate’, too many to be of practical use in this evaluation. A 2011 evaluation by the same authors sorted these many definitions into four broad definitional categories which suggest that for something to be innate it must be one or more of the following: heritable, genetic in origin, canalised, not learned.

Research on sexual orientation has not taken a coherent position on the definition of innateness; some research (Rice, Anderson, Risch, & Ebers, 1999) has sought to identify genes linked to homosexuality in order to prove or disprove its innate nature, while others (McFadden and Pasanen, 1999, for example) have searched for gay-straight differences that implicate neuroendocrine factors rather than learning. Indeed, nearly every possible definition of innateness has been used in research on homosexuality, and no single study has found evidence that suggests being gay is genetic, heritable, canalised and not learned. Given the confusion in the scientific community as to the definition of innate, I will evaluate research into the aetiology sexual orientation using all four of Mameli and Bateson’s broad variants of innateness.

Sexual orientation is often investigated through observational family studies. Given the similar environments that members of a family inhabit, it is often difficult to determine whether a discovered relationship is due to heritable, genetic factors or social learning within the family context. Nevertheless, family studies have provided valuable information about sexual orientation.  Pillard and Weinrich (1986) found that homosexual men were far more likely than heterosexual men to have homosexual or bisexual brothers – their most conservative estimate of an 18% incidence of homosexuality among brothers of homosexual men far exceeds the population incidence of homosexuality. This suggests that homosexuality may ‘run in families’, i.e. it is innate with a heritable genetic component; though the authors concede that shared environment could also be a contributing factor. Interestingly, no such correlation was found between lesbians and their sisters. Although birth order was not taken into account in this study, these findings could possibly be explained by the maternal immune hypothesis posited by Blanchard (2001).

Bailey and Pillard (1991) gathered data on the sexual orientation of pairs of gay men and their brothers, including monozygotic twins, dizygotic twins, non-twin brothers and adopted brothers. The concordance between biological brother pairs was found to be less than half of what would be expected if being gay was entirely genetically determined, but greater than what would be expected between randomly selected non-brother pairs. This study therefore provides some evidence towards a genetic explanation of homosexuality, not enough to validate a claim that being gay is entirely innate in the genetic sense. Indeed, when one considers that the adopted brother pairs in the same study had a greater concordance than would be expected, and that dizygotic twin pairs had greater concordance than non-twin brother pairs, it seems likely that the childhood environment these brother pairs shared had a significant influence on their sexual orientation. Similar results were obtained in a later study of familial incidence of female homosexuality (Bailey, Pillard, Neale, & Agyei, 1993), suggesting that for women as well as men, being gay is not entirely innate, but nor is it entirely environmentally determined or learned.

Family studies into sexuality have provided some evidence towards the innateness of being gay, but are not able to identify a ‘gay gene’, if indeed such a gene exists. Hamer, Hu, Magnuson, Hu and Pattatucci (1993) found that homosexual men had more homosexual maternal relatives than would be expected based on population averages, but no more homosexual paternal relatives than expected. From this, they inferred sex-linked transmission of sexual orientation, a finding that was backed up by their discovery of a correlation between the presence of markers at the Xq28 site of the X chromosome and maternal homosexuality. This appears to be evidence in support of the hypothesis that being gay is heritable and has a genetic component, although Hamer et al. only identified a region of interest, not a specific gene. Unfortunately, Hamer et al. did not investigate Xq28 markers in pairs of heterosexual men, so it is impossible to know if this linkage relates to male sexual orientation in general, or just to being gay. In any case, a follow-up study by Rice, Anderson, Risch and Ebers (1999) found no such linkage, throwing doubt on the validity of Hamer et al.’s findings.

The second major approach to researching the aetiology of homosexuality involves neuroendocrine mechanisms, particularly the effects of androgen exposure. Rodent studies show early castration followed by administration of oestrogen causes male rats to assume female sexual postures while female pups given testosterone at birth assume male sexual postures (Quadagno, Shryne, Anderson, & Gorski, 1972). These findings have been interpreted as evidence that perinatal androgens are required to masculinise rats, and that hormonally-induced sex-atypical mating behaviour is evidence of induced homosexuality. However, this only takes into account the rat assuming sex-atypical positions, not the second rat participating in the sexual behaviour.  Consequently, these results’ relevance to human sexual orientation are questionable. That said, they provide valuable background information on the role of androgens in sexual differentiation and gender roles, and a springboard for study into the effects of androgens on human sexual orientation.

An early human study found that a small sample of gay males responded to injected oestrogen as females would (Dörner, Rohde, Stahl, Krell, & Masius, 1975). A later study found that homosexual men secreted more luteinising hormone in response to artificially raised oestrogen levels than heterosexual male controls, but less than women (Gladue, Green, & Hellman, 1984).  These results suggest that male homosexuality is related to a feminised neuroendocrine response to androgens, which is not in keeping with the later work of Williams et al. (2000). However, a similar study by Gooren (1986) found that both men and women respond to applied oestrogen with a surge of luteinising hormone, that there is no significant difference between the response of homosexual and heterosexual women, and that there is significant person-to-person variation in male responses, which is not correlated to sexual orientation. As such, it seems that the differences noted by Dörner et al. (1975) and Gladue, Green & Hellman (1984) simply illustrate the natural variation in hormonal response, rather than providing evidence for a neuroendocrine theory of being gay. This line of research is interesting, but it does not provide a viable alternative to the hypothesis that being gay is learned, not innate.

One of the most robust findings in sexual orientation research is that right-handed males with older brothers are more likely to be gay than first-born males or males with older sisters (Blanchard, 2001). Blanchard suggests that fraternal birth order accounts of the sexual orientation of between 15% and 24% of gay males, and that there is also a correlation between birth weight, sexual orientation and fraternal birth order. This finding has been replicated in studies in several countries. It is hypothesised that mothers become increasingly immune to male-specific antigens with each male foetus they carry; though human data is not yet available, rodent studies support this view (McLaren, Simpson, Tomonari, Chandler, & Hogg, 1984). Obviously this antigen explanation is not applicable to women, and indeed no comparable female birth order effect has been noted. This supports the antigen explanation for male homosexuality, but is further proof that being gay is the result of a number of interacting factors, which may not be the same for gay men and women. Nevertheless, this finding is strong evidence that being gay is not innate in the canalised genetic sense; rather, a significant proportion of male homosexuality is due to prenatal environmental factors. Conversely, if innate’ is simply defined as ‘not learned’, Blanchard’s findings make it clear that, at least for a significant number of male homosexuals, being gay is innate.

Another factor linked to fraternal birth order is finger-length ratio. Men who have older brothers are likely to have a smaller difference in length between their second and fourth right hand digits (a more masculine finger-length ratio) than first-born males. On average, gay men have more masculine finger-length ratios and a greater number of older brothers than straight men, and gay women have more masculine finger-length ratios than straight women (Williams et al., 2000). Finger length is influenced by prenatal exposure to androgens, so perhaps being gay is also influenced by exposure to androgens. The results, though statistically significant, were not clear-cut; not all of the participants had the finger-length ratios expected of their gender and orientation, and there was a large amount of overlap between the groups. Nevertheless, Williams et al. (2000) provide some further evidence that being gay is not learned; rather prenatal exposure to hormones may have an important effect.

The development of the human auditory system is also mediated by foetal androgen exposure. Females exhibit stronger click-evoked otoacoustic emissions than males, but females with male twins (who are exposed to extra androgens in utero on account of their twin) display male-type otoacoustic emissions. Thus it has been hypothesised that the high androgen levels male foetuses experience reduce their otoacoustic emissions (McFadden, Loehlin, & Pasanen, 1996). McFadden and Pasanen (1999) found that gay females exhibited weaker click-evoked otoacoustic emissions than heterosexual females; indeed their responses fell in the expected male range. No such difference was found between gay and straight males. Sex differences are also apparent in some measures of auditory evoked potentials (AEPs). McFadden and Champlin (2000) have shown that lesbians exhibit masculinised AEPs, and gay males hypermasculinised AEPs. These results suggest that increased foetal androgens cause masculinisation, and are in keeping with findings regarding links between androgens, sexual orientation and finger-length ratio (Williams et al., 2000).

It is unlikely that the auditory system is directly influenced by sexual behaviour, but it is possible that a related lifestyle factor, for example exposure to a particular drug, is responsible for the observed differences in otoacoustic emissions. However, given the wealth of supporting research in this area, I would support the authors’ conclusion that sexual orientation and auditory differences are related to increased foetal androgen levels. It would be interesting to discover whether women with male twins are more likely to be gay than women with female twins, or no twins at all. If this were the case, then this would provide strong supporting evidence towards McFadden and Pasanen’s conclusion that being gay is linked to prenatal androgen exposure, which in turn would be evidence that being gay is not learned, but also not heritable. If ‘innate’ is taken as ‘not learned’, then the evidence correlating sexual orientation with androgens strongly suggests that being gay is innate – few would argue that a physiological response in utero is an example of learning. But if heritability, canalisation, and/or genetic origin are required, this line of evidence provides far less support for the innate hypothesis. In particular, the strong impacts of prenatal androgen exposure and maternal antibodies show that sexual orientation is not canalised.

The third major area of scientific research into sexual orientation focuses on physical differences between gay and straight people, particularly in brain regions linked to sexual behaviour. LeVay (1991) found that a cell group in the interstitial nuclei of the anterior hypothalamus (INAH 3) was larger in heterosexual than homosexual men, though there was significant overlap between the groups. Unfortunately, this study investigated only a small number of brains, most of which had been sourced from men who had died of AIDS. As such, the findings must be interpreted with caution, especially considering that they have not been replicated. Even assuming that there are sexual orientation-linked differences in the INAH3 region of the human hypothalamus, it is impossible to tell if these differences were present at birth, and caused the individual’s sexual orientation, or if they developed throughout the lifespan as a consequence of the individual’s sexual behaviour. In the absence of a large, inter-generational, longitudinal study monitoring the brains of individuals from birth to adulthood, it seems rash to draw conclusions about the aetiology of homosexuality from physical differences noted in adults.

None of the scientific evidence available at present provides a truly adequate explanation for homosexuality, or indeed heterosexuality. The clear significance of androgens suggests that any heritable, genetic aspects of sexual orientation are strongly influenced by the perinatal hormonal environment, but family studies suggest that both genetic factors and an individual’s socio-cultural environment are relevant (though no research supports the claim that people choose to be gay). The aetiology of male homosexuality may well be different to that of female homosexuality, and some of the most robust findings apply only to a proportion of the gay population. It seems unlikely, but perhaps future studies into sexual orientation using large samples and a battery of genetic, physiological and neuroimaging tests will determine one or more heritable genetic factor(s), not influenced by learning or environment, responsible for sexual orientation. Until then, although I agree that being gay is not a choice I cannot support the claim that being gay is innate.

References

Bailey, J., & Pillard, R. (1991). A genetic study of male sexual orientation. Archives of General Psychiatry, 48(12), 1089-1096.

Bailey, J., Pillard, R., Neale, M., & Agyei, Y. (1993). Heritable factors influence sexual orientation in women. Archives of General Psychiatry, 50(3), 217-222.

Blanchard, R. (2001). Fraternal birth order and the maternal immune hypothesis of male homosexuality. Hormones and Behaviour, 40(2), 105-114. doi: 10.1006/hbeh.2001.1681

Deverson, T. (1997). The New Zealand Pocket Oxford Dictionary  (2^nd ed.). Auckland, New Zealand: Oxford University Press.

Dörner, G., Rohde, W., Stahl, F., Krell, L., & Masius, W. (1975). A neuroendocrine predisposition for homosexuality in men. Archives of Sexual Behaviour, 4(1), 1-8. doi: 10.1007/BF01541882

Gladue, B., Green, R., & Hellman, R. (1984). Neuroendocrine response to estrogen and sexual orientation. Science, 225(4669), 1496-1499. doi: 10.1126/science.6089349

Gooren, L. (1986). The neuroendocrine response of luteinizing hormone to estrogen administration in heterosexual, homosexual, and transsexual subjects. The Journal of Clinical Endocrinology & Metabolism, 63(3), 583-588. doi:10.1210/jcem-63-3-583

Hamer, D., Hu, S., Magnuson, V., Hu, N., Pattatucci, A. (1993). A linkage between DNA markers on the X chromosome and male sexual orientation. Science, 261(5119), 321-327. doi: 10.1126/science.8332896

Kinsey, A., Pomeroy, W., & Martin, C. (1948). Sexual Behaviour in the Human Male. Philadelphia, United States of America: W. B. Saunders

Klein, F., Sepekoff, B., & Wolf, T. (1985). Sexual orientation: A multi-variable dynamic process. Journal of Homosexuality, 11(1 & 2), 35-49. doi: 10.1300/J082v11n01_04

LeVay, S. (1991). A difference in hypothalamic structure between heterosexual and homosexual men. Science, 253(5023), 1034-1037. doi: 10.1126/science.1887219

Mameli, M., & Bateson, P. (2006). Innateness and the sciences. Biology and Philosophy, 21, 155-188. doi 10.1007/s10539-005-5144-0

Mameli, M., & Bateson, P. (2011). An evaluation of the concept of innateness. Philosophical Transactions of the Royal Society B, 366(1563), 436-443. doi: 10.1098/rstb.2010.0174

McFadden, D., & Champlin, C. (2000). Comparison of auditory evoked potentials in heterosexual, homosexual, and bisexual males and females.  Journal of the Association for Research in Otolaryngology, 1(1), 89-99. doi: 10.1007/s101620010008

McFadden, D., Loehlin, J., & Pasanen, E. (1996). Additional findings on heritability and prenatal masculinization of cochlear mechanisms: Click-evoked otoacoustic emissions. Hearing Research, 97(1-2), 102-119. doi:10.1016/S0378-5955(96)80012-8

McFadden, D.,  & Pasanen, E. (1998). Comparison of the auditory systems of heterosexuals and homosexuals: click-evoked otoacoustic emissions. Proceedings of the National Academy of Sciences of the United States of America, 95(5), 2709-2713.

McLaren, A., Simpson, E., Tomonari, K., Chandler, P., & Hogg, H. (1984). Male sexual differentiation in mice lacking H-Y antigen. Nature, 312(5994), 552-555.

Pillard, R., & Weinrich, J. (1986). Evidence of familial nature of male homosexuality. Archives of General Psychiatry, 43(8), 808-812.

Quadagno, B., Shryne, J., Anderson, C. & Gorski, R. (1972). Influence of gonadal hormones on social, sexual, emergence, and open field behaviour in the rat (Rattus norvegicus). Animal Behaviour, 20(4), 732-740. doi: 10.1016/S0003-3472(72)80145-3

Rice, G., Anderson, C., Risch, N., & Ebers, G. (1999). Male homosexuality: absence of linkage to microsatellite markers at Xq28. Science, 284(5414), 665-667. doi: 10.1126/science.284.5414.665

Sell, R. (1997). Defining and measuring sexual orientation: a review. Archives of Sexual Behaviour, 26(6), 643-658. doi: 10.1023/A:1024528427013

Williams, T., Pepitone, M., Christensen, S., Cooke, B., Huberman, A., Breedlove, N., Breedlove, T., Jordan, C., & Breedlove, S. (2000). Finger-length ratios and sexual orientation. Nature, 404, 455-456. doi: 10.1038/35006555