Genes, Hormones, and Sexuality

Published in: January-February 2008 issue.


AS A WOMAN and a scientist, I am disturbed by careless over-interpretations of some of the scientific literature about the “math” gene, or the “female” brain vs. the “male” brain, or the “genetic” basis of intelligence. The comments made in 2005 by Lawrence Summers, then president of Harvard, about the “innate inability” of women to do math, aroused a firestorm among female scientists. As a lesbian, I feel the same way about the increasing number of publications contrasting the “homosexual” brain with the “heterosexual” brain.

The media lap these stories up; apparently the public can’t get enough of them. The issues rapidly become political, leaving any kind of objective view behind. Scientists design studies and collect data. But the interpretation of data and their practical implications quickly lapse into the social and political realm, and objectivity is lost. A few years ago The New Yorker magazine ran a cartoon, depicting a nurse showing a newborn baby to the father with the caption: “It’s a lesbian.” A joke, but I found it somewhat chilling. A play on Broadway entitled The Twilight of the Golds, later made into a movie, posited that there’s a gay gene that can be detected in utero, leaving the fetus vulnerable to abortion following amniocentesis. In the play, the elected abortion was botched and the expectant mother became sterile. Was this retribution for having the abortion, or for not wanting a gay child?

Contrary to these simplifications, the relationship between homosexuality and biology, if any, is certainly very complex. “Biology” as a causative agent can mean different things to different people: sex hormone levels in the adult; the hormones or other factors during early brain development of a fetus in utero or soon after birth; or the genes for some factor affecting the brain that’s involved with gender identity. A behavior pattern as complex and variable as homosexuality cannot possibly be due to a single altered gene or even several genes alone. Many women have lived happy heterosexual lives for a number of years before changing to a lesbian orientation. This is clearly non-deterministic, in contrast to the genetic factors that cause, say, cystic fibrosis or Tay-Sachs disease or Down syndrome. On the other hand, homosexuality as a way of life is so compelling to so many gay people, including myself, that it seems unlikely to this biologist that it can be due wholly to social factors completely divorced from biology.

My own discipline is endocrinology, the study of hormones. The growth of endocrinology as a discipline during the first half of the 20th century fascinated the public (Schwartz, 2001). The idea that an infinitesimal amount of a chemical circulating in the bloodstream could mean the difference between life and death, as in the case of insulin, or the difference between normal intelligence and cretinism, as in the case of thyroid hormone, was mind-boggling. The endocrine hypothesis for homosexuality started in this period when it was demonstrated that many of the overt characteristics that we call “male” or “female” in animals and in people (genitalia, fat distribution, height, hair distribution, etc.) are the result of the post-pubertal estrogen secreted during the menstrual cycle in the adult female or the testosterone secreted in the adult male. It was perhaps natural that people began assuming that homosexuality itself was due to the “wrong” hormones being present in the adult homosexual. To my knowledge, there are no definitive data that show this to be at all true. That is, there is no evidence that gay men have lower or higher testosterone or lower or higher estrogen levels than heterosexual men. Similarly, there are no reliable data on circulating sex hormone differences between lesbians and heterosexual women.

But several countries have prescribed blockade of testosterone secretion or its action on the body in male sex offenders, including rapists and homosexuals. Alan Turing, the renowned British mathematician who invented the first computer (the “Turing Machine”) and broke the German U-boat code in World War II—a gay man—was later arrested and charged with being a homosexual, which was illegal in England at that time. In a public trial he was found guilty and given the choice of receiving estrogen injections or going to jail. He chose the estrogen, a treatment that reduced his sex drive by suppressing the pituitary gonadotropic hormones, thereby suppressing the signal for testosterone secretion by the testes. Two years after the trial, he died of cyanide poisoning in his home and was presumed to have committed suicide. While castration, estrogen, and anti-androgen drugs have been prescribed for homosexual men under some conditions, these treatments would be expected to interfere with erections and with sex drive, and perhaps aggressiveness, but would not change either sexual orientation or gender identity.

Having failed to demonstrate clear abnormalities in adult hormone levels in homosexuals, investigators began to look for other explanations. In rodents such as mice and rats, the testis secretes hormones during prenatal life and immediately after birth. These androgens permanently alter the brain. Male rodents deprived of this early androgen grow up with some female endocrinological and behavioral characteristics. On the other hand, female rodent pups given early treatment with masculinizing hormones do not ovulate as adults or show typical female sexual behavior. Endocrinologists label these early developmental actions of hormones as “organizing” effects, since they appear to organize the brain permanently in a male direction. If the same thing is true for humans, then maybe it is the prenatal hormonal environment that determines homosexuality. Rodents have been marvelous models for the way gonads work and for studying the interactions between the pituitary and the gonads, but not for the study of complex psychosocial factors. Furthermore, in rodents the early secretion of androgens in neonatal males precludes the stimulating action of estradiol in eliciting a pre-ovulatory secretion of luteinizing hormone (LH) from the pituitary gland. (This positive feedback action of estrogen is what causes the estrous cycle to be maintained in female rodents or the menstrual cycle to be maintained in women.) But adult human males can be induced to release a surge of LH from the pituitary following estradiol treatment, under particular conditions, such as when they are being prepared for sex change surgery (Gooren, 1986).

Psychologists have done numerous studies looking for a heritable, genetic component in sexual orientation. The usual way the studies are done is to advertise for gays in the gay media and then check for homosexuality in siblings or twins. In studies of male twins, monozygotic twins (single egg) are more likely than are dizygotic twins (two eggs fertilized by two different sperm) to be concordant for homosexuality. But the degree of concordance is only about fifty percent at the highest. Data for lesbians also ranges near fifty percent concordance, but the studies are less consistent. Data on twins raised apart are few but tend to support these findings. One problem with these studies, readily discussed by the authors, is the way in which the respondents were recruited, namely from gay magazines, which allows for a good deal of self selection, so this is not a random sample of male twins. For example, the lingering stigma associated with homosexuality may have discouraged some pairs from volunteering. But even if we accept the figure of fifty percent concordance, it means that fifty percent of the monozygotic twins differ in their sexual orientation. This is hardly a definitive genetic etiology.

But, carrying this one step further, in 2001 my colleague Michael Bailey published a paper in which he discussed the “ethical” aspects of parental selection of sexual orientation in their unborn children. He concluded that “allowing parents to select their children’s sexual orientation would further parents’ freedom to raise the sort of children they wish to raise … and is unlikely to cause significant harm.” Shades of Twilight of the Golds. I was reminded of the Nobel Laureate Linus Pauling’s suggestion in the 1960’s that carriers of the genetically determined sickle cell disease be tattooed on the forehead for easy recognition.

In 1993 a study appeared in Science magazine by Dean Hamer and colleagues at NIH. They claimed that they had evidence for a gene on the X chromosome, the one inherited from the mother, that was linked to homosexuality in males. They recruited 76 subjects through HIV clinics. There were also 38 pairs of homosexual brothers and relatives recruited through homophile publications. The Kinsey scale was used to ascertain the degree of homosexuality. Hamer found that a high number of respondents who scored high on the scale had brothers and maternal uncles or maternal cousins (aunts’ sons) who were also homosexual, suggesting inheritance from a gene or genes on the X chromosome, which came from their mothers. They did a gene linkage analysis on families with two homosexual brothers, typing for 22 DNA markers on the X chromosome. They found a linkage between sexual orientation and the presence of an area in the X chromosome described as “distal Xq28” (which contains several hundred genes). This was not so in seven pairs of subjects. The study concluded, conservatively, that “one subtype of male sexual orientation is genetically influenced.” The media attention to this admittedly weak claim was overwhelming, demonstrating yet again how the public is attracted to stories about sex and genetics.

However, in 1999, a second study was published in Science (Rice) that received no media attention at all. This group of researchers recruited subjects by advertising in gay magazines for families in which there were two gay brothers. They had responses from 52 sibling pairs in 48 families. They measured linkages to four markers at the gene site Xq28 and found no significant linkage. They concluded that there is not an X-linked gene for male homosexuality. Unlike the Hamer study, they did not look for maternal transmission specifically.

If there is a “gay gene,” it cannot be deterministic in the way that eye color is: even studies that point to a genetic component indicate that the linkage is weak. The human genome contains only about 30,000 genes, suggesting that much of the differential development in individuals results from combinations of genes and interactions between gene expression and environmental factors. This makes the discovery of underlying genetic contributions difficult. Hamer himself later published an article in Science titled “Rethinking Behavior Genetics,” where he challenges “the assumption that the rich complexity of human thought and emotion can be reduced to a simple, linear relation between individual genes and behavior.”

For years there have been studies of sex differences in human brain size and morphology. I must confess that I tend to avoid this literature because with whatever good intentions a study starts, whatever is being measured, the data always seem to be interpreted as favoring the ruling, dominant group. If a brain nucleus (which is a local accumulation of nerve cells) is larger in males than females, then “bigger is better.” If it is smaller in men, then it favors “closer precision or finer thinking.” Now there is a growing literature comparing the brains of gay males with heterosexual males and females. The studies measure the area of a given nucleus whose function in humans is not well understood and then compare the size. Thus, for example, Simon LeVay compared a part of the hypothalamus of men who had died of AIDS and of men and women presumed to be heterosexual—and found the gay men’s organ to be smaller.

A new set of experiments has now surfaced from the laboratories of physiological psychologists examining other aspects of homosexual brain functions. Hearing acuity and response to pheromones are the most recent traits that appear to differ in homo- and heterosexual individuals. Just how these traits are causally related to homosexuality, if at all, is simply unknown.

Jean Wilson, a former president of the Endocrine Society and the author of a major textbook on clinical endocrinology, is an expert on the effects of hormones on sexual differentiation. In a masterful article, she analyzed the role of androgens in male gender role behavior. Wilson concentrated on gender role behavior in subjects demonstrating two genetically influenced enzyme abnormalities where the sex chromosomes are unambiguously male or female but the external genitalia are ambiguous. These were, of course, not homosexuals but intersexed individuals. The deficiencies in these enzymes make it impossible for the fetus to synthesize the androgenic steroid hormones responsible for external genitalia differentiation in a male direction, resulting in some form of intersexuality. In both types of cases, genetic (XY) males were born with female or ambiguous genitalia and were assigned to the female gender at birth. Many of these individuals remained female in their gender role, while some changed their sexual orientation later on. Wilson concluded that factors other than androgen—“social, psychological, biological—are of equal or greater importance in modulating human sexual behavior. Indeed, the sex of rearing may be more important than the endocrine milieu … it may not be a coincidence that many (although not all) of the instances of reversal of gender role behavior in these two disorders [to a male direction]have occurred in countries and/or ethnic groups in which men play a dominant role.”

Issues related to anatomical sex, XY and XX chromosomes, gonadal hormones, and sexual orientation are growing increasingly complex as more and more species are studied and new techniques are devised for probing these questions. For example, the whiptail lizard, which lives in the American Southwest, reproduces parthenogenetically (having no male contribution), and their colonies consist mostly of clones of females. David Crews has observed what appears to be “sexual” behavior between females, finding that the “courting” female has ovaries with fewer large follicles than the courted female. It is not clear whether this behavior has an adaptive value in the wild. The media labeled the story about these findings as “lesbian lizards,” surely a leap into a fantasy world! Another example is the recent work by A. P. Arnold, et al. (2004), suggesting that sex chromosomes per se may alter neural cells before the intervention of the sex hormones. He has suggested that in male fetuses some XX cells and XY cells in the brain and elsewhere differentiate even before the onset of gonadal hormone secretion at puberty. In rats, cultures of brain cells harvested before the occurrence of sexually dimorphic blood levels of testosterone show differences between cells of XX or XY composition. Since it remains controversial whether some homosexuals have altered XX and XY sex chromosomes, this aspect of genetic composition may or may not be relevant.

It is instructive to contrast the stance of feminists and gays with respect to theories of biological etiology. For members of both groups, the overt biological phenotypes are “normal,” and both groups have supported changes in the laws, urging equal treatment. But feminists have argued that women’s biology, acknowledged to be hormonally and genetically different from men’s, at least with respect to sexual phenotype, is not definitive and that the differences do not impact one’s ability to do certain types of work (like math!) or to run for political office. In contrast, many gays have argued that our biology is different and determinative, that gayness is no more a “choice” than is gender or eye color. Clearly, this might make a difference to some non-homosexuals.

As a scientist, I have not pursued the issue of homosexuality, having been more interested in questions concerning the regulation of female hormonal cycles. But I have always recognized that whichever way biological studies come out, the prevailing political culture will, to a certain extent, dictate the use to which the data are put. Homosexuality is a hot-button issue sociologically and politically. While studies can be designed to examine various aspects of homosexuality, interpretations inevitably will be constructed according to the biases of the interpreter. I despair that this is the case and only hope that having more women and gays in biology will contribute to a more rational interpretation of the data.

As political and social conservatives in 2007 step up their campaign against gay marriage and continue their frenzied attacks on abortion and stem cell research, I watch with bated breath to see how they will approach the possible dilemma of the prenatal diagnosis of homosexuality. If homosexuality is “genetic” and can be diagnosed prenatally, will they support legalized abortion of a “homosexual” fetus? Will they then drop their opposition to legalized abortion of unwanted fetus? Stay tuned.



Arnold, A.P., J.Xu, W.Grisham, X.Chen, Y-H. Kim and Y. Itoh. “Minireview: Sex chromosomes and brain sexual differentiation.” Endocrinology 145, 2004.

Crews, D. and K. T. Fitzgerald. “‘Sexual’ behavior in parthenogenetic lizards.” Proc. Natl. Acad. Sci. 77, 1980.

Fausto-Sterling, A. Sexing the Body: Gender Politics and the Construction of Sexuality. Basic Books, 2000.

Gooren, L. “The neuroendocrine response of luteinizing hormone to estrogen administration in the human is not sex specific but dependent on hormonal environment.” Journal Clinical Endocrinology and Metabolism 63, 1986.

Greenberg, A. S. and J. M. Bailey. “Parental selection of children’s sexual orientation.” Arch Sexual Behavior 30, 2001.

Hodges, A. Alan Turing: The Enigma. Simon and Schuster, 1983.

Rice, G., C.Anderson, N. Risch and G. Ebers. “Male homosexuality: Absence of linkage to microsatellite markers at Xq28.” Science 284, 1999.

Schwartz, N. B. “Perspective: Reproductive endocrinology and human health in the 20th century: A Personal Retrospective.” Endocrinology 142 (6), 2001.

Wilson, J. “The role of androgens in male gender role behavior.” Endocrine Reviews 20, 1999.


Neena B. Schwartz, PhD, is emerita professor of neurobiology at Northwestern University.


Read More from Neena Schwartz