by Jesse Bering
—Signed, no longer woefully yours, yet in perpetually ready and melodramatic sorrow, your once, and likely future, heartbroken gay friend, J.B.
Top Scientists Get to the Bottom of Gay Male Sex Role Preferences
It’s my impression that many straight people believe that there are two types of gay men in this world: those who like to give, and those who like to receive. No, I’m not referring to the relative generosity or gift-giving habits of homosexuals. Not exactly, anyway. Rather, the distinction concerns gay men’s sexual role preferences when it comes to the act of anal intercourse. But like most aspects of human sexuality, it’s not quite that simple.
I’m very much aware that some readers may think that this type of discussion isn’t proper science. But the great thing about good science is that it’s amoral and objective and doesn’t cater to the court of public opinion. Data don’t cringe; people do. Whether we’re talking about a penis in a vagina or one in an anus, it’s human behavior all the same. The ubiquity of homosexual behavior alone makes it fascinating. What’s more, the study of self-labels in gay men has considerable applied value, such as its possible predictive capacity in tracking risky sexual behaviors and safe-sex practices.
People who derive more pleasure (or perhaps suffer less anxiety or discomfort) from acting as the “insertive” partner are referred to colloquially as tops, whereas those who have a clear preference for serving as the receptive partner are commonly known as bottoms. There are plenty of other descriptive slang terms for this gay male dichotomy as well, some repeatable (“pitchers versus catchers,” “active versus passive,” “dominant versus submissive”) and others, well, not by a gentleman.
In fact, survey studies have found that many gay men actually self-identify as “versatile,” which means that they have no strong preference for either the insertive or the receptive role. For a small minority, the distinction doesn’t even apply, since some gay men lack any interest in anal sex and instead prefer different sexual activities. Still other men refuse to self-label as tops, bottoms, versatiles, or even gay at all, despite their having frequent anal sex with gay men. These are the so-called men who have sex with men (or MSM) who often have heterosexual relations as well and tend to see themselves as straight rather than bisexual.
Several years ago, a team of scientists led by Trevor Hart at the Centers for Disease Control and Prevention studied a group of 205 gay male participants. Among the group’s major findings were these:
1. Self-labels are meaningfully correlated with actual sexual behaviors. That is to say, based on self-reports of their recent sexual histories, those who identify as tops are indeed more likely to act as the insertive partner, bottoms are more likely to be the receptive partner, and versatiles occupy an intermediate status in sex behavior.
2. Compared with bottoms, tops are more frequently engaged in (or at least they acknowledge being attracted to) other insertive sexual behaviors. For example, tops also tend to be the more frequent insertive partner during oral intercourse. In fact, this finding of the generalizability of top/bottom self-labels to other types of sexual practices was also uncovered in a study showing that tops were more likely to be the insertive partner in everything from sex-toy play to verbal abuse to urination play (aka “water sports”).
3. Tops were more likely than both bottoms and versatiles to reject a gay self-identity and to have had sex with a woman in the past three months. They also manifested higher internalized homophobia—essentially the degree of self-loathing linked to their homosexual desires.
4. Versatiles seem to enjoy better psychological health. Hart and his coauthors speculate that this may be due to their greater sexual sensation seeking, lower erotophobia (fear of sex), and greater comfort with a variety of roles and activities.
One of the primary aims of this study was to determine if self-labels in gay men might shed light on the epidemic spread of the AIDS virus. In fact, self-labels failed to correlate with unprotected intercourse and thus couldn’t be used as a reliable predictor of condom use. Yet the authors make an excellent—potentially lifesaving—point:
Although self-labels were not associated with unprotected intercourse, tops, who engaged in a greater proportion of insertive anal sex than other groups, were also less likely to identify as gay. Non-gay-identified MSM [again, men who have sex with men] may have less contact with HIV-prevention messages and may be less likely to be reached by HIV-prevention programs than are gay-identified men. Tops may be less likely to be recruited in venues frequented by gay men, and their greater internalized homophobia may result in greater denial of ever engaging in sex with other men. Tops also may be more likely to transmit HIV to women because of their greater likelihood of being behaviorally bisexual.
Beyond these important health implications of the top/bottom/versatile self-labels are a variety of other personality, social, and physical correlates. Some psychologists point out that prospective gay male couples might want to weigh this issue of sex role preferences seriously before committing to anything long-term. From a sexual point of view, there are obvious logistical problems of two tops or two bottoms being in a monogamous relationship. But since these sexual role preferences tend to reflect other behavioral traits (such as tops being more aggressive and assertive than bottoms), “such relationships also might be more likely to encounter conflict quicker than relationships between complementary self-labels.”
Another intriguing study was reported in the Archives of Sexual Behavior by the anthropologist Matthew McIntyre. McIntyre had forty-four gay male members of Harvard University’s gay and lesbian alumni group mail him clear photocopies of their right hand along with a completed questionnaire on their occupations, sexual roles, and other measures of interest. This procedure allowed him to investigate possible correlations between such variables and the well-known 2D:4D effect, which I mentioned in my essay on gay men and navigational skills. Somewhat curiously, McIntyre discovered a small but statistically significant negative correlation between 2D:4D and sexual self-label. That is to say, at least in this small sample of gay Harvard alumni, those with the more masculinized 2D:4D profile were in fact more likely to report being on the receiving end of anal intercourse and to demonstrate more “feminine” attitudes in general.
Many questions about gay self-labels and their relation to development, social behavior, genes, and neurological substrates remain to be answered; indeed, they remain to be asked. That many gay men go one step further and use secondary self-labels, such as “service top” and “power bottom” (a pairing in which the top is actually submissive to the bottom), reveals even further complexity. For the right scientist, there’s a lifetime of hard work just waiting to be had.
Is Your Child a “Pre-homosexual”? Forecasting Adult Sexual Orientation
There are signs, some would say omens, glimmering in certain children’s demeanors that, probably ever since there were children, have caused parents’ brows to crinkle with worry, precipitated forced conversations with nosy mothers-in-law, strained marriages, and ushered untold numbers into the deep covenant of sexual denial. We all know the stereotypes: an unusually light, delicate, effeminate air in a little boy’s step, often coupled with solitary bookishness, or a limp wrist, an interest in dolls, makeup, princesses, dresses, and a strong distaste for rough play with other boys; in little girls, there is the outwardly boyish stance, perhaps a penchant for tools, a lumbering gait, a square-jawed readiness for physical tussles with boys, an aversion to all the perfumed, delicate, laced trappings of femininity.
So let’s get down to brass tacks. It’s what these behaviors signal to parents about their child’s incipient sexuality that makes them so undesirable; these behavioral patterns are feared, loathed, and often spoken of directly as harbingers of adult homosexuality. However, it is only relatively recently that developmental scientists have conducted controlled studies with one clear aim in mind: to accurately identify the earliest and most reliable signs of adult homosexuality
. In looking carefully at the childhoods of gay adults, researchers are finding an intriguing set of behavioral indicators that homosexuals seem to have in common. And, curiously enough, the age-old homophobic fears of many parents reflect some genuine predictive currency.
In their technical writings, researchers in this area simply refer to pint-sized prospective gays and lesbians as “pre-homosexual.” This term isn’t perfect: it manages to achieve an uncomfortable air of biological determinism and clinical interventionism simultaneously. But it is, at least, probably fairly accurate. Although not the first scientists to investigate the earliest antecedents of same-sex attraction, J. Michael Bailey, a psychologist, and the psychiatrist Kenneth Zucker published a seminal paper on childhood markers of homosexuality with their controversial article in Developmental Psychology in 1995. The explicit aim of this paper, according to the authors, “was to review the evidence concerning the possible association between childhood sex-typed behavior and adult sexual orientation.” So one thing to keep in mind is that this particular work isn’t about identifying the causes of homosexuality per se but instead about indexing the childhood correlates of same-sex attraction. In other words, nobody is disputing the likely genetic factors underlying adult homosexuality or the well-established prenatal influences. Instead, it is simply meant to index the nonerotic behavioral clues that best predict which children are most likely to be attracted, as adults, to those of the same sex and which are not.
By “sex-typed behavior,” Bailey and Zucker are referring to that long, now scientifically canonical list of innate sex differences in the behaviors of young males versus young females. In innumerable studies, scientists have documented that these sex differences are largely impervious to learning and found in every culture examined (even, some researchers believe, in youngsters of other primate species). Now, before that argumentative streak in you starts whipping up exceptions to the rule—obviously, there is variance both among and within individual children—I hasten to add that it’s only when comparing the aggregate data that sex differences leap into the stratosphere of statistical significance. The most salient among these differences are observed in the domain of play. Boys engage in what developmental psychologists refer to as “rough-and-tumble play,” which is pretty much exactly what it sounds like, whereas girls prefer the company of dolls to a knee in the ribs.
In fact, toy interests are another key sex difference, with boys gravitating toward toy machine guns and monster trucks, and girls orienting toward baby dolls and hyper-feminized figurines. Young children of both sexes enjoy fantasy—or pretend—play, but the roles that the two sexes take on within the fantasy context are already clearly gender segregated by as early as two years of age, with girls enacting the role of, say, cooing mothers, ballerinas, or fairy princesses and boys strongly preferring more masculine characters, such as soldiers and superheroes. Not surprisingly, therefore, boys naturally select other boys for playmates, and girls would much rather play with other girls than with boys.
So on the basis of some earlier, shakier research, along with a good dose of common sense, Bailey and Zucker hypothesized that homosexuals would show an inverted pattern of sex-typed childhood behaviors (little boys preferring girls as playmates and infatuated with their mothers’ makeup kits; little girls strangely enamored with field hockey or professional wrestling … that sort of thing). Empirically, explain the authors, there are two ways to investigate the relation between sex-typed behaviors and later sexual orientation. The first of these is to use a prospective method, in which young children displaying sex-atypical patterns are followed longitudinally into adolescence and early adulthood, such that the individual’s sexual orientation can be assessed at reproductive maturity. Usually this is done by using something like the famous Kinsey scale, which involves a semi-structured clinical interview about sexual behavior and sexual fantasies to rate people on a scale of 0 (exclusively heterosexual) to 6 (exclusively homosexual). I’m a solid 6; like Stephen Fry I wanted to get out of a vagina at one point in my life, but ever since then I’ve never had the slightest interest in going back into one.
Conducting prospective studies of this sort is not terribly practical, explain Bailey and Zucker, for several reasons. First, given that a relatively small proportion of the overall population is exclusively homosexual, a rather large number of pre-homosexuals are needed to obtain a sufficient sample size, and this would require a huge oversampling of children just in case a small subset turns out gay. Second, a longitudinal study tracking the sexuality of children into late adolescence takes time—around sixteen years—so the prospective approach is very slow going. Finally, and perhaps the biggest problem with prospective homosexuality studies, not a lot of parents are likely to volunteer their children. Right or wrong, this is a sensitive topic, and usually it’s only children who present significant sex-atypical behaviors—such as those with gender identity disorder—who are brought into clinics and whose cases are made available to researchers.
For example, the psychologist Kelley Drummond and her colleagues interviewed twenty-five adult women who were referred by their parents for assessment at a mental health clinic when they were between three and twelve years of age. At the time, all of these girls had several diagnostic indicators of gender identity disorder. They might have strongly preferred male playmates, insisted on wearing boys’ clothing, favored rough-and-tumble play over dolls and dress-up, stated that they would eventually grow a penis, or refused to urinate in a sitting position. As adults, however, only 12 percent of these women grew up to have gender dysphoria (the uncomfortable sense that one’s biological sex does not match one’s gender identity). Rather, the women’s childhood histories were much more predictive of their adult sexual orientation. In fact, the researchers found that the odds of these women reporting a bisexual/homosexual orientation were up to twenty-three times higher than would normally occur in a general sample of young women. Not all tomboys become lesbians, of course, but these data do suggest that lesbians often have a history of cross-sex-typed behaviors.
And the same holds for gay men, according to Bailey and Zucker. They revealed that in retrospective studies (the second method used to examine the relation between childhood behavior and adult sexual orientation, in which adults simply answer questions about their childhoods), 89 percent of randomly sampled gay men recalled cross-sex-typed childhood behaviors exceeding the heterosexual median. Some critics have questioned the general retrospective approach, arguing that participants’ memories (those of both gay and straight individuals) may be distorted to fit with societal expectations and stereotypes about what gays and straights are like as children. But in a rather clever study published in Developmental Psychology, evidence from childhood home videos validated the retrospective method by having people blindly code child targets on the latter’s sex-typical behaviors, as shown on the screen. The authors found that “those targets who, as adults, identified themselves as homosexual were judged to be gender nonconforming as children.”
Numerous studies have since replicated this general pattern of findings, all revealing a strong link between childhood deviations from gender role norms and adult sexual orientation. There is also evidence of a “dosage effect”: the more gender-nonconforming characteristics there are in childhood, the more likely it is that a homosexual/bisexual orientation will be present in adulthood.
But—and perhaps you’ve been waiting for me to say this—there are several very important caveats to this body of work. Although gender-atypical behavior in childhood is strongly correlated with adult homosexuality, it is still an imperfect correlation. Not all little boys who like to wear dresses grow up to be gay, nor do all little girls who despise dresses become lesbians. Many will be straight, and some, let’s not forget, will be transsexuals. Speaking for myself, I was rather androgynous, showing a mosaic pattern of sex-typical and sex-atypical behaviors as a child. In spite of my parents’ preferred theory that I was simply a young Casanova, Zucke
r and Bailey’s findings may account for that old Polaroid snapshot in which eleven of the thirteen other children at my seventh birthday party are little girls. But I also wasn’t an overly effeminate child, was never bullied as a “sissy,” and, by the time I was ten, was indistinguishably as annoying, uncouth, and wired as my close male peers.
In fact, by age thirteen, I was deeply socialized into masculine norms. In this case, I took to middle school wrestling as a rather scrawny eighty-pound eighth-grader, and in so doing, I ironically became all too conscious indeed of my homosexual orientation. Cross-cultural data show, actually, that pre-homosexual boys are more attracted to solitary sports, such as swimming, cycling, and tennis, than they are to rougher contact sports, such as football and soccer; they’re also less likely to be childhood bullies. (Prospective gay males who adapt too rigidly to the perceived gender norms as they grow older may, in fact, become hyper-masculinized to such a degree that, as we’ve seen already, they also become dangerously homophobic in the process.) In any event, I distinctly recall being with the girls on the monkey bars during second-grade recess while the boys were in the field playing football, and looking over at them thinking how it was rather strange that anyone would want to act that way.
Another caveat is that researchers in this area readily concede that there are likely multiple—and no doubt very complicated—developmental routes to adult homosexuality. Heritable, biological factors interact with environmental experiences to produce phenotypic outcomes, and this is no less true for sexual orientation than it is for any other within-population variable. Since the prospective and retrospective data discussed in the foregoing studies often reveal very early emerging traits in pre-homosexuals, however, those children who show pronounced sex-atypical behaviors may have “more” of a genetic loading to their homosexuality, whereas gay adults who were sex typical as children might trace their homosexuality more directly to particular childhood experiences. For example, in a rather stunning case of what we might call “say-it-isn’t-so science”—science that produces data that rebel against popular, politically correct, or emotionally appealing sentiments—recent controversial findings in the Archives of Sexual Behavior hint that men—but not women—who were sexually abused as children are significantly more likely than nonabused males to have had homosexual relationships as adults. Whatever the causal route, however, none of this implies, whatsoever, that sexual orientation is a choice. In fact, it implies quite the opposite, since, as we know from the rubber lover and foot fetishists we met earlier in this book, prepubertal erotic experiences can later consolidate into irreversible sexual orientations and preferences.