Implications of Evolutionary Theory for Human Homosexuality
In accordance with Kirsch and Rodman dominance hierarchies appear to be an important element to many primate proto-homosexual behaviors. However, these hierarchies should probably be seen as only one act in a longer play that begins with territoriality and deceptive "transvestite" tactics and ends with alliance formation and affection. This is a play about how animals come to cooperate with each other, and about the origin of society. I can see no reason for not adding an encore for human homosexuality. Humans are certainly much more cooperative than other animals, and our society is much more complex.
If the association between cooperation and homosexuality is correct, then we should expect humans to show more homosexuality than most other complex animals, with the possible exception of bonobos, who, as Wrangham and Peterson (1996) argue, may be even more cooperative and more homosexual. I think humans do practice more homosexuality, although an overconcentration on North European cultures may sometimes confuse the issue. As I pointed out elsewhere (Werner 1990), North European cultures are extremely unusual in restricting homosexual behavior to the more or less exclusively homosexual. In the rest of the world homosexual behavior is much more generalized. Normal men have sexual relations with the exclusive homosexuals, or with each other at different ages.
What kinds of evidence might support or refute the "dominance hierarchy/cooperation" argument for homosexuality? There are three questions where evidence may be forthcoming. First, the "dominance hierarchy/cooperation" argument makes predictions about deeply rooted feelings in all humans. Second this theory may also account for individual variation in all cultures. Third, the theory may help explain why cultures vary.
As pointed out above, more complex animals, seem to add new, more complex associations with homosexuality to the already existing repertoires of simpler species. Yet they still retain the older repertoires. We should, thus, find evidence of these older repertoires in humans. This makes sense in terms of the "tinkerish" econony of natural selection. As modern neuroscientists have pointed out (Damasio 1994; Vincent 1990; LeDoux 1996), the human brain is constructed in "layers." The phylogenetically inner layers are more conservative, varying less from one species to the next, in conformity with the great dictum of natural selection: "If it ain't broke, don't fix it".
The outer layers are more recent, and vary more from one species to the next and from one individual to the next (Cairns-Smith 1996). Thoughts or perceptions organized in the outer layers need to produce their effects by acting through the inner layers. For example we may reason our way (using the outer brain layers) into a fearful state (produced in the inner layers), or may similarly reason our way out of a fearful state. However, as Damasio (1994) points out, when the different regions of the brain are in disagreement, feelings may be less "convincing." When they all agree, they may be perceived as especially powerful or "raw."
The differences are seen even in language. For example, swearing, unlike other language capacities, has its origins in the phylogenetically older sub-cortical parts of the brain. Brain-damaged individuals may lose their capacity for virtually all language, but still retain the ability to swear (Pinker 1994). Psychoanalysts (e.g. Arango 1989) and anthropologists (Duerr 1993) have taken advantage of this phenomenon in order to analyze the sources of our most powerful emotions. In the case of dominant / submissive relations our swearing vocabulary is rich and revealing. Primate mounting behavior is revealed in expressions like "he wants our asses," or "up yours!" (The use of these "symbolic" expressions of dominance may be found in primates as well. Enmoto (1990) reports the case of one male bonobo expressing its dominance over another by using a gesture normally used to solicit sex from an estrous female.)
References to submissive individuals in different languages are consistently related to primate gestures of submission. In English we call overly submissive individuals "ass kissers" or "brownies." In German and Vietnamese, the expressions are only slightly different: "Arschkriecher", and "NimBa", respectively, expressing the act of crawling up to the dominant's behind. Spanish speakers say "lame culo", southern Slavs "Dupolizac", and Russians "Podliza" all referring to licking the dominant's behind.
The Brazilian term refers to another primate gesture puxa-saco (scrotum tugger), while Uruguayans say chupa medias (literally sock sucker), referring to what may be another scent marker. Turks use a milder "Dalkavuk", referring simply to the subordinate's bowed back, although they also have a stronger "Kiáimi yala" referring to ass licking. Other English primate-gesture expressions to humiliate subordinates include "rubbing it in", "smearing it in your face", "sucker." The reader can probably fill in more examples.
"Dirty words" are helpful in understanding attitudes that may be deeply rooted in all individuals. But there may be important differences between people as well. For example, in the United States some of the "cross-gender" characterstics that distinguish predominate male homosexuals from predominate male heterosexuals are not particularly "feminine" in a general sense. Rather they are simply non-aggressive.
That male homosexuals as children generally avoided rough and tumble play may have more to do with submissive behavior than with femininity. This may be true not only for predominate homosexuals, but for others as well. McConaghy et al. (1991) looked at homosexual feelings among males with predominant heterosexual attractions, and discovered that the degree of homosexual feelings is correlated most strongly with the childhood characteristic of disliking outdoor and contact sports. There were no correlations of homosexual feelings in this population with activities like cooking or playing with dolls as a child.
More systematic comparisons in other cultures may help to clarify these questions. Most anthropological studies have concentrated on general cultural norms, categories, and symbolic systems, not on individual differences. This is unfortunate, because despite the many studies of homosexuality in the world's more simple societies, we still know little about individual variation in homosexual activities for these societies. It is even difficult to affirm that exclusive homosexuals do not exist in some of these societies. Lindholm's (1982) study of the Swat Pashtan of Pakistan is revealing of the problems involved. In this society males typically sing praises to their homosexual lovers, and argue that sexual relationships with boys are less demanding and more pleasurable than relations with women. Yet all of the men get married, and their wives bear children.
At first glance it might appear that, in spite of the prevalence of homosexual activities, there are no exclusive homosexuals — all males engage "equally" in homosexuality. Yet the language has disparaging terms for passive homosexuals, and Lindholm documented cases where predominant homosexuals married women, but where it was their male lovers who inseminated their wives. The search for correlations between childhood and adult behaviors in these societies are very rare (for example Whitam 1983).
More importantly, practically no one has looked for correlations in these societies between degree of homosexual feelings among predominantly heterosexual men, and their psychological, developmental and biological characteristics. The only example I know of is that of Cardoso (1994). In his study of a Brazilian fishing community, Cardoso found that the men most interested in sex with the community's exclusive homosexuals were more likely (statistically significant) to have avoided soccer playing as children than were other men who had fewer relationships with the homosexuals.
But these men were not different from other men in the more clearly cross-gender behaviors (like playing with dolls). In sum, anthropological studies of the symbolic associations, identities and even practices of homosexuals in different cultures are quite welcome, but without studies of individual variation, they cannot answer the basic question of whether similar genetic or developmental factors result in homosexual feelings in different societies. To answer this question we need more systematic studies of how much, and why males within a given society differ psychologically and biologically.
I think evolution can also help us understand why cultures vary in the ways they organize homosexuality. But first we should recognize that there are many aspects to cross-cultural variation in homosexuality - how cultures define sexual identities, how much activity is permitted, associations of homosexuality with valued or undesired characteristics, etc. So there are many different things that need to be explained with regard to cross-cultural variation. An explanation that works for one aspect of homosexuality (e.g. attitudes about homosexual activities, per se) may not work for other aspects (e.g. identity formation, or the psychological characteristics of individuals).
Also, in all societies there may be individuals with naturally different degrees of homosexual feelings. Culture may affect the orientations and practices of those in the middle of this continuum, but have less effect on the endpoints. The cross-cultural variation in homosexuality is great. If we wanted to, we could show that each culture has its own unique identities, roles, values, and symbolic associations with regard to homosexuality. But it is also possible to make a few generalizations. Elsewhere (Werner 1990) I suggested that the world's societies might be categorized into only a few types according to who has sex with whom in same-sex relationships.
The first type ("gay" culture) is defined by the fact that predominant homosexuals have sexual relations primarily with other predominant homosexuals. This is the system of Northern Europeans and their descendants, and may be spreading with the processes of economic globalization. However, it is actually the most rarely found ethnographically. In an examination of the HRAF records I have been unable to find a single non-Northern European "traditional" (i.e. not influenced by Europeans) culture with this system.
he second system, probably the most widespread, might be labeled the "bicha/bofe" type (using Brazilian terms). In this system, more or less exclusive homosexuals (not always, but very often transvestite or at least effeminate males) have sexual relationships with men who are not culturally distinguished from other men. This system is found among such diverse peoples as southern Europeans and their descendants (e.g. Brazilians), many Native North American Indian societies (Navaho, Sioux, Mohave, Utes, Zuni, Apaches, Shoshones, Yurok, Pomo, Pawnees, Mandan, Crow, Fox, Bellacoola, Aleuts, etc.), many Native South American societies (Guayaki, Kanela, Tupinamba, Tapirape, Warao, Tucano, Toba, Tehuelche, pre-contact Mapuche, Aymara, and Cuna, etc.), Far East and Asian cultures (Bengali, Burusho, Kashmir, Karen, Burma, Philippines, Semai, Malaysians, Indonesians, China, lower-class Japan, Chuckchee, Koryak, etc.), a few Oceanic societies (Iban, Toradja, Belan, Makassar, Pukapuka, Marquesans, Tonga), and a few Middle Eastern and African cultures (Oman, Teda, Amhara, Hausa, Twi, Tanala, Zulu, etc.).
The third system (the "age-grade" system), somewhat less widespread, consists in homosexual relations between older men and younger boys or men, often with a "master/apprentice" type of relationship. This system is found in traditional Europe (ancient Greeks), Africa (Azande), the Middle-East (Siwans), Asia (Badaga, Tibetan monasteries, Samurai/aristocratic caste Japanese), and Oceania (Etoro, Sambia, Malekula, Aranda, etc.). The fourth type might be labeled the adolescent-sex system, and consists in homosexual relationships between adolescents but which disappear after marriage. This system is found in many oceanic societies (Lau, Manus, Wogeo, Ifugao, Marquesans, Tikopia), in some African societies (Ngonde, Hottentot, Shona, Mongo), and in some South American societies (Nambikwara, Yanomamo, Araucanians).
A few societies might be classified into more than one of these types - especially the adolescent-sex type may also occur with the bicha/bofe type. In addition, a single society (like the Brazilians or the Japanese) may have different systems within different classes. Still, most societies are characterized primarily by one or the other systems. This does not mean that there are not other important differences between societies. For example, it is very important to the people involved if they are treated with respect, ridiculed or disdained, have high status or are slaves, are harshly punished, or given rewards, and privileges, etc. Some of the subtleties of these differences have been documented by Williams (1986) in his analysis of the differences between Sioux winkte and American "gays". Besides questioning how gays could have sex with other gays, the Sioux also lamented that gays, unlike winkte, seemed to have lost their special spiritual side.
The idea that male homosexuality is related to dominance hierarchies/cooperation may help explain why cultures adopt different homosexual systems. A first point is that greater and closer cooperation should generally be associated with more homosexual feelings and practices among the general population. For example, work in the fishing community studied by Cardoso required men to observe each other carefully and cooperate closely during work (pulling up nets for example) as well as to spend time together.
In addition, and probably more importantly, getting ahead or even surviving in this society depends more on "personal alliances" than on unbiased, abstract laws, applied equally to all. The adolescent and bicha/bofe systems permit a great deal of homosexual behavior between potential allies, not only at the start of adult life, but also afterwards, during "parties," when men can have collective sex with the local bichas. Among the Tupinamba, for example, the local transvestites set up their houses outside the main ring, and received visits from groups of men. The Tapirape would take a preferential homosexual along on their hunting trips to service the men. Brazilian men often do their wildest partying at a bicha's residence.
The age-grade systems may be related to other forms of cooperation and power based on age rather than personally negotiated alliances. Australian aborigenes, like the Aranda, for example, have often been described as "gerontocratic" and boys are required to have homosexual relations with mentors who could potentially be their fathers-in-law.
Finally, where cooperation is replaced by more individualized work and goals, and where access to power is determined by more abstract criteria, like curriculum vitae, or professional competence, homosexuality may take the form of the gay system. The importance of individualism in the Protestant ethic of northern Europe, coupled with the rise of a more professional capitalist economy, may have discouraged the deepening of same-sex alliances, and consequently of homosexuality. Most men simply adapted to this by renouncing homosexual relationships. This meant that the more homosexually inclined were limited to sexual relationships with others like themselves. This eventually led to the formation of gay ghettos and the "gay" culture.
The type of hierarchy in which men spend their lives (which, in complex societies would be a smaller unit than the total society) may also affect other aspects of homosexuality. For example, where hierarchies are established by personal ties between individuals, but are unstable, the distinctions between what the submissives do and what the dominants do should be accented (e.g. in prison situations). In this light, Mendes (1997) carried out a series of experiments in which she asked male students at the Federal University of Santa Catarina to complete a short comic strip story about a prison. Mendes varied the stories slightly to ascertain the effects of these variations on student responses.
Stories that included "intimate visits with women" significantly reduced the likelihood of ending stories with a homosexual rape scene. But even with these visits, 40% of the subjects still ended their stories with rape (other choices were physical fight, non-sexual friendship, and friendly sex). When Mendes contrasted prisons organized on the basis of personal loyalties, versus an abstract evaluation system, there was a slight (non-significant) tendency to cite rape more often in the "personal loyalties" situation (53.3% versus 33.3% of 60 students). The statistically most significant differences, however, occurred when Mendes combined the "personal loyalties" condition with either "stable" or "unstable hierarchies" (few or frequent changes of cellmates). Where hierarchies were more unstable, respondents were much more likely to end their stories with a rape scene (68.5% versus 26.7% of respondents).
These findings suggest it may be useful to include questions of hierarchy in general studies of homosexuality. For example, consider the finding of Adams, et al. (1996) that homophobic men are more sexually excited by homosexual pornography than are non-homophobes. The standard psychoanalytic interpretation is that homophobia results from repressed homosexual desires. But another possibility is that having an unstable status based on personal loyalties causes both greater homosexual excitement and more concern about what role one plays in these relationships – with the passive role being denigrated. Many other relationships between forms of male/male cooperation/hierarchy and homosexuality might be hypothesized. But I hope these examples are sufficient to show the value of an evolutionary perspective in suggesting what kinds of relationships to look for.