Benjamin Dykes. Journal of Homosexuality. Volume 38, Issue 3. 1999.
This paper has two purposes: first, to offer criticisms of a growing trend in gay studies: that of sorting examples of male homosexuality throughout the world and history into typologies of homosexuality that claim to tell how many kinds of homosexuality there are. Since it has long been common to point out homoeroticism’s diverse meanings in disparate cultures, it seems only natural that the typologies of homosexuality should come under scrutiny, especially since the posited kinds are not generally treated as merely heuristic devices in the cross-cultural literature. Like any serious method of classification, the typologies represent attempts to make metaphysical distinctions between kinds. Yet when one takes a close look at the literature, the kinds tend to come ready-made, and numerous presuppositions are left unexamined.
After criticizing some conflations and methods in the typologies, in broad strokes I shall offer a programmatic interpretation of male homosexuality molded around these typologies and the relevant cross-cultural literature, and indicate some new directions for research. Unfortunately, since the cross-cultural record is largely silent on women’s relationships, I must restrict myself to men even though I think something of a parallel case could be made for women. By way of preview, I will argue that homosexuality is a specific activity and development of ‘‘male consciousness’’: a cognitive and social consciousness of oneself as male, and as standing in relationship to other males. Homosexuality is the male-oriented expression of this, the expression of male consciousness coming into relation to itself. The so-called ‘‘kinds’’ of homosexuality are the functional moments of this consciousness. Put simply, my interpretation is dialectical, historical, and organic. I want to give subjective comportment, interpersonal relationships, and broader cultural meanings each their proper due, while trying to decode the dialectical patterns among what now pass as distinct, fixed ‘‘kinds’’ of homosexuality.
Because it is the most widely used (as opposed to Roscoe 1988 or Donaldson and Dynes 1990, e.g.), I will refer most often to Barry Adam’s typology (Murray 1992:xv-xvi), which categorizes male homosexuality into four kinds: (a) age-structured, usually exemplified by ancient Greek pederasty; (b) genderdefined, usually exemplified by the Native American berdache or ‘‘two-spirit’’ people; (c) profession-defined, which is a contested category used to cover a wide range of traditions and practices, from cross-dressing shamans to dancing boys; and (d) egalitarian/‘‘gay,’’ which is found with modern gay Americans.
Conflations and Problematic Methods of Sorting Kinds
Common Conflations
The first problem that arises when examining the typologies is what might be called a problem of the ‘‘genus’’: properly conceiving the universal whose kinds one seeks. Both in the cross-cultural literature and in everyday life, at least two striking and related conflations in the use of the term ‘‘homosexuality’’ complicate the matter: (1) using ‘‘homosexuality’’ to mean simply a same-sex object; and (2) using it to refer merely to external acts, even if something like ‘‘orientation” is recognized or ostensibly meant. This concern is more than a call for linguistic clarity, though: the conflations reflect confusing and inadequate concepts of homosexuality. When the concept is clarified, clearer language will follow.
The first conflation is well-known: using ‘‘homosexual’’ or ‘‘homosexuality’’ as a differential term used to describe the sexual object. For example, one might speak of ‘‘homosexual rape’’ in prisons, when we only mean a ‘‘same-sex’’ rape: it may very well be that the parties are heterosexual or bisexual, in orientation. In the cross-cultural literature one often finds this pattern of using ‘‘homosexual’’ when only ‘‘same-sex’’ is meant or needn’t be emphasized: male youths will ‘‘engage in reciprocal homosexual relations’’ (Adam 1986:21) and ‘‘homosexual promiscuity’’ (Herdt 1993:183) together. But surely the use of ‘‘homosexual’’ to mean the same-sex object is superfluous and muddies the waters: the males certainly couldn’t have had ‘‘heterosexual promiscuity’’ together, and one often finds that such males lust after women as well. Murray’s brief description of the ‘‘boy-wives’’ of the Sudanese Azande is another case in point (Murray 1992: xvii). Among the Azande and similar groups, older male warriors took boys as ‘‘wives.’’ The boys performed tasks typical of female roles and were sexually engaged with their older ‘‘husbands.’’ Finally, the boys graduated to ‘‘husband’’ roles themselves when older, and some of these boys later married and had sex with women (Adam 1986:23). On the basis of their role as ‘‘wives’’ Murray claims that the boys are ‘‘accustomed to exclusive receptive homosexuality.’’ Strictly speaking though, he really can only mean they are ‘‘accustomed exclusively to being anally penetrated by other males,’’ since his reason for the label ‘‘homosexuality’’ is based solely on the existence of same-sex sexual activities.
The conflation becomes clearer if we remember that ‘‘bisexuality’’ is rarely found in the literature at all, much less the typological treatments of what are even called ‘‘homosexualities.’’ As a friend of mine once opined, male bisexuals are both heterosexual and homosexual–that is, instead of having both ‘‘samesex/gender’’ and ‘‘opposite-sex/gender’’ interests, they are homosexual when intimate with men, and heterosexual when with women. On this view, it would appear one is only bisexual during a menage-a-trois with both men and women!
Whereas the first conflation holds for same-sex objects even in thought, in the second conflation ‘‘homosexuality’’ is used specifically for external acts, even when we know how subjective orientation may cohere with or diverge from the act performed. For example, Herdt reports that among the Sambia of Papua New Guinea, young boys are required to fellate teenage youths and ingest their semen. This is believed to initiate puberty and to help ‘‘grow’’ them into strong men, since masculine power manifests itself as semen. Females and feminine power are poisonous and must be shunned during this period (and throughout much of life). Yet despite the fact that Herdt notes some boys display a very high and long-lasting interest both in active and passive fellatio with other males, while others have very low interest (Herdt 1994: 252)–i.e., shades of ‘‘homosexual’’ and ‘‘heterosexual’’–the Sambia are always said in this work to exemplify ‘‘age-graded homosexuality.’’ That is, the subjective interests of the participants are put aside and the whole institution glossed with ‘‘age-graded homosexuality’’ based upon outward behavior.
Herdt later recognized this conflation and improved it in Ritualized Homosexuality in Melanesia (Herdt 1993: ix) when he said it was more accurate to represent ‘‘age-structured homosexuality’’ in Melanesia as ‘‘boy-inseminating rites.’’ But what does this mean for the place of Melanesia in terms of homosexuality? Is ‘‘boy-inseminating’’ the practice involved in the kind ‘‘ritualized homosexuality’’? Or is the whole thing merely ritualized ‘‘age-structured homosexuality’’? In a sense it’s true that the new phrase is less ‘‘inelegant and unreflective’’ (Herdt 1993: ix) than before, but framing the issue in terms of ‘‘boy-insemination’’ completely tears us from the typologies of homosexuality.
The upshot of these conflations is that homosexuality qua orientation, qua same-sex object, and qua social system generally are trussed together, and usually in favor of external acts with a same-sex object. This results in rather odd conclusions: on the one hand, same-sex rape and fellatio among males who are interested in both men and women, are examples of homosexuality. On the other, a closeted and married gay man who is only sexually active with his wife has no place in the typologies at all! The consequence for all of this can hardly be overrated: one never knows if we are really describing a kind of homosexuality, or a kind of same-sex behavior. Sticking to just one sense of homosexuality in every case would force a whole re-evaluation of these typologies which would turn out to be something quite different. Yet as I stated above, what really needs to be corrected is not so much the language, but the concepts. The concept of homosexuality needs to have these distortions removed from it.
Procedures and Assumptions in Sorting the Kinds
No definite criteria for differentiating the kinds. This is a perennial problem of any typology: determining the right principles of differentiation between kinds of a thing. With the typologies of homosexuality, one finds little clear reason for the kinds involved, and theoretically there is an unlimited number of kinds. Not only are very different, far-flung cultures and practices all called ‘‘homosexuality,’’ but each could exemplify a special kind of it. Yet it isn’t enough simply to assert one has picked out the most important features of the institutions, because the rejoinder is always that it is we who have picked them out according to our own biases.
Take for example Adam’s typology. First, note that this typology leaves out another kind which David Greenberg substantiates in great detail throughout The Construction of Homosexuality: class-structured homosexuality (Greenberg 1988). Greenberg refers age-structured, gender-defined and egalitarian/‘‘gay’’ kinds of homosexuality to kinship-based societies, whereas class-structured homosexuality occurs with the dawn of archaic civilizations. So in at least one typology, a whole realm of same-sex sexual relations appear to be left out or subsumed without explanation.
Second, even some supporters of Adam’s typology disagree whether profession-defined is a genuine kind. In all cases, the examples of profession-defined homosexuality either straddle two other kinds or fit best in one other. In Korea for example, traveling all-male troupes called Namsadang provided entertainment for rural villages even into this century: the entertainment included acrobatics, music, puppets, and boy members, who sexually gratified both the men in the villages and the older troupe members. Boy members were dressed effeminately and with makeup, while older ones dressed as gender-typical men. Dynes compares them to ‘‘butches’’ and ‘‘queens’’ (Dynes 1992: xiii).
Yet although the Namsadang is included in profession-defined homosexuality, it’s clear that its chief features fit both age-structured and gender-defined homosexuality (Murray 1992: xx-xxi): the boys begin as the penetrated, younger members and graduate to insertor roles (age-structured), and they also are ‘‘expected to enact some aspects of the feminine gender role’’ (gender-defined). The fact that these two halves happen to be joined in a profession doesn’t seem to warrant its being considered an independent kind. Other groups such as the Azande boy-wives also feature characteristics of age-structured and gender-defined homosexuality (they are the wives of older men until a certain age when they take boy-wives of their own), but Murray (1992: xx) and Adam (1986: 23-24) do not grant the boy-wives their own category: they are stuck firmly in age-structured homosexuality. And female-impersonating Japanese prostitutes are definitely engaged in a profession, yet they fall under gender-defined.
Third, the terminology employed by typologists is often dropped, blurred, or added to without explanation. Even when presenting Adam’s gender-defined homosexuality, Murray (1992: xxi) introduces it as ‘‘gender-crossing/ mixing/defined organization of homosexuality,’’ though each of these alleged synonyms carries a distinct meaning. This blurring of terms suggests that only a gist is meant: this kind has something to do with gender, but we are not sure exactly what. In fact, in the preface to Oceanic Homosexualities Murray and Dynes suddenly introduce terms for many more kinds than Adam’s typology, which is incidentally the one they themselves use: ‘‘androphile/gynecophile’’; ‘‘ritualistic’’; ‘‘shamanistic’’; ‘‘aristocratic warrior ephebophilia’’; ‘‘contemporary institutionalized prostitution,’’ and so on (Murray 1992: xi). The introduction gives a few others (Murray 1992: xvi). Some of these kinds seem to be synonymous with Adam’s, but this is not explicitly stated and the extra kinds are mentioned casually without being explained. In fact, in Murray and Roscoe’s most recent work it is difficult to tell how ‘‘age-structured’’ is significantly different from ‘‘gender-defined’’ (Murray and Roscoe 1998:7).
Finally, some kinds may be said to have sub-kinds: Adam (1986:22-29) notes that not one but two different models are represented in age-structured homosexuality. One, which includes Greek pederasty and the Azande boywives, is the ‘‘ancient model.’’ In societies represented by this model, ‘‘homosexuality is a medium for the transmission of folklore contained by the masculine gender and is a second stage of parenting that succeeds the motherchild relationship.’’ It is therefore a means to ‘‘the social reproduction of male culture.’’ The other is the ‘‘Melanesian model,’’ which broadly involves a mandatory sexual relationship between young boys and older males for purposes such as physical growth: each boy must ingest the semen of older males in order to grow to full adulthood. Males must also be protected from feminine pollution, and boys rescued from women’s influence. The Sambia and others are represented by this model. But again, in the absence of a firm principle of differentiation, it seems we could equally abolish such distinctions, or make these sub-kinds into their own full-fledged kinds.
The kinds are usually chosen based upon formal aspects. On the whole, societies are sorted into kinds based upon one implicit criterion: the purely formal circumstance of the relative social position of the individuals involved. The individuals concerned are noted to be of disparate ages, disparate genders, disparate classes, or equal position. Or, one notes what position each person takes in sexual acts, i.e., ‘‘active’’ or ‘‘passive.’’ These formal aspects are then taken to be the defining features of the particular kind in a given society. The individuals’ own motives, society’s and their beliefs about who they are and what they are carrying out, are not factors in the sorting of the kinds. In sum, what the kinds mean on any deeper level is rarely identified.
In practice this results in deeply different cultural institutions and behaviors that share similar surface features being collected together under one kind. In Gary Leupp’s excellent book Male Colors for example, the pre-Tokugawa period in Japan (lasting roughly until 1603) is said to have contained ‘‘gender-structured homosexuality’’ (Leupp 1995:55-57). This was characterized by effeminate young males adorned with cosmetics and/or fine clothing, who were involved in various sexual and love relationships with older men. In order to place this tradition in a global perspective, Leupp explicitly links these to the berdache and the Indian hijras. The hijras are males (and occasionally hermaphrodites) who, in their devotion to the goddess Bahuchara Mata, have their penis, testes and scrotum entirely cut off (Nanda 1990:26-29; Sharma 1989:2-3). Living as women in separate communities outside the caste system, they are linked in Hindu thought with fertility, auspiciousness, and magical power. Furthermore, they often prostitute themselves to non-castrated males.
Yet it is clear that only the most extreme selectivity based upon formal surface features could link the Japanese tradition with the two latter groups. How hijra castration and magical powers are necessarily linked with homosexuality, or could relate to the motivations for the Japanese tradition, is missing. Such a formal approach restricts any inner relationship linking the cultures and traditions together from coming to light. For it may be that the societies gathered under the umbrella of gender-defined do belong together, but it is not simply because they are gender-defined. Certainly the age-structured and egalitarian/‘‘gay’’ kinds of homosexuality are heavily determined by conceptions of gender, too. Although age, gender, and class are prominent features of these kinds, which suggests there is something correct about identifying these factors, the kinds need to be recast in better conceptual terms that address their content.
Homosexuality as a universal is severed from the particular kinds. The above two problems of defining homosexuality coherently, and sorting the kinds properly, tends to result in a methodological separation of homosexuality the universal, or the substance of what one is dealing with, on the one hand, from its kinds on the other. In other words, the universal is separated from the particulars. Each side tends to be considered in abstraction, as though one could solve definitional problems and settle the question of homosexuality as the universal on one side, and separately match up abstract properties from disparate cultures to form the typological kinds on the other. And since (understandably) no one wants to grant a Platonic ‘‘form’’ of homosexuality, ‘‘homosexuality itself’’–because considered as something separate–is judged to be only nominal, a representation, and not real in any interesting sense. The view I shall pursue below judges the matter differently, namely that the universal is expressed in and through the particulars which constitute its dialectical moments: each allegedly separate ‘‘kind’’ is always expressing the universal through some determinate shape.
Given these unjustified situations and methods, following are a couple of ways the reality of homosexuality itself ends up being denied or affirmed. First its reality might be affirmed, because we simply define a priori what homosexuality is, and then look for its kinds. But this method ends up appearing arbitrary and involves an unjustified appeal to other definitions. Secondly then we might deny its reality, since off to the side it is meaningless: we empty homosexuality of all meaning and content, and only grant reality to the kinds of it.
General Results of Current Typologies
‘‘A priori’’ definitions. This is both a method and a result. Here, homosexuality per se definitely is something and has a meaning: it is either whatever we explicitly define it to be, or it is construed according to a presupposed, external criterion. As Roscoe (1988:10) points out, static a priori agents and causes are usually adopted in order to simplify the matter, often result in the rejection of historical understanding. For example, if homosexuality is consistently reduced to an unhistorical ground, then one treats all of history as repetitive and ‘‘only a version of the present.’’ Or, if it is understood in terms too narrowly tied to the present, then ‘‘the present is the only history lesbians and gay men have.’’
In the typologies this result manifests itself as two general problems: (a) an unjustified limitation of the study to one or a few aspects of human beings, usually only behaviors. This is very common, but usually hidden in the text. For example, Murray states that ‘‘the [Adam] typology seems to encompass the empirical variance that has been observed in societal schema of male-male sexual relations’’ (Murray 1992: xvi, emphasis mine). Yet only by defining homosexuality in terms of behaviors could one reach this conclusion. Note how homosexuality is equated with the sexual object (male-male) as well as with external acts (sexual relations), two conflations I discussed earlier. So instead of merely showing a correct correspondence between Adam’s typology and the cross-cultural record, Murray is really saying: ‘‘Adam’s typology is a typology of homosexuality; I define homosexuality in terms of behaviors; therefore Adam’s typology encompasses the empirical variance in male-male sexual relations.’’ In another place (1992: xiii) Murray notes that ‘‘[H]omosexual behavior probably occurs everywhere … [but] sexual identity is not a domain for everyone… ’’ Again, since homosexuality is equated with behaviors, he proceeds to disregard any question of identity at all and moves into discussing homosexuality (per the typologies) in terms of how it is organized as observable behaviors. The fact that some societies do take notice of ‘‘sexual identity,’’ and how the selfconceptions of individuals cross-culturally may be structurally related, is not considered.
The second problem is really something we are forced to do when we adopt this method: (b) a constant referral back to other unjustified definitions. The result usually goes like this: an author may decide to discuss only behaviors, either since homosexuality itself does not exist, or because only behaviors can be reliably evaluated. This immediately turns the typology into one not of homosexuality, but of behaviors. But what kinds of behaviors? Homosexual behaviors … which simply repeats the need for the proper point of departure. If the author understands ‘‘homosexual’’ as a characteristic common to all societies with such behavior (e.g., ‘‘two anatomic males having sex’’), this begs the question of homosexuality’s ontological status: it implies that ‘‘homosexuality itself’’ is actually a formal characteristic distilled from a number of societies, with the varied cultural details shorn from it. This, too, is a claim that needs justification. This method and result never has a proper starting point, because we have to keep going back to justify the present term.
The ‘‘dustbin of entities.’’ This phrase is taken from Herdt (1993: ix), and represents the ultimate way of expressing this common, nominalist conclusion of all the tendencies we have seen so far. He says, ‘‘[T]he received category ‘homosexuality’ … must now be represented as one of several different sociocultural types.’’ On this view, homosexuality is a bare name, a dustbin: and it is whatever we put into it or attribute to it–without regard, say, to the relations between the entities we dump in, or what justifies our choice of entities.
The technique here is basically this: suppose that in the world we have societies A through L. Comparing common formal characteristics gives us Kind 1 (represented by A and B); Kind 2 (C, D, E); Kind 3 (F, G); Kind 4 (H, I, J); Kind 5 (K, L). In this example we have then five kinds of homosexuality. When we ask what homosexuality is, the answer is that ‘‘Homosexuality is: Kind 1 + Kind 2 + Kind 3 + Kind 4 + Kind 5.’’ ‘‘Homosexuality’’ refers to, and is the same as, whatever kinds are assigned to it, and it is only represented in terms of these kinds.
For the proponents of this view, it appears that a step forward has been made, since one has cast off the shackles of a clumsy, monolithic homosexuality. And to be sure, we do need to be rid of presupposed, clumsy conceptions of homosexuality. But in my view, rather than being much of an advance, such an account both extols and falls prey to all of the theoretical shortcomings I have outlined. Not only does it give up on homosexuality as a universal, and invite any arbitrary collection of ‘‘entities’’ to be thrown into the dustbin, not to mention any arbitrary way of organizing them into typologies, but it is uncritical of what might be called its ‘‘analytic’’ way of forming static kinds, or construing cultures and people as being composed of isolable properties and ‘‘entities.’’ On the contrary, the broad conclusion I want to draw from this methodological section is that analytic social interpretations are insufficient.
Discussion and Suggestions
At the beginning of this paper I stated that the solution I offer rests on an understanding of homosexuality as a certain activity of consciousness. But first I should say a few words about the social constructionism/essentialism debate and the role of ‘‘consciousness.’’ For, aside from the fact that speaking of ‘‘consciousness’’ might sound slightly mystical, it might also strike some as clearly essentialist. On the contrary, interpretations of ‘‘kinds’’ of people are always implicitly consciousness-based models and, strictly speaking, what is usually meant by essentialism is not a theoretical tendency one could consistently maintain. Briefly, my own view on the debate is that one needn’t take sides on it, because it is based on a false dichotomy and its issues need to be fleshed out in completely different ways, without unhelpful references to ‘‘constructions.’’
Throughout the 1980s and early 1990s social constructionists developed accounts of sexual orientation in opposition to theoretical tendencies that were labeled ‘‘essentialist.’’ After many years of battle over the complex issues involved, Edward Stein (1992) made an important contribution to the debate when he offered a bare-bones definition of each side’s position: social constructionism of all sorts holds that homosexuality and sexual orientation are ‘‘culture-dependent,’’ ‘‘relational,’’ and ‘‘not objective.’’ That is, they are somehow related to larger cultural arrangements in society. Essentialists were supposed to believe just the opposite, namely that homosexuality and sexual orientation are non-relational properties that one has without any relation to any other people or objects in the world. Now these standpoints are generally argued for straightforwardly based on these assumptions. But understood in another, more useful way, these definitions mean the debate was over whether homosexuality and sexual orientation are mediated (i.e., related to the individual’s awareness of himself or herself within a whole network of social conceptions and practices), or immediate to the exclusion of any mediation (i.e., that one could be gay without any relation to other people or things whatsoever).
Understood this way, the essentialist side is clearly in trouble. Essentialism is simply incoherent because there is no property or person that does or could exist in such an immediate, non-relational way: in what could only be termed a void. Any account of homosexuality and sexual orientation must include an explanation of the relationship between the subjectivity of individuals and the greater social conceptions of sexuality, masculinity, et cetera. The sexual subject is mediated on all sides, and also within himself or herself by complex judgments, interests, and his or her own history. Essentialism is not a view anyone could seriously hold–but note this doesn’t mean that only social constructionism is true, because looked at in this way everybody has trivially to be a social constructionist.
The important lesson to take from this debate is that meaningful differences between theories of sexuality are really a matter of the direction and character the social and natural mediation takes, not talk about constructions. Any account of homosexuality and sexual orientation must include an explanation of the relationship between the subjectivity of individuals and the greater social conceptions of sexuality, masculinity and femininity, et cetera. This mediation of the individual is implicitly recognized, as it must be, in every serious account of what homosexuality is–whether in ethnology, social constructionism, biology, Kinsey, and so on.
Moreover, such accounts must include the mediating role of consciousness, since categories of social discourse are at least partly an outcome of the activity of human minds (no matter their biological ground, for example), but they have their effects by being recognized, confronted, and absorbed by other human minds. So, far from excluding a consciousness based model of causes, theories of social kinds already presuppose such a model. To drive this point home, one may refer to two other accounts of social kinds that have tried to describe a more complex interplay of human intentions and social categories. One such model is Ian Hacking’s ‘‘dynamic nominalism’’ (Hacking 1986), which views each category as coming from a unique mixture of both the autonomous activity of individuals and the labeling activity of experts, which creates a ‘‘reality some people make their own’’ (Hacking 1986:234). Thus these categories do not cause identities, but create ‘‘new ways for people to be’’ (Hacking 1986:223). From his descriptions of the people who comprise the two vectors of activity we can see that categories such as sexual orientation are to a degree created out of, and received by, our consciousness, and are immanent in our intentions.
Another approach is Claudia Card’s suggestion that categories of sexuality comprise a Wittgensteinian ‘‘family resemblance’’ or ‘‘open concept’’ (Card 1995:30-1, 45-6). On such a view, the various cultural expressions of homosexuality are connected, but in the manner of family members sharing overlapping features that show a resemblance only (i.e., not everyone in the family has the same large ears or strong chin). That is, there is no single feature held in common by all the kinds, solely in virtue of which they all may be categorized as homosexuality. Instead they form something like clumps, where features of some kinds criss-cross and overlap, with other features connecting them to other clumps. In Card’s account too, we are not just dealing with abstract categorizing but with discovering the overlapping ways in which actual people in actual cultures have cognized themselves and their situations: therefore we are back to recognizing a consciousness-based model.
These two theorists show there are many possible creative theories of the interplay between individuals and categories, of the relationship between various cross-cultural expressions of them, and that all implicitly involve a consciousness-based model of some sort. That said, I think the following points are crucial for a proper rendering of what are now treated as the kinds of homosexuality.
First, neither in a cultural interpretation nor in a systematic cross-cultural exposition should we begin right away with ‘‘homosexuality.’’ As we have seen, ‘‘homosexuality’’ is too often unexamined or taken for granted. The richness of the cross-cultural material suggests that homosexuality is too complex to act as an initial, free-standing concept to which we instantly attribute kinds. Otherwise, ‘‘homosexuality’’ as a ready-made concept is nothing more than an unjustified, abstract name. Instead, we should treat the universal, that is to say the concept of homosexuality, as a result and goal. Our methodology and metaphysics should be geared to show it both as something internally mediated and complex, a building up and transformation of more immediate and basic relations, feelings, and so on; and as a specification of broader cultural grounds. For example, it seems that homosexuality must at least be grounded in relationships with other males, which means one could begin with such relationships and investigate how these provide a ground for homosexuality as a concrete phenomenon. Only then could we be justified in trying to speak of something like more broad-scale ‘‘kinds’’–but note that such kinds would simultaneously be developed accounts of relationships between men, rather than simply a monolithic, presupposed homosexuality.
Homosexuality is most fully understood when we construe it organically: as subjective comportment, as interpersonal relationship, and as social system, all with their dialectical and internal complexity. Each has its own nuances and role, and when we say ‘‘homosexuality’’ we may mean one of these aspects over the other. So it is not ‘‘wrong’’ to say homosexuality is an orientation, but it is incomplete: we haven’t said anything about orientation’s subjective, constituent moments. Similarly, that one finds oneself only liking (say) men, or both men and women, is important and crucial, but it goes little beyond simple feeling, without inquiring into the structure of this phenomenological relationship. Hence, only a systematic philosophy of homosexuality would be the cognition of what homosexuality is, and would offer the precision and thoroughness we need to replace current methods. But at the beginning, homosexuality is strictly abstract and cannot be automatically introduced.
Second, I argue that homosexuality should be construed in terms of a specific activity and development of what I call ‘‘male consciousness,’’ the simplest formulation of which is just ‘‘an awareness of oneself as male and in relation to other males.’’ Put differently, homosexuality is always intimately linked to the general male consciousness expressed in, and undergirding, males in a given society. As such, it has as its background both male consciousness generally and homosociality more particularly–but again, it is a specific activity and development within this milieu. Let me say a few more words about what I mean by this ‘‘male consciousness,’’ to assure you that it is nothing mystical–in fact I could compare it to Hegel’s conception of ‘‘spirit,’’ which is a type of activity that is productive of and conditioned by social relations.
Instead of ‘‘male consciousness’’ I might have said something like ‘‘intragender sympathy,’’ if such a phrase weren’t so purely affective. What I mean is an activity of bodies, affects, and cognition that is (a) immanent in male individuals’ psychology, their comportments and attitudes; (b) a common spirit or ethos of male social groups embodied in their customs, norms, and relationships between their members; and (c) somewhat distinct from what is construed as feminine, though obviously it is conditioned by it, and interacts with it. Now as consciousness, it is generated by human minds and so not a mere matter of bodies interacting, but of course it is not wholly distinct from them: for example, it is well-known that the penetrability of male bodies plays a role in male culture, whether its penetrability represents a weakness or source of pleasure; and so-called ‘‘age-structured’’ homosexuality emphasizes the young male body’s immaturity and growth. As such, male consciousness permeates and conditions the social milieu and individuals qua male. Since it is an activity of bodies, affects and cognition, we may retain both behaviors and comportments like sexual orientation as objects of cultural interpretation, since orientation is a form of consciousness.
So far I’ve given some general features of male consciousness, and not much detail. But this is because, like homosexuality, male consciousness mustn’t be considered as static and given beforehand: I argue that it, too, must be seen as process, development, and activity. Its more general nature undergirds males in a culture, and has to do with bodies, interpersonal relationships, and cultural meanings; but such is only its abstract nature, while its expression and development actually takes place in existent cultures. In some cases, the male consciousness of a culture may have elaborate provisions for sex between men, or for certain sorts of friendship, while granting little time to the cultivation of individual subjectivity. In our interpretations we have to see how such elements of male consciousness fit together.
On this view, the ways in which men comport themselves affectively, erotically, in friendship, and so on, are also the ways in which they express and cultivate male consciousness, and how they have knowledge of themselves and it. Yet while the general activity of male consciousness undergirds all males, and is the backdrop for and embraces all orientations, preferences, and so on, not all men have the same erotic relationship to it. For individual men are also erotically related to what is construed as feminine, and how this comportment is expressed will make the difference between how males of different orientations cultivate male consciousness and how they cultivate themselves for it, like a sort of ethical life. From this point of view, the difference between sexual orientations and other subjective comportments will be largely a matter of understanding the erotic dialectical relationships individual men have with male consciousness and with what is construed as feminine.
This gives us a provisional way to lay out the differences between sexual orientations which goes beyond the usual view of a compass needle stuck upon a single person, and move to a more dialectical and interpersonal view. In the heterosexual orientation, masculinity and femininity are taken to be more or less absolute opposites which are incomplete without the other and are meant to be united. So the relationship to other men is essentially incomplete, or carries an order to turn away from it, since true unity lies across a posited gulf, with femininity as embodied in women. In the bisexual orientation, a direction both toward the masculine and toward the feminine are combined and experienced as an absence of limitation or negative freedom. And in the homosexual orientation, male consciousness is male-oriented, becomes its own erotic object, and so encompasses both subject and object. Contrary to the heterosexual orientation, unity is not found outside oneself, for one begins with something at hand; it is fertile out of its own resources and enjoys the repose of being nourished by what is male. Thus a true typology of homosexuality would attempt to show how male-oriented male consciousness–through different cultures–eventually comes into complete repose with itself.
In practice, this yields more meaningful comparisons than usually found in the literature. Take two examples of age-structured homosexuality: the Sambia and relationships between older Japanese monks and their younger acolytes in the pre-Tokugawa era. Among the Sambia, we have a shared maleoriented consciousness via the exclusion of feminine influences as embodied by women, and therefore in such a way that by having such a conscious dread of the feminine, it is constantly affected and conditioned by it. It stands in conscious, locked opposition to it. In the Japanese example, the shared maleoriented consciousness was not a fortress against feminine poison and debilitating influences. It involved an explicit delight in boys, but it was not conditioned by a conscious withdrawal from the feminine as embodied by women. There was rather a self-cultivating delight that disregarded it. Both cultures formally had something like age-structured homosexuality, but in light of the analysis I’m advancing, on a deeper level they consisted in somewhat different things. The Sambia relationship has a polarized dependence upon the existence of the feminine, while the Japanese relationship has relative independence from it. These characteristics were moreover recognized by members of these societies, though not using this language (Herdt 1994; Leupp 1995; Saikaku 1990; Watanabe and Iwata 1989). With this sort of analysis we can see how terms like age-structured homosexuality are inadequate.
The type of systematic study I suggest would be in two parts, roughly as follows. In the first, more phenomenological and conceptual part, one would develop the full concept of homosexuality (which would have been grounded in detailed cross-cultural study). This would (a) describe the dialectical development of types of subjective comportments such as desire, taste, preference, and orientation, and what type of relationship they set up between subjectivity and the ‘‘other.’’ Homosexuality qua orientation would be one form of the general phenomenological construal of orientation. Next (b) one would show how these subjective comportments are expressed in actual interpersonal relationships such as friendship, love, sexual activity. Finally, (c) one would show how these comportments and interpersonal relationships generate and are reflexively conditioned and informed by, broader social practices, institutional norms, conceptions of the masculine or gender, et cetera. This third moment of the first part would be the place to show the structure and dialectical interrelationship between what are now called the ‘‘kinds’’ of homosexuality. Here then, we might distinguish homosexuality qua orientation from the full homosexuality qua social and interpersonal system, which latter might alternately be called ‘‘complex fraternity.’’ This last moment is what I propose to flesh out further in what follows, since it is what the typologies aim at describing.
The second part would investigate one by one the regions and cultures where these more formal moments are actually being expressed, and perform cultural interpretations of their personal and cultural elements as I have outlined. In such a way, raw ethnological research gives us the material for the interpretation, and cultures are reciprocally reinterpreted to give what may well be a richer portrait of their practices and experiences. Furthermore, it might guide ethnologists to look for types of relationships (for example) which hadn’t been investigated before.
Seen from this point of view, what typologies try to describe are really the ways in which this male consciousness that is aware of itself as male, and oriented toward what is male, has experienced itself. That is, they are self-articulating stages or moments generated through the activity of this process of male consciousness orienting toward itself and being nourished by itself. These are the moments through which it comes into complete relation with itself–both conceptually, and actually in the historical world, through individual societies’ traditions. Current typologies have something right, but the ‘‘dustbin of entities’’ view is wrong. Rather, the concept of homosexuality is involved in and preserved in its moments, not severed from them. The two sides occur together: in gender-defined homosexuality (though this is currently a defective category), the homosexuality side should be seen as a male-oriented activity and development of male consciousness, and the gender-defined side as an internal mode of it that describes something about its form and content and how it is expressed in societies. We never leave the ‘‘universal’’ (the concept of homosexuality) behind, because it is always being actualized through its moments, and brought into existence with them. It is brought out of the conceptual realm, out of the realm of potential, and actualized through sex and love between living men by means of these functions.
These propositions open up a number of possibilities: for example, using analyses like the one above, we might see that individual societies exemplify stages within a particular function–age-structured homosexuality might be seen as progressing from the more dependent and non-self-sufficient Sambia to the more independent Japanese monasticism. Or, each function itself may exemplify one stage of the concept of homosexuality. Murray (1992: xxvi) says a succession occurred in the West from age to gender to egalitarian homosexuality, and that Japan experienced a similar transformation from age to gender. Greenberg too, notes similar changes in the kinds from kinship-based societies to class-stratified civilizations.
Indeed, conceptualizing homosexuality as I am suggesting gives insight into such a view of transitions and development between moments. Let’s examine Murray’s succession: what do these kinds really mean, and what is their relationship to one another? Using my propositions we can see not only a historical but a conceptual pattern emerge, involving at least two elements: (a) the relation between sex and gender in this male consciousness; (b) the relationship to the feminine as embodied in women. They should be described, though, using conceptual terms, rather than the convenient yet misleading labels usually given to the kinds.
The first stage is said to be age-structured homosexuality. This is better described as: (a) a unification of the male sex and gender, and (b) the externalization of the feminine. In this functional mode of homosexuality, the male sex and gender are unified in each person, and the feminine is conceived as something external, residing only in women or children. Masculinity is a prize or mystery to be kept away from women, and males are the proper receivers of masculinity and male attentions.
The next stage is said to be gender-defined homosexuality. This is better characterized as: (a) an internal split between sex and gender, and (b) an importing or expression of the feminine, but still in an external way. Now, one partner is recognized to have stereotypic feminine or ambiguous gender traits. These feminine or simply ‘‘non-masculine’’ traits (depending upon the particular culture) are seen as natural to him. The other partner is conventionally masculine. The feminine is now something that can be adopted by males, or expressed in males. However, because it involves the formalism of one male possessing masculinity, while the feminine is externalized onto the other male, this sort of institutionalized relationship is still external.
The third stage is said to be egalitarian/‘‘gay’’ homosexuality. This more accurately involves: (a) a reunification of sex and gender, but in a better way than before, and (b) a non-dependence on the external feminine. The previous two involved a dependence on external femininity because the male consciousness was defined in terms of a mutual opposition. Here, the male sex and gender are reunited in each partner, but masculinity doesn’t need the external feminine to help define it. This function of homosexuality involves a mutual and equal enjoyment of the male partners, and the recognition that the masculine has its own emotional depths and inner resources that don’t need to reference external factors.
The word ‘‘better’’ in the third stage may be misleading, because it gives the impression that even contemporary relationships between older men and younger men are somehow ‘‘worse’’ than contemporary egalitarian relationships. This misconception arises because we tend to lump all age-structured relationships together merely because of the formalism of their involving different ages. But it would be wise to forget the idea that all age-structured relationships are the same. By ‘‘better’’ I mean it overcomes the dependence mentioned in the first and second stages. And clearly modern relationships that are structured by age are better in this way, because unlike the Sambia or Japanese monasticism, they really involve the improved relationship described in Stage Three, except they happen to include differently-aged partners. This makes even modern age-structured relationships good candidates for inclusion under Stage Three.
It might also be useful to explore the idea of permutations, or of functions within functions. For example, under the convenient yet misleading heading ‘‘age-structured’’ it might be more accurate to say we have all three functions as stages: first we would have groups like the Sambia, where masculinity is considered more a substance to be given to boys, and femininity a poisonous substance to be kept away. This correlates with the first stage. Then we would have age-structured societies where the younger boys are considered feminine, like boy-wives, which passes away as they get older. Thus this still involves the formalism of femininity being a substance, but one that can be adopted by or expressed in males–the second stage. Third, we would have modern same-sex relationships that are age-structured. They are no longer dependent upon the external feminine. Both partners are simply male, but in a better way than before because that bifurcation is overcome. This gives a few rough ideas of the affinities, continuities, and distinctions between what now pass as ‘‘kinds of homosexuality.’’
In conclusion, the above points lay out a different answer to the problem of homosexuality as a universal and suggest how it should be further developed in a systematic philosophy of homosexuality. As the gaps in the historical record continue to be filled, I think that only a dialectical, activity-centered interpretation grounded in male consciousness along the lines I have suggested will preserve any typological schema as a viable tool, albeit with corrections. On a local level, this would allow us to do proper justice to all levels of cultural interaction, while appreciating their interconnections. On a global and philosophical level, this would allow us to explain the broad historical continuities and common patterns that we find, while avoiding other unhelpful techniques.