![[Document Home]](./../../../images/42x53/doc-home-button.gif)
![[Bottom Page]](./../../../images/42x53/bottom-button.gif)
![[Prev Page]](./../../../images/42x53/prev-button.gif)
![[Next Page]](./../../../images/42x53/next-button.gif)
I'd like to begin this essay on language and gender/sexuality by drawing attention to the slash in the title. If I were computer literate enough to do so, I might drop the slash and instead write the title so that the word `gender' appeared with a line or an X through it, putting it sous rature, as deconstructionists say: under erasure -- absent but still present. However, since my computer skills are not quite up to the task, I make my point with a slash: "Language and Gender/Sexuality". And in fact, that slash there is just fine, quite appropriate. It separates sexuality from gender, almost like the Lacanian bar. But not quite, for it also slashes gender, wounds it, punctures its integrity and inserts sexuality as different from, but also implicated in gender.
Images of erasure and slashing are both apt ones, because in this essay, I want to try to develop for the study of language the relevance of the insight, first articulated with clarity by Gayle Rubin in 1984, that sexuality is importantly different from, and not reducible to, gender. In her well-known article "Thinking Sex" Rubin argued that "although sex and gender are related, they are not the same thing, and they form the basis of two distinct arenas of social practice" (1993:33). Building on Michel Foucault's thesis that sexuality as an arena of knowledge, feeling and practice only arose during the last three hundred years, Rubin argues that "[t]he realm of sexuality has its own internal politics, inequities, and modes of oppression" (4). Sexuality cuts across other modes of social inequality, such as racial, class, ethnic or gendered inequality, and it sorts individuals and groups according to its own intrinsic dynamics (22). "Modern Western societies appraise sex acts according to a hierarchical system of sexual value" (11), Rubin explains; a hierarchy in which monogamous married reproducing heterosexuals are at the top, and promiscuous homosexuals, transvestites, sex-workers, sadomasochists and others cluster around the bottom. Rubin points out that this sexual stratification is not merely theoretical; it is profoundly social and political -- it results in only "good" sex acts being accorded moral complexity, it results in public demonization of sexual variation, and it results in very concrete, and very harsh, legislation that terrorizes and penalizes individuals who engage in consensual sexual practices that do not meet normative standards of approval.
The main theoretical contribution for which Rubin's essay is known is her suggestion that feminism, which is essentially a theory of gender oppression, is not necessarily the privileged site of a theory of sexuality (32). "Feminist conceptual tools were developed to detect and analyze gender-based hierarchies", Rubin explained. "To the extent that these overlap with erotic stratifications, feminist theory has some explanatory power. But as issues become less those of gender and more those of sexuality, feminist analysis becomes misleading and often irrelevant. Feminist thought simply lacks angles of vision which can fully encompass the social organization of sexuality" (34).
Rubin's essay, and the theoretical developments that it helped set into motion the development of what has come to be known as queer theory. Like feminism, queer theory is not a singular entity. Arising out of a combination of political activism and poststructuralist approaches to identity and language, queer theory consists of a number of more or less coherent perspectives that interrogate sexuality as historically and culturally situated sets of values, knowledges, structures, institutions and identities.
By concentrating on sexuality in this way, queer theory has managed to wrest it from purely psychological attention and constitute it as an entirely new object of socio-cultural inquiry. And just as whole new avenues of inquiry and opportunities for understanding were opened when feminist scholarship began pointing out that society is systematically structured in ways that give men privileges and protect their dominance generated profound new insights into the nature and workings of society, or when postcolonial critics began insisting that the knowledge Western societies produces about non-Western people both legitimates and facilitates our domination of those people, so has the queer assertion that sexualities are not so much natural identities as they are culturally generated subject positions made it possible to analyze the workings of society and culture in ways that have not been imaginable until now.
As I hope my comments have indicated, queer theory, despite the name, is not so much concerned with homosexuality, as it is concerned with all forms of sexuality. For this reason, queer theory has rapidly made an impact on a large number of disciplines, mostly humanistic disciplines like film studies, cultural studies, history and philosophy; but it has begun to be felt even in the social sciences. Here work has largely concentrated on sexual minorities, and sociologists working with sexual minorities and political scientists engaged in thinking about political recognition struggles have begun to turn to and develop queer theory, as have anthropologists like myself who work with gender variant groups.
The one humanistic or social science discipline where queer theory has yet to make any impact whatsoever is linguistics. This in itself need not be so surprising -- linguistics is notoriously slow to take account of developments outside its own discipline, as the recent severe criticisms of sociolinguistics by scholars like Deborah Cameron (1990) and Kathryn Woolard (1985), have noted. However, in the case of queer theory, the lack of impact on linguistics is truly remarkable, because one of the foundational concepts of queer theory is a linguistic one -- namely the notion of performativity, as it was introduced by J.L. Austin, and later critiqued and modified by Jacques Derrida.
Performativity, in the sense of acting on the world and calling something into being through the felicitous use of language, was famously appropriated and reinvigorated by the feminist philosopher Judith Butler in her 1990 book Gender Trouble. In that book, Butler argues that gender is performative -- it is a discursive production; the result of the very actions that are usually thought to result from it. Butler flips cause and effect, making what is essentially an ethnomethodological point that gender is not an ontological state from which practice flows -- it is, instead, the effect of practice; a practical accomplishment. Furthermore, because the locus of gender lies not in ontology but in the actions that performatively call it into being, those actions must be repeated, in order to "produce the appearance of substance, of a natural sort of being" (1990:33).
As the Australian literary scholar Annamarie Jagose so aptly phrased it, Gender Trouble put the concept of performativity into a kind of hypercirculation (1996:86). From having been a concept of concern to philosophers of language and a few deconstructive theorists, suddenly the concept was everywhere. Everywhere, that is, except linguistics.
With one important exception; the reason for this essay. The concept of performativity, and the concerns of queer theory have recently been appearing in work that is emerging now on language and sexuality. Concern with language and sexuality is not, strictly speaking, an entirely new phenomenon. There are works dating back to the 1930s and 40s which document the argot of gay men (e.g. Read 1977 [1935]; Legman 1941; see also Giallombardo 1966 for what seems to be the earliest scholarly discussion of lesbian language). The 1980s saw a brief attempt to articulate a coherent perspective on language and sexuality -- the publication in 1981 of James Chesebro's edited volume Gayspeak seemed to promise scholarly legitimization of the topic. This, however, didn't happen, and as Anna Livia and Kira Hall recently noted in their review of this period, "the field itself failed to emerge" (1997b:5).
A lot has happened since Gayspeak, however, and for the past few months, I been reviewing the burgeoning literature on the topic of language and sexuality that has begun appearing in the nineties for a review article on queer language that I have been commissioned to write for the Annual Review of Anthropology. My completely unexpected and rather dispiriting conclusion is that despite an explosion in publishing on this topic, the field still hasn't emerged. Furthermore -- and this is even worse -- I don't see that it will emerge unless the linguistic work on language and sexuality radically changes direction and level of theoretical sophistication.
For the rest of this essay, I want to explain what I mean by that. But rather than documenting my claims by presenting scores of short summaries of the many articles I have been reviewing, what I am going to do here is cut straight to the chase by concentrating on two works that I think constitute the most significant contributions to this area of study. They are significant in the sense that they are read and referred to by everyone working with the topic of language and sexuality. By examining what I see as quite serious problems in these two works, I hope to illustrate the fundamental difficulties that lie at the base of all current work on language and sexuality, and I hope to use those difficulties to develop some suggestions about what must happen if the field of language and sexuality is ever to develop in a productive and challenging way.
The first work I want to discuss is a short article by William Leap. Leap is a linguist who works at the American University in Washington D.C.. He is an important figure in the study of language and sexuality, first of all because he publishes very prodigiously -- in addition to being the author of Word's out, which is the only monograph devoted to the study of the language of a group that is defined in terms of its sexuality, Leap has also edited the volume Beyond the lavender lexicon, and a special issue of the journal World Englishes on queer language. He also has published numerous articles on the topic of what he calls "Gay English". In addition, Leap founded and still hosts an annual event called the Conference on Lavender Languages and Linguistics, in which scholars, most of whom are graduate students, present papers on topics related to language and sexuality.
The particular work by Leap that I want to examine here is a short one entitled `Can there be a gay discourse without gay language?' (Leap 1996a). This paper has the same title as, and is pretty much identical to, the first chapter in Leap's monograph Word's out: gay men's English (1996b). I have chosen `Can there be a gay discourse without gay language?' because it is a very succinct summary of the theoretical orientation that buoys up Leap's entire body of work.
The title of the paper `Can there be a gay discourse without gay language?' asks whether there can be such a thing as a "gay" way of speaking if there is not something identifiable as a "gay" language. Now given that question, one might be forgiven for wanting to know how Leap defines the key terms "gay" and "language". Unfortunately, they never are defined -- not in this paper, nor anywhere else in Leap's work. It is thus perpetually unclear what exactly a gay speaker is, and who exactly adjudicates the boundaries and the content of "gay". Also unclear is the basis of Leap's formalist understanding of "language"; one that appears to assume an a priori distinction between `language' and `discourse'.
Now lack of clarity is not always a bad thing -- undecidability can often be mined and deployed in ways that brightly illuminate the workings of language and social relations -- this is, indeed, the whole basis of deconstructionist theory. But Leap is not a deconstructionist. And even though he never defines his key terms, he seems to know what he means by them, and he seems to assume that we do too.
Some of his assumptions can be gleaned from his description of his data. The speakers whose language Leap analyzes, he explains, are "some gay men in the United States -- primarily urban white gay men from academic or professional backgrounds and male college students with similar home/family profiles" (399). Note though that even in this description, "gay" is not defined. One is left to imagine, therefore, that what makes all these men gay is self-ascription. But one is also left to imagine that the precise denotational content of that self-ascription might vary among the different individuals even within this small group.
In any case, what happens next is that this very particular group of urban white professional self-identified gay men becomes synonymous with generic gay men. So the men in Leap's social milieu become a metonym for a larger collective. The men become the unmarked "gay men" in Leap's work, and their linguistic behavior becomes the similarly unmarked `Gay English' (with capital letters).
Having, in this way, created an object of inquiry, Leap then explains that his research has identified two linguistic characteristics that he sees as fundamental to Gay English. The first is: "gay men follow any number of linguistic strategies to ensure that conversations with other gay men are co-operative, not exclusionary or antagonistic" (399). The second is that gay men use euphemism, code words and innuendo to signal their own sexuality and to ascertain the sexuality of other men in settings in which a question like "Are you gay?" would be inappropriate and even dangerous.
Now since conversation analysts have shown us that conversations, even explicitly adversarial ones, are by definition co-constructed, and that co-construction involves a wide range of carefully choreographed linguistic strategies and moves, one might wonder what is so special about gay men engaging in co-operative talk. One might also wonder in what way the use of euphemism and code-words are in any way particularly gay. But this is what Leap wants to argue, and this is where we reach the heart of the problem.
This is the way Leap himself defines the issue: Are the features of linguistic interaction that he identifies as being habitually used by gay men (i.e. what he calls co-operative discourse and language of risk) -- are these linguistic features "linguistic properties or [are they] properties that are expressed in linguistic terms but are derived entirely from elements of social discourse external to that of the linguistic system" (399-400). He goes on to outline the consequences of both of those propositions: "Under the first alternative", he writes:
co-operative discourse and language of risk are in fact linguistic properties -- that is, part of a speaker's grammar (his knowledge of language) or part of the rules for text-making that enable a speaker to make use of grammatical knowledge in situated social exchange. If this is the case, the presence of such features in gay men's knowledge of English is substantially different from the knowledge of English maintained by heterosexual persons. This difference in turn argues in favor of the uniqueness of Gay English and raises a series of additional claims about gay men's socialization and identity construction.
Bearing in mind that we still have no definition of what "gay" means, what Leap seems to be saying here is that gay men may somehow have a different grammatical competence than non-gay individuals. However, since he never defines how exactly gay men's use of co-operative discourse or innuendo differs from anyone else's, other than in terms of vocabulary and social situation, we are left wondering what exactly the "knowledge of language" is that is supposed to be so different. In addition, because Leap nowhere explains which theory of grammatical competence he is drawing on to permit him to claim that differences in pragmatic language usage reflect actual differences in grammatical competence (something I have no doubt that Chomsky would be the first to reject), the idea that gay men might have a particular grammatical competence is linguistically nonsensical.
Leap goes on to say that there is also another, alternative, explanation of what "Gay English" might be. "[I]t is possible", he continues,
that co-operative discourse and language of risk are not implicit in gay men's knowledge of language at all but are reflections, in linguistic terms, of the more general interplay of power and social process that informs gay discourse at all sites....Gay English, under this alternative, is still a viable component of gay culture [upon reading that, one wonders under what circumstances would Gay English not be a viable component of gay culture?], but the contributions of Gay English to the integrity and authority of gay culture are now somewhat scaled down. The uniqueness of Gay English (if in fact this is a unique variety of English at all ) now derives from sources external to linguistic knowledge -- speaker identity, speech context, content or topic of discussion. Under this formulation gay language becomes a specialized vocabulary (or, at best, an inventory of idiom and metaphor), so that researchers now have no reason to take the details of linguistic form into account when exploring gay culture or communication. Finally, the absence of a uniquely Gay English suggests that gay resistance to heterosexual oppression is also not linguistically dependent; hence any variety of English can be a suitable format for such resistance, and any speaker of any variety of English can participate knowledgeably and with authority in its construction" (400; emphasis added)
Here we have all the shibboleths that a number of sociolinguists have been roundly and repeatedly criticizing since the mid-1980s. First, Leap tells us that if Gay English is not part of grammatical competence, then it must be a kind of language that is a mere reflection of society, one whose meaning derives entirely from sources external to itself. If this is true, he continues, "researchers now have no reason to take the details of linguistic form into account when exploring gay culture or communication" -- a formulation that implies that the exploration of linguistic form is somehow unimportant to an understanding of culture and communication; a formulation that, moreover, also invalidates all of Leap's own research efforts. Finally, the seemingly crucial insight that "any speaker of any variety of English can knowledgeably and with authority" oppose homophobic discourse and contest heterosexual oppression, is abandoned and left unexplored. Leap seems to feel that the importance of Gay English is somehow diminished if it is possible for it to be appropriated and used by just anybody. But what language is not available for appropriation and use by just anybody?
Leap himself says that he is "unclear" on which of his two alternatives for Gay English is more accurate (399). However, the tone of his comparisons, and the tenor of his other work (e.g. 1995, 1996b) -- which celebrates what he calls "authenticity" in Gay English -- make it pretty clear which horse he's putting his money on.
Because Leap does not succeed in identifying any feature of language except lexical items that might be particular to gay men, his work is tautological. The answer to the question "What is Gay English?" is: English spoken by gay men. What makes it gay? The fact that gay men speak it. Why do gay men speak it? Because they are gay men. And so on, round and round we go. This is a dog chasing its tail. Furthermore, Leap's investment in having Gay English be "authentic", a condition that he defines as "optimal, valuable and life-cherishing" (401), belies a very specific (and dated) political agenda, one seemingly rooted in the "Glad to be Gay" and "Gay is Good" movement of the 1970s. What is theoretically damaging about this agenda is that it first of all it forces Leap to assume the adjudicating role of someone who must somehow differentiate "authentic" gay speech from "inauthentic" gay speech. And second, his view that "authentic" Gay English is somehow the property of gay men blocks all inquiry into the ways in which the linguistic features that compose it are resources that are available to anyone to use for any purpose, regardless of their sexuality.
If Leap's tautological and linguistically untenable understanding of language and sexuality were unique to his own work, I would not have devoted so much space to discussing it. As it happens, however, the basic problems that Leap's work exemplifies are also present in the overwhelming majority of research on language and queer sexualities. In that work, sexual categories are assumed, not explained or interrogated; the only people whose language is analyzed are people who explicitly self-identify as queer, so we start out by "knowing" the identities whose very constitution ought to be precisely the issue under investigation; and commonplace interactional features, such as successful turn taking or the co-construction of topical coherence are identified as being characteristic of queer language, because people we already know to be queer employ them in their speech, and this queerness is said to be "reflected" in their language. The spinning dog reappears.
I have been using the word "queer" rather carelessly in reference to the publications of Leap and others whose research resembles his. Leap himself uses this word extremely rarely -- his work is firmly anchored in an epistemology of the closet; one in which `gayness' is an identity that one `has' and chooses to either reveal to or conceal from others -- and he refers to queer theory almost not at all. Work on language and sexuality that has been appearing within the last couple of years differs here from Leap's, in that some of it explicitly and frequently invokes queer theory as its conceptual frame. But does this different theoretical positioning make a difference in the way in which language and sexuality is conceived or analyzed? To investigate this, I want to turn now to the second work I will be examining, namely the introductory essay to the recent volume entitled Queerly Phrased: language, gender, and sexuality (Livia & Hall 1997a). If there is a state-of-the-art work to which we might turn to see how scholars are examining language and sexuality, this is it. Therefore, the introduction, by Anna Livia and Kira Hall (Livia & Hall 1997b), is particularly relevant to our concerns here, because it explicitly sees itself as outlining an approach to the study of language and sexuality, and as setting an agenda for future research.
Livia and Hall begin their essay with the Foucauldian axiom that our contemporary understandings of sex, and our contemporary sexual categories, are historically generated and culturally specific. This means categories like `gay language' are meaningless outside particular Western contexts, because it is far from certain that people like `gays' even exist as a social and ontological category in the way they have come to do here. But rather than see this limitation as a possibility to be exploited (after all, one can wonder if the theoretical insights produced, and the empirical data examined, might not have been much richer in the field of language and gender if researchers had explicitly recognized and continually insisted that the data they were analyzing was not "women's language", so much as it was the language of white, middle class professional women produced in particular contexts), Livia and Hall regard Foucault's caution about projecting our own sexual categorizations onto others as a difficulty to be overcome. In other words, even though they recognize the problem of studying gay, lesbian, bisexual or transsexual discourse cross-culturally or transhistorically (since those very concepts and identities are modern North American and northern European ones), Livia and Hall want to be able to stake out a field of inquiry that can do exactly that.
The solution to this dilemma, they claim, lies in the Austinian notion of performativity, as it has been developed by Judith Butler. By looking at the way in which language performs actions on the world and calls identities into being through its own felicitous pronouncement, Livia and Hall suggest that linguists can "bring performativity back to its disciplinary origins" (13) and use it to examine the ways in which the language used by sexually variant people calls them into being, creating, in this process "its own object of research" (12).
While this inventive suggestion makes good intuitive sense, I want to point out two quite significant problems with this approach. The first is that in the notion of the performative favored by Livia and Hall, Austin's distinction between felicitous and infelicitous performatives plays no role. All of their examples are of successful, happy, performatives; ones that always work. This lack of attention to the ways in which performatives can fail, coupled with their view of queer language as inherently performative, leads them to make the quite startling claim that queer language is inherently intentional. "An utterance becomes typically lesbian or gay only if the hearer/reader understands that it was the speaker's intent that it should be taken up that way", assert Livia and Hall. "Queerspeak should thus be considered an essentially intentional phenomenon..." (14).
Perhaps the most surprising thing about this statement is that it occurs in an essay that freely invokes Austin, Butler and Derrida to argue its principle points. But the main point of Derrida's criticism of Austin; the point elaborated in great detail in the 1977 paper `Signature Event Context' (in Derrida 1991), and in his lengthy essay Limited Inc (Derrida 1988), is that that performatives work not because they depend on the intention of the speaker, but because they embody conventional forms of language that are already in existence before the speaker utters them. Performatives work, and language generally works, because it is quotable. This is the meaning of Derrida's famous example of the signature, with which he concludes `Signature Event Context'. In order for a signature to count as a signature, Derrida observes, it has to be repeatable; it has to enter into a structure of what Derrida calls iterability, which means both `to repeat' and `to change'. Signatures are particularly good examples of iterability, because even though one repeats them every time one signs one's name, no two signatures are ever exactly the same. The main point, however, is that in order to signify, in order to be authentic, one's mark has to be repeatable -- if I sign my name XCFRD one time and CDFRE the next time, HTGFRT the next time, and so on, it won't mean anything -- it will not be recognized as a signature -- as a meaningful mark. To be so recognized, the mark has to be repeated. But if something is repeatable, this means that it simultaneously becomes available for failure (if I am drunk, my signature may not be recognized, it will fail and my check won't be cashed). It also becomes available for misuse and forgery. This availability for quotation without my permission, untethered to any intention I may have, is what Derrida means when he says that failure or fraud are not parasitical to language, exceptions, distortions (as Austin maintains). On the contrary, quotability is the very foundational condition that allows language to exist and work at all. The fact that all signs are quotable (and hence, available for misrepresentation) means that signification cannot be located in the intention of speakers, but, rather in the economy of difference that characterizes language itself. In this sense, failure and misuse are not accidental -- they are structural.1 Derrida's point is that a speaker's intention is never enough to anchor meaning, to exhaustively determine context. Language constantly evokes other meanings that both exceed, contradict and disrupt the language user's intentions.
In some of her most recent work, Butler (1997) uses Derrida's arguments about the fallacy of intentional language to develop her own perspective on hate speech, arguing that hate speech, which is perhaps the best example of how speech is always cited from elsewhere (hate speech works, after all, by virtue of having been repeated and being repeatable), does not always work in ways intended by speakers, and that it is in the gap between intention and effect (in other words, in the gap between illocution and perlocution) that effective responses to hate speech can be formulated.
In any case, the point that both Derrida and Butler are at pains to elaborate is that no speech is ever "essentially intentional". Quite the opposite -- while intention does have a place in a theory of language, it cannot, Derrida explains, "govern the entire scene and the entire system of utterances" (Derrida 1991:104). What all this means is that any attempt to define a queer linguistics through appeals to intentionality is hopelessly flawed from the start because it is dependent on precisely the fallacy of intention that Derrida definitively dispensed with twenty years ago.
In addition to deploying the notion of performativity in this dubious manner, the second significant problem with Livia and Hall's use of the notion is the way their use of performativity elides the specifically sexual dimension of the language under scrutiny. After all, even if we grant the performative dimension of language, queers are not the only ones to use language performatively. The language of numasticists, sommeliers and oncologists certainly calls individuals who use the language into being as coin collectors, wine connoisseurs or medical experts on cancer. But are those individuals doing exactly the same thing as gays or lesbians or hijras or mollies? If they are not, then in what ways is their language different? If it isn't different, in what way is the original observation that language constitutes identity not simply a banal platitude?
Livia and Hall do not broach this question in their essay, and my point here is that given the language that they use to delineate their field of inquiry, they can't broach it. The main problem is that despite the fact that the goal of their essay is to draw up an agenda for research on language and sexuality, Livia and Hall nowhere have even a word to say about what `sexuality' might be. Indeed, at the precise moment in their text when they would seem forced to define their object of inquiry, they vaporize sexuality into gender.
This moment occurs at that point I have already mentioned, when Livia and Hall have explained that we can't assume that gays, lesbians, bisexuals and transsexuals exist everywhere, but that we still want to be able to think about sexuality and language. At this juncture, when a definition of sexuality becomes unavoidable, the authors invoke Judith Butler, and they tell us that "[w]ith the theory of gender performativity we move away from the social construction of sexuality [i.e. we move away from the vexing problem of projection that Foucault identified] to the discursive construction of gender" (11). This is the move that the field of language and sexuality must make, Livia and Hall tell us, in order to be able to constitute an object of inquiry. But note the glissement, the semiotic slippage, that occurs in that short sentence. "We move away from the social construction of sexuality to the discursive construction of gender". Note how sexuality here collapses into gender; becomes gender. In stark opposition to the very queer theory that Livia and Hall animate throughout their essay, they do not separate sexuality and gender; instead, they amalgamate them.
Once sexuality becomes gender, we are back to where Gayle Rubin started. We are back to a view of sexuality that sees it as analyzable in the same terms as gender. And since gender has a strong tendency to be analyzed in terms of mutually exclusive identity categories (namely "man" and "woman") the risk looms large that an analysis of sexuality will also be framed in terms of mutually exclusive identity categories -- only this time the categories will be "gay" or "lesbian" or "bisexual", instead of "man" and "woman". And instead of naive generalizations about "men's language" and "women's language", what we will produce instead are naive generalizations about "Gay English" or "lesbian language" (e.g. Moonwomon 1995; Morgan and Wood 1995). What is very noticeably left out of this picture entirely, what is nowhere mentioned by Livia and Hall, or Leap, or any of the other authors I have read, is everything that arguably makes sexuality sexuality -- namely, fantasy, desire, repressions, pleasures, fears, and the unconscious.
It is precisely these phenomena, I would like to argue, that the field of language and sexuality will need to confront if it is ever to constitute itself as a field of inquiry. For it is precisely these phenomena, however one ultimately wishes to explain them, that in many senses make up sexuality, compose it. Or do we think that we explain sexuality if we limit our analysis to overtly claimed identities and explicitly intended expressions? How, in that case, would we ever explain the sexuality of a boyfriend of a Brazilian travesti, who self-identifies as heterosexual, but who regularly has sex with his girlfriend, who has a penis? How would we explain the sexuality of a woman who can't get excited unless her girlfriend dresses up as a boy, takes out her strap-on, and addresses the woman as `faggot'? Or a heterosexual man who secretly imagines himself as a woman each time he has sex w ith one? Or who imagines his women as men? Or even one who imagines his women as "women"? If we limit ourselves only to an analysis of what these different individuals say and overtly intimate about their sexuality, we would surely be describing something. But is that something the sum total of "sexuality"? I submit that this is as misguided as thinking we have explained "language" if we limit ourselves to making claims about the grammatical competence of an ideal speaker/hearer.
The big problem here, of course, is that linguistics has no tools with which to approach the unconscious. Throughout the history of modern linguistics, the unconscious has played a significant role; indeed, I think it could easily be argued that the unconscious is the very resource of all linguistic analysis; even when it is dismissed, as when Bloomfield argued that linguists "must study the way people talk -- without bothering about the mental processes that we may conceive to underlie or accompany these habits" (1922), or even when it is denied, as in Conversation Analysis.
It could be thought, for example, that Conversation Analysis doesn't need a theory of the unconscious, since it can explain many features of linguistic interaction in structural interactional terms. However, a cornerstone of CA is the notion of preference, expressed, for example, in the crucial concepts of preferred and dispreferred second turns. In his exposition of CA, Levinson (1983:333) argues that "the notion of preference...is not intended as a psychological claim about the speaker's or hearer's desires, but as a label for a structural phenomenon very close to the linguistic concept of markedness, especially as used in morphology" Levinson then cites Bernard Comrie's definition of markedness, which reads as follows: "The intuition behind the notion of markedness in linguistics is that, where we have an opposition between two or more members..., it is often the case that one member is felt to be more usual, more normal, less specific than the other" (333; emphasis added). Levinson goes on to define preferred and dispreferred seconds in terms of "a rule for speech production that says "try to avoid the dispreferred action" (333; emphasis added). Note here that despite Levinson's assertion that no psychological claims are being made, in fact, psychological assumptions about avoidance and feelings underlie both his own, and Comrie's, claims about markedness. The unconscious is raised here, only to be denied. Or, perhaps, repressed.
It is this idea of repression that most starkly differentiates linguistic understandings of the unconscious from the idea of the unconscious as it is understood and elaborated in psychoanalysis. Even when the unconscious is explicitly accorded a central role in linguistic theory, as it is in work on conversational inference, or as I think one might agree it is in all versions of generative grammar, repression plays no role. Gumperz's or Chomsky's unconscious is not one in which desire or fantasy play any role (I'm referring, of course, to their theoretical unconscious, not their personal ones). This unconscious is seen entirely in terms of cognition, of knowing. It is more accurately thought of as a `non-conscious'. Even linguistic research that explicitly takes its cue from Freud (such as the work by Victoria Fromkin and others on parapraxes, or slips of the tongue [e.g. Fromkin 1973, 1980]) brackets out all concern with repression or desire to concentrate only on what language reveals about underlying grammatical knowledge.
What I am calling for is a study of language that works with a more supple and sophisticated notion of the unconscious. The obvious place to start would be for linguists to examine more closely the work of Jacques Lacan, who famously argued that the unconscious is structured like a language: like language, it has its own combinatory patterns, its own syntactical and semantic operations. This kind of understanding of the unconscious; one that essentially sees it as discourse, and, not only that, but as what Lacan called "the discourse of the Other" (1977a:172), is clearly amenable to linguistic theorizing.2
The problem for anyone working with Lacan is his specific explanations of the topology of the unconscious. Lacan is, after all, a psychoanalyst, one who consistently finds his inspiration in what he calls a "return to Freud" (1977b). And even though Lacan dramatically modifies Freud, sometimes beyond recognition, he still maintains an absolute investment in an understanding of sexuality that sees it in relation to castration (which Lacan theorizes as `lack'), and the male genital organ (which Lacan draws on Saussure to theorize as `the signifier of signifiers': `the phallus'). There is an enormous secondary industry around the writings of Lacan; and there one can read often quite esoteric debates about the nature of jouissance, or whether the phallus is necessarily the penis, or whether lesbians, for example, can have it too.
Now there are a great many problems with all this, not least of which is that the debates about Lacan are usually carried out without regard for one shred of empirical evidence -- instead, discussions about Lacan tend to be quibbles about the specifics of his terminology (what is the precise relationship between the Symbolic and the Real?), or rarefied philosophical debates about the dialectics of the symptom. Furthermore, as Eve Sedgwick points out, despite psychoanalytic theory's "almost astrologically lush plurality of its overlapping taxonomies of physical zones, developmental stages, representational mechanisms and levels of consciousness" what always happens, in the end, is that all the complexity becomes "sleeked down" to a singled-minded concern with "the mother, the father, the preoedipal, the oedipal, the other" (1990:23-24).
For all these reasons, I am not suggesting that linguists have an interest in trying to refashion themselves into psychoanalysts. However, what I am suggesting is that the questions raised by psychoanalysts are highly provocative and in many cases inspirational. And more than that, they are questions that are necessary to ponder if we think we might have anything of interest to say about the relationship between language and sexuality. We don't have to swallow Lacan's phallus, as it were, but we should respect the fact that he raises it, and we should at least try to understand the role it plays in his theorization of sexuality. We should look back again at the questions that he and Freud asked. But we don't have to be particularly respectful of their answers. We should acquaint ourselves with their work at least enough to be familiar with their theories of the unconscious, and to attempt to identify what we object to, and what we would like to see rethought in different terms.
In doing this, phrasing our inquiry in terms of sexuality might be counterproductive. In fact, although the topic of this essay is language and sexuality, the suggestion with which I shall finish is that we dump sexuality altogether and reformulate our concern in other terms -- for example, in terms of language and desire. To speak of desire rather than sexuality would break open our analysis and free it up to ask questions and do work that is not being done by the work that takes sexuality as its starting point.
Reformulating questions of `language and sexuality' in terms of `language and desire' would compel us to do several things. First of all, it would compel us to decisively shift the ground of our inquiry from identity categories to culturally grounded semiotic practices. The desire for recognition, for intimacy, for erotic fulfillment -- none of this, in itself, is specific to any particular kind of person. What is specific to different kinds of people are the precise things they desire and the manner in which particular desires are signalled in culturally codified ways. For example, sexual desire of a man for a woman is conveyed through a range of semiotic codes that may or may not always be conscious, but that are recognizable as conveying desire because they are iterable signs that continually get recirculated in social life. The iterability of the codes is what allows us to recognize desire as desire. This means that all the codes are resources available for anyone -- be they straight, gay, bisexual, shoe fetishists or anything else -- to use. This also means that desire cannot best be thought of in terms of individual intentionality. Because it relies on structures of iterability for its expression, desire is available for appropriation and forgery; as we know from cases where men invoke the desire of the Other to claim -- ingenuously or not -- that they thought that a woman whom they raped desired them; or that they thought that a man whom they killed was coming on to them. Researchers interested in language and desire need to be able to explain this too -- they need to explain not only intentional desire, but also forged desire.
Second, a focus on desire rather than sexuality would compel us to define desire -- that is, it would compel us to be explicit about the theory or theories of desire that lie at the base of our investigations. What desire are we talking about? Whose desire? Directed at what? Why? All of those questions would have to be answered before we could proceed with any research. And to answer them, we would have to be precise. Do we mean desire in Lacan's terms as that which is lacking -- as that which must always exceed our grasp; as the Sisyphusian attempt at retrieval of the pre-Oedipal plenitude that can never be regained? Or do we mean desire in Deleuze's terms, as an endlessly proliferating force that continually produces new objects with which to satisfy itself? Do we mean Foucauldian desire, which highlights the relation between historically generated constellations of power and knowledge, and historically generated practices through which individuals come to know themselves as subjects? Or do we mean something else altogether?
Finally, a focus on desire rather than sexuality would allow analysis expanded scope to explore the role that fantasy, repression and unconscious motivations play a role in linguistic interactions. Here is where we might want to return to Lacan's development of Freud's concepts of `condensation' and `displacement' in terms of metaphor and metonymy. In his essay on `The agency of the letter in the unconscious' (Lacan 1977a), Lacan reinterprets `condensation' and `displacement' in linguistic terms. He inverts Saussure's diagram of the sign, placing signifier above the signified, thereby giving primacy to the sound pattern in the genesis of the concept. Lacan also asserts that the signified is simply another signifier; one that occupies another position, one `below the bar' within signification. The ultimate point of this exercise is to argue that the unconscious consists of signifiers which have fallen below the bar. That is, they have undergone repression and are prevented from traversing the bar and gaining access to consciousness (Grosz 1990:96). In Lacan's model, this repression is thought of in terms of metaphor, which is a relationship in which one term takes the place of -- `occults' or submerges -- another. Metonymy, on the other hand, is a relation of horizontal movement in which one term substitutes for another (as when `crown' substitutes for `king'). In this movement of substitution, Lacan sees the movement of desire, which "is based on a chain of substitution whereby the first (lost) object of desire generates a potentially infinite chain of (only partially satisfactory) substitutes" (Grosz 1990:100). The relevance of all this for us is the space for intervention that Lacan offers linguists when he uses metalinguistic terms to analyze the structure and workings of the unconscious. Lacan continually stresses that there is a natural affinity between linguistics and psychoanalysis. And there is: it is not mere coincidence that Ferdinand de Saussure's son Raymond became a psychoanalyst.
In conclusion, I hope I have made it clear that I think that the study of language and sexuality -- or, better, the study of language and desire - poses quite a radical challenge to linguistic understandings of language.3 A focus on desire forces us to face up to the normally overlooked fact that linguistics has no real understanding of the unconscious, and no real means with which to analyze how something as fundamental as desire is conveyed or what role language plays in the constitution of something as human as sexuality. Perhaps it is because of the radical implications of asking these kinds of questions, that the field of language and sexuality has still not emerged, despite over two decades of research on language and gender. As Gayle Rubin argued and as queer theory maintains, gender and sexuality, while they are inextricably implicated in one another, they are not exactly the same thing. Therefore, an analytic focus on the one rather than the other will yield particular research questions, particular data, and particular insights. Even though it has suffered from serious theoretical shortcomings, research on language and gender has consistently enriched our understanding of the way in which language constitutes individuals as particular kinds of people. Newer work in language and gender that explores the linguistic dimension of gender performativity (e.g. the articles in Hall & Bucholtz 1995) promises to push our understandings of these issues firmly and productively into the next millennium. So with this underway, perhaps the time has come now to seriously turn our attention to language and sexuality; not only to find out more about sex, but also to find out what sex can tell us about linguistics. By turning our attention to language and desire, my hope is that we will not only articulate a linguistics of sex. We might also develop a truly sexy linguistics.