Two recent books, Language and sexuality, by Deborah Cameron and Don Kulick, and Talking gender and sexuality (TGS), edited by Paul McIlvenny, seek to elucidate the social construction of gender and sexuality by combining the methods of interactive discourse analysis with the theoretical insights of poststructuralism, feminism, and queer theory. Guided by the basic ethnomethodological question “How are gender and sexuality ‘done’?”, the authors of both volumes are motivated, at least implicitly, by a political commitment to deconstructing and opposing sexism and heteronormativity (the ideologically enforced assumption that all people are or should be straight). In one way or another, therefore, both books also ask, paraphrasing McIlvenny, “By what linguistic-interactional means might normative gender and sexual identities be ‘undone’?” Although McIlvenny's coauthors all share his theoretical concerns at a general level, the chapters in TGS focus primarily on the means whereby gender and sexual categories are positively instantiated and indexed in talk; only a few explicitly engage how these modes of categorization might be resisted or transformed. Cameron & Kulick take their theorizing one step further, questioning the analytical and political utility of the concept of sexual identity itself.

Drawing heavily on Judith Butler's theory of gender performativity, both McIlvenny and Cameron & Kulick approach gender and sexuality as distinct but related modes of organizing social inequality. In their third chapter Cameron & Kulick argue that, if linguists want to understand the social conditions that oppress lesbians, gay men, and other sexual minorities, they need to pay critical attention not only to the language practices of those groups but also to linguistic constructions of heterosexuality. The point of such research is not to give heterosexuals equal time, but to elucidate how heterosexuality is linguistically constructed and oriented to as a social norm for queer and straight people alike. Because heterosexuality is rarely topicalized as such, its situational relevance can be difficult to detect. Nevertheless, employing linguistic-anthropological notions of indexicality, Cameron & Kulick re-present four case studies from the literature on language and gender, all of which highlight the formative role of heteronormativity in the linguistic construction of conventional gender identities. These studies also show how gender and sexuality are performatively coarticulated with other identities and activities, such as age, class, gossip, and leisure. This chapter is by far the book's strongest and will be extremely useful in university courses on language and gender, where heterosexuality is often insufficiently problematized.

Cameron & Kulick's primary goal, in terms of the amount of space they devote to it, is to criticize other linguists' emphasis on sexual identity (chaps. 1 and 4) and to propose an alternative theoretical framework focusing on “language and desire” (chap. 5). Chap. 4 reviews the history of social-scientific research on language and homosexuality and reiterates Kulick's previously published criticisms of contemporary scholarship on the language practices of gay men, lesbians, and other sexual minorities. In contrast to chap. 3, which cites the work of other scholars in order to build a positive theoretical argument, the point and tone of this chapter are overwhelmingly negative. With few exceptions, other linguists' research is treated as fundamentally flawed because it focuses too narrowly on “identity” and thus “leaves unexamined everything that arguably makes sexuality sexuality: namely, fantasy, repression, pleasure, fear and the unconscious” (p. 104). By emphasizing the shortcomings of even the “better” scholarship in so-called lavender and queer linguistics, Cameron & Kulick clearly seek to depict their own theoretical framework as a radical and necessary departure from that body of work.

Yet Cameron & Kulick's proposals display some of the same logical weaknesses that they attribute to their colleagues, and the distinctiveness of their framework is more apparent than real. Thus, while it is true that some scholars construe “sexual identity” tautologically, as both cause and effect of linguistic practice, Cameron & Kulick's definition of “sexuality” is no less tautological, consisting of “ways of being sexual” and “not only whom one desires [sexually] but also what one desires to do” (10). In addition, whereas Cameron & Kulick criticize other scholars for reducing “sexuality” to “sexual identity,” their own treatment of “desire” effectively reduces it to “sexual desire.” The latter is especially surprising given their favorable citation of Gilles Deleuze and Félix Guattari, who insist that “desire” is manifest in such ostensibly nonsexual events as walking, sleeping, winter, and spring. Finally, although Cameron & Kulick's elaborate engagement with psychoanalytic theory distinguishes them rhetorically from the majority of their linguistic colleagues, their theoretical proposals are substantively similar. In particular, after repeatedly calling for a theory of language and sexuality that goes “beyond” identity or leaves it “behind,” they proceed to advocate a theory of “identification” that focuses on the emergent, processual nature of (sexual) identity formation. The continuity of this approach with that of other language and sexuality scholars is obvious, yet Cameron & Kulick's acknowledgment of that fact is sporadic at best.

A more fruitful – and collegial – approach would acknowledge the shared psychoanalytic heritage of both “identity” and “desire,” and would review the extant literature on language, gender, and sexuality in order to problematize and operationalize both concepts for empirical discourse-analytic research. It would also consider the extent to which an emphasis on “identity” continues to pervade not just the study of language and sexuality, but sociolinguistics and anthropology at large, along with the historical and cultural-ideological reasons for that emphasis. Cameron & Kulick allude to such an approach in their second and final chapters, where they endorse ethnographic approaches to language and sexuality and briefly discuss, inter alia, the contested emergence of the concept of “sexual addiction,” the coarticulation of sexuality and race, and the linguistic dimensions of political struggles over rape and AIDS.

Among Cameron & Kulick's most intriguing theoretical proposals are their call for discourse analysts to attend to what is “unsaid” as well as said, and their passing references to the ways sexuality is “materialized” in discourse. The former is explored with reference to the discursive psychologist Michael Billig's treatment of “repression,” especially his point that prohibitions against certain statements, images, or actions are likely to incite transgressive desires for those very experiences. Cameron & Kulick consider several examples of this process, most of which focus on interactions that transgress conventional boundaries between private and public domains, such as personal ads, sadomasochistic role-play, and pornography. Many of these examples are taken from the mass media, and their contents are often sensational. Although Cameron & Kulick acknowledge that these are ripe for analysis with respect to the mutual imbrication of sexuality with political economy, ideology and the state, their analyses focus primarily on the ways these interactions construct pleasure, fear, and other emotions in “intimate” settings. In so doing, Cameron & Kulick pass up several opportunities to theorize the “materialization” of sexuality in a way that takes the materiality of discourse into account. (Another unexplored proposition is their call for distinguishing “performance” from “performativity,” which Cameron & Kulick mention twice as if it were a main theme of the book, and without citing its venerable linguistic-anthropological precedents, e.g., the collaborative work of Richard Bauman and Charles Briggs.)

Investigating the relationship between said and unsaid is a project that unites many of the chapters in TGS. Billig's public debate with the conversation analyst Emmanuel A. Schegloff over the politics of defining “text” versus “context” figures prominently in this regard, especially as it relates to the problem of observing social inequality in talk. As noted by Cameron & Kulick, because sexism, heterosexism, and other forms of inequality are rarely oriented to as such, they are opaque – and therefore seemingly irrelevant – to the methods of sequential conversation analysis (CA). Accordingly, several chapters also make use of membership category analysis (MCA), a framework proposed (but later abandoned) by Harvey Sacks, in order to ascertain the normative “cultural knowledge” that is implicitly indexed by participants' use of social-identity categories in talk. For example, in analyzing a group discussion involving social-science researchers and young British adults, Elizabeth H. Stokoe & Janet Smithson consider the “heterosexist” implications of one participant's statement about generic households where “the woman's staying at home with the kids” (98), the normative power of which is reinforced by the other participants' unmarked response to it. Liisa Tainio analyzes how, in the course of narrating their youthful courtship, two elderly Finnish interviewees (“the husband” and “the wife”) use various grammatical means to portray themselves and each other as sexually agentive or passive. Tainio says these utterances “go against” certain Western and/or Finnish stereotypes about the sexuality of men and (old) women. Sigurd D'hondt analyzes the comments made (in Kiswahili) by three male adolescents in Dar es Salaam, Tanzania, with regard to a teenage girl, Julie, who has just passed by. Although much of the extract consists of the boys' speculation over whether Julie has recently had an abortion, even more intriguing is the way the boys' own (hetero)sexuality figures in their moral negotiation over the normative category of “Muslim.”

Although the insights afforded by MCA in these chapters seem sound, none of the authors satisfactorily addresses the reasons that led Sacks to abandon MCA and that continue to make conversation analysts suspicious of it. Particularly suspect is the assumption by Stokoe & Smithson and Tainio that members' “cultural knowledge” can be accessed straightforwardly, insofar as they include analysts (i.e., themselves) in the category of “member.” They thus overlook the fact that claims to membership are often contested, and that members' cultural knowledge is likely to vary, along with their ideological commitments. Some engagement with ethnographic method would help rectify this problem. Other analytical choices seem inconsistent with the authors' commitment to an empirically rigorous MCA. For example, Tainio's preoccupation with age and heterosexuality leads her to ignore the male interviewee's obvious orientation to class, as indexed by his self-inclusion in the membership categories of “farm help” and “workers.” D'hondt likewise overlooks the cultural implications of giving Julie a (non-Muslim) pseudonym, while identifying the male participants solely as E, F, and N.

Andrew Fish uses MCA to reanalyze the telling of a dirty joke by three teenage American boys, originally analyzed by Sacks. In Sacks's analysis, this joke, which was previously told to one of the boys by his younger sister, ironically comments on the cultural knowledge that 12-year-old girls are and are not supposed to have about sex. However, Fish's analysis of the joke's retelling suggests that the boys' references to the absent sister's presumed sexual knowledge index their own “fragile” identities as “masculine” and “grown up.” Fish describes this as “the repressed on parole,” a pun that brilliantly highlights the fact that what is unsaid (repressed) is performatively indexed and incited by what is said (cf. Saussure's parole). As a sophisticated example of the kind of analysis advocated by Cameron & Kulick, Fish's approach is notable for the way it attends to the mutual constitution of desire and identity.

Drawing heavily on Billig's work, Fish situates his use of MCA within the larger analytical framework of discursive psychology (DP). Other authors also employ DP, but they align themselves more closely with Schegloff and sequential CA, and they make the utility of their methods their primary topic. For example, Susan A. Speer & Jonathan Potter seek to demonstrate the benefits of an “anticognitivist” version of DP – consisting essentially of CA plus rhetorical analysis – for illustrating the practical implications of Judith Butler's abstract theories of performativity, heteronormativity, and hate speech. Their analysis of how heterosexism is “managed” in three different texts – a television talk show featuring an antigay political activist; a social-scientific interview with a “liberal” straight man; and a television documentary about the Gay Games – yields a number of suggestive insights about the sexual politics of contemporary British society. Alexa Hepburn explores the compatibility of a similar version of DP with Jacques Derrida's methods of textual deconstruction. In her analysis of interviews with Scottish teachers, Hepburn notes that the teachers' “gendered accounts” of school bullying reproduce a “binary logic” whereby girls and boys are presented as having essentially different personalities. However, the rhetorical variability of the teachers' accounts also reveals a Derridean “logic of supplementarity” and reliance on metaphor that destabilize that binary. These chapters would benefit from some consideration of the practical relevance of the authors' findings to their progressive political goals. For Hepburn, this would entail a discussion of the literature on bullying and gender inequality in schools. In the case of Speer & Potter, their analysis of a gay male sports announcer's “resistance” to heterosexism needs to be considered in light of the actual or potential effects of his remarks, at least for the audiences who witnessed them, if not for (British) society at large.

Celia Kitzinger's proposal for a “feminist conversation analysis” addresses the political challenges of CA head on. The empirical focus of her investigation is university seminars in which some participants “came out” as nonheterosexual “without anyone noticing.” Kitzinger's approach to coming out and (not) noticing as interactional accomplishments is classic CA, but when it comes to analyzing the political implications of these actions, she employs other research methods, such as interviews, quasi-statistical comparisons across interactional events, and her own member's knowledge as a lesbian living in a heterosexist world. Although these methods are part of the standard toolbox of sociolinguistics and linguistic anthropology, Kitzinger characterizes them as “feminist CA” in honor of Schegloff, whom she identifies as her teacher, and as a rebuttal of Billig, whom she criticizes for ignoring work (including her own) that combines CA methods with feminist principles. Disciplinary labels aside, one aspect of member's knowledge that Kitzinger would do well to consider is the institutional structure of university seminars, where professors, teaching assistants, and students have unequal access to such conversational resources as topic-choice, turn-allocation, and the freedom to comment on (i.e., “notice”) other speakers' contributions.

Jenny Sundén uses a similar combination of methods to analyze conversational texts from an online discussion forum where a command called “@gender” allows participants to assume a variety of gender identities besides the conventional female and male. This chapter is unique among the papers in TGS in a number of ways. Methodologically, Sundén employs ethnographic participant observation and interviews systematically, and she presents the insights gained from these methods on a relatively equal footing with her discourse analysis. (The textual nature of her fieldsite surely facilitated this decision.) In addition, although Sundén cites a range of research on language, gender, and sexuality, she does not situate her work relative to CA, DP, or ethnomethodology per se. She is thus able to concentrate on the theoretical questions with which McIlvenny frames the volume, pertaining to the “doing” and “undoing” of gender and sexuality. She also adduces conversational evidence for a notion of “identity” that both incorporates and informs “desire.” Her most important finding is that, far from facilitating a utopian transcendence of conventional gender categories, nonnormative uses of the “@gender” command actually seem to have incited participants to pay more explicit attention to one another's real gender/sex identities, not less. Accordingly, while Sundén accepts the poststructuralist argument that gender and sex are discursively constructed, her data also compel her to recognize that discourse itself is both materially and ideologically constrained.