Developmental Psychology 1995, Vol. 31, No. 1,95-104 Copyright 1995 by the American Psychological Association, Inc. Transitions From Heterosexuality to Lesbianism: The Discursive Production of Lesbian Identities Celia Kitzinger Loughborough University Sue Wilkinson University of Hull This article explored the discursive production of a major disjuncture in sexual identity in adult life: women's accounts of transitions to lesbianism after a substantial period of heterosexuality. Eighty semistructured interviews with self-identified lesbians, all with at least 10 years prior heterosexual experience (plus additional materials drawn from published autobiographical sources), were ana- lyzed within a social constructionist framework. The article examined the creation of contexts in which sexual identity transitions become possible, explored how such transitions are defined and marked, identified the consequences, and detailed the continuing development of lesbian identity posttransition. In conclusion, the article reflected on the status and salience of such data in support- ing the social constructionist position, particularly in the face of the continuing popularity of essen- tialist theories of sexual identity development. After more than two decades of social constructionist ap- proaches to sexual identity (e.g., Gagnon & Simon, 1973; C. Kitzinger, 1987; Mclntosh, 1968/1992; Plummer, 1981, 1992; Weeks, 1977), biological models of lesbianism and male homo- sexuality are becoming increasingly popular, both in the scien- tific literature (e.g., Ellis & Ames, 1987; LeVay, 1991) and in the media. Although biological and early socialization models may present homosexuality as either a natural variation or an un- natural deviation from the "norm" of heterosexuality, "caused" variously by brain structure or function, genetic or hormonal influences, or early childhood experiences, they in- variably assume heterosexuality as a natural, unproblematic category (see, e.g., Money, 1988, p. 11). Such an analysis fails to recognize that the category homosexual can only exist in rela- tion to the category heterosexual (part of the social construc- tionist argument) and is in direct contradiction to those radical feminist analyses of heterosexuality that present it as neither natural nor normal, but rather as a coercive patriarchal institu- tion (e.g., Adams, Lenskyj, Masters, & Randall, 1990; Rich, 1987; Wilkinson* Kitzinger, 1993). There is now a large body of work predicated on the assump- tion that lesbianism and male homosexuality are essences— core, fundamental ways of being that are determined prenatally or in early childhood. Key researchers in thefieldconcur on this point: According to Bell, Weinberg, and Hammersmith (1981), adult homosexuality stems from homosexual feelings experi- Celia Kitzinger, Social Sciences Department, Loughborough Univer- sity, Loughborough, United Kingdom; Sue Wilkinson, Health Studies Research, Institute of Nursing Studies, University of Hull, Hull, United Kingdom. We thank Corinna Petre (Institute of Nursing Studies) for her sterling secretarial support. Correspondence concerning this article should be addressed to Sue Wilkinson, who is now at Social Sciences Department, Loughbor- ough University, Loughborough, Leicestershire LE11 3TU, United Kingdom. enced during childhood and adolescence; according to Money (1988, p. 124), "the most important formative years for homo- sexuality, bisexuality, and heterosexuality are those of late in- fancy and prepubertal childhood." Within this framework, identifying oneself as homosexual (which is referred to by the lesbian and gay community as coming out) is merely a process of learning to recognize and accept what one was all along: In- deed, the very expression coming out suggests that the lesbian has always been inside, awaiting debut (Hollander & Haber, 1992). This idea, that homosexuality and heterosexuality are fixed early on and persist relatively immutable to change, is reflected in the literature documenting the failure of conversion therapies (e.g., Halderman, 1991). It is frequently suggested that essen- tialism of such theories is a "straw man" erected by the social constructionists as a foil for their own perspective (cf. Stein, 1990, p. 326), but as has been argued elsewhere (C. Kitzinger, in press), such essentialist models of lesbianism and gay male identity development are, in fact, the norm, reflecting and per- petuating popular theories about homosexuality. Essentialist arguments of this type fail to address the experi- ence of many women. Recent research on heterosexuals who become lesbian (Golden, 1987) and on lesbians who become heterosexual (Bart, 1993) emphasizes the "fluctuating," "fluid," and "dynamic" nature of sexuality for "protean" women (pp. 246-247). Women's sexual fluidity has long been apparent in the psychological and sexological literature, but it is often sub- merged in the data rather than explicitly theorized. Drawing on data from Kinsey, Pomeroy, Martin, and Gebhard (1953), Mclntosh (1968/1992) pointed out that [I]t is interesting to notice that although at the age of 20 far more men than women have homosexual and bisexual patterns (27% as against 11 %), by the age of 35 thefiguresare both the same (13%). Women seem to broaden their sexual experience as they get older, whereas more men become narrower and more specialised, (p. 154) Most women who self-identify as lesbian do so only after an earlier period in their lives during which they identified as het- 95