Feminist film theory attempted to exit from the inversion model in the context of queer theory and yet to retain a (psychoanalytic) account of gender processes. Psychoanalytic models narrate femininity as an insubstantial category produced in language as the negation of the category of masculinity. Desire (lack of satisfaction) is signified by the phallus which positions 'the woman' as its object. A non-phallic model of (lesbian) desire is thus rendered inarticulable. De Lauretis' (1988) noted this dependence placed on 'man' as the dominant term in sexual difference (gender) and argued that a lesbian-feminist politics based on this concept fails to escape from a heterosexualising (inversionary) frame. The structural paradox of identity politics could be re-stated in slightly different terms:
It thus appears that 'sexual difference' is the term of a conceptual paradox corresponding to what is in effect a real contradiction in women's lives: the term, at once, of a sexual difference (women are, or want, something different from men) and of a sexual indifference (women are, or want, the same as men). And it seems to me that the racist and class-based practices legitimated in the notion of 'separate but equal' reveal a very similar paradox in the liberal ideology of pluralism where social difference is also, at the same time, social indifference. (De Lauretis, 1988a: 142)
De Lauretis effected a realignment of lesbianism by drawing a distinction between 'hom(m)o-sexuality' (or 'sexual indifference') and 'homosexuality' (signifying a lesbian sexual difference from heterosexuality). In this way, lesbian desire can be refigured not as an appropriation of 'the phallus' but as a refusal of (hetero)sexuality by which gender is articulated as a (differentiating) relation to the (same) phallus.
Case hybridised a psychoanalytic model of 'desire' with a gay model of 'camp' in order to 'denaturalise' the inversionary model, resituating the butch-femme relationship in the performative context of lesbian theatre. A lesbian performance utilises deconstructive readings of popular culture to underscore the fictional nature of gender:
The identification with movie idols is part of the camp assimilation of dominant culture. It serves multiple purposes: (1) they do not identify these butch-femme roles with 'real' people, or literal images of gender, but with fictionalized ones, thus underscoring the masquerade; (2) the history of their desire, or their search for a sexual partner becomes a series of masks, or identities that stand for sexual attraction in the culture, thus distancing them from the 'play' of seduction as it is outlined by social mores; (3) the association with movies makes narrative fiction part of the strategy as well as characters [...] they are fictional at their core in the camp style and through the butch-femme roles. (Case, 1988: 302)
This lesbian subject comes into being as she foregrounds the fictional nature of the heterosexual role-models which she incorporates to a performance which is re-referred to a lesbian context. Thus, whether 'the lesbian' speaks as butch or femme, she speaks from an undifferentiated subject position — that of 'the lesbian.' In this way, Case articulated a modality of lesbian desire without fixed poles of gendered identification. However, besides considerations of the multiple discourses potentially at work in the production of lesbian subjectivity, Case effectively relied on a presumed universality of the butch/femme diad to lesbian cultures.
[B]utch/femme may be a widely recognized set of conventions within lesbian culture, but it must not be assumed to be a constitutive or obligatory identification. (Lamos, 1994: 97)
Studlar (1991) used an updated Freudian conceptualisation of the masochistic scenario to explore lesbian pleasure in the popular cinematic text but, again, in a desubjectivising mode. Rather than directly addressing an elusive location for the lesbian in the visual relay, she argued that ultra-feminine codings or transvestite dress in Dietrich's films does not function fetishistically to cover for the missing phallus for the male spectator as suggested by Mulvey (1975). Instead, Dietrich represents a pre-Oedipal fantasy figure, signifying a disavowal of the father's phallic power and its redistribution onto the authoritative pre-Oedipal (phallic) mother, in a playful, multiple, sensuality. Dietrich's cross-dressing thus undermines the naturalisation of sexual difference and of heterosexual desire.
Studlar went on to suggest that the relay of the gaze is changed by the frequent absence of the diegetic male protagonist in Dietrich's films. According to Mulvey (1975), in the absence of a diegetic protagonist, the male spectator is brought into direct erotic rapport with the fetishised spectacle of the female star. Studlar pointed out that the lesbian spectator might also experience a direct erotic rapport with the image. In contrast to the separation of the voyeuristic and objectifying moment of the look from the narcissistic and identificatory, there is a 'subversive' female-to-female looking in which identification is disseminated between desire for, and identification with, the authoritative and eroticised image of the woman. In Freudian terms, such pleasure is 'perverse' (non-object-oriented) rather than expressing an inverted object choice.
De Lauretis (1991) developed her (1988a) position in a reading of the lesbian-made film She Must Be Seeing Things (1987) in which neither diegetic character emerges as singular protagonist, and each represents 'the lesbian.' The function of voyeurism is re-articulated through an unstable shifting of subject-object configurations between lesbianised intra-diegetic and extra-diegetic spaces. These shifts take place in the diegetic relationship between the lesbian characters, in which the gaze of the lesbian spectator is also implicated. The diegetic femme is identified with the filmmaker (and thus with the controlling phallic gaze of the camera and the extra-diegetic investigator-narrator). Diegetically, the butch occupies the position of protagonistic investigator, with whom the enunciative position appears complicit. The subjectivity of the spectator is thus distributed across contradictory diegetic subject-object positions, and intra- and extra-diegetic enunciative positions (the investigator investigates the investigator). Thus both can be seen as occupying the subject position of 'the lesbian' together in a way which is ultimately gender-undifferentiated, but different from heterosexual desire.
This dissemination also forms a connection between the lesbian as spectator and the diegetic scenario of lesbian fantasy. This process itself is re-contextualised and rendered self-conscious in the film-within-a-film (the 'Catalina' story). In its most basic (primal) form, the subject of a fantasy is present in the syntactical organisation of the scenario rather than 'present' in the sense of occupying a specified position in the fantasy narrative. She Must be Seeing Things represents a lesbian fantasy scenario in this film-within-a-film. The autoerotic fantasy (the 'Catalina' story) is framed by the symbolic (representation of the process of cinematic articulation of the fantasy). Thus, She Must Be Seeing Things uncovers the structure of lesbian spectatorial identification and involvement in the film fantasy, mediated by the involvement of the two lesbian subjects, Agatha and Jo. The denaturalising effect of the use of masquerade in She Must be Seeing Things assists the self-reflexivity of lesbian spectatorship:
It is in that space between the fantasy scenario and the self-critical, ironic lesbian gaze — a space the film constructs evidently and purposefully — that I am addressed as spectator and that a subject-position is figured out and made available in terms of a sexual difference that is not a difference between woman and man, between female and male sexuality, but a difference between heterosexual and lesbian sexuality [...] (De Lauretis, 1991: 251)
De Lauretis [1988a] had objected to an extra-textual lesbian spectator being brought into the fray as putative controller or guarantor of lesbian-specific readings of popular culture. This lesbian spectator herself would inescapably remain the subject of identificatory fixity in the relay as an identificatory subject of 'sexual indifference.' The involvement of the lesbian spectator in this scenario of desire, therefore, implicitly depends on regarding lesbian strategy for reading film along the same lines as she had proposed elsewhere (1988b) for women's film. That is, on a mobilisation of oppositional (lesbian) readings in collusion with a feminist (lesbian) filmmaking practice. The conventional relay of popular cinema's IMR would render such a 'pact' between feminist practitioner, reader, and text, impossible, or at least incomplete.
Traub (1991) pointed out the diversity of lesbianisms "in relation to the shifting fortunes of gender ideologies and conflicts, erotic techniques and disciplines, movement politics, fashion and consumer trends, media representations, and paradigms of mental illnesses [...] to name just a very few" (305), but yet immediately reconstituted this diversity within the ubiquitous terms of the psychoanalytic paradigm.
Yet, despite the subsequent deligitimization of the psychoanalytic paradigm by the feminist and gay liberation movements [...] [I]n novels and films, mainstream and alternative - and more importantly in our reactions to them — 'lesbian' continues to be thought through and within a psychoanalytic nexus of signification [...] [and lesbians are] unwitting reproducers of it. (Traub, 1991: 306 – emphasis hers)
Although, for Traub, the extra-textual (historicised) lesbian spectator might effect "a refusal of those gender dichotomies that organise erotic desire" (Traub, 1991: 316) 'the lesbian' nevertheless continues to signify homogeneously by/as "desire" and not as the unstable and differentiated product of discursive contestations.
These revisions of the psychoanalytic model of spectatorship were intended to open out and supersede the difficulties of locating a lesbian subject position in the heterosexualising relay. Furthermore, lesbian practices could be figured as different from dominant practices (as actively 'subversive' in their denaturalising effects) rather than as being entirely overdetermined by them as in Mulvey's (1975) model. On the other hand, whilst lesbian diversity has been acknowledged, the issues have been evaded rather than addressed. If desire is refigured as multiplicitous, it nevertheless remains the universal referent of its diverse modalities. Indeed, this model remains dichotomous rather than diverse since it refigures lesbianism as that which is (homogeneously) different from heterosexuality.
In effect, Traub (1991: 305) rapidly reabsorbed an acknowledged lesbian diversity to the universalising construct of 'desire.' De Lauretis' (1993: 152) argument that lesbianism cannot be treated as a universal category breaks off at the point of observing that, nevertheless, "the discourses, demands, and counter-demands that inform lesbian representation are still unwittingly caught in the paradox of socio-sexual (in)difference . . ." (1993: 155). It would seem that queer theory's mobilisation of 'sexuality' as key referent of its resistance needs to be decentred and reframed in all the specificities of its agendas and articulations. If, indeed, it is the centrality of techniques of 'sexuality' in the production and control of everyday life in the 20th century which has produced and prioritised 'perverse' strategies of resistance, then a discourse of 'sexuality' is always-already assimilated. Grosz thus cited Foucault's (1976: 157) recommendation that "[t]he rallying point for the counterattack against the deployment of sexuality ought not to be sex-desire, but bodies and pleasures" (Grosz: 155). A queer project of aligning lesbianism with a contradictory or 'perverse' content of desire (whether linked to identity or not) can only be seen, in feminist as well as Foucauldian terms, as a depoliticising move which reinscribes lesbianism within the controlling discourse of 'sexuality' where it can be 'disciplined,' commodified, and re-circulated in media not as politicised resistance but as an alternative content of desire. In order not only to resist the disciplines of 'sexuality' and the evacuation of gender from its discourse, but also to take account of lesbian diversity rather than (a singular) lesbian difference from heterosexuality, discursive constructs of 'the lesbian' need to be adequately historicised.