Copyright Catherine Grant 2000
This final draft was published as ‘Secret Agents: Feminist Theories of Women’s Film Authorship’, Feminist Theory Vol. 2, No. 1, April 2001, pp. 113-130.

Secret Agents: Feminist Theories of Women’s Film Authorship

Catherine Grant

Authorial Directions
Virtually all feminist critics who argue in defence of female authorship as a useful and necessary  category assume the political necessity for doing so. (Mayne, 1990: 97).

It’s already clear that the old categories and ways of thinking will not work well enough for us. (Rich, 1998: 83)

Unlike many other words referring to the activities of particular kinds of cultural producers  (‘writer’, ‘painter’, ‘dramatist’), the term ‘author’ raises intrinsic questions about authority and about whether the individual is the source or the effect of that authority.  Despite the deconstruction of traditional understandings of the ‘author-as-subject’, the ‘author-as-source-of-meaning’, and of individualist ideologies in general, especially during the latter part of the twentieth century (Barthes, 1968; Foucault, 1969), these kinds of questions concerning authorial authority, as well as the institution of authorship, have remained fairly central ones for feminists in theorizing and teaching about women’s activities in the field of cultural production, because of their connections with broader feminist debates about different kinds of subjectivity and agency under patriarchy (Miller, 1986; Watts, 1992). In this paper, I will present an overview of feminist theoretical debate, from the early 1970s to the present, on the subject of women’s film authorship. Given that my tour will be, of necessity, highly selective, I have opted to concentrate here on feminist theorizations of women’s agency in film authorship. While in early contributions to feminist film theory, this concept was frequently implied but did not always dare to speak its name openly, for reasons I shall go on to explore, more recent theoretical studies almost invariably reveal explicit explorations of agency and agent-hood. I will attempt to analyse these developments primarily by revisiting key overviews of this field, ones which not only recapitulated on the issues around film authorship but also attempted to move the debate on in new ways, an objective I share.

Perhaps unsurprisingly, the benefits for feminist theory of asking authorial questions of women’s interventions into filmmaking have never seemed as self-evident as they have with literary authorship; nor have they proved quite as resistant to post-structural critique. By contrast with most literary and artistic endeavours, film production is, of course, usually understood to be collective, collaborative, even ‘industrial’, especially in its dominant commercial modes. By no means has it been taken for granted, then, that ‘authorship’ can or should be attributed to an analogous, solitary ‘artist-figure’ in the film production process (cf. Gaut, 1997). The routine ascription of ‘authoritative’ creative agency in filmmaking may actually vary between, or be shared among a number of potential ‘actors’ in the filmmaking process (for instance, the scriptwriter, the producer, the studio, or any star performers). Nonetheless, the idea or ‘function’ of the author (Foucault, 1969) has emerged and persisted as a discursive category in film culture largely in the person of the film director who, in conventional narrative cinema, normally ‘puts the script on film by co-ordinating the various aspects of the film medium’ (Bordwell and Thompson, 1993:13).

It is important to note that the birth of this idea of the director as film author, or auteur, has been traced back by most cultural historians to the late 1940s and early 1950s, and to the debates which took place in French, British and US film magazines about the relative artistic value of cinema, compared with the much longer-established arts. As John Caughie writes:

Within its distinguishable currents […] auteurism shares certain basic assumptions: notably, that a film, though produced collectively, is most likely to be valuable when it is essentially the product of its director […]; that in the presence of a director who is genuinely an artist (an auteur) a film is more than likely to be an   expression of his individual personality; and that this personality can be traced in a thematic and/or stylistic consistency over all (or almost all) of his films. (Caughie, 1981: 9)

This kind of voluntarist and Romantic understanding of the agency of film authorship as encapsulating the possibilities for expression of an (especially male) artist’s ‘personality’ was immediately co-opted by film commerce, for the purposes of which the name of the author came in the post-war period, outside and inside Hollywood, to ‘function as a “brand name”, a means of labelling and selling a film and of orienting expectations and channelling meaning and pleasure in the absence of generic boundaries and categories’ (Neale, 1981: 36).  Yet, while commercial and socio-historical aspects of the emergence of the author-function in film have usually been set aside by film theorists,  the formal or textual assumptions of early auteurism have continued to provide an important critical focus. From the 1950s onwards, academic and non-academic film studies often concentrated on expertly teasing out the putative traces of authorial subjectivity in film texts. In this way, an implied or imagined ‘textual’ author/director (Caughie, 1981, following Booth, 1961), gradually began to be foregrounded, often unconsciously or inadvertently, on the basis of ‘a textual indeterminacy which [took] shape in the reading [or critical] process’ (Stoddart, 1995: 47).

Although film critics have continued to use directors’ known biographies to produce authoritative interpretations or to detect consistent ‘signatures’ across a body of work, many post-1970s film theorists have been ‘at pains to distinguish cinema’s enunciating agency from the figure of the director or scriptwriter’ (Silverman, 1988: 11),  as they took up the challenges set by anti-humanist critiques of the concept of authorial intentionality (following Wimsatt and Beardsley, 1946). Structuralist film theorists ‘recast’ for their own purposes (Bordwell, 1985: 23) Benveniste’s (1971) linguistic theories of ‘enunciation’, thus evacuating cultural agency of individual human origins; it was the system which ‘spoke’, and not the author (Barthes, 1968; Metz, 1981). From the late 1970s onwards, post-structuralist film theory largely moved away from questions of directorial authorship to pay greater attention to other aspects of cinematic enunciation. In particular, it set about investigating ‘the way [the film text] says “you”’ (Casetti, 1998: 15), by focusing on the productivity of spectating, or film ‘reading’, an agency which provides the ‘one place where [textual] multiplicity is focused’ but, once again, usually to be examined ‘without history, biography, psychology’ (Barthes, 1968: 148).

The reason why I have felt it important to sketch out the development of mainstream academic conceptualizations of film authorship up to the 1980s is that these have been highly formative of key aspects of the feminist theoretical work which I shall now move on to examine in detail (for example, their routine conflation of, and sometimes confusion between ‘real’ and ‘implied’ directorial and spectatorial agencies in the processes of meaning production, as well as the preference for explorations of various kinds of authorial and spectatorial avatars in the film text). Until quite recently, as I shall attempt to show, feminists’ reluctance to move beyond the film text in their explorations of women’s authorial agency left many of them ill-equipped to answer convincingly at least one simple question: what exactly were the feminist objectives of studying women’s cinema within the conceptual frameworks they inherited?

Women’s Cinema
The image of women in the cinema has been an image created by men. The emergent women’s    cinema has begun the transformation of that image. These notes explore ideas and strategies     developed in women’s films. (Johnston, 1973)

As Claire Johnston’s words from the front cover of her 1973 edited pamphlet Notes on Women’s Cinema testify, feminist theorizations of film authorship did at least begin with radical political concerns about women’s limited presence in (or routine absence from) the male dominated cultural sphere. In the 1970s, Johnston’s own pioneering contribution to this newly emerging field, and especially her important, and highly influential essay ‘Women’s Cinema as Counter-Cinema’, included in the pamphlet (1973: 24-31), was prominent among early work born of a feminist activism, and of a perceived need to ‘advise’ a contemporaneous generation of feminist filmmakers on questions of praxis, or ‘ideas and strategies’.

What was new about Johnston’s work, other than its rejection of so-called ‘sociological’ models of feminist film criticism in favour of ones derived from structuralist film theory,  was its primary focus on Hollywood cinema, rather than on the avant-garde, or art-cinema practice of women filmmakers which had more usually garnered the attention of feminists interested in ‘counter-cinema’. In her study, Johnston valorises the example of the work of two women directors, in particular: Dorothy Arzner who made eighteen films in the Hollywood studio system of the late 1920s, 1930s and 1940s, and Ida Lupino, the British-born actress who produced and directed seven films (and numerous television dramas) in the USA in the late 1940s, 1950s and 1960s.  For Johnston, both women used conventional stories, genres and styles of the time to articulate critiques of prevailing sexist ideology: head on in the case of Arzner’s heroine Judy (played by Maureen O’Hara) who, in her 1940 film Dance, Girl, Dance, turns on her own vaudeville audience of ogling men, berates them for their voyeurism, and tells them how she sees them;  while more subtly in Lupino’s films which reveal ‘reverberations within the narrative’, generated by the ‘convergence of two irreconcilable strands’ – ‘Hollywood myths of women versus the female perspective’ (Johnston, 1973: 29-30). Johnston wanted future feminist filmmaking to learn from these examples of entertainment films, ‘in which the feminine “voice”, by formal means, breaks through (ruptures) the patriarchal discourse’, as Janet Bergstrom noted (1988: 81).

Descriptions of what a ‘feminine “voice”’ in cinema was, as well as how and from where it was articulated remained only very fuzzily traced in Johnston’s polemic. Nonetheless, her ‘rupture thesis’ (Bergstrom, 1988: 81) was enabled by another original aspect of her approach, one which lent her work at least the potential to deal close up with the specificity of the medium of film. Instead of dismissing the concept of auteurism, as other feminist theorists had, on the grounds of its sexist cult of the male personality, Johnston warmly embraced it (or at least a particular version of it ) for its interpretative potential:

Further elaborations of the auteur theory […] have stressed the use of the theory to delineate the unconscious structure of the film. As Peter Wollen says, ‘the structure is associated with a single director, an individual, not because he has played the role of artist, expressing himself or his vision in film, but it is through the force of his preoccupations that an unconscious, unintended meaning can be decoded in the film, usually to the surprise of the individual concerned’. In this way, Wollen disengages both from the notion of creativity which dominates the notion of ‘art’ and from the idea of intentionality. […] What Peter Wollen refers to as the ‘force of the author’s preoccupations’, […] is generated by the psychoanalytic history of the author. This organised network of obsessions is outside the scope of the author’s choice. (Johnston, 1973: 27, citing Wollen, 1972)

What Johnston cannot explain, as Helen Stoddart asks of Wollen’s original conceptualization of the director as ‘a neutral agent (rather than agency), through which wider social meanings are simply refracted’, is why should it be that the director ‘remains the chosen [unconscious] catalyst figure’, rather than the other component parts of any film’s production (Stoddart, 1995: 47)? Furthermore, Wollen’s phrase ‘the force of the author’s preoccupations’, to which Johnston returns on a number of occasions, is an interesting one for a feminist to deploy when it refers only to preoccupations generated by ‘psychoanalytic history’. How can feminist preoccupations (ones with, for example, creating a ‘counter-cinema’) be usefully conceptualized, say, as ones ‘outside the scope of the author’s choice’?

I would argue that some of the difficulties for the contemporary feminist reader of Johnston’s work result from her fascinating, if contradictory imbrication, at least in theorizing the processes of cinematic organization, of a largely pre-determined (in a Marxist sense) and unconscious (in a psychoanalytic sense) directorial version of ‘authorship’ with a voluntarist form of feminist spectatorship, able expertly to detect ‘organised networks of obsessions’.  While Johnston argues that ‘it is instructive to look at films made by women within the Hollywood system which attempted by formal means to bring about a dislocation between sexist ideology and the text of the film’ (Johnston, 1973: 29), it is difficult to see how, considering her arguments elsewhere, she can use the verb phrase ‘attempted… to bring about’, given the purposeful agency it should normally imply. If the ‘unconscious structures’ of a film text cannot usefully be shaped by their directorial authors, why aim to advise women filmmakers at all? On a number of levels, then, Johnston’s 1973 essay remained vague as to how and why women, given all these constraints, might produce feminist ‘counter-cinema’, or even just ‘different’ cinema. In the final lines of her essay, she notes that ‘Voluntarism and utopianism must be avoided if any revolutionary strategy is to emerge’ (1973: 31). But, even before she begins to move towards this conclusion, Johnston has reflected that ‘Polemics for women’s creativity are fine as long as we realise they are polemics […]’ (1973: 28): a partial and fleeting recognition, perhaps, that there are lots of cakes that feminist theorists cannot have and eat at the same time.

Enunciating ‘Woman’ and Women
Present categories of [female authorship in the cinema] are undoubtedly much more useful in analyzing the configurations of “woman” on screen than in coming to terms with the ways in which women directors inflect cinematic practice in new and challenging ways. (Mayne, 1990: 97, responding to Silverman, 1988)

Some fifteen years after Notes on Women’s Cinema appeared, Kaja Silverman published the chapter ‘The Female Authorial Voice’ in her 1988 book The Acoustic Mirror. In this chapter, Silverman assumes the political necessity of the category of female authorship, urging ‘that the gendered positions of libidinal desire within the text be read “in relation to the biological gender of the biographical author, since it is clearly not the same thing, socially or politically, for a woman to speak with a female voice as it is for a man to do so, and vice versa,”’ (Mayne, 1990: 97, citing Silverman, 1988: 217). Despite this declaration, however, from a contemporary perspective Silverman’s piece appears very much to be another feminist example in a trend, detected by John Caughie (1981: 242), towards the ‘dissolution [in film theory in the late 1970s and early 1980s] of questions of authorship into questions of enunciation and subjectivity’.

In her chapter, Silverman admirably surveys the fifteen years of feminist debate on film authorship since Johnston’s pioneering contribution, and in particular reads this debate through the highly complex post-structural and post-modern reconfigurations of authorship more generally, arguing that the author is ‘constructed in and through discourse’ (1988: 209). Following structuralist appropriations of Benveniste’s work on enunciation (Benveniste, 1971), Silverman recommends the close examination of how the ‘enunciator’ (here, the ‘discursively-constructed’ director) is inscribed in the film text. For her, this requires detailed analysis across a whole range of textual features, ranging from the film’s ‘thematic preoccupations, to the designation of a character or group of characters as a stand-in for the author, to the various enunciative strategies (sonoric as well as visual) whereby the film auteur’s presence is marked (whether explicitly or implicitly), to the “fantasmatic scene” [identified (with) as the “author’s desire” by the spectator] that structures a director’s work’ (Mayne, 1990: 97, citing Silverman, 1988: 212-217). Like Johnston, Silverman takes up Wollen’s conceptualization of the psychoanalytic ‘pattern of energy cathexis’ in authorship (Wollen, 1972: 170): this notion ‘forces a further reconceptualization of the author “outside” the text, and of his relationship to the author “inside” the text’ (Silverman, 1988: 197).

But eventually Silverman must ask what of the ‘textual status of a female author’ (my emphasis: 212)? Though it is difficult to see where the ‘difference’ might reside for her. Unlike Johnston, she concerns herself not with an overtly feminist or even ‘classically “feminine”’ female-directed cinema (ibid.), but instead with the example of Italian filmmaker Liliana Cavani, director of fourteen art-films from the 1960s onwards. She concludes of the work of this director that within the ‘libidinal economy’ (233) of her films, a key role is played ‘by the recurring figure of the marginal male subject, […] who functions as a kind of nodal point for the authorial dream, and who casts onto the director herself the image of what she would like to be’ (ibid.). This ‘pattern of energy cathexis’ establishes for Silverman the position which the viewer will come to occupy through identifying with this subject (ibid.). At one key point in her argument, she refers to Cavani’s ‘condensations in her extra-cinematic discourse’ (comments in interviews) through which she repeatedly draws attention to her ‘intense investment in the subject-positions occupied by certain of her male characters’ (220). She concludes her book by noting that her own fascination with Cavani’s marginal male characters suggests ‘that I in some way participate in the desire [that circulates in the films], and that my own authorial subjectivity reflects or replicates the one those texts project’ (233-234).

Agential connections between a text’s ‘psychic structures’ and the extra-textual ‘psychic structures’ of its ‘authors’ (for Silverman, as we have seen, this seems to include both director and spectator) are, of course, by definition arguable rather than straightforwardly demonstrable. While the argument for each set of structures may have some merits, as in this case, with reasonably compelling supporting data put forward by Silverman (internal and external textual evidence, and an interpretation of these based on Freudian, and especially Lacanian categories of analysis), I would argue that the correspondence posited between these internal and external ‘structures’ and the gender of the film director is based simply on vague assertion: a change in gender would seemingly produce little ‘difference’ (even within the terms of her own argument, her focus on masochism and the pre-Oedipal phase offers a model of cinema ‘producing meaning and pleasure without the mediations of castration, sexual difference and feminine lack’ [Kuhn, 1994: 195-196]).  Furthermore, while it would seem that Silverman’s model of critical spectatorship is considerably less voluntarist than Johnston’s, it is evident, nonetheless, that when she writes that ‘it is clearly not the same thing, socially or politically, for a woman to speak with a female voice as it is for a man to do so, and vice versa,’ (Silverman, 1988: 217), she is much less interested in any female authorial voice on the ‘outside’ than she is in the putative ones detected by the spectator inside the ‘libidinal economy’ of the film text. The relevance of many of these matters to the feminist project of theorizing women’s film authorship is, once more, difficult to deduce: it is not clear from her work to what extent ‘the fact of female authorship gives a particular or distinct inflection to the representation of female desire’ (Mayne, 1990: 100).

In her 1990 book The Woman at the Keyhole: Feminism and Women’s Cinema, based on a 1981 essay, Judith Mayne writes that, while ‘reading against the grain’ of traditional cinematic representations of women has been a central feminist strategy, even after many years of feminist study, surprisingly little comparable attention has been paid

to the function and position of the woman director. Central to a theorizing of female authorship in the cinema is an expanded definition of textuality attentive to the complex network of intersections, distances, and resistances of “woman” to “women”. The challenge of female authorship in the cinema for feminist theory is the demonstration of how the divisions, overlaps, and distances between “woman” and “women” connect with the contradictory status of cinema as the embodiment of omnipotent control and individual fantasy. (Mayne, 1990: 98)

Despite the first few words of this quotation, Mayne’s project here still sounds primarily like a theoretically-informed formalist one (and indeed her book contains excellent formal analysis). Nonetheless, due to the constraints on ‘broader’ kinds of discussion, of which Mayne is very aware in her book (given her disquisition on the subject of essentialism), in the continuing wake of (post-)structuralist inspired ‘suspicion of any kind of biographical information’ (Mayne, 1990: 104, citing Halprin, 1984: 32), it should not be surprising that, in her chapter entitled ‘Female Authorship Reconsidered’ Mayne prefers to examine the textual ‘signatures’, especially irony, of her favoured female director, Dorothy Arzner.  In choosing this methodology, she is self-consciously following in the footsteps of a generation of feminist literary critics such as Nancy Miller and Margaret Homans, for whom ‘female authorship is analyzed not in terms of simple categories of agency and authority, but rather in terms of complex textual and cultural processes which dramatize and foreground women’s relationships to language, plot, and the institutions of literature’ (Mayne, 1990: 91; Homans, 1986; Miller, 1988).

In spite of the rather too straightforward opposition between ‘simple’ agency  and ‘complex’ textuality which she sets up here, Mayne does have in mind a limited theorization of Arzner as an agent when she returns to Claire Johnston’s discussion of Dance, Girl, Dance. Whereas Johnston had privileged female to male looking relations in this film as evidence of resistance to patriarchal representational norms, isolating in particular the confrontational moment when Judy returns her male audience’s gaze, Mayne suggests that ‘female authorship acquires its most significant contours in Arzner’s work through relations between and among women’ (Mayne, 1990: 101). Her argument is based on the observation that, in Dance, Girl, Dance, and in Arzner’s films more generally, relationships between and among women ‘account for much more narrative and visual momentum than do the relations between men and women’: ‘Then one begins to wonder about the perspective that informs these preoccupations’ (Mayne, 1990: 104). In this regard, Mayne makes a limited, and largely homologous, appeal to the ‘extra-textual’, pointing out that Arzner’s evidently lesbian persona provides another ‘text’ which ‘mediates the relationship between the director, her films, and their reception’ (104). Such mediations are not explored further in this study, however. Once their potential relevance is asserted, with brief reference to what would now be called Arzner’s butch-lesbian ‘self-fashioning’ in a number of publicity photographs which Mayne reproduces, she returns to the text(s) to examine in more detail the abundant evidence of a lesbian authorial ‘signature’.

While her approach in The Woman at the Keyhole had a great deal to recommend it, from a number of feminist theoretical and political perspectives (not least its challenge to homophobia in feminist film theory itself; Mayne, 1990: 117), it still ran a number of methodological risks. While Johnston’s argument slipped between determinism and voluntarism, and Silverman’s fused ‘implied’ authorial and spectatorial subjectivities, bracketing the ‘historical author’ entirely, aside from relevant ‘condensations’ in interviews, Mayne’s methodology combined close textual analysis (deployed, on the whole, within a psychoanalytic frame of reference) with a largely asserted, and at times metaphorical, correspondence with hazily sketched, authorial ‘personae’. But homology and juxtaposition could only take her, and feminist film theory, so far with questions of women’s authorship. Despite her stated desire not to conflate the works of women filmmakers with the project of a transhistorical and transcultural feminist aesthetics of the cinema (1990: 7), Mayne would need the historical, and by extension the biographical, to play a greater part in her self-consciously non-universalizing desire for ‘“local” analysis of films by women’ (1990: 224-225), when she returned to study the work of Dorothy Arzner.

Theorizing Agent/Text Mediations
Authorship is the principle of specificity in the world of texts. So far from consolidating the notion of a universal or unitary subject, the retracing of the work to its author is a working back to historical, cultural and political embeddedness. (Burke, 1998: 202)

Have feminist film theorists been overly anxious about the wrong kind of ‘essentialism’ in all these years of considering women’s film authorship? As John Caughie pointed out some twenty years ago, ‘the danger […] in placing the author as a fictional figure inside the text, [is that] we remove the most accessible point at which the text is tied to its own social and historical outside: the danger, that is, of constructing the text as an ideal essence’ (Caughie, 1981: 3). And, as Séan Burke writes more recently, ‘the need to ground authorship should be felt most intensively within political forms of cultural criticism’, such as feminist ones (Burke, 1998, 202).

While some pre- and early 1990s feminist theorists,  like Mayne, occasionally attempted to redress the ‘disembodiment and disembeddedness’  of their inherited modus operandi, these kinds of ‘outside’ concerns usually proved difficult to incorporate in ‘theoretical’ work as it was then constituted. They often continued, therefore, to be bracketed in favour of various kinds of textual evidence, a move defended by the tacit understanding that this was how ‘Film Theory’ should be performed.  This does not mean that other kinds of theoretically informed work did not continue to be carried out on women directors, just that such work did not usually attempt to pass itself off as ‘Theory’. This is the case with a number of valuable monographs which have appeared during the 1990s on individual female directors. Impressive studies such as Annette Kuhn’s edited collection of work on Ida Lupino (1995), and Alison Smith’s Agnès Varda (1998) on France’s major woman director of the last forty years are clearly informed by post-structural and post-modern feminist theories, but combine their major emphasis on formal analysis with an important awareness that film directors, male or female, are but one significant element ‘in a complex process of film production and reception which includes socio-economic and political determinants, the work of a large and highly skilled team of artists and technicians, the mechanisms of production and distribution, and the complex and multiply determined responses of spectators’ (Holmes and Ingram, 1998: VII). While these studies, and others like them, do not retreat from examining questions of agency, either at the formal level of ‘textual energies’ and ‘preoccupations’, or at that of analyzing what can be known of the actions actually performed by their women directors, their conceptual frameworks for assessing anything other than the most manifest of these, or the more obvious forms of social and cultural constraints on them, are rather limited. In the final part of my discussion I would like to turn to two more ambitious pieces of feminist work which make an important contribution to these and other aspects of theorizing authorship.

Judith Mayne’s book Directed by Dorothy Arzner (1994) is a very comfortable continuation and expansion of her earlier work on that director, one which attempts a more complex theoretical synthesis. Perhaps in part because of the different publishing demands of a ‘larger audience’ for a film-director monograph, as she indicates herself (1994: 2-3), in this book, ‘biographical’ methodologies, within their obvious limits (‘Lives are never so accessible’; 1994: 3), can be more fully and, for the most part, unembarrassedly utilized. As Directed by Dorothy Arzner was not intended to be a ‘Theory’ book, so Mayne can revel in her role as the pre-eminent historian of Arzner’s career and films. While the book’s actual theoretical complexity quickly becomes apparent, nonetheless, it does not attempt what I would regard as a thoroughgoing conceptualization of Arzner’s agency. Mayne still holds back from drawing explicit cause and effect connections regarding certain text-author correspondences at which she nevertheless allows her narrative to hint. Despite this reticence, the representations made of Arzner’s life, as well as of her films, are subjected to an analysis performed through the optic of Judith Butler’s positing of gender not as ‘a stable identity or locus of agency from which various acts follow; rather [as] an identity tenuously constituted in time, instituted in an exterior space through a stylized repetition of acts’ (Butler, 1990: 140; cited by Mayne, 1994: 5). For Mayne, this formulation does help her to link more fully than in earlier work Arzner’s ‘performance as a lesbian’ with the representation, in her films and in her life, of masculinity and femininity as denaturalized performances. Importantly, Mayne’s self-declared ‘focus throughout […] is on the various texts and images which bear the imprint “directed by Dorothy Arzner”’ (Mayne, 1994: 6):

I never wanted to write a complete biography, definitive or otherwise. This book is intended, rather, as a study in portraiture, in the literal and figurative senses of the term. I am interested in what kinds of films Arzner made, in how those films and Arzner herself were written about, and in how Arzner was portrayed during her career. (Mayne, 1994: 6-7)

With this project Mayne considerably extended the range of what it is legitimate to theorize when studying women’s film authorship, from their films to their biographies in a broad, and highly suggestive sense. Her study also demonstrated the way forward to a much more complex understanding of the mediations between agent and cultural product, enabling the earlier text/author impasses to be broken down.

A similar, though less sustained contribution to the same kind of project is made by Susan Martin-Márquez in her book Feminist Discourse and Spanish Cinema (1999). As her title indicates, Martin-Márquez is less explicitly concerned with individual directors than she is with the ‘discursive’ relationship of large social formations, such as gender, with the possibilities for feminist cultural agency more generally (she studies questions of female/feminine subjectivity in cinema directed by women and men). Still, the first half of her book is given over to three chapter-long historical studies on the agency of three woman directors from Spain’s past (‘Pioneering Agency: Rosario Pi’, ‘Negotiating Agency: Ana Mariscal’, and ‘Imaging Agency: Pilar Miró’), while her final chapters examine the work of Spain’s numerous contemporary women directors. Though she engages in a good deal of formal analysis, much of it influenced by Silverman,  like the later Mayne she usefully deploys agency as a ‘reiterative or re-articulatory practice, immanent to power, and not a relation of external opposition to power’ (Butler, 1993: 15), when discussing the actions of women cultural producers and their socio-cultural ‘reception’ in a broad sense. Following Carol Watts’s work on feminist literary theory (1992), which in turn was influenced by Butler, agency emerges out of localized struggles within the cultural construction of gender:

Watts argues that a woman author, working inside the tradition of genre in literature, may opt to take up a number of different subject positions through her writing ‘depending upon how her choice is constrained and the ways in which the forms are already culturally and institutionally defined and internalized by her’. Where Silverman speaks of an unconscious authorial desire as mirror and moulder of history, Watts sees a conscious agency, bounded, it is true, by the limits of available subject positions as well as by unconscious processes, but no less a producer of new ‘cultural forms’. Watts’s formulation, carefully avoiding ‘an unreflective retreat to notions of authorial intentionality, does allow for individual intervention in the social realm’ through a transgressive manipulation of artistic conventions. In short, if the theoretical pendulum has swung back, it has done so ‘with a difference’, responding to the combined pull of post-structuralist and feminist forces. (Martin-Márquez, 1999: 46; citing Watts, 1992: 89, and Silverman, 1988)

I shall return to this last assertion of Martin-Márquez’s shortly. But I would point first to the fact that, whether or not one agrees with Martin-Márquez and Watts on the precise theoretical model of women’s agency they posit here (which differs in any case: Martin-Márquez’s version relies in part on conceptualizing ‘unconscious processes’ in a psychoanalytic sense, whereas for Watts these are processes by which social and cultural conventions are internalized ‘automatized’, or naturalized), it is clear that for both of them these formulations operate as a kind of ‘reverse discourse’ on women’s agency, enabling agency, after decades of embarrassed deconstruction, finally to be subjected to analysis in the form of its textual, biographical traces, alongside more conventionally ‘legitimate’ activities for feminist cultural theorists, such as applying theories to ‘primary’ literary and film texts in formal ‘readings’.

Furthermore, the subtitle of Martin-Márquez’s book Sight Unseen points to what surely ought to have been a key object for feminist cultural theories had they not been quite so dictated to by more transcendental forms of ‘Theory’. Claire Johnston had written in Notes on Women’s Cinema that ‘The notion of women’s creativity per se is as limited as the notion of men’s creativity. [Creativity] is basically an idealist conception’ (1973: 28). Yet what she was not able to conceptualise, given her methodology, was that these two kinds of ‘creativity’ are ‘idealisms’ which are not ‘seen’, or conceived of in discourse, in the same ways.  Investigating how the particular forms taken by women’s creative agency have often been invisible, or at the very least seen, and judged, in a very different, often negatively inflected way compared with that of men, is one of Martin-Márquez’s most fruitful objectives in her book, especially in her chapter tracing the successful but highly troubled directorial career of Pilar Miró from the 1970s to the 1980s. Here, as with Mayne on Arzner, Martin-Márquez examines not just the films and the ‘facts’ of the director’s life, but also interviews, film reviews and academic studies, and photographic and film records of Miró’s public and private lives. She uses this broad selection of cultural ‘texts’ to enable her analysis of the many mediations at work in ‘imaging’ Miró’s authorial, and especially her putative auteurial status, which at times approximated that of her male auteurist contemporaries in the last years of the Franco regime, or since, but which was never a given in the same way theirs was.

In each of the studies I have been considering in this section of my paper there has been a reasonably confident return to considering various aspects of directorial ‘authors’ as agents: female subjects who have direct and reflexive, if obviously not completely ‘intentional’ or determining relationships to the cultural products they help to produce, as well as to their reception; ones which, moreover, will often repay explicit feminist investigation, on their own or as part of a broader examination of ‘elite’ and other forms of cultural agency and agent-hood available under patriarchy to particular women at particular times and in particular places. My own view is that such work can, and should go further. For example, to date, work on women directors has been carried out only on a very narrow range of agents (often the same names recur), primarily from Europe, North America, and other Anglophone countries, or on avant-garde and art film directors from a handful of non-western or southern-hemisphere cinemas. This work should also be extended to other kinds of female authorship in cinema, as it has begun to be in star studies (Stacey, 1994).

Although formalism was an important phase for feminist film theory, enabling it (albeit it in a limited way) to try to come to terms with the specificity of film as a medium on its own terms, I would argue that, in its glorious isolation of the text, it has not been all that helpful with regard to many other pertinent questions on female authorship. While certain theorists paid too much credence to auteurist methodologies in the early days of feminist film theory, they were unable to predict the ways in which this kind of work could be easily co-opted in the academic version of a much wider commerce in film auteurism which has taken hold in the last ten to fifteen years, with increasingly reificatory and commodifying processes (Grant, 2000: 101, 108). I would argue that feminists do not have to stop paying attention to formal structures or to the style of texts. Like Mayne and Martin-Márquez, they might simply have to broaden their notion of what constitutes a ‘primary text’ in film studies, and adopt more rigorous methods for ‘interactional’ and ‘inter-subjective’ analysis, such as those suggested by Timothy Corrigan (1991: 101-136), one of an increasing number of film theorists and critics influenced by Giddens’s structuration theory and the sociological tradition of symbolic interactionism, as well as by the analytic philosophical writing on agency of, amongst others, Charles Taylor.

Martin-Márquez wrote that ‘if the theoretical pendulum has swung back, it has done so “with a difference”, responding to the combined pull of post-structuralist and feminist forces’ (Martin-Márquez, 1999: 46; citing Watts, 1992: 89). She may well be correct. But, to abuse her metaphor a little more, if we are to swing the pendulum back any further, we could well find ourselves in the realm of (a distant relative to) the kinds of ‘sociological’ approaches to film studies summarily dismissed by Claire Johnston, and all other ‘right-minded’ feminist theorists in the 1970s and 1980s. This would be ironic, of course, and would be far from welcomed by most feminist film theorists. These days, however, when film studies as a whole is rather less defensive about its own authorized status as an academic discipline, having found a reasonably comfortable and distinctive niche in the academy, alongside media and cultural studies, perhaps this interdisciplinary luxury is one we could well afford to explore further.

Notes
1    Here I am paraphrasing Donald Pease’s discussion of literary authorship (1988: 106).
2    Steve Neale posits that the post-war European attribution of special artistic significance to film directors or auteurs helped to stabilise the categorisation of the highly eclectic output of art cinema (Neale, 1981: 33). In a recent article for Screen, I build on Neale’s work in order to examine some contemporary commercial aspects of auteurism in ‘globalized’ film culture and to argue for the importance of these in theorizing film authorship (Grant, 2000).
3    This usually unacknowledged disavowal of socio-historical or economic aspects by theorists raises similar questions about what gets to ‘count as theory’ as those     raised by the articles which appeared in the ‘Interchanges’ section of Feminist Theory 1(1): 96-118.
4    Except for theoretically informed considerations of the artisanal context of avant-garde film practice (see Cook, 1977).
5    Johnston’s main critique of ‘sociological perspectives’, such as those of the first feminist film journal Women and Film which appeared in 1972 is that they take as their ‘starting point the manipulation of women as sexual objects by the media [and are] derived from a view of the media rather than from a consideration of the specificity of cinema’ (Johnston, 1973: 3). Elsewhere in the introduction to her pamphlet, she  naturalizes her theoretical perspective as follows: ‘A study of cinema would inevitably have to proceed from an analysis of the functioning of ideology (Marx) and, taking into account that any cultural discourse is largely determined by forces totally beyond the individual artist’s control, it would also have to make use of the theory of the working of the unconscious as developed by Freud. Seen in this perspective, an analysis of the functioning of signs within the discourse assumes enormous importance for film-critics and film-makers if a counter-cinema is to be constructed’ (Johnston, 1973: 3. My emphasis)
6    Johnston also edited a collection of essays on Arzner’s work (Johnston, 1975).
7    This explanation is also hinted at by Judith Mayne: ‘While Johnston’s analysis seems to stress equally the importance of auteurism and of “symptomatic readings” [of film texts], her work is read today far more in the context of the latter’ (Mayne, 1990: 96). Interestingly, Janet Bergstrom criticizes Johnston on her lack of attention to film text (Bergstrom,1988: 83): ‘Johnston’s criticism of the sociological critics was that they didn’t take account of film’s specificity, of film as a signifying system. But in fact, one finds very little attention to the signifier in these articles. Her arguments operate almost exclusively on the level of the narrative signifieds, these signifieds being equivalent here to highly interpreted narrative events’ (1988: 84).
8    These comments of Caughie’s help to introduce an essay by Sandy Flitterman-Lewis (1981). Flitterman-Lewis has made a major contribution to feminist film theories of authorship with her 1996 book To Desire Differently, which, despite the much greater attention it admirably affords to historical discussion of women film directors, is similar in its theoretical approach to Silverman’s (1988); I have therefore opted not to assess it here for reasons of space. A few years before Caughie’s reader, Janet Bergstrom’s sympathetic challenge to Johnston’s work on women’s film authorship appeared (Bergstrom, 1979), to which I have already referred.
9    Smelik argues a related point: ‘Silverman’s model of the double “authorial voice” inside and outside the text clarifies the question of desire and identification [but ends up] confirming the Freudian apparatus and consequently fails to account for feminist attempts to move beyond it’ (1998: 48).
10    Just as Silverman makes only one limited appeal to the words of an author ‘outside’ a film text, in an attempt to provide some biographical evidence to back up a formalist point she makes, Mayne is almost as careful with deploying quotations from the director, or other biographical information: see for example her rhetorical ‘dis-ease’ around discussing what was known about Arzner’s lesbian sexual preference and identity (Mayne, 1990: 104). This last point derives from some unpublished work I carried out jointly with Núria Triana Toribio in 1996 on Spanish and Latin American women’s film authorship: I gratefully acknowledge her contribution.
11    Maggie Humm has also indicated what she sees as the advantages for feminist film theory of paying attention to feminist literary theory on the subject of authorship. In her book chapter ‘Author/Auteur: Feminist literary theory and feminist film’ she outlines her understanding of the range of textual signatures and specificities in women’s film authorship, arguing for a greater attention to detail than has hitherto been achieved in work to date. (Humm, 1997: 90-110). While, in a similar way to some of Mayne’s work, it is not always clear what the status of the film director’s agency is vis-à-vis the ‘results’ of her textual readings (she examines the work of Dutch filmmaker Marleen Gorris, who has had a very high degree of authorial control, especially in her first films), Humm’s textual analyses are very well achieved. I am less convinced by her use of Showalter’s ahistorical categories of women’s ‘literary differences.’(Humm, 1997: 100; Showalter, 1986).
12    As well as other ‘politically motivated’ theorists who, for example, worked in the late 1980s and early 1990s on the film authorship of lesbians and gay men, and people of colour (see Dyer, 1990, 1991). Dyer’s work has often gone beyond textual criticism to investigate forms of directorial agency.
13    Concepts borrowed from McNay’s discussion of the relevance of Bourdieu’s work for exploring the disembodiment and disembeddedness of the subject of masculinist thought (McNay, 1999: 95).
14    Anneke Smelik’s recent book chapter ‘In Pursuit of the Author: On Cinematic Directorship’ is interesting in this regard (Smelik, 1998: 28-55). It is one of the more self-consciously theoretical studies of feminist cinema which have appeared in the last few years; though it adds little to the general picture, in its close attention to film rhetoric based on Branigan’s (1984) analyses of point of view in cinema, it does clarify some of the issues around ‘implied’ authorship in particular, and so is worthy of attention.
15    In particular, apart from her use of a similar psychoanalytic framework to Silverman, Martin-Márquez examines Marvin D’Lugo’s discussion of ‘allegories of authorship’ which draws on Silverman’s authorial categories: these ‘include onscreen representations of the director or his/her double. For Silverman, examination of this figure leads to the tracing out of an authorial image refracted through lenses of gender, ideology and history’ (Martin-Márquez, 1999: 48; citing D’Lugo, 1992, and Silverman, 1988).
16    Quite apart from the not insignificant historical detail that there are different material and structural constraints which prevent female film ‘creativities’ from being allowed to exist in the same way as male ones (women’s traditional lack of access to film schools, industry unions, etc.).
17    Rita Felski has successfully incorporated some of these approaches in her work on feminist literary culture (see especially Felski, 1989). Also, see Andrew Tudor (1998) for some interesting, contemporary perspectives on sociology and film.

References
Barthes, R. (1968)  ‘The Death of the Author’ trans. Stephen Heath, pp. 142-148 in Barthes, R. (1977) Image-Music-Text. London: Fontana.
Benveniste, E. (1971) Problems in General Linguistics trans. Mary Meek. Coral Gables: University of Miami Press.
Bergstrom, J. (1979) ‘Rereading the Work of Claire Johnston’, Camera Obscura 3-4;
also pp. 80-88 in Penley (ed) (1988), Feminism and Film Theory. London: BFI
Publishing/Routledge.
Bergstrom, J. (1988) ‘Enunciation and Sexual Difference’, pp. 159-185 in Penley (ed), Feminism and Film Theory. London: BFI Publishing/Routledge.
Booth, W.C. (1961) The Rhetoric of Fiction. Chicago: University of Chicago Press.
Bordwell, D. (1985) Narration in the Fiction Film. London: Routledge.
Bordwell, D. and Thompson, K. (1993, 1979) Film Art: An Introduction. New York: McGraw-Hill.
Branigan, E. (1984) Point of View in the Cinema: A Theory of Narration and Subjectivity in Classical Film. New York: Mouton.
Butler, J. (1993) Bodies That Matter: On the Discursive Limits of Sex. London:
Routledge
Casetti, F. (1998) Inside the Gaze: The Fiction Film and its Spectator, trans. by Nell     Andrew with Charles O’Brien. Bloomington and Indianapolis: Indiana University     Press.
Caughie, J. (ed) (1981) Theories of Authorship. London: Routledge/BFI.
Cook, P. (1977) ‘The Point of Self-Expression in Avant-Garde Film’, reprinted pp. 271-281 in Caughie, J. (ed) (1981) Theories of Authorship. London: Routledge/BFI.
Corrigan, T. (1991) A Cinema Without Walls: Movies and Culture After Vietnam. London: Routledge.
D’Lugo, M. (1992) ‘Authorship and the Concept of National Cinema’, Cardozo Arts &  Entertainment Law Journal 10: 591-606.
Dyer, R. (1990) Now You See It: Studies on Lesbian and Gay Film. London: Routledge.
Dyer, R. (1991) ‘Believing in Fairies: The Author and the Homosexual’, pp. 185-201 in Fuss, D. (ed) Inside/Out: Lesbian Theories and Gay Theories. London:     Routledge.
Felski, R. (1989) Beyond Feminist Aesthetics: Feminist Literature and Social Change. Cambridge, MA.: Harvard University Press.
Flitterman-Lewis, S. (1978) ‘Woman, desire, and the look: feminism and the enunciative apparatus in cinema’, Ciné-Tracts 2(1): 63-68.
Flitterman-Lewis, S. (1996) To Desire Differently: Feminism and the French Cinema expanded edition. New York: Columbia University Press.
Foucault, M. (1969) ‘What is an author?’ lecture trans. Josué V. Harari and reprinted in Foucault, M. (2000) Essential Works of Michel Foucault 1954-1984, Volume 2: Aesthetics, Method, and Epistemology edited by Faubion, J., pp. 205-222. London: Penguin.
Grant, C. (2000) ‘www.auteur.com?’, Screen 41(1): 101-108.
Gaut, B. (1997) ‘Film Authorship and Collaboration’, pp. 149-172 in Allen, R. and Smith, M. (eds) Film Theory and Philosophy. Oxford: Oxford University Press.
Halprin, S. (1984) ‘Writing in the Margins (Review of E. Ann Kaplan’s Women and Film: Both Sides of the Camera)’, Jump Cut 29.
Holmes, D. and Ingram, R. (1998) ‘Series Editors’ Foreword’, pp. VII-VIII in Smith, A. Agnès Varda. Manchester: Manchester University Press.
Homans, M. (1986) Bearing the Word: Language and Female Experience in Nineteenth-Century Women’s Writing. Chicago: University of Chicago Press.
Humm, M. (1997) Feminism and Film. Edinburgh: Edinburgh University Press.
Johnston, C. (ed) (1973) Notes on Women’s Cinema. London: Society for Education in     Film and Television.
Johnston, C. (ed) (1975) The Work of Dorothy Arzner: Towards a Feminist Cinema. London: British Film Institute.
Kaplan, E. A. (1983) Women and Film: Both Sides of the Camera. London and New York: Routledge.
Kuhn, A. (1994, 1982) Women’s Pictures: Feminism and Cinema. London: Routledge.
Kuhn, A. (ed) (1995) Queen of the Bs: Ida Lupino Behind the Camera. Trowbridge: Flicks Books.
McNay, L. (1999) ‘Gender, Habitus and the Field: Pierre Bourdieu and the Limits of Reflexivity’, Theory, Culture & Society 16 (1): 95-117.
Martin-Márquez, S. (1999) Feminist Discourse and Spanish Cinema: Sight Unseen. Oxford: Oxford University Press.
Mayne, J. (1990) The Woman at the Keyhole: Feminism and Women’s Cinema. Bloomington and Indianapolis: Indiana University Press.
Mayne, D. (1994) Directed by Dorothy Arzner. Bloomington and Indianapolis: Indiana University Press.
Metz, C. (1981) The Imaginary Signifier, trans. by Celia Britton et al. Bloomington and Indianapolis: Indiana University Press.
Miller, N. (1986) ‘Changing the Subject: Authorship, Writing , and the Reader’, pp. 102-120 in de Lauretis, T. (ed) Feminist Studies/Critical Studies. London: Macmillan Press.
Miller, N. (1988) Subject to Change: Reading Feminist Writing. New York: Columbia University Press.
Neale, S. (1981) ‘Art Cinema as Institution’, Screen 22 (1): 11-39
Pease, D. (1988) ‘Author’, pp. 105-117 in Lentricchia, F. and McLaughlin, T. (eds) Critical Terms for Literary Study. Chicago and London: Chicago University Press.
Rich, B. R. (1998) ‘ In the Name of Feminist Film Criticism’ pp. 62-84. Chick Flicks: Theories and Memories of the Feminist Film Movement. Durham and London: Duke University Press.
Showalter, E. (ed) (1986) The New Feminist Criticism: Essays on Women, Literature and Theory. New York: Pantheon.
Silverman, K. (1988) The Acoustic Mirror: The Female Voice in Psychoanalysis and Cinema. Bloomington and Indianapolis: Indiana University Press.
Smelik, A. (1998) And the Mirror Cracked: Feminist Cinema and Film Theory. London: Macmillan Press.
Smith, A. (1998) Agnès Varda. Manchester: Manchester University Press.
Stacey, J. (1994) Star Gazing. London: Routledge.
Stoddart, H. (1995) ‘Auteurism and Film Authorship’, pp. 37-57 in Hollows, J. and Jancovich, M. (eds) Approaches to Popular Film. Manchester: Manchester University Press.
Tudor, A. (1998) ‘Sociology and Film’, pp. 190-194 in Hill, J. and Church Gibson, P. The Oxford Guide to Film Studies. Oxford: Oxford University Press.
Watts, C. (1992) ‘Releasing Possibility into Form: Cultural Choice and the Woman Writer’, pp. 83-102 in Armstrong, I. (ed) New Feminist Discourses. London and New York: Routledge.
Wimsatt, Jnr, W.K. and Beardsley, M.C. (1946) ‘The Intentional Fallacy’, Sewanee Review 54 (3): 468-488. Revised edition pp. 3-18 in Wimsatt, Jnr, W.K. (1954) The Verbal Icon: Studies in the Meaning of Poetry. Lexington: University of  Kentucky Press.
Wollen, P. (1972) ‘The Auteur Theory’, Signs and Meanings in the Cinema. London: Secker and Warburg.

Secret Agents – Feminist Theories of Women’s Film Authorship – Copyright Catherine Grant 2000