5. “Primitive” Narration

The period of film history known as the “primitive” era is framed by the first public exhibition of films by the Lumière brothers at the Grand Café in Paris in 1895, and by the development, some ten to fifteen years later, of the codes and conventions of nascent, classical narrative cinema. Although the term primitive cinema is used interchangeably with early cinema in describing this period of film history, I will nonetheless use it in quotation marks in order to foreground its problematic implications—particularly since the women’s films which will be examined in the next chapter are concerned precisely with those implications. For many years, the early cinema was regarded by film theorists and historians, if not as a vast wasteland then at the very least as an uninspiring collection of awkwardly put-together scenes, with little structure and even less art. For most modern spectators, early films are difficult to watch, either because “nothing happens,” or because of seemingly endless repetitions of the same events (particularly in chase films), or because of monotonous shots with little or no variation.

The “primitive” style of filmmaking assumes a frontal camera, an unchanging distance in medium long shot, and no camera movement. In other words, the techniques of focusing and centering to which viewers of the classical cinema are accustomed are absent in the early cinema.¹ In what might initially appear as something of a paradox, the “primitive” style also refers to the ways in which early films broke away from these standard characteristics—through a change in camera distance or angle, or through camera movement. When such departures are present, they are virtually always motivated. Early versions of close-ups, for instance, are literally produced in “primitive” films, whether by showing a child gleefully peering through a magnifying lens and closeup, isolated images of the objects of his vision (Grandma’s Reading Glass, 1900), or by showing a man with a telescope peering eagerly at a woman passerby and a shot of the woman’s ankle in close-up (As Seen through a Telescope, 1900).

Recent research on the first fifteen years of motion pictures has challenged many assumptions about the evolution of film narrative. Films and filmmakers previously disregarded as mere preludes to the “real” beginnings of cinema—the films of D. W. Griffith—have been reevaluated as providing different formulations of cinematic possibilities.² The strangeness or unreadability or many early films has been taken as a sign, not of their irrelevance but of our own conditioning by the apparatus of the classical cinema. Noël Burch locates an “otherness” of the early cinema in what he calls the “acting out, at the level of narrative, of gesture, at the iconographic, scenographic levels, the symbolism of those fundamental strategies which were to develop over the next quarter of a century (ca. 1907 to 1932) and which came to constitute what is still (abusively) called The Language of Cinema.”³ For Burch, early cinema offers bold demonstrations of the appeal of film. In early films a character (usually male) might look through a keyhole at another character (usually female), whereas this voyeuristic relationship would later be incorporated into camera movement, editing, and so on. Study of the early cinema, from this vantage point, is a kind of return of the repressed of the classical cinema.

Tom Gunning pursues the “otherness” of the early cinema in a different direction, arguing that “primitive” films explored areas which would have no place in the classical narrative fiction film—the playfully unique cinematic space of Georges Méliès’s films populated by fantastic creatures and disappearing acts, the direct address to the spectator eventually prohibited in the name of a self-enclosed narrative universe, the sheer pleasure of camera movement in early panorama films. These practices were abandoned in the evolution of mainstream film, but they find an echo in the works of avant-garde filmmakers. Gunning acknowledges that the relation between early and avant-garde cinema is not a simple one: “Certainly early filmmakers envisioned no aesthetic project like that of the avant-garde filmmaker. It is dubious that any of these films were thought about aesthetically at all. However, they display quite nakedly new relations to the representation of space that the camera made possible. Some of these possibilities were rediscovered by the avant-garde.”⁴ However different the approaches of Gunning and Burch may be in other ways, they share a definition of the “otherness” of the “primitive” cinema as something that would be marginalized in the evolution of classical film narrative.

Feminist explorations into the early cinema have followed both of the lines of development suggested in Burch’s and Gunning’s work. Studies by Linda Williams and Lucy Fischer on the early films of Melies have demonstrated, from quite different perspectives, fundamental ways in which the emerging forms of cinematic representation affirm the polarity of male subject and female object; hence, Méliès’s films offer bold demonstrations of the appeal of the classical cinema, along the lines of Burch’s argument.⁵ Lynne Kirby has argued, however, that early cinema offers, rather, a spectator who is “ ‘undone,’ uncoded, a subject whose sexual orientation vis-à-vis spectatorship is broken down, put into crisis—hystericized.” Kirby’s specific point of reference is early train films, and the heterogeneous modes of pleasure and identification she finds there are quite close to Gunning’s designation of the alterity of early cinema: “Early train films are as often involved in the undoing of sexual difference, of a set of anchors for sexual identity that floats, comically, in an age of mechanical production.”⁶ To some extent, of course, the conclusions one draws from early cinema depend upon which films one takes as representative of the era. Williams’s and Fischer’s analyses of Méliès’s films, on the one hand, and Kirby’s analyses of train films, on the other, are equally convincing, and their opposing conclusions suggest, most obviously, the need to understand the “primitive” era of filmmaking as composed of different and competing notions of “otherness.”

Put another way, much of the attraction of early film lies in its simultaneous evocation of, yet absolute distance from, the classical cinema. Following Gunning’s argument, a significant body of work on the early cinema has explored how the films of avant-garde filmmakers, especially but not exclusively male American filmmakers of the late 1960s and early 1970s, may be read in relationship to the early cinema.⁷ An equally compelling argument can be made for the works of some women filmmakers. In the next chapter, I will examine a group of films by women directors which revise and redefine the “primitive”; in the process, they unpack the different resonances that the very term primitive implies. This chapter will lay the groundwork, as it were, for that analysis. Beginning with an examination of the “primitive” as it applies not just to the early cinema alone but also and especially to how that era has been conceptualized in contemporary film studies, I will move to an examination of how different modes of address in the early cinema are informed by gender dynamics. While some early films offer bold representations of the feminist thesis that the man possesses the authoritative gaze in cinema (therefore adding fuel to the argument that the cinematic apparatus is shaped by one unbroken history of the objectification of the female body), my purpose in this chapter is not to examine how the early cinema set the stage for the classical cinema that was to come, but rather to examine why women filmmakers would be drawn to modes of representation so strongly affiliated with the period.

Many early films, and not just the magic films of Méliès, appear to confirm the widely held claim that the cinematic apparatus, emergent or otherwise, is made to the measure of male desire. Consider, for instance, the representation of the human body, and in particular the fragmentation of the body evident in early versions of close-ups. While the difference is not absolute—men’s bodies were fragmented by the camera too—there is little question that the display of a body part is much more characteristic of the representation of the female body.⁸ John Hagan has noted the frequency with which women’s legs and feet function as fetish objects in the early cinema, citing It’s a Shame to Take the Money (1905), in which a policeman and a shoeshine boy get to look at a woman’s legs; and Female Crook and Her Victim (1905), in which a woman falls against a man in order to rob him, and raises her dress to hide the man’s wallet. Her victim is duped, but the payoff is the glimpse of leg offered for erotic contemplation.⁹ An isolation of feet occurs in The Gay Shoe Clerk (1903), in which the shoe clerk’s stroking hand motivates a close-up of the foot (although the “woman” in this case is a female impersonator).

In the classical cinema, the close-up is associated, in its most conventional form, with the face; but close-ups of the face in the early cinema—unlike those of legs and feet—tend to be more identified with a mode of enunciation, the direct address to the spectator. In so-called facial films, conscious mugging and contortions are directed at the camera. In Goo Goo Eyes (1903), a female impersonator, made up to look stereotypically ugly, grimaces and contorts his/her face. The sexual ambiguity which often is disguised or displaced in the early cinema—the fact that female roles were frequently assumed by men—is here put on display. Another variation on a similar theme occurs in Wrinkles Removed (1902), in which a woman’s neck and head are being massaged by a man who attempts to turn her head to get a better look at her. Each time he does so, the woman grimaces and contorts her face. These close-ups of “women’s” faces are radically different from those of women’s feet and legs, and are far more evocative of the alterity of the early cinema of which Gunning speaks. The direct address to the camera would constitute one of many taboos of the classical cinema, and the relative autonomy of the image—that is, the fact that it is not defined completely within the look of another, a designated spectator within the film—suggests an ambiguous status.

A Subject for the Rogue’s Gallery (1904) draws on both uses of the close-up. A woman prisoner mugs for the police photographer and for the audience simultaneously, and the camera moves forward to frame her face. The film ends when the woman, now at close range, bursts into tears. Unlike the mugging in Goo Goo Eyes or Wrinkles Removed, here the woman’s face is contained within a structure marked by the separation of subject and object, the climax of which is the conquest of the woman, a conquest marked by the close-up of the face bursting into tears. The production of the close-up here situates the woman’s face within the more clearly defined fetishism of It’s a Shame to Take the Money or The Gay Shoe Clerk, but with the important difference that the woman initially resists the act of being photographed, and her resistance suggests a thrill of conquest in the production of the close-up.

Representations of the female body in its entirety in early film tend to have clothing play a central role. In both Annabelle Butterfly Dance and Annabelle Serpentine Dance (1895), for instance, a woman dances for the camera. She is dressed in billowing fabric which swirls about her, almost engulfing her at times. That the woman is performing for the camera is clear, but the movements of her body and the swirling fabric do not seem to be anchored in a specific context. The film could be taking place on a stage or in a studio. There is an excess in the film, an excess of flowing movement, so that there is little distinction between the movements of the body and the movements of the fabric. The film screen, like the fabric, conceals the female body and displays it simultaneously. The simultaneous display and concealment can be read as one of the fundamental conditions of representation of the female body. A film such as Trapeze Disrobing Act (1901) suggests in bold terms that condition. Two men, seated in a theater box, watch a woman, seated on a trapeze, strip. There is little doubt as to the intended audience of this spectacle, and there is little excess in the movements of the woman’s body.

Just as A Subject for the Rogue’s Gallery demonstrates the mechanism of the close-up of the woman’s face, What Happened on 23rd Street, New York City (1901) suggests, if more subtly, the relationship between concealment and display. The film shows a city street, with people walking toward and away from the camera. As one woman walks toward the camera, her skirt billows upward as she steps onto a subway grate, an action marking the end of the film, and providing a rudimentary narrative structure of exposition and punchline. In this film, a city street is framed in such a way as to suggest that the film screen is located on a boundary line central to many of the pleasures of early film, between a “real” street, bustling with the activity of everyday life, and the voyeur’s position suggested in Trapeze Disrobing Act. The woman’s body, now clad in street clothes, neither dances nor strips; she does nothing more.than walk toward the camera. These seemingly natural movements appear restricted when compared to the movements of the woman’s body in Annabelle Butterfly Dance and Annabelle Serpentine Dance. Similarly, the billowing fabric which at times overwhelms the woman’s body in the dance films is suggested here as well, but now in the single, restricted movement of lifting the woman’s skirts. The source of the movement is no longer the body itself but a subway grate. The woman’s body is not just the object of spectacle; it is narrativized, set in place in a chain of motions not her own. And the onlooker, who is part of the scene in Trapeze Disrobing Act, here is invisible, contained within the camera and mise-en-scéne.

Figure 23. Annabelle Serpentine Dance (Museum of Modern Art)

Figure 24. Trapeze Disrobing Act (courtesy of the American Federation of Arts Film/Video Program, from “Before Hollywood” exhibition; Patrick G. Loughney, photographer)

Figure 25. What Happend on 23rd Street, New York City (Museum of Modern Art)

There is a significant difference between a film such as Goo Goo Eyes, on the one hand, and A Subject for the Rogue’s Gallery, on the other, or Trapeze Disrobing Act and What Happened on 23rd Street. Goo Goo Eyes, with its direct look at the camera and its designation of a “female” face as a disguise, suggests a mode of representation that would have no place in what would become the classical mode of representation of the female face. A Subject for the Rogue’s Gallery, however, evokes the classical polarity of man as subject, female as object. Similarly, the difference between the clothed female bodies in Annabelle Butterfly Dance and What Happened on 23rd Street indicates how, in the early cinema, the female body acquired a representational function by being placed within a rudimentary narrative structure. Goo Goo Eyes and Butterfly Dance seem more evocative of the early cinema as embodying a sense of sheer visual discovery (as in Tom Gunning’s argument), and A Subject for the Rogue’s Gallery and What Happened on 23rd Street the “acting out” of the pleasures of the classical cinema of which Burch speaks.¹⁰

I have characterized these different kinds of representation of the female face and the female body as speaking to the increasing narrativization of the woman’s body. Many researchers of the early cinema are reluctant, however, to assign narrative functions to the devices of early film, largely because such a search for narrative smacks of the desire to see early film not on its own terms but as the mirror image—even if a negative one—of the classical cinema; to see narrative, that is, as the manifest destiny of cinema, evidenced from the very outset. For some, narrative wish-fulfillment speaks to a larger problem in film studies, the tendency to impose the (verbal and written) categories of narrative and narration imported from literature, to explain filmic narration with which it may, in fact, have little in common.

In response to these potential confusions, Andre Gaudreault defines narrativity in film as the intertwining of monstration (showing) and narration (telling), with monstration occurring at the level of the shot, and narration at the level of editing.¹¹ Tom Gunning distinguishes between the cinema of attractions, focusing on visual display and exhibitionism, which dominated early film until 1906-07, and storytelling film, concentrating rather on diegetic absorption, which became dominant later, and which would prefigure the classical cinema of identification and narrative. This is not an absolute distinction, since, as Gunning argues, “the cinema of attraction does not disappear with the dominance of narrative, but rather goes underground, both into certain avant-garde practices and as a component of narrative films, more evident in some genres (e.g., the musical) than in others.”

Gunning refers to two of the films discussed previously as examples of the identical phenomenon of “attraction”:

The close-up cut into Porter’s The Gay Shoe Clerk (1903) may anticipate later continuity techniques, but its principal motive is again pure exhibitionism, as the lady lifts her skirt hem, exposing her ankle for all to see. Biograph films such as Photographing a Female Crook (1904) [of which A Subject for the Rogue’s Gallery is a variation] and Hooligan in Jail (1903) consist of a single shot in which the camera is brought close to the main character, until they are in midshot. The enlargement is not a device expressive of narrative tension; it is in itself an attraction and the point of the film.¹²

While Hooligan in Jail employs the same camera movement as Photographing a Female Crook, the frame of the movement is quite different. The camera movement in Hooligan in Jail traces the prisoner’s delighted reaction to food that has been brought to him. To be sure, Hooligan and the female crook are both examples of the marginal figures favored in many early films, and in both cases the camera captures a childlike reaction; but there is nonetheless a significant difference between the two films.

If Hooligan “exhibits” a childlike response, what the female crook “exhibits” first and foremost is the fact of her female identity, and of her difference from the men, and the male agency of the camera, that surround her. Both films may be said to be exhibitions, not of the scenes they capture but of the cinema itself, of its capacity to create a closer view; but in this respect, the difference between them is even more striking. For the camera movement in Photographing a Female Crook and A Subject for the Rogue’s Gallery produces a reaction, whereas the camera movement in Hooligan in Jail records one. If “exhibitionism” is an appropriate word to use to describe the difference of early cinema—different, that is, from the voyeurism ostensibly dominant in storytelling film—then it needs to account for, rather than obscure, different kinds of display at work.

Gunning himself uses the distinction between early cinema as exhibitionist versus narrative cinema as voyeurist:

Contrasted to the voyeurist aspect of narrative cinema analyzed by Christian Metz, this is an exhibitionist cinema. [. . .] From comedians smirking at the camera, to the constant bowing and gesturing of the conjurors in magic films, this is a cinema that displays its visibility, willing to rupture a self-enclosed fictional world for a chance to solicit the attention of the spectator.¹³

However, Metz’s reading of the classical cinema as voyeurist does not replicate the classic duality of voyeur and exhibitionist, but rather displaces it. The classical cinema does not accentuate voyeurism at the expense of exhibitionism, but puts into place a whole other dialectic:

cinema manages to be both exhibitionist and secretive. The exchange of seeing and being-seen will be fractured in its centre, and its two disjointed halves allocated to different moments in time: another split. I never see my partner, but only his photograph. This does not make me any less of a voyeur, but it involves a different regime, that of the primal scene and the keyhole. The rectangular screen permits all kinds of fetishisms, all the nearly-but-not-quite effects, since it can decide at exactly what height to place the barrier which cuts us off, which marks the end of the visible and the beginning of the downward tilt into darkness.¹⁴

The argument that cinema operates according to a scopic economy of voyeurism and fetishism, and not the classic economy of voyeurism/exhibitionism, attains, of course, its most succinct form in Laura Mulvey’s classic thesis that cinema “works” precisely by its simultaneous appeal to scopophilia and fetishism.¹⁵

Even if exhibitionism is not easily placed in opposition to voyeurism insofar as the cinema is concerned, one could still argue that the early cinema is characterized precisely by an acting out of exhibitionist displays which would return in the classical cinema only in the form of spectacular displays subordinated to the dialectic of voyeurism and fetishism. Gunning’s argument is based in part on the exhibition context for early motion pictures, where fairgrounds and sideshows would appear to have created a quite literal exhibitionist frame for motion pictures. But one could argue in just the opposite way as well. It could be said that motion pictures marked the erasure of popular entertainment as performance; that the positions of visual identification offered by early films were characterized by the displacement of exhibitionism by fetishism, by the separation of subject and object, and by distance between performer and onlooker.

Although Gunning does not spell this out in detail, the implication in his argument is that gender is not really a category with great signifying authority in early film. Of a 1904 film, The Bride Retires (which is cited in Noël Burch’s film about the early cinema, Correction Please), Gunning notes that it “reveals a fundamental conflict between this exhibitionistic tendency of early film and the creation of fictional diegesis. A woman undresses for bed while her new husband peers at her from behind a screen. However, it is to the camera and the audience that the bride addresses her erotic striptease, winking at us as she faces us, smiling in erotic display.”¹⁶ That it was possible in early cinema for a woman to solicit directly the attention of the audience may well be a function of the alterity of the early cinema, but this kind of exhibitionism seems to me quite different from other examples of visual display, since it evokes the register of fetishism, of the classic representation of the female body as capable of possessing a look only when that look solicits the attention of a male viewer. To use John Berger’s phrase, this seeming “exhibitionist” is more convincingly described as a “surveyed female.”

In the context of the present discussion, it is crucial, then, to account for the different ways in which the mode of address of early cinema occurs, and as I have suggested, gender accounts for some rather significant differences between different practices. I would like to move, then, to a discussion of the gendered components in some of the most dominant modes of address, those that employ human figures to embody both the visual fascination and the rudimentary narrative structures of early film. First, however, it is necessary to address another problem common to film studies and particularly relevant to early film: the extent to which one can speak of an actual “narrator” in film. There is a well-founded reluctance in narrative studies to use the term narrator except in the most specifically defined instances.¹⁷ The bracketing of the term narrator in contemporary discussions of film narrative has evolved in part from the desire to focus on narrative as a process, on narration not as contained in an individual but rather as a structure, a function linking a variety of subject positions. The desire to designate a cinematic “narrator” may appear retrograde at best, whether the product of naive equations between film and categories of literary narrative (which themselves are problematic, since the distinction between “third-person” and “omniscient” is not so clear-cut, for instance), or of an equally naive assumption that the individual human figure is the measure of all narrative authority.

In some studies on the early cinema, this question of the narrator is examined in terms of narrative omniscience. Thus, in his discussion of temporal structures in early cinema, and of the tendency of many early filmmakers to use temporal overlap, André Gaudreault distinguishes two modes. The first relies on a succession of autonomous tableaux, not unlike the narrative mode characteristic of medieval literature. The storyteller, Gaudreault writes, “lacked omniscience, perhaps because he had not sufficiently developed a consciousness of his craft. . . .” The second mode found its most persuasive representative in D. W. Griffith. Here, the “ubiquitous camera” had learned to “go through walls.”¹⁸ Gaudreault refers to the power of omniscience associated with the development of narrative techniques such as cross-cutting and camera movement.

Discussion of film narrative in the early years is haunted by omniscience, and the specter virtually always leads, sooner or later, to the figure of D. W. Griffith. Griffith’s status has undergone considerable reevaluation in recent years. Griffith may not have “invented” a cinema in which a story “tells itself,” but he is nonetheless the single figure most connected to the myth of a self-enclosed narrative universe, understood from the point of view of ideology as well as genius. It is often assumed that Griffith’s style represents the dream of classical cinema, but the way in which his presence as narrator was built into the structure of his films was not necessarily successful. As Tom Gunning points out, “This invisible but sensed hand will reach its apogee in Griffith’s commercial disaster Intolerance. The ‘uniter of here and hereafter’ will prove an obstacle to much of his audience, a frustration rather than a guide.”¹⁹ One of the earliest forms of narration in early cinema was that offered by lecturers who frequently accompanied film programs. They literally demonstrated to spectators the way through film, whether by summarizing the plot of a novel on which a film was based, or by pointing out the central action of a shot. That lecturers disappeared suggests, of course, that their function had become integrated into the developing conventions of film narrative, whether through intertitles or the “invisible hand” guiding the spectator through Griffith’s films.²⁰

Figure 26. The Living Playing Cards (Museum of Modern Art)

The narrating function fulfilled by the lecturer, by intertitles, and by the “invisible hand” of Griffith’s films is primarily one of linear continuity and referential information. Other, more limited kinds of narrating functions are fulfilled by those figures who appear to direct, mediate, or otherwise act out the visual pleasures of the cinematic scene. I will refer to these characters as “primitive narrators,” recognizing that they are neither omniscient narrators nor the absolute agents of “primitive” narration—i.e., they are objects of the camera’s view at the same time that they act out the emerging visual and narrative capacities of the film medium. The most obvious example of the “primitive narrator” is the figure of the conjuror or magician, most common in the films of Méliès. In Le Mélomane (The Melomaniac [1903]), Méliès appears as a magical music teacher who repeatedly takes off his head and throws it on a staff to represent musical notes; in Les Cartes vivantes (The Living Playing Cards[1905]), Méliès portrays a magician who transforms playing cards into living human beings. Here, as in other of his films, Méliès looks directly at the camera, accentuating his role as solicitor of the audience’s attention.

Another related category of the “primitive narrator” is the photographer, or image-maker. In Getting Evidence (1906), a detective armed with a camera pursues a couple and is frustrated in his repeated attempts to capture them on celluloid; when at last he does manage to photograph them, he discovers he has been following the wrong couple. In The Story the Biograph Told (1903), a boy surreptitiously films a man and his secretary kissing in the Biograph office, and the film is screened at a motion-picture show attended by the man and his wife. In one of Méliès’s most interesting films, Une Chute de cinq étages (The Photographer [1906]), Méliès plays a photographer about to take a picture of a young couple in his studio. The background screen falls on them, however, and the camera falls out the window onto a concierge sitting outdoors. As she struggles to remove herself from the camera and dropcloth, a passerby takes her for a bull and attempts some amateur bullfighting.

While films featuring magicians and conjurors persisted alongside those with image-makers, in many ways the role of the image-maker subsumed the role of the magician. The difference between Méliès’s Une Chute de cinq étages and his many magic films is instructive in this regard. As he adopts the role of image-maker, Méliès’s omnipresence and omnipotence as master of ceremonies diminish, giving way to the narrative and spectacular possibilities of the apparatus itself. For in the second shot of Une Chute de cinq étages, the camera assumes the role of narrator, which in the previous shot was performed by the photographer.

Other “primitive narrators” function simultaneously as spectators upon whom are bestowed the spectacular and sometimes infantilizing possibilities of the medium. The dream film, popular in particular in the pre-1907 era, offers the dreamer as both the agent and the victim of the power of the dream.²¹ In Let Me Dream Again! (1900), a man flirts with a young woman who is wearing a mask; he then awakens to find himself embracing his wife in bed. And the Villain Still Pursued Her, or The Author’s Dream (1906) shows a writer who falls asleep at his desk and imagines himself courting a young woman; the two are pursued by her unwanted suitor. A Midwinter’s Night Dream (1906) portrays a child who falls asleep in the snow, and a fade depicts a scene in which a girl invites him into a house where holiday festivities occur.

Given the extent to which contemporary analysis of cinema has focused on the power of the gaze, “primitive narrators” in early voyeur or keyhole films offer particularly telling embodiments of the equivalence between cinematic pleasure and voyeuristic fantasies. Pull Down the Curtains, Suzie (1903), like the previously mentioned Trapeze Disrobing Act (1901), is a single-shot film in which a woman undresses, observed by a male spectator. In other films, the voyeur’s gaze motivates more sophisticated structures of the look, alternations between the voyeur and the object of the look.²² In Peeping Tom (1901), a hotel porter looks through the keyholes of a series of rooms, with keyhole masks imitating his vision. A similar structure occurs in The Inquisitive Boots (1905), in which a bootblack, motivated by nothing more than the desire to look, moves down a hotel corridor looking through keyholes. In both of these films, there is rudimentary punishment of the voyeur: Peeping Tom is chased away by a man who opens a door, and Inquisitive Boots is squirted at through the keyhole.

Figure 27. Pull Down the Curtains, Suzie (courtesy of the American Federation of Arts Film/Video Program, from “Before Hollywood” exhibition; Patrick G. Loughney, photographer)

What serves as punishment in these two films takes on another function in A Search for Evidence (1903), where the look through the keyhole is motivated not by mischief but by the ostensibly worthy cause of finding an errant husband. An angry wife and a detective proceed down a hotel corridor in search of her adulterous mate. The woman peers into one room after another, and the camera imitates her look. When she finds her husband, the detective is shown peering through the keyhole, followed by a shot of the interior of the hotel room. The detective in this film acquires a function that is far more “narrative” in the classical sense than any of the other modes of address discussed thus far, since his look authorizes the cut to the final shot of the film, from the hallway to a lateral, interior shot of the room. He crosses the threshold dividing hallway from room interior, the threshold dividing subject and object of the look.²³ The detective is thus a “primitive narrator,” but his function is more complex in that he transcends the role of voyeur and crosses the threshold separating subject and object of the look. His narrative function is one which will be assumed in a variety of ways by the evolution of cinematic technique itself. In A Subject for the Rogue’s Gallery, the look of the camera and the look of the (invisible) police photographer are joined, and the police officers who restrain the woman become mediators for that conjunction. The integration of the function of a “primitive narrator” like the detective in A Search for Evidence into the moving camera in A Subject for the Rogue’s Gallery is significant in the development of the capacity of film narrative to present itself as pure story, to capture the viewer in the movement of the camera and hence to deaccentuate the marks of discourse.

Put another way, the absorption of the “primitive narrator” into the movement of the camera across the threshold is emblematic of how classical film narration would envelop such bold figures of visual authority and fascination and render them invisible through the apparently seamless narrative of linear, novelistic film narrative. Other early films expand and develop the role of the “primitive narrator” in other ways. Terrible Ted (1907) portrays a child who, in the family living room and in the presence of his mother, reads a book. As soon as she leaves the room, he turns his attention to what we assume to be a forbidden text, a volume of the Wild West Weekly entitled, appropriately enough, “Terrible Ted.” The image on the cover reveals a hold-up, and a close-up of it shows a woman being held captive. Ted reads for a while from the magazine and then folds it up into his back pocket. He takes a gun from a drawer and goes outside, where he begins shooting at policemen on the street. He chases them and begins leading a group of children. Then there is a cut to the countryside, where a hold-up of a stagecoach and a rescue of a damsel in distress recreate the cover of the magazine. Ted suddenly appears on the scene, kills the villains, and begins a series of adventures: he shoots a villain in a saloon, stabs a bear about to kill a Native American woman picking flowers, is captured and tied to a stake, is rescued by the woman, is chased by a group of men who kill the woman, discovers the camp of Native Americans and stabs them all, then scalps them. After a triumphant display of the scalps, we return, suddenly, to the living room, where Ted is asleep. His mother strikes him and the child cries.

Terrible Ted incorporates the figures of both the dreamer and the conjuror. But neither the dream frame nor the enactment of “magic” is clear-cut.²⁴ The cut to a countryside and the reenactment of a magazine cover are clearly fantasy, but the initial crossing of the threshold clouds the boundary between interior and exterior space, between the real and the imagined, between the dreamer’s conciousness and the printed text. With Ted, the viewer crosses a threshold, is located in an ambivalent territory. Only with the return to the living room and the mother’s angry gesture—a simple configuration of the oedipal scenario which lurks over the cinema from its outset—is the spectator separated from the dream, given a position of detached observation. In Terrible Ted the role of the “primitive narrator” is considerably more complex than the functions of dream and voyeurism, magic and image-making, visible in other films. Here, Ted is located in an ambivalent space between subject and object; he embodies and acts out a narrative desire. In Terrible Ted, the child has incorporated the narrative authority of the adult into his own fantasy, in a series of displacements that lead from mother to book to gun to police to various villains.

If Terrible Ted is then exemplary of the function of the “primitive narrator” in early film, it would appear that he confirms what has become common wisdom in film theory: that the subject of enunciation in the cinema is male, infantile perhaps, but male nonetheless, emphasized in particular by the parade of “others”—women, Native Americans—as well as the mother’s punishing gesture as a reality principle. There are female “primitive narrators,” but they appear so rarely as to suggest that they are the proverbial exceptions that prove the rule. Conjurors and magicians are virtually always male, but one curious counterexample discussed by Lucy Fischer is A Pipe Dream (1905), in which a woman, smoking a cigarette, conjures a tiny man in her palm. Unlike the conjuror in Méliès’s films, this woman is taken aback by her magical powers. A Pathe remake of Méliès’s Le Mélomane features a woman music teacher and conjuror. Whereas Méliès’s magical powers involved putting replicas of his own head on the musical staff as notes, the woman in the 1904 remake removes the heads of men to create her musical notes.

Fischer suggests that such isolated examples of the female conjuror are significant for the more complex view they create of the male-female relationship in early cinema.

In the cases where women magicians exist, they are figures of awe and dread. This makes clear the fact that woman is not always perceived as powerless—a passive prop. Rather woman’s power is often acknowledged, but it is viewed as perilous and perverse. Perhaps, the male magician is not only performing tricks upon the female; he is preventing her from performing more dangerous tricks upon him.²⁵

The way in which women’s power is acknowledged, however, is fully consonant with a scenario of male desire. I stress this not in order to return to a feminist vision of film history that has been criticized throughout this book—a vision, that is, that would stress one single, unbroken narrative of male dominance over women—but rather to suggest that for those women filmmakers who have reexamined a “primitive” mode of representation, the issue is not the rediscovery of what would amount to a moment of bliss prior to the imposition of the conventions of classical narrative, a moment then to be “excavated” in much the same way the pre-oedipal has been mined for its possibilities of other modes of relation and signification.

Figure 28. What Happened in the Tunnel (Courtesy of the American Federation of Arts Film/Video Program, from “Before Hollywood” exhibition; Patrick G. Loughney, photographer)

Some instances where women do function as “primitive narrators” in early film seem to build upon the resistance of the woman to being filmed in A Subject for the Rogue’s Gallery, but “resistance” here should not be read as an alternative female-centered vision. An instructive example in this context occurs in What Happened in the Tunnel (1903). Two women, a white woman and her black maid, are riding in a train. When the train passes through a tunnel—a passage marked by a darkened screen—a male passenger attempts to steal a kiss from the white woman. Once the train has left the tunnel, he realizes, much to his chagrin, that he has kissed the black woman. The man looks, and wants to possess what he sees; but he kisses the “wrong” woman, the inappropriate object of spectacle. The two women in this film are objects of the male look, but they return the look, by laughing at the man. This “primitive narrator” crosses a threshold into potentially dangerous territory, marked by the laughter of the women which is the punchline to the film. But if that laughter suggests a resistance to the authority of the male look, it is a resistance that is nonetheless locked into the hierarchy of subject and object, a hierarchy doubly inscribed through sexual and racial difference. The women’s laughter is possible only within a firmly established structure of self and other.

Other films which feature female “primitive narrators” continue to define this process of “resistance” as fully part of the emerging scenarios of the classical cinema. In the 1909 film Choosing a Husband, a young woman is bewildered by the number of young men who want to marry her. Four men propose to her, and she tells each one that she will make her decision the next day. She then hides behind a screen, and leaves an attractive woman in the drawing room to greet them. Predictably, each suitor arrives, finds the other woman, and makes advances to her. The woman comes out from behind the screen and dismisses the four would-be husbands. At the conclusion of the film, the woman finds true love at last, when another man returns from a trip abroad. The woman in Choosing a Husband occupies the position of the husband in The Bride Retires, but the direct engagement of the spectator’s look in that film is here replaced by an exchange between a man and a woman. The woman does possess an active look, one grounded in her own curiosity, a curiosity based in its turn on a principle of resistance to the authority of the male subject; but her curiosity is stripped of any threat it might represent by being channeled into the conclusion of marriage which would become the classical form of resolution par excellence.

A 1913 film, The Innocent Bridegroom, traces a plot that evolves from the complications of a woman’s curiosity. A widow is suspicious of the man she is about to marry, and hires a private detective to follow him. In the meantime she does some sleuthing of her own, and once she is convinced of her suitor’s good faith, they marry. The woman forgets, however, to call off the private detective, who follows them on their honeymoon. Hence the film depicts the clash between two “primitive narrators,” the widow and the detective. How Men Propose (1913) is a particularly imaginative rendition of the narrative implications of resistance to the equation between woman and image of the male look. We see a young man embrace a photograph depicting a woman’s face. A second young man, and then a third, discover that they all are courting the same woman, for each possesses the same photograph. Having distributed identical photographs of herself, the woman sends each man an identical letter. “I’m returning your ring,” she writes. “I’m writing an article on how men propose. You can keep the ring.” How Men Propose offers one of the most interesting forms of female “primitive” narration, since in this film the woman actually responds to her status as image.

One of the most interesting examples of a female “primitive narrator” is in a 1912 film, The Mind Cure. Professor Conner, a hypnotist, attempts without success to frighten away the numerous men who court his daughter. Frustrated by the failure of his paternal authority, he writes in his journal that he is going to cure his daughter, a “confirmed flirt,” through hypnotic treatment. Unbeknownst to him, daughter Pearl reads his journal, so that when he hypnotizes her (“you will dislike all young men”), she resists. A man enters the professor’s office, and Pearl immediately chases him away to convince her father of the success of his treatment. The professor’s next patient is Chester, Pearl’s suitor, who comes to be treated for bashfulness. He follows Pearl’s example, but now embraces Pearl to prove that the treatment has worked. In the concluding shot of the film, Chester and Pearl present their marriage license to the professor.

Professor Conner is a narrator who fails—fails as a principle of paternal authority, fails as a conjuror-hypnotist. A crucial moment in terms of Pearl’s own function as narrator occurs when the father attempts to hypnotize his daughter. Pearl faces the camera, her eyes closed in accordance with the taboo against actors’ looking directly into the camera.²⁶ The audience shares in her faking of hypnotic treatment. The conjuror-hypnotist has retreated to the background of the image, with a figure of playful resistance now designated as source of the narration. Pearl’s rebellion is limited: Her desire, the audience discovers with the arrival of Chester on the scene, is not to remain a “confirmed flirt” but to marry. While it is the daughter’s joke which moves The Mind Cure along and gives it a narrative direction, here, as in all of these films which feature a woman’s desire, the woman’s status within a scopic economy is confirmed.

As I hope is obvious by now, these female “primitive narrators,” and their playful resistance to the polarity of male subject/female object, do not constitute a simple position of alterity from which one can imagine other formulations of cinematic desire. Rather, they depend and rely on the classical polarities of (male) subject and (female) object. There are occasional moments where the female narrator acquires a function, not of resistance to be quickly channeled into resolution, but of a playful refusal that seems to offer another possible scenario, like the woman’s ironic reduction of marriage to a photograph and a ring in How Men Propose, and even in the faking of hypnotic treatment in The Mind Cure. But by and large, these designations of female “primitive narrators” continue the narrativization of the female body that one sees in such earlier films as Subject for the Rogue’s Gallery and What Happened on 23rd Street. At the beginning of this chapter, I suggested—following the different feminist arguments that have been made about the early cinema—that the early cinema needs to be seen as simultaneously related to yet radically other than the classical cinema. However, virtually every example I have cited thus far leads to the dismal sense of a master narrative in place from the outset—the master narrative, that is, of a cinematic apparatus seemingly destined to represent the polarity of male subject and female object.

Ironically, the figure of “primitive” narration which seems less saturated with the rigid polarity of gender is voyeurism. Given the extent to which voyeurism is the shorthand term so frequently used to evoke the specter of cinema as a supremely patriarchal form, such a claim may seem ludicrous. But while voyeurism and fetishism are, in Laura Mulvey’s analysis, the two intertwining components of sexual polarity in the classical cinema, in early cinema the one does not necessarily presuppose the other. Some of the films discussed previously, including A Subject for the Rogue’s Gallery and What Happened in the Tunnel, suggest voyeurism and fetishism simultaneously. But the actual and literal representation of voyeurism in early film is not so firmly linked to the emerging structures of Mulvey’s definition of visual pleasure. While female voyeurs are in somewhat short supply in early cinema, they do not always function as mere props for the fantasies of the male subject, as is the case in the Pathé remake of the Méliès film mentioned earlier.

This is not to say that insofar as voyeurs are concerned, there is no gender differentiation whatsoever; and I certainly do not want to claim a “progressive” status for female voyeurism in early film! Indeed, it would constitute a kind of regressive utopianism to discover in any single figure of the early cinema a position from which to define an untainted source of female cinematic pleasure. What I am suggesting, rather, is that in early film the interdependence of voyeurism and fetishism, while clearly at work in a nascent way in some cases (such as A Subject for the Rogue’s Gallery), is not always the condition of visual pleasure that it is in Laura Mulvey’s analysis of the classical cinema.

Although the argument could be made that the voyeurs of early cinema are class types more than gender ones, and that voyeurs are so infantilized as to be practically undifferentiated in sexual terms, there are significant differences between female and male voyeurs. Previously I referred to two voyeur films, Peeping Tom and The Inquisitive Boots, which follow a pattern similar to most films of the type. A man, usually a servant (or, in the case of The Inquisitive Boots, a bootblack), moves from one keyhole to another in a corridor, the film thus alternating between a shot of him looking and a shot—frequently through a keyhole mask—of what he is looking at. The voyeur usually mimics what he sees before moving on to the next keyhole, and the conclusion of the film is a rudimentary punishment of the voyeur. In Peeping Tom, the voyeur spies on several women (including one female impersonator). When his gaze lands upon an isolated man in a room, the door opens and the voyeur is chased away.

The 1904 film Un coup d’oeil par étage (Scenes on Every Floor) offers a somewhat more sophisticated pattern. Here, a male servant delivers mail to each of several flats in a building, and the film alternates between his looking through the keyhole and the scene that he witnesses, although here the scenes are presented without the frame of the keyhole mask. As in Peeping Tom, the servant faces the camera and mimics what he has seen on each floor—a businessman making a frantic phone call, three children in a pillow fight, a woman (yet another female impersonator) fondling a cat. On the top floor, the servant discovers a fire, quickly opens the door, and rushes to summon firefighters. In both of these voyeur films, it is not only the relationship between male voyeur and the object of his look that is significant, but also and especially that his look initiates a crossing of the threshold that allows a rudimentary form of resolution. In some voyeur films, the crossing of the threshold actually motivates a reverse angle of the inside of the room, thus prefiguring one of the most standard representations of the look in classical cinema.

In most voyeur films, the individual scenes witnessed by the voyeur are a series of unconnected, often peculiar, fragments.²⁷ In A Search for Evidence, for instance, which differentiates so forcefully between the male and the female look, and inscribes the look of the woman as lacking the authority to cross the threshold, the serial and peculiar character of the scenes witnessed is strong (including a man rocking a baby, and a man who attempts to light a chandelier). A 1905 film, Die Rache der Frau Schultze (Mrs. Schultze’s Revenge), is particularly interesting in this respect, since it offers not only a glimpse of a female voyeur but also the production, by the woman, of a scene worthy of the most peculiar vignettes glimpsed through a corridor keyhole. The film portrays a set divided into two rooms that share a wall. On the right is Frau Schultze’s bedroom; on the left, the apartment of her neighbor, a musician. Frau Schultze goes to bed, only to be awakened by her neighbor, who returns late to his apartment and begins playing his piano.

When a fly lights on Frau Schultze’s nose, she comes up with a plan. She peers through the keyhole of the door that separates her room from her neighbor’s, and through a keyhole mask she and we see the musician. Frau Schultze then sends the fly through the keyhole, and it proceeds to torment the musician. He tries to swat it, but misses consistently. He even covers himself and his music with spattered ink (from trying to swat the fly when it is positioned on a bottle of ink), and he wreaks havoc on many objects in his apartment. Finally, a group of men are summoned to put an end to the chaos, and among them is a triumphant Frau Schultze.

In this film, female voyeurism engages not only the pleasure of peering through keyholes, but also and especially the desire to manipulate the scene. In La Fille de bain indiscrète (The Indiscreet Bathroom Maid [1902]), a female bath attendant moves down a hallway, peeking through transoms in much the same fashion as Peeping Tom. She too mimics the gestures of her clients in various stages of undress—first a woman, then a man, and finally a man with his servant. While the alternating structure is identical to that of other voyeur films, the resolution is not. When the man in the last “scene” pushes his servant into the bathtub, this voyeur dissolves in laughter, allowing neither the punishment, the climax, nor the reverse angles to room interiors motivated by other scenarios of male spectatorial desire. In this film, female voyeurism does not inspire a crossing of the line separating subject and object of the look.

Although it is impossible, with such limited examples drawn from only two early films, to draw anything resembling a firm conclusion, it is nonetheless notable that in both cases, female voyeurism does not require the punishment or scolding of the voyeur. To be sure, a film such as A Search for Evidence suggests that a female voyeur can peek, but does not possess the authority necessary to penetrate the room—a privilege reserved for the detective. But perhaps what the female voyeur does possess is, precisely, a desire that would have no place in the emerging codes of narrative cinema laid out so neatly in A Search for Evidence. Neither of the female voyeur films contains prescient paradigms of the classical cinema. But exhibitionism does not seem to be the right term to describe the activities of the female voyeurs, either.

I have expressed some reservations about Tom Gunning’s designation of what he calls the cinema of attractions as “exhibitionist,” as opposed to the presumably “voyeurist” pleasures of classical narrative, particularly insofar as the term exhibitionist can mask a series of effects that are more properly described as fetishistic, and which are more connected to the gender polarity of the classical cinema than the word exhibitionist would suggest. In the case of female voyeurism, however, I think early cinema does offer a suggestion of another kind of pleasure in the cinema which would “go underground,” as Gunning puts it, only to emerge in some films by women directors.

Although most researchers into the early cinema are careful to avoid any simple idealization of the radical difference that early films represent, there is a perhaps inevitable tendency, if not to idealize the early cinema then to simplify and monolithize the classical cinema to which it contrasts. The tendency is perhaps most apparent in Burch’s work, as the very phrase “Institutional Mode of Representation” suggests. One of the most common assumptions in contemporary film theory of the last fifteen years has been that the classical Hollywood cinema is, in Raymond Bellour’s words, a “machine of great homogeneity,” a system which serves to affirm white, patriarchal, Western values through a self-enclosed narrative universe of crisis and resolution.²⁸ It is beyond the scope of the present chapter to engage in the long-standing debate in film studies on the homogeneity of classical film narrative.²⁹ But given the extent to which so much feminist film theory supports the monolithic view of classical film narrative, it is important to specify that neither the “primitive” narration I am describing in this chapter nor the figures of the “primitive” in women’s films to which I will turn in the next chapter can be read as rigidly distinct or absolutely excluded from the realm of classical film narrative. Many studies of early film, while attentive to the dangers of overromanticizing the “primitive” era, seem nonetheless to regard the classical cinema as completely saturated with the laws of continuity, fictional coherence, and diegetic illusion.

Tom Gunning is extremely attentive to how the “cinema of attraction” dominated the early cinema and acquired other forms with the advent of classical narrative cinema. But in his discussion of point of view and early voyeur films, Gunning differentiates between early and classical film by appropriating both Metz’s and Mulvey’s descriptions of the latter. Early keyhole films, says Gunning, “find their heir in Cocteau’s Hotel des Folies-Dramatiques sequence from Blood of a Poet with its dream-like discontinuities and sense of wonder, rather than in the highly narrativized Rear Window.” Continuing the distinction, Gunning writes:

While these films [early keyhole films] involve voyeurism, the spectator they address is still far from the voyeur spectator of classical narrative film, that invisible witness who watches unobserved by any of the inhabitants of the film’s world, secure in his isolation in the dark. The classical spectator is constructed within a fantasy of a powerful invisible gaze able to insinuate itself into the most private of dramas.³⁰

I assume that Rear Window exemplifies for Gunning, as it did for Laura Mulvey, the construction of the “classical spectator.” Tania Modleski’s rereading of Rear Window is intended primarily as a revision of the widely held view that this film, like most of Hitchcock’s works, exemplifies the polarity of male subject and female object that most theories of the classical spectator take as a given. Hence Modleski demonstrates that the film, far from positing an unquestioned authority of the male gaze, both problematizes male voyeurism and investigates the components of female spectatorship.³¹ Equally interesting in the present context is the way Modleski, borrowing from Susan Stewart’s analysis of “miniatures,” describes the scene of voyeurism in Rear Window—the apartment building across the courtyard—as a dollhouse, a miniature which in Stewart’s words “tends toward tableau rather than narrative.”³² Although the contexts are obviously quite different, there is nonetheless a strong connection between this “tableau” and the figures of display which Gunning demonstrates as so central to the “other” visual pleasure of early film. The point I want to stress, then, is that figures of “primitive” narration may be far more influential and reverberate more strongly than most notions of classical narrative or the classical spectator would lead us to believe.

I am thinking, in this context, less of how dominant figures of narrative authority may be seen—following Burch’s argument—in relationship to their infantile counterparts, and more of how different kinds of narrative and visual agency may coexist in a single film. In Josef von Sternberg’s 1932 film Blonde Venus, for example, Helen Faraday’s (Marlene Dietrich) son, Johnny (Dickie Moore), acts as a “primitive narrator” and recalls Terrible Ted’s entry into the cinematic. The various male protagonists of Blonde Venus desire Dietrich but cannot possess her. Their desire finds a constant echo in the child’s desire to possess the woman. It has often been said that Blonde Venus never manages to portray convincingly the dual role of Dietrich—as mother and as performer.³³ In narrative terms, however, the function of the child as “primitive narrator” is to make visible the fragile connection between the two roles. There is a scene in the film where Johnny’s function occasions a very peculiar mise-en-scène. He and his mother are framed and theatricalized by the gauzy curtains in an archway of their apartment, and their discussion is quite literally decentered by the grotesque mask which Johnny wears perched sideways on the top of his head, suggesting just as literally another scene of representation.³⁴

In William Wyler’s 1940 film The Letter, two worlds are depicted, one “Western” and one “Asian,” with the character of Leslie Crosbie (portrayed by Bette Davis) located in a transgressive space between the two. The lawyer who defends Mrs. Crosbie at her trial for having murdered her lover functions as the most obvious principle of narrative authority in the film. But in the Asian sphere of the film, there is another controlling figure, the Chinese wife of Leslie Crosbie’s victim. The Asian woman is a “primitive narrator” in that she quite literally does not speak the language of the film, and embodies a mysterious power akin to that of a conjuror. She exercises an authority which the forces of Western civilization are never quite successful in dispelling. She may not play a joke on the lawyer, as Pearl does on her father in The Mind Cure, but she does function to question the neat hierarchy drawn in the film between East and West.

The function of the Asian woman in The Letter suggests, of course, that the notion of a “primitive narrator” has to do with Western constructions of the so-called primitive that extend much further than the early history of motion pictures, into the realm of cultural and racial difference. This could be the result of mere coincidence—the term primitive having “stuck,” where any one of a number of other words might have done as well. Not all historians of the period are happy with the word primitive, but it has become the most common shorthand phrase to refer to a variety of aspects of the early cinema.³⁵ This is not to say, however, that ideological reverberations of the “primitive” have been completely ignored by scholars of early film. But like most convenient shorthand terms, the word primitive is sometimes used without further attention to its implications.

Tom Gunning has observed that “the term primitive persists . . . partly out of inertia, but also because it cradles a number of connotations that stand in need of further examination and critique.” Gunning notes two such connotations, a “childish mastery of form in contrast to a later complexity,” and a “period of lack in relation to later evolution,” particularly insofar as the absence of editing is concerned. While the former, simplistic view of the primitive cinema is on the wane, the latter, more sophisticated view, Gunning says, presents its own series of problems: “Even those who maintain the uniqueness and value of early film within a nonlinear view of film history have a hard time avoiding a description of early cinema as a sort of degree zero in the evolution of montage.”³⁶ Indeed, whatever the degrees of difference between “childish mastery” and “lack,” both take as their point of reference and return the classical cinema, and the assumption that the early years of filmmaking both departed from and laid groundwork for the codes of linear continuity, narrative identification, and patterns of opposition and closure characteristic of the classical Hollywood cinema. As Gunning puts it, early cinema is “simultaneously different from later practices—an alternate cinema—and yet profoundly related to the cinema that followed it.”³⁷ To a considerable extent, then, how one defines the “primitive” in the early years of film history will rely on assumptions, both implicit and explicit, about the classical Hollywood cinema.

Still, one wonders about the extent to which a definition of the “primitive” cinema is inflected by the cultural meanings associated with the word primitive. Kristin Thompson, analyzing the relationship between the “primitive” and classical cinema, and alluding to the problems that inhere in the term primitive, writes:

The term “primitive” is in many ways an unfortunate one, for it may imply that these films were crude attempts at what would later become classical filmmaking. While I use the word because of its widespread acceptance, I would prefer to think of primitive films more in the sense that one speaks of primitive art, either produced by native cultures (e.g., Eskimo ivory carving) or untrained individuals (e.g., Henri Rousseau). That is, such primitive art is a system apart, whose simplicity can be of a value equal to more formal aesthetic traditions.³⁸

Thompson’s validation of the “primitive” cinema, as different from but not necessarily inferior to the classical cinema, draws explicitly on the modernist appropriation in visual arts of so-called primitive cultures as a source of inspiration and challenge to Western traditions.³⁹ The frame of reference here is stylistic, and Thompson does not really pursue the analogy in any depth. But again, one wonders if the “primitive” in “primitive” cinema has persisted, not only in terms of history and style but in terms of cultural preoccupations as well. The “scene” of “primitive” cinema is informed by a fascination with otherness, with the exotic, with all that is seemingly alien to Western culture and subjectivity. It has been an observation of long standing that the early cinema borrowed extensively, both in form and in subject matter, from popular arts of the time—vaudeville, magic shows, cartoons, stereoscope cards, dime fiction.

A popular theme in these arts was the encounter, both between the West and its colonized other, and between whites and blacks. It is no surprise that early film would replicate such encounters. The popularity of the Lumiére brothers’ scenes of village life in Africa suggests that film was particularly well suited to the circulation of racist stereotypes where the native “other” is portrayed as the object of a view. The relationship between the white woman and the black woman in What Happened in the Tunnel taps the construction of racial, and racist, difference as well. The cinematic representation of people of color as “primitives”—whether through the pernicious racial stereotypes of early films or the tourist’s view of “native life”—suggests, of course, that the frame of cinematic fantasy served to perpetuate and rationalize the distinction between a white Western self and its other.

But the “primitives” of early cinema include servants peeking through keyholes as well as black youths stealing watermelons; naive spectators such as Uncle Josh as well as Native Americans of comic-strip proportions. One common denominator to all of these representations of class, racial, and ethnic difference is childlike naivete. That some film historians have focused on modes of spectatorship in their revision of the early cinema suggests that here, too, an implicit assumption has been made that early film viewers brought with them only a “primitive” capacity to distinguish between illusion and reality. The supposedly terrified reaction of early viewers to the Lumière film of a train arriving in a station has remained a persistent anecdote of early cinema, suggesting that some film historians have understood not only early films but early viewers as captured within a naive, childlike state of reception.⁴⁰

The infantilism that inheres in the word primitive has a particularly strong psychoanalytic resonance in French, where “scène primitive” translates as the “primal scene.” While Noël Burch frequently refers to the early cinema as so many “primal scenes” of the cinema, his psychoanalytic terminology functions in a more general metaphoric sense, whereby the early cinema is a kind of “id” to the “ego” of classical cinema. For some psychoanalytic critics in France, however, the early cinema functions very literally to stage the primal scene.

For instance, Michel Marie takes as his point of departure Metz’s assertion that the cinema is made to the measure of reactivations of the primal scene. Marie refers to characters in early films, such as the voyeurs in Un coup d’oeil par etage and The Inquisitive Boots, as “metaphoric expressions of the infantile spectator, of the regressive adult, a perverse scopophiliac who escapes the universe of ‘normal’ sexuality . . . he is apparently not I, a lucid and conscious spectator, but the Other, the eccentric and repressed aspect that every adult rediscovers in going to the cinema.”⁴¹ It is true that early film offers some bold representations of classic psychoanalytic scenarios. Uncle Josh tears down the material of the screen (typical primal scene material) only once the parental couple have appeared on screen; Georges Méliès’s control of the female body speaks quite obviously to the threat of femininity. However, the possibility of an equivalence between film and the primal scene is based on an originary moment of cinematic identification, an assumption most prevalent in Metz’s work, which has been criticized strongly within psychoanalytic film theory.⁴²

Gunning’s analysis of the early cinema as a cinema of attraction establishes a context for a discussion, not only of the alterity of the early cinema but also of the project of those avant-garde filmmakers whose work is inspired by the thrill of visual discovery evident in early film. If the notion of exhibitionism is central to Gunning’s distinction between early and classical film, it also informs his reading of avant-garde practice. Drawing on Sergei Eisenstein’s concept of a “cinema of attractions,” based on principles of visual shock and confrontation, Gunning suggests that like early cinema, the avant-garde film inspired by it develops a relationship to the spectator grounded in “exhibitionist confrontation rather than diegetic absorption.”⁴³

The preoccupation with the “primitive”—understood in several senses—in the works of some women filmmakers seems to me related to what Gunning suggests, but different at the same time. Related, to the extent that women filmmakers have been drawn to the sense of visual discovery evident in early film; but different, to the extent that the reverberations of the “primitive” in the films I will discuss are not purely formal, and can in no way be described as an “exhibitionism” that downplays “voyeurism,” or a direct confrontation that circumscribes narrative. Rather, the inquiry into “primitivism” is very much connected, not to the dismantling or bracketing of narrative but to its reconceptualization. A number of women filmmakers have taken up the interrogation of the “primitive” on several different levels. In the next chapter I turn to an examination of the ways in which the “primitive” has been explored, revised, and recontextualized in different ways in women’s films.