ROSALIND KRAUSS, From Cindy Sherman: Untitled (1993) Some people have told me they remember the film

that one of my images is derived from, but in fact I had no film in mind at all. -Cindy Sherman1 Here is a curious story: an art critic writes an account of Cindy Sherman presenting her work to an art school audience. She shows slides of her "film stills"the black and white photographs in which as both director and actress she projects a range of 1950s screen images and next to each, he reports, she presents the real movie stills on which her own images were based. What emerges through this comparison, he says, is that "virtually every detail seemed to be accounted for: right down to the buttons on the blouses, the cropping of the image, even the depth of field of the camera."2 Although he is upset by what this comparison reveals about the slavishness of Sherman's procedure-the stroke- for-stroke meticulousness of the copy, so to speak- he is certain that what Sherman is after when we encounter these Stills is in any case a recognition of the original, although not as a source waiting to be replicated, but rather as a memory waiting to be summoned. So he speaks about the viewer of the normally unaccompanied Sherman Still "starting to recall the original film image." And, he says, "if it wasn't the actual film" the viewer recalled, "then it was an ad for it; and if not that, then it was a picture from a review in a newspaper. On its face this story is amazing. Because in a Sherman Film Still there is no "original." Not in the "actual film," nor in a publicity shot or "ad," nor in any other published "picture." The condition of Sherman's work in the Stills and part of their point, we could say- is the simulacral nature of what they contain, its condition of being a copy without an original. The structure of the simulacrum, then, along with Sherman's exploration of it, is something that needs to be examined. But even before doing so, it is worth staying with the story of the slide show and its putative unveiling of an "original," which is to say the story's blatant, screaming, Rashomon-like, misrecognition. Did Sherman ever show real movie stills next to her own work? And if so, to what end? Since her own images manage their projection of a whole array of stereotypical Hollywood or New Wave heroines, along with the very atmospheres through which they are castthe film noir's hard-bitten denizen of the night, one of Hitchcock's plucky but vulnerable career girls, the B-movie's small- town innocent swamped by Metropolis, a New Wave vehicle of alienated despair, and so onand yet do all of this from within a kind of intense, generalized memory, what would a comparison of, say, a still from a Douglas Sirk film and a Cindy Sherman mean? Could it indicate that the sense that the two images intersect-no matter how distant their actual details might be-derives from the way both Sherman and Sirk (in addition to Sirk's actress) are each imaginatively focused on a remembered fantasy-the same remembered fantasy-of a character, who is "herself" not only fictional, but, like Emma Bovary, the creature as well of fiction, a character woven from the tissue of all the romances she has ever consumed? Could it mean that with the stereotypes projected by these fictions, with regard to the creatures of this fantasized romance, could it mean that these boxes-within-boxes of seeming "memory" always produce what appears to be an authentic copy, even though there is no "real" original to be found? So that Sirk's copy and Sherman's copy uncannily overlap like two searchlights probing

through the night toward the same vaguely perceived target? Let's speculate that this is why Sherman would show her own image and, say, Sirk's. Why, then, would the critic misrecognize the comparison, making one a copy and the other an original: Sherman, the artist, copying the "real" of the Hollywood film? Roland Barthes, the critic, would have a word with which to explain this strange hallucination; and that word would be myth: the art critic who "saw" the comparison as replication-Untitled, Film Still = image taken from real filmwas in the grip of myth, consuming it, Barthes would say. Barthes would, of course, be using the term myth in a somewhat limited, rather technical way. And if it is useful to explain how he deploys the term, it's because myth is also what Sherman herself is analyzing and projecting in Untitled Film Stills. Although not as a myth consumer, like the critic; but rather as a mythographer, like Barthesa demystifier of myth, a de-myth- ifier. To consume a myth is to buy a package along with the salesman's pitch. The salesman's pitch names it, and the buyer, never looking under the hood, accepts the name, is satisfied (or suckered) by the pitch. The somewhat more technical analysis involves the terms signified and signifier, or form and content. It goes like this: a schoolchild reads in a grammar book quia ego nominor leo. 3 The signifiers of this string of words are the letters-the material component of the composite through which each sign (as here, each word) is made up; the signified is the lion and its namethe ideational content that is articulated by the units cut out by the signifiers: "because my name is lion." At the level of the individual sign the relation between signifier-letter-and signified- idea and their conjunction would look like this: Sd /Sr = Sign. But this sign, or string of signs, is found in a grammar book and thus "because my name is lion" is not left at what could be called the denotational level where it is pointing to lions, to their habitats, to their strength, as in, let us say: "If I have taken the prey from my weaker fellow animals, it is, among other things, because my name is lion." Rather the Latin phrase is being used as an example, a mere instance of the grammatical agreement between subject and predicate. And as such an instance, the richness of the signthe lion, its strength, its habitat, etc.is itself divided from within. And a second layer, parasitical on the first meaning, is installed. This second layer is formal; it is the subject/ predicate structure of the sentence in which grammatical agreement is at stakeany instance of agreement, lions, snakes, butterflies, no matter. This formal layer constituting the phrase as "mere" example is thus empty. But it preys on the fullness of the layer of the sentence understood as meaning. And Barthes's argument is that for myth to work, it must prey on it. So what is myth? Myth is depoliticized speech. Myth is ideology. Myth is the act of draining history out of signs and reconstructing these signs instead as "instances," in particular, instances of universal truths or of natural law, of things that have no history, no specific embeddedness, no territory of contestation. Myth steals into the heart of the sign to convert the historical into the "natural"something that is uncontested, that is simply "the way things are." In the case of "because my name is lion," the myth is the combination of meaning and form into the content that reads: "this is the principle of agreement in Latin." But beyond that the mythical content conveys the importance of order and regularity in the structure of Latin, as well as one's sense, as reader, of belonging to a system of schooling in which many children like oneself are also learning this principle, and the idea that this principle is addressed to oneself, meant for oneself. "See! This is what `grammatical agreement' looks like." This is what Barthes calls the interpellant aspect of mythical speech. 4 It is addressed to its readers, calling out to them, asking

them to see and agree to the way this example confirms this principle, at one and the same time fading before the principle's authoritythis is just an exampleand filling that authority with a kind of subservient but needed specificitysee! nature is brimming with just the thing this means: "because my name is lion." The more famous example Barthes uses in his analysis of mythical speech is one that is closer to Sherman's Film Stills, since it is not composed of letters and words but of a photograph and its depictions. It is a magazine cover of Paris Match in which a black soldier is shown giving the French salute. The photographas physical object, with its areas of dark and lightis the signifier; the depicted elements are the signified. They combine into the sign: a black soldier giving the French salute. That combination then becomes the support for the mythical content that is not just a message about French imperialism"France is a global nation; there are black subjects who also serve it"but a message about its naturalness, as the signified of the first order of the mythic support is called up as an example to fill up and instance its mythic contention: "Imperialism is not oppressive; it is natural, because we are all one humanity; you see! examples of how it works and the loyalty it engages can be found everywhere, anywhere, for example, in this photograph where a black soldier gives the French salute." The "you see!" part of the message is, of course, the interpellant part. It is the myth summoning its consumer to grasp the meaningfulness of the first order signthe photograph-as-signifiedand then to project his or her conviction in that unitary, simple meaning, onto the more complex, hazy, insinuating level of the contents of the myth. So let's go back to Sherman and the Rashomon- factor: the critic sitting there in the darkened auditorium of the School of Visual Arts, looking at a set of slide comparisons and believing something about their replicative relationship, believing this to be the case because after all Sherman's work, he is certain, takes us back in any event to the real film we remember. What is crucial here is that he has bought the pitch and never thought to look under the hood. He has taken the first order sign as a composite, a signifier and signified already congealed into a finished meaningactress X in film Yand he has completed the mythical content. Here it would be something like: Cindy Sherman is an artist and artists imitate reality (Universal Truth No. 1), doing so through their own sensibilities, and thus adding something of themselves to it (Universal Truth No. 2). The formula we come out with was penned by Emile Zola. It goes: Art is important; it gives us a piece of nature seen through a temperament. Nature in the Sherman case would be of a somewhat technological kind, namely, the original film role, which Sherman would pass through the temperament of her own memory and projection; she would externalize this observed and felt bit of the world, and her work of artthe externalization of these emotionswill be her expression, with which we as viewers can empathize. Art = Emotion relayed through nature. That's the myth and that's why the critic has to produceno matter through what process of self-deception or hallucination-the "original," the bit of nature, the filmic heroine in her role. That's what it's like to be a myth consumer. To buy the pitch. To fail to look under the hood. What, then, is under the hood? What is always under the hood is the signifier, the material whose very articulation conditions the signified. And further, working away under the hood, either on or with the signifier, is the effort perhaps to limit the possibility that it might produce a multiplicity of unstable signifieds and promote a sliding among them or, on the other hand, to do the reverse and welcome, even facilitate such sliding. Limitation is the work of realism in novels and films:

to every signifier one and only one signified. 5 Conversely, sliding and proliferation has always interested the antirealist (what used to be called the avant-garde) artist. 6 Work on the signifier is perfectly available for observation in Sherman's Untitled Film Stills. Take the group of images that includes #21, #22, and #23. In all three, Sherman wears the same costume, a dark, tailored suit with a white collar and a small straw cloche pulled onto a mop of short blond curls. But everything else changes from one still to the next: as in the first, #21, the register is close-up taken at a low angle; in the second, #22, a long shot intricates the character amid a complication of architectural detail and the cross fire of sun and shadow; and the last, #23, frames the figure in medium shot at the far right side of the image against the darkened emptiness of an undefined city street and flattened by the use of a wide-angle lens. And with each reframing and each new depth-of- field and each new condition of luminosity, "the character" transmogrifies, moving from type to type and from movie to movie. From #21 and the Hitchcock heroine to #23 and the hardened film noir dame, there is no "acting" involved. 7 Almost every single bit of the character-which is to say, of each of the three different charactersis a function of work on the signifier: the various things that in film go to make up a photographic style. It was just this that Judith Williamson, one of the early feminist writers on Sherman's work, described when she said that in the stills, "we are constantly forced to recognize a visual style (often you could name the director) simultaneously with a type of femininity. The two cannot be pulled apart. The image suggests that there is a particular kind of femininity in the woman we see, whereas in fact the femininity is in the image itself, it is the image."8 This fact that there is no free-standing character, so to speak, but only a concatenation of signifiers so that the persona is released, conceived, embodied, established by the very act of cutting out the signifiers, making "her" a pure function of framing, lighting, distance, camera angle, and so forth, is what you find when you look under the hood. And Sherman as demythifier is specifically allowing us, encouraging us to look under the hood, even as she is also showing us the treme ndous pull to buy into the mythwhich is to say, to accept the signified as finished fact, as free-standing figure, as "character." Thus there is the tendency when speaking of the film stills to enumerate their personae, either as the roles"a woman walking down a dark street at night; another, scantily clad, with martini in hand, peering out the sliding glass door of a cheap motel" 9 or as the actresses who project them: Gina Lollabrigida, Monica Viti, Barbara Bel Geddes, Lana Turner ... That neither the roles nor the actresses are free-standing, that all are, within representation, effects outcomes, functions of the signifiers that body them forth is what Barthes labored to demonstrate in his extraordinary book S/Z, an analysis of the inner workings of literary realism. Showing that each "character" is produced through a concatenation of separate codessome the signifiers or operators of difference, whether of gender (male/female) or age (young/old) or position (rich/poor); others the operators of references to general knowledge keyed into the text by the merest aside ("as in the Arabian Nights"); still others the operators of the puzzle that drives the narrative forward toward its Truth (who is? what is?)what Barthes makes clear is that when a naive finally arrives to refer to or denote a character, that na me is buoyed up, carried along, by the underlying babble of the codes."10 The name is thus the signifiedthe characterthat the author slides onto the codes to produce realism's appearance that for every name there is a referent, a denotation, a unified emp irical fact. What is being masked is that the name, rather than pointing to a primary entity in the "real," is an effect of the vast already-written, already-heard, already-read of the codes; it, the denotation, is merely the

last of these codes to be slipped into place. The consumer of realist fiction, however, buys the pitch and believes in the "character," believes in the substance of the person from whom all the rest seems to follow as a set of necessary attributes, believes, that is, in the myth. Most of those who write about the Film Stills acknowledge that Sherman is manipulating stereotypes and that though these are being relayed through a generalized matrix of filmic portrayals and projections, there is of course no real film, no "original," to which any one of them is actually referring. So the myth consumer of my opening anecdote is something of an exception and in that sense a straw man. And yet we have not far to look to find other versions of myth consumption, or the direct connection to the signified-as- instance. One form of this that can be found in the mountainous literature on Sherman's work is to assume that each of these signifieds is being offered as an instance of Sherman's own deeper self the artist (as in Universal Truth No. 2, above) becoming the vehicle through which the fullness of humanity might be both projected and embraced in all its aspects. Peter Schjeldahl, for example, understands the individual Film Still's signified to be Sherman's "fantasy of herself in a certain role, redolent usually of some movie memory," with all the different characters resonating together to form the totality of the artist's selfhood in her oracular role as "our" representative: "Sherman's special genius has been to locate the oracle not in the `out there' of media bombardment but in the `in here' of her own partly conditioned, partly original minda dense, rich sediment of half- remembered, half-dreamed image tones and fragments.... She has mined this sediment for ideas, creating an array of new, transpersonal images that spark across the gap between self and culture."11 The mythic content Schj eldahl then consumes from these instances of the self-as-oracle is that it is in the nature of the artist to organize "messages that seem to tell us our nature and our fate." Another forin of myth consumption is to continue to buy into the finished signified of the role, the "character, "but to see the multiplicity of these as various forms of what Arthur Danto seems to like to call The Girl. He provides his own roll call of these variants: The Girl in Trouble, The Girl Detective, The Girl We Left Behind, Daddy's Brave Girl, Somebody's Stenog, Girl Friday, The Girl Next Door, The Whore with the Golden Heart . . . But his point is that "The Girl is an allegory for something deeper and darker, in the mythic unconscious of everyone, regardless of sex.... Each of the stills is about The Girl in Trouble, but in the aggregate they touch the myth we each carry out of childhood, of danger, love and security that defines the human condition."12 Although Danto turns here to the term myth, he uses it not in the manner of the de-myth- ifier, but as the unsuspicious myth consumer: buying into the signified of every variant of The Girl as an instance of the myth that there is a shared space of fantasy, or what he himself provides by way of mythic content as "the common cultural mind."

... it is necessary to fly in the face of Sherman's own expressly non-, even anti-, theoretical stance. Laura Mulvey13 Not surprisingly, given the fact that Sherman's Film Stills focus exclusively on women, on the roles women play in films, on the nature of those roles as preset, congealed, cultural clichshence their designation as "stereotype"and by implication, on the pall that the realworld pressure to fill these roles casts over the fates of individual women, feminist writers have embraced Sherman's art, seeing it as "inseparable from the analysesand the challengeof

feminist work on representation." But even as they have done so, they have been disgusted by its consumption as myth. For such consumption, they point out, inverts the terms of Sherman's work, taking the very thing she is holding up for critical inspection and transposing it into the grounds of praise. 14 Arguing that there is, however, a logicno matter how perversebehind such a transposition, Abigail Solomon-Godeau sees a mechanism at work there to recut Sherman's art by exchanging what is dismissed as the narrow, somewhat threadbare cloth of feminist investigation for the more noble garments that drape the artist who addresses the humanity- ingeneral of "the common cultural mind." This, she reasons, is necessary to the art world's promotion of Sherman to the status of major artist, something incompatible with a feminist understand ing of her enterprise. Therefore, as an apparatus of promotion (in both the media and museums) has supplanted other kinds of writing about Sherman, the mythical reading of the meaning of her work has followed. And thus it is no accident that Danto would ne ed to recast the import of the Film Stills by insisting that they "are not in my view merely feminist parables."15 But it must be said that within feminism itself the import of the Stills has also been recast, a recasting that is articulated in Solomon-Godeau's essay, although buried in its footnotes. For if Judith Williamson's early treatment of the Film Stills appeared under the title "Images of Woman," Solomon-Godeau has now, eight years later, transposed this to "woman-as-image," and signaled to the reader the importance of this distinction. 16 Indeed, almost two decades of work on the place of woman within representation has put this shift into effect, so that a whole domain of discourse no longer conceives of stereotype as a kind of mass- media mistake, a set of cheap costumes women might put on or cast-aside. Rather stereotype itself rebaptized now as "masquerade," and here understood as a psychoanalytic term is thought of as the phenomenon to which all women are submitted both inside and outside representation, so that as far as femininity goes, there is nothing but costume. Representation itself films, advertisements, novels, and so forth would thus be part of a far more absolute set of mechanisms by which characters are constructed: constructed equally in life as in film, or rather, equally in film because as in life. And in this logic woman is nothing but masquerade, nothing but image. As Laura Mulvey has described this shift: "The initial idea that images contributed to women's alienation from their bodies and from their sexuality, with an attendant hope of liberation and recuperation, gave way to theories of representation as symptom and signifier of the way problems posed by sexual difference under patriarchy could be displaced onto the feminine."17 It was Mulvey's own 1975 text, "Visual Pleasure and Narrative Cinema," that most formatively set out that latter argument in which woman is constructed as spectacle and symptom, becoming the passive object of a male gaze. Which is to say that in her essay a relation is set up among three terms: (1) the observation that there are gender distinctions between the roles that men and women play in films males being the agents of the narrative's action; females being the passive objects or targets of that narrative, often interrupting the (masculine) action by the stasis of a moment of formal (feminine) opulence; (2) the conception that there is a gender assignment for the viewers of films, one that is unrelentingly male since the very situation of filmic viewing is structured as voyeuristic and fetishistic, its source of pleasure being essentially an eroticization of fetishism: "the determining male gaze projects its fantasy onto the female figure, which is styled accordingly," she writes; and (3) that these assignments of role are a function of the psychic underpinnings of all men and women, since they reflect the truths about the unconscious construction of gendered identity that psychoanalysis has brought to light:

"Woman ... stands in patriarchal culture as signifier for the male other, bound by a symbolic order in which man can live out his fantasies and obsessions through linguistic command, by imposing them on the silent image of woman still tied to her place as bearer of meaning, not maker of meaning."18 In that last sentence, which slides from the domain of filmic representations to the universal condition of how "woman stands in patriarchal culture," there are packed a large number of theoretical assumptions that knot together around concepts about the unconscious, castration, and the import of structural linguistics for psychoanalysis. Insofar as Sherman's work is implicated in those assumptions and the analysis about woman-as-image that flows from them the Film Stills, for example, repeatedly presented as either a text to be explained by this analysis and/or a consequence of itit is necessary to unpack these assumptions, no matter how schematically. The psychic economy that drives men to activity and speech and women to passivity and silence is an economy that also separates looking from being looked at, spectator from spectacle. And that economy is organized, according to this reading of psychoanalysis, around castration anxiety, which is to say in terms of an event through which the child is made aware of sexual difference and, in one and the same moment, socialized by being subordinated to parental law. And if difference and the law converge in a single psychic configuration, they do so in relation to a visual event in which the possibility of absence is verified in the body of the "castrated" mother, the woman from whose genitals the phallus can be seen to be absent. 19 Siding with the paternal law, the child chooses speech, for which the master signifier is now the emblem of difference itself the phallic signifier, the signifier as phallus. It is in this sense that Mulvey refers to the male as maker of meaning in contrast to woman as bearer of meaning, a bearer now because the lack she is seen as manifesting on her own body, insofar as it sets up the phallus as signifier which is to say a differential function through which the play of meaning now operates this lack is necessary to the social system of order and sense to which Mulvey gives, following Jacques Lacan, the name Symbolic. 20 Thus she writes, "An idea of woman stands as linchpin to the system: it is her lack that produces the phallus as a symbolic presence, it is her desire to make good the lack that the phallus signifies."21 If the economy of sexual difference sets up a division of labor in relation to language, it also produces a separation of roles, it is argued, in relation to vision. On the one hand cinematic pleasure is scopophilic, voyeuristic: it wants to see and to control its objects of sightbut at a distance, protected by its own remove in the dark and at a point of vantage that perspective triangulates for it, the occupant of this point guaranteed, through this visually unified position of control, a sense of its own (phallic) mastery. On the other hand this pleasure is put in jeopardy by the very image of the woman it wishes to master insofar as that woman is marked as well as the bearer of the threat of castration. Thus it is necessary for this spectator to convoke the psychic mechanism of denial, for which the classic psychoanalytic instance is fetishism: the male child entering a perversion in which he sees the proof of sexual difference but continues nonetheless to believe in the woman as "whole," not-castrated: the phallic mother. The fetish constructed through this mechanism of denial thus restores to her body what is known to be "missing." If film works constantly to re-create woman as a symptom of man's castration anxiety thus silencing herit also works, and here even harder, to situate her as eroticized fetish: the image of lack papered over, the emblem of wholeness restored. Woman is in this sense skewered in place as an image that simultaneously establishes her as other than manthe Truth that it is he

who possesses the phallusand at the same time the fetishized image of the whole body from which nothing is missing. Stephen Heath describes this visual scenario from the point of view of the gazing male subject "Everything turns on the castration complex and the central phallus, its visibility and the spectacle of lack; the subject, as Lacan puts it at one point, looks at itself in its sexual member and then for the consequences for the woman secured as spectacle: What the voyeur seeks, poses, is not the phallus on the body of the other but its absence as the definition of the mastering presence, the security, of his position, his seeing, his phallus; the desire is for the other to be spectacle not subject, or only the subject of that same desire, its exact echo.... Fetishism too, which often involves the scopophilic drive, has its scenario of the spectacle of castration; and where what is at stake is not to assert that the woman has the penisphallus but to believe in the intact, to hold that the woman is not castrated, that nothing is lost, that his representation, and of him, works. Always, from voyeurism to fetishism, the eroticization of castration. 22 It is with this theoretical armature in place, then, that Laura Mulvey herself looks at the Film Stills, understanding them to be rehearsing this structure of the male gaze, of the voyeurist constructing the woman in endless repetitions of her vulnerability and his control: "The camera looks; it `captures' the female character in a parody of different voyeurisms. It intrudes into moments in which she is unguarded, sometimes undressed, absorbed into her own world in the privacy of her own environment. Or it witnesses a moment in which her guard drops as she is suddenly startled by a presence, unseen and off screen, watching her."23 And yet, we could say, it is this very theoretical armature that operates in such a description to put a mythic reading of the Film Stills in place, one that is not taking the trouble, indeed, to look under the hood. Judith Williamson had seen the constructed role emerge in the Stills as a consequence of the signifiers through which any filmic image must be built"the two cannot be pulled apart," she had said; Laura Mulvey, on the other hand, is buying into a signified-as-instance, a congealed sign, the semantic totality that reads "woman-as- image," or again, "woman as object of the male gaze." Sherman, of course, has a whole repertory of women being watched and of the camera's concomitant construction of the watcher for whom it is proxy. From the very outset of her project, in Untitled Film Still #2 (1977), she sets up the sign of the unseen intruder. A young girl draped in a towel stands before her bathroom mirror, touching her shoulder and following her own gesture in its reflected image. A doorjamb to the left of the frame places the "viewer" outside this room. But what is far more significant is that this viewer is constructed as a hidden watcher by means of the signifier that reads as graininess, a diffusion of the image that constructs the signified /distance/, a severing of the psychic space of the watcher from that of the watched. In Untitled Film Still #39 (1979), it is not so much the grain of the emulsion that establishes the voyeuristic remove, with its sense that one is stealing up on the woman, as it is a kind of nimbus that washes around the frame of the image, repeating in the register of light the sense of barrier that the door frame constructs in the world of physical objects. But in Untitled Film Still #81 (1979) there is a remarkably sharp depth of field, so that such /distance/ is go ne, despite the fact that doorways are once again an obtrusive part of the image, implying that the viewer is gazing at the woman from outside the space she physically

occupies. As in the other cases, the woman appears to be in a bathroom and once again she is scantily dressed, wearing only a thin nightgown. Yet the continuity established by the focal length of the lens creates an unimpeachable sense that her look at herself in the mirror reaches past her reflection to include the viewer as well. Which is to say that as opposed to the idea of /distance/, there is here the signified /connection/, and what is further cut out as the signified at the level of narrative is a woman chatting to someone (perhaps another woman) in the room outside her bathroom as she is preparing for bed. The narrative impact of these images tends to submerge the elements through which it is constructed, elements such as depth-of-field, grain, light, etc. which, it would seem, are too easy to dismiss as merely "formal" integers, whereas they function as signifiers crucial to the semantic effect. That Sherman is concentrated on these aspects is made very palpable in the one Film Still that seems inexplicable within the series as a whole: #36 (1979). Of all the Stills this one is so severely backlit that nothing can be seen of the character's face and almost nothing of her body beyond its silhouette. Standing in front of a curtain through which the powerful backlighting is dramatically diffused, she extends one of her arms upward almost out of frame; the other bends to grasp the elbow of the first in what could be a gesture of washing but remains radically ambiguous. As pattern her body reads black on the white of the ground, and her garmentsthe bodice of her slip and the stiffened film of a crinoline parted slightly from her body, create the only area of modulation or middle tone in the image. To a far greater degree than almost any other in the series, this work is deprived of narrative implication. A few months prior to the making of this Still, an imageor rather two images remarkably like it were published: two photographs by Edgar Degas, of a ballerina dressed in a low-cut bodice, her skirt a diaphanous crinoline, standing in front of a luminous curtain and reaching with one arm upward, her other arm bent inward at the elbow. These photographs, published by a critic who just a few months later would launch Sherman in an essay called "Pictures," an article providing the first serious critical context for her work (Sherman's first solo exhibition was still one year away), are related to one another through an extraordinary ambiguity with regard to light. 24 For having solarized the negative of his photograph to create reversals between negative and positive areas within the image, Degas then created both a negative and a positive print. And the dark/light reversals that arise from this treatment constitute the dancer as a phantom whose existence can be located nowhere. As Douglas Crimp described it: In the print in which the right arm and torso of the dancer appears to be normally positive, the shadow of the arm on the wall she grasps appears as a streak of light. Her face, also apparently in shadow, and her "dark" hair are registered as light. At this point, obviously, language begins to fail. How can we any longer speak of light and dark? How can we speak of a white shadow? a dark highlight? a translucent shoulder blade? When light and dark, transparency and opacity, are reversed, when negative becomes positive and positive, negative, the referents of our descriptive language are dissolved. We are left with a language germane only to the photographic, in which the manipulation of light generates its own, exclusive logic. 25 And in the publication of the twinned Degas photographs, the same dancer turns to confront her own mirror image as, flipped from negative to positive, she is also flipped left and right. Folded

in a way almost impossible to imagine around the axis of her own body, that body is folded as well around a ghostly condition of luminosity that produces it now as solid, now as if in X-ray. Shrman's Untitled Film Still #36, in its condition of being hors serie, has also the aura of this impossibly folded Degas dancer, turning in a light that has no focus, and indeed no possible external point of view. Perhaps the Still was addressed, imaginatively, to Crimp; but such an address has nothing in it of the theorization of the male gaze and the psycho-politics of sadistic control. Further, as we will see, this kind of backlighting, and all that it does to fragment the gaze, will emerge as a crucial element or signifier in Sherman's work of the early 1980s. But that is to anticipate somewhat, getting ahead of our story. NOTES:
1. Lisbet Nilson, "Q & A: Cindy Sherman," American Photographer (September 1983): 77. 2 Richard Rhodes, "Cindy Sherman's `Film Stills,"' Parachute (September 1982): 92 3 Roland Barthes, Mythologies, trans. Annette Lavers (New York: Hill and Wang, 1972): 109-159 4 Rosalind Coward and John Ellis, Language and Materialism (London: Routledge and Kegan Paul, 1977), 43-44, 47. 5 Ibid., 125. 6 In invoking the metaphor of the used car saleman and the buyer who does or doesn't look under the hood, I am perhaps implying that the myth's manipulation of signifiers and signifieds is somehow concealed. But it is important to emphasize that it is wholly visibile, out in the open. As Barthes says: "This is why myth is experienced as innocent speech: not because its ntentions are hidden-if they were hidden, they could not be efficacious-but because they are naturalized" (Mythologies, 131). 7 Another similar series-not sequential in a narrative sense but simply grouped around the same costume-is comprised of stills # 17-20. 8 Judith Williamson, "Images of `Woman,"' Screen 24 (November 1983): 102; she quotes Jean-Louis Baudry, "The Mask," Afterimage 5 (Spring 1974): 27 9 Lisa Phillips, Cindy Sherman (New York: Whitney Museum of American Art, 1987): 14. 10 Roland Barthes, S/Z, trans. Richard Howard (New York: Hill and Wang, 1974): "What gives the illusion that the sum is supplemented by a precious remainder (something like individuality, in that, qualitative and ineffable, it may escape the vulgar bookkeeping of compositional characters) is the Proper Name, the difference completed by what is proper to it. The proper name enables the person to exist outside the semes, whose sum nonetheless constitutes it entirely" (191). 11 Phillips, Cindy Sherman, 8. 12 Arthur Danto, Untitled Film Stills: Cindy Sherman (New York: Rizzoli, 1990), 14. 13 Laura Mulvey, "A Phantasmagoria of the Female Body: The Work of Cindy Sherman," New Left Review 188 (July/August 1991): 137. 14 Abigail Solomon-Godeau, "Suitable for Framing: The Critical Recasting of Cindy Sherman," Parkett 29 (1991) : 112. 15 Danto, Untitled Film Stills, 14. 16 Solomon-Godeau, "Suitable for Framing," 115. 17 Mulvey, "Phantasmagoria of the Female Body," 139. 18 Laura Mulvey, "Visual Pleasure and Narrative Cinema," Screen 13 (Autumn 1975): 6-18; republished in Laura Mulvey, Visual and Other Pleasures (Bloomington: University of Indiana Press, 1989) : 15. 19 The sequence of texts in which Freud develops this scenario begins with "Infantile Genital Organization of the Libido" (1923), "The Passing of the Oedipus-Complex" (1924), and "Female Sexuality (1931). In the 1925 essay, "Some Psychological Consequences of the Anatomical Differences between the Sexes," the scenario takes a different form; it stresses the sense in which meaning does not arise in the presence of the visual field but is only retrojected on it as a result of a verbal prohibition: "When a little boy first catches sight of a girl's genital region, he begins by showing irresolution and lack of interest; he sees nothing or disowns what he has seen ... It is not until later, when some threat of castration has obtained a hold upon him, that the obervation becomes important to him: if he then recollects or repeats it, it arouses a terrible storm of emotion in him and forces him to believe in the reality of the threat."

20

. In their introductory essays, Juliet Mitchell and Jacqueline Rose present the development from the scenic event described by Freud to its subsequent semiological elaboration by Lacan. See Mitchell and Rose, eds., Feminine Sexuality: Jacques Lacan and the Ecole Freudienne (New York: Norton, 1982). 21 Mulvey, "Visual Pleasure," 14 22 Stephen Heath, "Difference," Screen 19 (Autumn 1978): 89. 23 Mulvey, "Phantasmagoria," 141. 24 Pictures was the title of an exhibition organized in the fall of 1977 by Douglas Crimp for Artists Space, New York, which focused on work structured around the issue of replication, work that thereby could bring notions of representation into question. The five artists included Troy Brauntuch, Jack Goldstein, Sherrie Levine, Robert Longho, and Philip Smith. Crimp's connection to these issues continued and led to an essay that enlarged the circle of "pictures" artists to include Cindy Sherman. See Douglas Crimp, "Pictures," October 8 (Spring 1979): 75-88. 25 Douglas Crimp, A Note on Degass Photographs, October 5 (Summer 1978): 99.