Representation THE WORLDS WE REPRESENT In the second chapter we saw that the seemingly simple substratum of the cinema, visual perception, is an immensely complicated and disputed concept. In contrast, the issue of representation which stands before us now as the next level to be treated has never been thought ot as simple by anyone and has been an explicit battleground lor competing theories of the cinema. It will be even less possible here to present a satisfactory summary of views and arguments surrounding this issue, so vast is it. touching even upon the nature of thinking itself. Hut we can highlight and isolate the special conditions ot representation which govern the cinema and the peculiar questions which the cinema raises as questions of representation. Amidst all the varying types of experiments with perception, barely outlined in Chapter 2. there dominates a nearly univocal belief in the importance of '•attention" in visual life. Only acts of cognitive expectation permit our eyes to move and focus in such a way that we see images. D. W. Hamlyn. berating all mechanistic discussions of perception, including even Gestalt psychology, demands that we study not just the eye. the stimuli, and the neural patterns of the brain, but the general conditions at play in any moment of perception.' Our eyes work differently in different circumstances, literally forming different images depending on the expectations which guide their use. 38 Concepts in Film Theory Given this framework, we would have to say that the general circumstances ot perception lor the cinema spectator seem quite limited and specitic in the lirst instance. We enter a theater and stare in front of us at a two dimensional screen for two hours. Yet within this strait jacket our eyes expect to coagulate film grain into shapes, objects, actions, and scenes: more important they expect to do so in ways which mimic the nearly unlimited viewing circumstances of life in the world. Cinema perception is a mode of "seeing as" wherein we see an array of light and shadow as a particular object and we see several hundred fragments of a full film as a particular world. Far from being a rare occurrence in perception, or a particularly devious one, cinema here joins myriad other instances of "seeing as." instances in which we notice an oscillation between what our senses deliver to us and how we identify this. Certainly the most startling cases of this involve illusions, but as 1Z. H. Gombnch, Nelson Goodman, and others have stressed, this structure of experience is ubiquitous.: In daily life we are prone to identify geometric patterns ot stimuli (an oblong, lor example) as objects named by a different geometrical ligure (a round table, set obliquely to our eves). If this is the case for veridical perception, how much more pervasive is "seeing as" lor explicitly judgmental visual acts which organize percepts into coherent wholes. We identity a set of varied stimuli not onlv as human beings, but as a group we call "the class" and oppose it to another blend of stimuli which we name "the teacher." Our experience, in short, does not merely add to our perception, it makes perception possible, for we perceive inferentially. Goodman has pursued the consequences ot these observations to the end. arriving at a pluralistic and nominalist philosophy which makes explicit use of art. There is no primary real world which we subsequently subject to various types of representation, he contends.' Rather it makes far more sense to '-peak of multiple worlds which individuals and groups construct and live within. Worlds are comprehensive systems which comprise all elements that lit together within the same horizon, including elements that are before our eyes in the foreground of experience, and those which sit vaguely on the horizon forming a * background. These elements consist of objects, feelings, associations, and ideas in a grand mix so rich that only the term "world" seems large enough to encompass it. Goodman is fond of using art as an explanatory model for his notion of "world." We step into a Dickens novel and quickly learn the types Representation 39 of elements that belong there. The plot may surprise us with its happenings, but every happening must seem possible in that world because all the actions, characters, thoughts, and feelings represented come from the same overall source. That source, the world of Dickens, is obviously larger than the particular rendition of it which we call Oliver Twist. It includes versions we call David Copperfield and The Pickwick Papers too. In fact, it is larger than the sum of novels Dickens wrote, existing as a set of paradigms, a global source from which he could draw. Cut out from this source are anachronistic elements like telephones or space ships, and elements belonging to other types of fiction (blank verse, mythological characters, and even accounts of the life of royalty). It should be clear that even such a covering term as "The World of Dickens" has no final solidity or authority. A young reader of David Copperfield and Oliver Twist might consider these texts to be versions of a world of education and family relations which concern him outside of literature. The Dickens scholar naturally would consider these texts to be pan of the complete writings of Dickens. What thev represented for Dickens himself, who lived within them during the years of their composition, no one can say. One goal of interpretation has always been to make coincide the world ot the reader with that of the writer. Although not a futile enterprise, the difficulties of accomplishing it. or ot knowing that it has been achieved, arc obvious. Artworks are indeed suitable examples of worlds and worldmaking. for they are cut off in time and space from our everyday life. Not only is "The Woman Weighing Gold" a world within a frame which can hold a viewer's entire attention, so also is the Vermeer room in a museum featuring his work. The museum itself is a kind of world that we enter and leave bringing with us expectations, memories, particular codes of behavior, and a very special type of perception. But out on tne bustling street we likewise live in a world divided by comprehensive types of interest. For most of us the world of politics exists as a separate sphere to which we occasionally attend. This is an immense world frequently represented for us on the news or in papers. The New York Times editorial on "detente" is a version of part of this world as is the rebuttal of this version printed the next day in Pravda. Whatever encompasses our attention is a world we have constructed to live within. Whatever organizes our sense of that world or of some portion of it is a version: and versions we call representations. 40 Concepts in Film Theory THE WORLD OF AND IN FILM Goodman's formulation makes it possible to speak of standard sense perception as "representational" in that each percept consists not only of its own quality but also of an indication pointing to the world to which it belongs. "This is a chair in the dining room" or "this is a swarm of molecules" is an equally true statement pertaining to a single ocular impression which the physicist had as he came down to breakfast. The first statement tits into his domestic world and the second into his professional world. Nor can we say that one statement is tmer than the other, if both are in fact true to the worlds in which they belong. The philosophical issues here go back centuries and can hardly be solved in this chapter. Does the Eskimo actually live in a world of multiple cold, white substances that we identify grossly and simply as snow? Goodman refuses to accord priority to the world of the chemist for whom such substances are particular definable states of the H20 molecule.4 Whether we agree with him or not. it is enough that recent philosophy has provided us with the room and the terms to permit a subtle description of the processes and effects of an in general and ol the cinema in particular. Fortuitously, the relevant issues that crystallize around the notion of "world" derive not uist from Ansilo-Amencan language philosophers like Goodman but 1mm continental phenomenology. Sartre's writings on the imagination. Alfred Schutz's sociology of "life-worlds." and Mikel Dutrenne's "Phenomenology ol the Aesthetic Experience" give weight to the common parlance of film critics who have always been comfortable with phrases like "Chaplin's world" or "The World of Citizen Kane."'' Instinctively we have cut off from our other experiences the special sensibility, gestures, and objects that belong to Chaplin's lilms or that tit into the kind at sepulchral space exemplifed by Citizen Kane. More generally theorists and the average spectator have cut off from ordinary life the world that exists within the movie theatre. "The World of Film" suggests the mechanism by which anything reaches the screen and. on reaching it. affects us. Instead of being a catalogue of things appearing on the screen (as in the Chaplin and Kane examples) "the world of film" is a mode of experience, rather like "the world of imagination." How does the cinema represent anything for us? In trying to answer this question Goodman advises us not to measure the ade- Representation 41 quacy of our representations against some supposed "reality" existing beyond representation but to isolate and analyze the peculiarities that make up the representational system of the cinema and that make its effects distinctive. Now the first elements of cinematic representation are perceptual. Earlier we discussed the tension of belief and unbelief in cinema as equivalent to the oscillation between looking and seeing or seeing and recognizing which is the integral structure of perception in sieneral. It is this equivalence that permits the casual, though philosophically naive, claim that "reality" is rendered in cinematic perception. More accurately we should say that the structure of cinematic perception is readily translated into that of natural perception, so much so that we can rely on information we construct in viewing tilms to supplement our common perceptual knowledge (which is also, as we have often noted, constructed knowledge). This explains the confidence that jurors place in cinematic records submitted by a lawyer, or that astronomers have in video images sent back from Mars, or that ethnologists have in footage brought back by explorers to distant lands. In all these instances cinematic information supplements what we know about one or another ol the worlds we inhabit. To some degree the tension between belief and doubt operates in even,' iconic Mgn system: the cinema, still photography, drawing, painting, and so on. In each of these an image strives to produce the effects of natural perception through a process quite different from natural perception. We effectively recognize our friend in an image processed by Kodak. II cinema heads our hierarchy of such sign systems, so that the jurv accepts a filmed record of the murder but rejects a drawing by an eyewitness and even a still photograph, it is due to cinema's mechanical and temporal aspects. The automatic registration of light on celluloid involves us in squinting at the image to "make out" the object in the glare and the grain (whereas a drawing could be much more clear). And the temporal flow which throws us from one imaae to the next demands that we adjust our recognition of what we see to the overall image which organizes itself gradually before us. But it is just this work that makes us assent to the film image, for ordinary perception involves precisely the same types of work even if the actual visual cues (the stimuli) are somewhat different. So at its basis cinema may be said to represent the numerous objects signified in light and shadow 42 Concepts in Film Theory over the course of an hour or two. But cinematic representation is more than a sequence of photographs, for the thousands of photoqrammes meld into pictures of scenes enduring over time. Instinctively we strive to put disparate scenes together so that the entire projection coheres. Thus, from the automatic operation of the phi phenomenon which produces movement out ot static and separated phoio^rammes to the classification of an entire lilm. the mind actively constructs images trom the light that stimulates it. At the tirst level the percepts we identify in the llowmg grain depend in a major way on our expectation that they will contribute to the larger representation which is at stake in the lilm. These still images then become animated and begin to pull us through the lilm along what Béla Balázs called a current of induction" toward a linal representation. It is this ultimate sense of a developing representation that makes the individual ph(tio\>rainmes readable and that likewise assures their smooth linkage in montage. Vet what is this linal representation other than a construct built up of the individual fragments it supposedlv makes comprehensible? Just as the basic percept ot cinema is a unit constructed out of light and shadow on lilm grain, so the entire cinematic representation is a major unit our mind puts together. More important, the structure of cinematic representation Irom beginning to end is one ol process, where fragments are ruled by the wholes thev add up to. and where belief and unbeliet keep our eyes on ihe screen while our mind glides into the world of the representation. Quite siinplv the oscillation at the heart of all instances of "seeing as" becomes in the cinema a vacillation between belie! ami doubt. Hie cinema fascinates because we alternately take it as real and unreal, that is. as participating in the familiar world of our ordinary experience yet ihen slipping into its own quite different screen world. Only an unusually strong act of attention enables us to focus on the light, shadow, and color without perceiving these as the objects they image. And. on ihe other side, only an equally strong hallucinating mode of attention can maintain from beginning to end the interchangeability of what we perceive and the ordinary world, negating all difference of image and referent. Cinema would seem to exist between these two extremes as an interplay between '"the real and the image." The lilm experience in general and every instance of viewing a lilm can be analyzed in terms of a ratio between realistic perceptual cues and cues which mark an effort and type of abstraction. Representation 43 Contributing to the sense of reality (of immediate apperception and non-mediation i are at least four elements, some of which Christian Metz outlined in his earliest writing. 1. Experimental preconditions, such as the darkened auditorium. 2. Analogical indices such that the image of an objeci shares actual visible properties with its referent. 3. The psychological imitation which cinematic tlow provides of the actual How of reality. Importantly, movement in the cinema is actual movement, not represented movement, and our mind is brought alive by it. 4. finally, ihe lure of sound, which establishes a second sense to verily ihe tirst and which analogically is more exact than image representation.7 All ol these characteristics tend to put us in front of a turned imatfe as if we were in front of a real scene in life. What keeps us from accepting the image as life is a lissure which we sometimes leap, sometimes reluse to leap, and most often straddle. Consisting of such experiential counters as bodily immobility, of nonanalogic aspects such as foreshortening, and of the more basic fact that the scene has been put before us by another, these anti-illusionistic elements lead us to treat the lilm not as life but as an image in the Sartnan sense, as a presence of an absence.' All films present themselves to us as real/image according to various ratios. To move across ihe bar is to shift intentionality in a manner not unlike what happens in iigure/ground experiments. Reality is here taken to be a type of consciousness characterized by certain indices of appearances and a certain mental activity. To shift to the imaginary is to move, as in daydream, to another "realm" while still adhering to manv of the phenomena associated with our reality state. The crucial marker of this particular experience of oscillation is the frame itself. The frame is the physical embodiment of the bar between image/reality and it marks as well the case that this experience is presented to me by another. I must attend "there" to the frame and not elsewhere. Classically slated, the screen as "window" is a place of perception: as "frame" or border it delimits and organizes perception tor signification. Jean Mitry saw this long ago." The frame keeps us off our guard. We search the screen as we search any perceptual Held, yet we feel the force of "this particular" disposition of objects and shapes. The superfluity of the facts of the visible 44 Concepts in Film Theory world imprints itself on every image, but the frame demands selectivity and motivation. We are given over to the world, yet we are given over to signification. Nor is this the end of it. for the image changes before our eyes; both the film and the world move on. The fact of movement introduces the category of narrative or. at least, its possibility. For while the framed image dissolves betöre us and the vibrant life of perception is reaffirmed, this How engages a narrative íntention-ality marked by retraining and shot changes. Although we perceive the dissolution of every scene, we group scenes into events that are not allowed to fall away but are held together as on a chain. From the angle of phenomenology, narrative refers to a type of consciousness into which audiences lock themselves when attending to the chain of movement in a tilm. It involves a particular form of image processing wherein sensations are read as significant in their temporal and causal interrelation. The study of narrative in cinema ought there-lore to begin with a determination of our relationship to the images and to the current of induction which runs through them, pulling us alter it. Such determinations would amount to genre studies if we formalized their results, since they would name and describe the customary relation into which spectators lapse (or against which they struggle) with regard to the lilmed material and its organization. If every film is a presence of an absence, we are still obliged to differentiate the types ol imaginary experience possible within various ratios of this relationship. A lilmed image may be considered the presence ot a reterent which is absent in space (live TV coverage) or in time (home movies). It may also be taken to be an image which is non-existent or whose existence is not in question one way or the other. Consciousness immediately makes decisions about the status of the image and from these decisions it processes the lilmic How in different ways. If the absent referent is deemed nonexistent we attend to the peculiarities of the image, necessarily striving to give existence to an unknown. If. on the other hand, the absent reterent has solidity for us (as a Iriend or a public figure in whose existence we believe), we may utilize our recognition ol the image to launch our consciousness into a state which calls up a mise-en-scene of the imaginary, producing nostalgia, desire, and the like.'" In this way we can consider our relation to the tlow of various types of movies. In the home movie situation each point interests us not as an accumulation of a past (retention) throwing us into a necessary fu- Representation 45 ture (protention), but only as a potential triggering device allowing a shift of consciousness. We wish to transcend the home movie by means of one or two of its images and attain a more private state. In other words, the intention of "conjuring up the past" lords it over the basic intentionality of "movement," using the life of movement to restore the dead past. Our frequent recourse to still-frame and creep-speed protection techniques certifies this hierarchy. Documentaries achieve a variety of ratios of presence/absence or image/referent. Since in most cases we know and believe something about the referent and its world, the documentary can sometimes serve the imaginary function already described in relation to home movies. We use and discard a hundred minutes of the Rolling Stones in order to recognize those live minutes that are sufficient to launch us into a reverie. The sound track in such a tilm already guarantees this sort of response. But if the tilm is about an obscure woodcutter of the Northwest, we must attend to the specifics of the image and try to buiid a sense of a world about which we know little even though we may have '"faith"' in it. Every documentary relies on our faith in its subject and. more important, utilizes our knowledge of it. Barbet Schroeder's portrait of Idi Amin" summarizes a good deal of data through voice-over narration in its first five minutes, but otherwise forces us to process the images of Idi within a field of consciousness already full of the Idi story. Indeed like many documentaries. Schroeder's film was under little compunction to achieve formal closure since his subject would continue to survive and his spectators would in fact have a greater understanding of the denouement of his tilm than he possibly could have had in 1973, not knowing Idi's final atrocities. Every fictional film likewise relies on some substratum of spectator understanding of the type of world that becomes the subject of the film. We bring our own sense of boxing to Rocky and of the strictures of bourgeois life to any Douglas Sirk film. But the fictional film, at least in most of its genres, quickly transfers our interest to the world of the image, calling on. but not playing to. our knowledge of its referent. In the fiction film all moments become significant as we construct a reterent whose absence is determinant, not merely accidental or logistical. Movement in fiction film is coterminous with the film itself. The viewer is asked to swim in a time stream, and he cannot look away without the fiction threatening to disappear. As Hugo Münsterberg noted fifty years ago.12 our mental flow coincides with the filmic flow in those 46 Concepts in Film Theory fictions that produce the strongest mental events. Whereas the techniques and codes that construct the illusion of the continuity oť movement in the fiction film may be the product of history and labor (mav change from era to era), the mode of consciousness by which spectators have always participated in the construction of a fiction is ahistor-ical and transcendental to the degree that it stems from certain conditions of perception and cognition operating in the everyday life world (conditions such as retention, protention, filling in. and so forth). It is for this reason that those filmmakers who break the cinematic flow (Godard. for instance) need to labor to do so, for they thwart the mind in its act of seizing something that seems to disappear for it when stopped. Among fiction films themselves we can categorize different ratios of perception to signification and begin to list genres and styles as we do so. Nashville and ľaisa affirm an overbrimming perceptual flux out of which certain stones have eddied. The Third Man and Rosemarx's Babv, on the other hand, construct tight networks of signification which wither all but certain perceptual possibilities. In all fictional cases we appropriate the situation of the narrator by succumbing to the film flow in the proper wav. Propriety vanes from genre to genre, from ľaisa to The Third Man. but the demands of narrative consciousness remain— demands that include its drive toward totalization, identification, explanation—even while these demands operate in different ways for each genre. Some of the differences amongst genres and films can be catalogued as functions of the imagination. The supplying of background information is negligible in the standard Western for our minds instantly fill the horizon of these films with the appropriate atmosphere, landscape, and props. Iíut in a film like Wind Across the Everglades or Dersu Uzala. both of which depend crucially on the relation of atmosphere and landscape to character and both of which are set in landscapes unfamiliar to most lilmviewers. the filmmakers must continually offer background shots, through composition in depth, pans away from action, and descriptive exposition. The film noir. to take another genre and another aspect of film construction, frequently employs both voice-over narration and returns to past action. The viewer is asked to gauge the action represented on the screen in relation to an overall judgment which is. so to speak, simultaneously present with the action. In standard gangster films, on the Representation 47 other hand, the straightforward, third person approach to the action asks us to project the end of the film (the death of the gangster) in the actions he sequentially institutes. The film noir hero, on the contrary, not only appeals to us through first person address, but speaks from a point where the action has reached its end. More modernist narratives like