Nội dung

The Tutor-Code of Classical Cinema Author(s): Daniel Dayan Source: Film Quarterly, Vol. 28, No. 1 (Autumn, 1974), pp. 22-31 Published by: University of California Press Stable URL: http://www.jstor.org/stable/1211439 Accessed: 16/06/2010 18:00 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ucal. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. University of California Press is collaborating with JSTOR to digitize, preserve and extend access to Film Quarterly. http://www.jstor.org 22 THETUOR-CODE F CLASSIALCINEM 22 THE TUTOR-CODEOF CLASSICALCINEMA clusions about statistical style analysis can be arrivedat. However, the resultsso far are based on more objective facts than have ever been used in the field of style comment before. The methods used can obviously be applied also to sections of a film when one is consideringthe interactionsbetween, and relationsof, form and content. And they can decide questionsof attribution, such as who really directed The Mortal Storm, Borzage or Saville? A few hours with a film on a moviolais alwaysmore instructivethan watching a second screeningof it, and then retiringto an armchairand letting one's imagination run riot. NOTES 1. H. B. Lincoln (ed.), The Computer and Music, Cornell, 1970; Dolezel and Bailey (eds.), Statistics and Style, Elsevier, 1969. 2. A. Sarris, The Primal Screen. Simon & Schuster, 1973, p. 59. 3. American Cinematographer,December 1972. DANIEL DAYAN The Tutor-Code of Semiology deals with film in two ways. On the one hand it studies the level of fiction, that is, the organizationof film content. On the other hand, it studies the problemof "filmlanguage," the level of enunciation. Structuralistcritics Classical Cinema languageis to literature. Linguisticstudies stop when one reaches the level of the sentence. In the same way, the system analyzedbelow leads only from the shot to the cinematographic statement. Beyond the statement, the level of enunsuch as Barthes and the Cahiers du Cinema of ciation stops. The level of fiction begins. "Young Mr. Lincoln" have shown that the level Our inquiry is rooted in the theoretical work of fiction is organizedinto a language of sorts, of a particular time and place, which must be a mythicalorganizationthroughwhich ideology specified. The political events of May 1968 is produced and expressed. Equally important, transformed reflection on cinema in France. however, and far less studied,is filmic enuncia- After an idealist period dominated by Andre tion, the system that negotiates the viewer's Bazin, a phenomenologist period influenced by access to the film-the system that "speaks"the Cohen-Seat and Jean Mitry, and a structuralist fiction. This study arguesthat this level is itself period initiated by the writings of Christian far from ideology-free.It does not merelyconvey neutrallythe ideology of the fictional level. As we will see, it is built so as to mask the ideological origin and nature of cinematographicstatements. Fundamentally,the enunciationsystem analyzed below-the system of the suturefunctions as a "tutor-code."It speaks the codes on which the fiction depends. It is the necessary intermediary between them and us. The system of the suture is to classical cinema what verbal Brian Henderson collaborated in writing this article from a previous text. Metz, several film critics and theorists adopted a perspective bringing together semiology and Marxism. This tendency is best represented by three groups, strongly influenced by the literary review Tel Quel: the cinematographic collective Dziga Vertov, headed by Jean-Pierre Gorin and Jean-Luc Godard; the review Cinethique; the new and profoundly transformed Cahiers du Cinema. After a relatively short period of hesitation and polemics, Cahiers established a sort of common front with Tel Quel and Cinethique. Their THE THE TUTOR-CODE TUTOR-CODE OF CLASSICALCINEMA OF CLASSICAL CINEMA 23 23 program, during the period which culminated between 1969 and 1971, was to establish the foundationsof a science of cinema. Defined by Althusser, this required an "epistemological break"with previous, ideological discourseson cinema. In the post-1968 view of Cahiers,ideological discoursesincluded structuralistsystems of an empiricistsort. In seeking to effect such a break within discourse on cinema, Cahiers concentrated on authors of the second structuralistgeneration (Kristeva,Derrida, Schefer) and on those of the first generation who opposed any empiricist interpretation of LviStrauss'swork. The point was to avoid any interpretationof a structurethat would make it appearas its own cause, thus liberatingit from the determinations of the subject and of history. As Alain Badiou put it, therefore not to search for latent meanings,but to look for that which causes or determinesthe structure. Given the Cahiers project of a search for causes, what means were availableto realize it? As Badiou points out, two systems of thought propose a structural conception of causality, Louis Althusser'sMarxismand JacquesLacan's psychoanalysis. Althusser'stheses massivelyinfluencedthe Cahierstheoreticalproductionduring the period in question. His influence was constantly commented on and made explicit, both within the Cahierstexts and by those who commented on them. Less well understood is the influence on Cahiers of Lacanian psychoanalysis,that other systemfrom which a science of cinemacould be expectedto emergeby means of a critiqueof empiriciststructuralism. For Lacan,psychoanalysisis a science. The structuralist activity was defined a few years ago as the construction of a "simulacrum of the object," this simulacrum being in itself nothing but intellect added to the object. Recent theoretical work conducted both in the Marxist field and in the psychoanalytic field shows that such a conception of structure should be completely rejected. Such a conception pretends to find inside of the real, a knowledge of which the real can only be the object. Supposedly, this knowledge is already there, just waiting to be revealed. (Cited by Jean Narboni in an article on Jancso, Cahiers du Cinema, #219.) Lacan's first word is to say: in principle, Freud founded a science. A new science which was the science of a new object: the unconscious . . . If psycho- Unable to understandthe causes of a structure, what they are and how they function, such a conceptionconsidersthe structureas a cause in itself. The effect is substitutedfor the cause; the cause remainsunknownor becomesmythical (the "theological"author). The structuralism of Cahiersholds, on the other hand, that there is more to the whole than to the sum of its parts. The structure is not only a result to be described, but the trace of a structuringfunction. The critic's task is to locate the invisible agent of this function. The whole of the structure thus becomes the sum of its parts plus the cause of the structureplus the relationshipbetweenthem, throughwhich the structureis linked to the context that produced it. To study a structureis analysisis a science because it is the science of a distinct object, it is also a science with the structure of all sciences: it has a theory and a technique (method) that makes possible the knowledge and transformation of its object in a specific practice. As in every authentically constituted science, the practice is not the absolute of the science but a theoretically subordinate moment; the moment in which the theory, having become method (technique), comes into theoretical contact (knowledge) or practical contact (cure) with its specific object (the unconscious). (Althusser,Lenin and Philosophy[MonthlyReview Press,New York, 1971],pp. 198-199.) Like ClaudeLevi-Strauss,Lacandistinguishes threelevels within humanreality. The firstlevel is nature,the third is culture. The intermediate level is that in which natureis transformedinto culture. This particularlevel gives its structure to humanreality-it is the level of the symbolic. The symbolic level, or order, includes both language and other systems which produce signification, but it is fundamentally structured by language. Lacanianpsychoanalysisis a theory of intersubjectivity,in the sense that it addresses the relationship(s) between "self" and "other"in- 24 24 dependentlyof the subjectswho finally occupy these places. The symbolic order is a net of relationships.Any "self"is definableby its position within this net. From the moment a "self" belongs to cultureits fundamentalrelationships to the "other"are taken in charge by this net. In this way, the laws of the symbolic order give their shape to originally physical drives by assigning the compulsory itineraries through which they can be satisified. The symbolicorder is in turn structuredby language. This structuring power of language explains the therapeutic function of speech in psychoanalysis. The psychoanalyst'stask is, through the patient's speech, to re-link the patient to the symbolic order,from which he has receivedhis particular mentalconfiguration. Thus for Lacan, unlike Descartes,the subject is not the fundamentalbasis of cognitive processes. First, it is only one of many psychological functions. Second, it is not an innate function. It appearsat a certain time in the development of the child and has to be constitutedin a certain way. It can also be altered, stop functioning, and disappear. Being at the very center of what we perceive as our self, this function is invisible and unquestioned. To avoid the encrusted connotationsof the term "subjectivity," Lacan calls this function "the imaginary." It must be understoodin a literal way-it is the domainof images. The imaginarycan be characterizedthrough the circumstancesof its genesis or throughthe consequencesof its disappearance. The imaginaryis constitutedthrough a process which Lacancalls the mirror-phase.It occurs when the infant is six to eighteen months old and occupies a contradictorysituation. On the one hand, it does not possess mastery of its body; the various segments of the nervous system are not coordinatedyet. The child cannot move or control the whole of its body, but only isolated discrete parts. On the other hand, the child enjoys from its first days a precocious visual maturity. During this stage, the child identifies itself with the visual image of the mother or the person playing the part of the CINEMA OF CLASSICAL THE THE TUTOR-CODE TUTOR-CODEOF CLASSICALCINEMA mother. Through this identification,the child perceives its own body as a unified whole by analogywith the mother'sbody. The notion of a unified body is thus a fantasy before being a reality. It is an image that the child receives from outside. Through the imaginaryfunction, the respective parts of the body are united so as to constitute one body, and thereforeto constitutesomebody: one self. Identityis thus a formalstructure which fundamentallydepends upon an identification. Identity is one effect, among others, of the structurethroughwhich images are formed: the imaginary. Lacan thus operates a radical desacralization of the subject: the "I," the "ego," the "subject"are nothing but images, reflections. The imaginaryconstitutesthe subject througha "speculary"effect common to the constitutionof all images. A mirroron a wall organizes the various objects of a room into a unified, finite image. So also the "subject"is no more than a unifying reflection. The disappearanceof the imaginary results in schizophrenia. On the one hand, the schizophrenicloses the notion of his "ego"and, more generally, the very notion of ego, of person. He loses both the notion of his identity and the faculty of identification. On the other hand, he loses the notion of the unity of his body. His fantasiesare inhabitedby horriblevisions of dismantled bodies, as in the paintingsof Hieronymus Bosch. Finally, the schizophrenicloses his mastery of language. The instance of schizophrenia illuminatesthe role of language in the functioningof the imaginaryin general.Because this relationship language-imaginaryis highly importantfor our subject,the role of the imaginary in cinema, we will pursuethis point in some detail. The role of the imaginaryin the utilizationof language points to an entire realm of inadequacy, indeed absence, in traditional accounts of language. Saussure merely repressed or avoided the problem of the role of the subject in languageutilization.The subjectis eliminated from the whole field of Saussurianlinguistics. CLSSCA THE OF CLASSICAL CINEMA TUTO-COD THE-TUTOR-CODE OFC-~---INEMA 25~- This elimination commands the famous oppositions between code and message, paradigm and syntagm, language system and speech. In each case, Saussure grants linguistic relevance to one of the terms and denies it to the other. (The syntagm term is not eliminated, but is put under the paradigms of syntagms, i.e., syntax). In this way, Saussure distinguishes a deep level of linguistic structures from a superficial one where these structures empirically manifest themselves. The superficial level belongs to the domain of subjectivity, that is, to psychology. "The language system equals language less speech." Speech, however, represents the utilization of language. The entity which Saussure defines is language less its utilization. In the converse way, traditional psychology ignores language by defining thought as prior to it. Despite this mutual exclusion, however, the world of the subject and the universe of language do meet. The subject speaks, understands what he is told, reads, etc. To be complete, the structuralist discourse must explain the relationship language/subject. (Note the relevance of Badiou's critique of empiricist structuralism to Saussure.) Here Lacan's definition of the subject as an imaginary function is useful. Schizophrenic regression shows that language cannot function without a subject. This is not the subject of traditional psychology: what Lacan shows is that language cannot function outside of the imaginary. The conjunction of the language system and the imaginary produces the effect of reality: the referential dimension of language. What we perceive as "reality" is definable as the intersection of two functions, either of which may be lacking. In that language is a system of differences, the meaning of a statement is produced negatively, i.e., by elimination of the other possibilities formally allowed by the system. The domain of the imaginary translates this negative meaning into a positive one. By organizing the statement into a whole, by giving limits to it, the imaginary transforms the statement into an image, a reflection. By conferring its own unity and continuity upon the statement, the subject organizes I 25 it into a body, giving it a fantasmatic identity. This identity, which may be called the "being" or the "ego" of the statement, is its meaning, in the same way that "I" am the meaning of my body's unity. The imaginary function is not limited to the syntagmatic aspect of language utilization. It commands the paradigms also. A famous passage by Borges, quoted by Foucault in The Order of Things, illustrates this point. An imaginary Chinese encyclopedia classified animals by this scheme: (a) belonging to the emperor; (b) embalmed; (c) tamed; (d) guinea-pigs; (e) sirens; (f) fabulous; (g) dogs without a leash; (h) included in the present classification. According to Foucault, such a scheme is "impossible to think," because the sites where things are laid are so different from each other that it becomes impossible to find any surface that would accept all the things mentioned. It is impossible to find a space common to all the animals, a common ground under them. The common place lacking here is that which holds together words and things. The paradigms of language and culture hold together thanks to the perception of a common place, of a "topos" common to its elements. This common place can be defined at the level of history or society as "episteme" or "ideology." This common place is what the schizophrenic lacks. Thus, in summary, the speculary, unifying, imaginary function constitutes, on the one hand, the proper body of the subject and, on the other, the limits and the common ground without which linguistic syntagms and paradigms would be dissolved in an infinite sea of differences. Without the imaginary and the limit it imposes on any statement, statements would not function as mirrors of the referent. The imaginary is an essential constituent in the functioning of language. What is its role in other semiotic systems? Semiotic systems do not follow the same patterns. Each makes a specific use of the imaginary; that is, each confers a distinctive function upon the subject. We move now from the role of the subject in language use 26 26TETTR-OEO to the role of the subject in classical painting and in classical cinema. Here the writings of Jean-Pierre Oudart, Jean-Louis Schefer, and others will serve as a guide in establishingthe foundationsof our inquiry.* We meet at the outset a fundamentaldifference between language and other semiotic systems. A famous Stalinianjudgmentestablished the theoretical status of language: language is neither part of science nor part of ideology. It representssome sort of a thirdpower, appearing to function-to some extent-free of historical influences. The functioningof semiotic systems such as painting and cinema, however, clearly manifests a direct dependency upon ideology and history. Cinema and painting are historical products of human activity. If their functioning assigns certain roles to the imaginary, one must considerthese roles as resultingfrom choices (conscious or unconscious) and seek to determine the rationale of such choices. Oudarttherefore asks a double question: What is the semiological functioning of the classical painting? Why did the classical painters develop it? Oudartadvancesthe following answers. (1) Classicalfigurativepaintingis a discourse. This discourse is produced according to figurative codes. These codes are directly produced by ideology and are therefore subjectedto historical transformations. (2) This discoursedefines in advancethe role of the subject,and therefore pre-determinesthe readingof the painting. The imaginary (the subject) is used by the painting to mask the presence of the figurative codes. Functioningwithout being perceived,the codes reinforcethe ideology which they embodywhile the paintingproduces"an impressionof reality" (efIet-de-reel). This invisible functioning of the figurativecodes can be definedas a "naturali*See Jean-Louis Schefer, Scenographie d'un tableau (Paris: Seuil, 1969); and articles by Jean-Pierre Oudart, "La Suture, I and II," Cahiers du Cinema, Nos. 211 and 212 (April and May, 1969), "Travail, Lecture, Jouissance," Cahiers du Cinema, No. 222 (with S. DaneyJuly 1970), "Un discours en defaut," Cahiers du Cinema, No. 232 (Oct. 1971). THE TUTOR-CODEOF CLASSICAL CINEMA LSSCLCNM zation":the impressionof reality producedtestifies that the figurative codes are "natural" (instead of being ideological products). It imposes as "truth"the vision of the world entertained by a certain class. (3) This exploitation of the imaginary,this utilization of the subject is made possible by the presence of a system which Oudartcalls "representation."This system englobesthe painting,the subject,and their relationshipupon which it exerts a tight control. Oudart'sposition here is largelyinfluencedby Schefer's Scenographie d'un tableau. For Schefer, the image of an object must be understood to be the pretext that the painter uses to illustratethe systemthroughwhich he translates ideology into perceptual schemes. The object representedis a "pretext"for the painting as a "text"to be produced. The objecthides the painting's textuality by preventing the viewer from focusing on it. However, the text of the painting is totally offered to view. It is, as it were, hidden outside the object. It is here but we do not see it. We see throughit to the imaginaryobject. Ideologyis hiddenin our very eyes. How this codificationand its hiding process work Oudartexplainsby analyzingLas Meninas by Velasquez.* In this painting,membersof the court and the painter himself look out at the spectator. By virtue of a mirrorin the back of the room (depicted at the center of the painting), we see what they are looking at: the king and queen,whose portraitVelasquezis painting. Foucaultcalls this the representationof classical representation,because the spectator-usually invisible-is here inscribedinto the painting itself. Thus the paintingrepresentsits own functioning,but in a paradoxical,contradictoryway. The painter is staring at us, the spectatorswho pass in front of the canvas; but the mirrorreflects only one, unchanging thing, the royal couple. Throughthis contradiction,the system of "representation" points toward its own functioning. In cinematographicterms, the mirror representsthe reverse shot of the painting. In *Oudart borrows here from ch. 1 of Michel Foucault's The Order of Things (London: Tavistock, 1970). TKE THE TUTOR-CODEOF CLASSICALCINEMA theatrical terms, the painting represents the stage while the mirror representsits audience. Oudart concludes that the text of the painting must not be reduced to its visible part; it does not stop where the canvas stops. The text of the painting is a system which Oudart defines as a "double-stage."On one stage, the show is enacted; on the other, the spectatorlooks at it. In classicalrepresentation,the visible is only the first part of a system which always includes an invisible second part (the "reverseshot"). Historicallyspeaking, the system of classical representationmay be placed in the following way. The figurativetechniques of the quattrocento constituteda figurativesystem which permitted a certain type of pictorial utterance. Classicalrepresentationproducesthe same type of utterancesbut submitsthem to a characteristic transformation-by presenting them as the embodiment of the glance of a subject. The pictorialdiscourseis not only a discoursewhich uses figurativecodes. It is that which somebody sees. Thus, even without the mirror in Las Meninas, the other stage would be part of the text of the painting. One would still notice the attention in the eyes of the painting's figures, etc. But even such psychologicalclues only reinforce a structurewhich could function without them. Classical representationas a system does not depend upon the subject of the painting. The Romanticlandscapesof the nineteenth century submit nature to a remodeling which imposeson them a monocularperspective,transforming the landscape into that which is seen by a given subject. This type of landscape is very differentfrom the Japaneselandscapewith its multiple perspective. The latter is not the visible part of a two-stage system. While it uses figurativecodes and techniques, the distinctive feature of representationas a semiological system is that it transforms the painted object into a sign. The object which is figuredon the canvas in a certainway is the signifierof the presenceof a subjectwho is looking at it. The paradox of Las Meninas proves that the presence of the subject must be signified 27 27 but empty, defined but left free. Reading the signifiers of the presence of the subject, the spectatoroccupies this place. His own subjectivityfills the emptyspot predefinedby the painting. Lacan stressesthe unifying function of the imaginary,throughwhich the act of reading is made possible. The representationalpaintingis alreadyunified. The paintingproposesnot only itself, but its own reading. The spectator's imaginarycan only coincide with the painting's built-in subjectivity. The receptive freedom of the spectator is reduced to the minimum-he has to accept or reject the painting as a whole. This has importantconsequences,ideologically speaking. When I occupy the place of the subject, the codes which led me to occupy this place become invisible to me. The signifiersof the presence of the subjectdisappearfrom my consciousness because they are the signifiersof my presence. What I perceive is their signified: myself. If I want to understandthe painting and not just be instrumentalin it as a catalystto its ideological operation, I must avoid the empiricalrelationship it imposes on me. To understandthe ideology which the painting conveys, I must avoid providing my own imaginaryas a support for that ideology. I must refuse that identification which the painting so imperiouslyproposes to me. Oudartstressesthat the initialrelationshipbetween a subjectand any ideologicalobject is set up by ideology as a trap which prevents any real knowledgeconcerningthe object. This trap is built upon the propertiesof the imaginaryand must be deconstructed through a critique of these properties. On this critique depends the possibilityof a real knowledge. Oudart'sstudy of classical painting provides the analyst of cinema with two important tools for such a critique: the concept of a double-stageand the concept of the entrapmentof the subject. We note firstthat the filmic image considered in isolation, the single frame or the perfectly static shot, is (for purposes of our analysis) equivalent to the classical painting. Its codes, 28 28TETTR-OEO even though "analogic"rather than figurative, are organized by the system of representation: it is an image designedand organizednot merely as an object that is seen, but as the glance of a subject. Can there be a cinematographynot based upon the system of representation?This is an interestingand importantquestion which cannot be explored here. It would seem that therehas not been such a cinematography.Certainly the classical narrative cinema, which is our presentconcern, is founded upon the representationsystem. The case for blanketassimilation of cinema to the system of representation is most stronglyput by Jean-LouisBaudry,who argues that the perceptualsystem and ideology of representationare built into the cinematographic apparatus itself. (See "Ideological Effects of the Basic CinematographicApparatus,"in Cinethique#7-8.) Cameralenses organize their visual field according to the laws of perspective,which thereby operate to render it as the perceptionof a subject. Baudrytraces this systemto the sixteenthand seventeenthcenturies, during which the lens technology which still governs photographyand cinematography was developed. Of course cinema cannot be reduced to its still frames and the semiotic system of cinema cannot be reducedto the systemsof paintingor of photography. Indeed, the cinematic succession of images threatensto interruptor even to expose and to deconstruct the representation system which commands static paintings or photos. For its succession of shots is, by that very system, a successionof views. The viewer's identificationwith the subjective function proposed by the painting or photographis broken again and again during the viewing of a film. Thus cinema regularlyand systematicallyraises the question which is exceptional in painting (Las Meninas): "Who is watching this?" The point of attack of Oudart'sanalysisis precisely here-what happens to the spectator-imagerelation by virtue of the shot-changespeculiar to cinema? The ideological question is hardly less important than the semiological one and, indeed, THE TUTOR-CODEOF CLASSICAL CINEMA LSSCLCNM is indispensableto its solution. From the standpoint of the imaginary and of ideology, the problem is that cinema threatensto expose its own functioning as a semiotic system, as well as that of painting and photography. If cinema consists in a series of shots which have been produced, selected, and ordered in a certain way, then these operations will serve, project, and realize a certain ideological position. The viewer'squestion, cued by the system of representation itself-"Who is watching this?" and "Who is ordering these images?"-tends, however, to expose this ideologicaloperationand its mechanics. Thus the viewer will be aware (1) of the cinematographicsystem for producing ideology and (2) therefore of specificideological messagesproducedby this system. We know that ideology cannot work in this way. It must hide its operations, "naturalizing"its functioning and its messagesin some way. Specifically, the cinematographicsystem for producing ideology must be hidden and the relation of the filmic message to this system must be hidden. As with classical painting, the code must be hidden by the message. The message must appear to be completein itself, coherentand readable entirely on its own terms. In order to do this, the filmic message must account within itself for those elements of the code which it seeks to hide-changes of shot and, above all, what lies behind these changes, the questions "Who is viewing this?" and "Who is ordering these images?"and "For what purpose are they doing so?" In this way, the viewer's attention will be restrictedto the message itself and the codes will not be noticed. That systemby which the filmicmessageprovidesanswersto the viewer's questions-imaginary answers-is the object of Oudart'sanalysis. Narrativecinema presentsitself as a "subjective" cinema. Oudartrefers here not to avantgarde experimentswith subjectivecameras,but to the vast majorityof fiction films. These films proposeimages which are subtlydesignatedand intuitively perceived as correspondingto the point of view of one characteror another. The point of view varies. There are also moments THE OF TUTOR-CODE TUTOR-CODE TH CINEMA OF~CLASSICAL CLSICLCNEA2 - ~ when the image does not represent anyone's point of view; but in the classical narrative cinema, these are relatively exceptional. Soon enough, the image is reassertedas somebody's point of view. In this cinema, the image is only "objective"or "impersonal"duringthe intervals between its acting as the actors'glances. Structurally, this cinema passes constantly from the personal to the impersonal form. Note, however, that when this cinema adopts the personal form, it does so somewhatobliquely,ratherlike novelistic descriptions which use "he" rather than "I" for descriptionsof the central character's experience. According to Oudart, this obliquenessis typical of the narrativecinema: it gives the impressionof being subjectivewhile never or almost never being strictly so. When the cameradoes occupy the very place of a protagonist, the normal functioning of the film is impeded. Here Oudart agrees with traditional film grammars. Unlike them, however, Oudart can justifythis taboo, by showingthat this necessary obliquityof the camerais partof a coherent system. This system is that of the suture. It has the function of transforminga vision or seeing of the film into a readingof it. It introducesthe film (irreducibleto its frames) into the realm of signification. Oudart contrasts the seeing and the reading of a film by comparingthe experiences associatedwith each. To see the film is not to perceive the frame, the camera angle and distance, etc. The space between planes or objects on the screenis perceivedas real, hence the viewermay perceive himself (in relation to this space) as fluidity, expansion, elasticity. When the viewer discovers the frame-the first step in reading the film-the triumph of his former possession of the image fades out. The viewer discoversthat the camera is hiding things, and therefore distrustsit and the frame itself, which he now understandsto be arbitrary. He wonders why the frame is what it is. This radically transforms his mode of participation -the unreal space between charactersand/or objects is no longer perceivedas pleasurable. It is now the space which separates the camera 29 from the characters. The latter have lost their quality of presence. Space puts them between parenthesesso as to assertits own presence.The spectatordiscovers that his possession of space was only partial,illusory. He feels dispossessed of what he is prevented from seeing. He discovers that he is only authorizedto see what happens to be in the axis of the glance of another spectator,who is ghostly or absent. This ghost, who rules over the frame and robs the spectator of his pleasure, Oudart proposes to call "the absent-one"(l'absent). The description above is not contingent or impressionistic-the experiences outlined are the effectsof a system. The systemof the absentone distinguishescinematography,a systemproducing meaning, from any impressed strip of film (mere footage). This system depends,like that of classicalpainting,upon the fundamental opposition between two fields: (1) what I see on the screen, (2) that complementaryfield which can be defined as the place from which the absent-oneis looking. Thus: to any filmic field definedby the cameracorrespondsanother field from which an absenceemanates. So far we have remained at the level of the shot. Oudart now considers that common cinematographicutterance which is composed of a shot and a reverse shot. In the first, the missing field imposes itself upon our consciousness under the form of the absent-onewho is looking at what we see. In the second shot, the reverse shot of the first, the missing field is abolishedby the presenceof somebodyor something occupying the absent-one'sfield. The reverse shot representsthe fictional owner of the glance correspondingto shot one. This shot/reverse shot system orders the experience of the viewer in this way. The spectator's pleasure,dependentupon his identification with the visual field, is interruptedwhen he perceives the frame. From this perception he infers the presence of the absent-one and that other field from which the absent-oneis looking. Shot two reveals a characterwho is presented as the owner of the glance correspondingto shot one. That is, the characterin shot two occupies 30 30 the place of the absent-one correspondingto shot one. This characterretrospectivelytransforms the absence emanating from shot one's other stage into a presence. What happens in systemic terms is this: the absent-oneof shot one is an element of the code that is attractedinto the message by means of shot two. When shot two replacesshot one, the absent-oneis transferredfrom the level of enunciation to the level of fiction. As a resultof this, the code effectively disappearsand the ideological effect of the film is thereby secured. The code, which produces an imaginary,ideological effect, is hidden by the message. Unable to see the workings of the code, the spectatoris at its mercy. His imaginary is sealed into the film; the spectator thus absorbs an ideological effect withoutbeing awareof it, as in the very different system of classical painting. The consequencesof this systemdeservecareful attention. The absent-one'sglance is that of a nobody, which becomes (with the reverse shot) the glance of a somebody (a character present on the screen). Being on screen he can no longer compete with the spectator for the screen's possession. The spectator can resume his previous relationshipwith the film. The reverse shot has "sutured"the hole opened in the spectator'simaginaryrelationshipwith the filmic field by his perceptionof the absent-one. This effect and the system which produces it liberates the imaginaryof the spectator,in order to manipulateit for its own ends. Besides a liberation of the imaginary, the sys- tem of the suture also commands a production of meaning. The spectator's inference of the absent-oneand the other field must be described more precisely: it is a reading. For the spectator who becomes frame-conscious, the visual field meansthe presenceof the absent-oneas the owner of the glance that constitutesthe image. The filmic field thus simultaneouslybelongs to representation and to signification. Like the classicalpainting, on the one hand it represents objects or beings, on the other hand it signifies the presenceof a spectator. When the spectator ceases to identify with the image, the image THETUTOR-CODE OF CLASSICAL CINEMA CINEMA~~~~~~ necessarilysignifiesto him the presence of another spectator. The filmic image presentsitself here not as a simple image but as a show, i.e., it structurallyasserts the presence of an audience. The filmic field is then a signifier; the absent-one is its signified. Since it represents another field from which a fictional character looks at the field correspondingto shot one, the reverse shot is offered to the film-audienceas being the other field, the field of the absent-one. In this way, shot two establishesitself as the signified of shot one. By substitutingfor the other field, shot two becomesthe meaningof shot one. Within the system of the suture, the absentone can thereforebe defined as the intersubjective "trick"by means of which the second part of a given representativestatementis no longer simply what comes after the firstpart, but what is signified by it. The absent-one makes the differentpartsof a given statementthe signifiers of each other. His strategm: Break the statement into shots. Occupy the space between shots. Oudart thus defines the basic statement of classical cinematographyas a unit composed of two terms: the filmic field and the field of the absent-one. The sum of these two terms, stages, and fields realizesthe meaningof the statement. Robert Bresson once spoke of an exchange between shots. For Oudart such an exchange is impossible-the exchangebetweenshot one and shot two cannot take place directly. Between shot one and shot two the other stage correspondingto shot one is a necessaryintermediary. The absent-one represents the exchangability between shots. More precisely, within the system of the suture,the absent-onerepresentsthe face that no shot can constituteby itself a complete statement. The absent-onestands for that which any shot necessarily lacks in order to attain meaning: anothershot. This bringsus to the dynamics of meaning in the system of the suture. Within this system, the meaning of a shot depends on the next shot. At the level of the signifier,the absent-onecontinuallydestroysthe balance of a filmic statement by making it the