JUMP CUT
A REVIEW OF CONTEMPORARY MEDIA

Formalism and its discontents

by Dana B. Polan

from Jump Cut, no. 26, December 1981, pp. 63-66
copyright Jump Cut: A Review of Contemporary Media, 1981, 2005

Noel Burch, To the Distant Observer: Form and Meaning in the Japanese Film. Berkeley: University of California Press, 1979. 387 pp. $19.95 ($9.75 paperback).

The Soviet Marxist critic, V. N. Voloshinov, in his book, Marxism and the Philosophy of Language (trans. Ladislaw Matejka, New York: Seabury Press, 1973), directed a telling criticism against formalist theory, against the assumption that particular phenomena have their truest existence within categories, within general forms which seem to embody those phenomena. Voloshinov uses the example of hunger to suggest that even so seemingly natural, seemingly nonsocial, and seemingly generalizable a category is, as category, as general form, an inadequate way to understand real instances and existences of phenomena. There is no such thing, Voloshinov suggests, as the real existence of hunger as category; rather, there are only particular hungers — their particularity to a large extent a consequence of the particular historical moment. How one experiences hunger, for example, relates to one's ability (dependent on social privilege) to satisfy that need in particular ways.

Voloshinov's critique is a powerful reminder of the historical rootedness of human phenomena (which is not at all to say that such phenomena are fixed, that their history can't be changed). The dangers which he predicts for any approach which tries to substitute rigid categories for an engagement with the social complexity of phenomena are precisely the dangers which come to plague the work of Noel Burch. For example, Burch's work gives evidence of a drive toward reification — toward the assumption, that is, that certain structures or forms (the forms of film, in Burch's case) inevitably have certain meanings or effects. [1] The central premise of idealist thinking  — the split between things and people — reasserts itself as a myth of objectivity: an attempt to freeze meanings in a palpable, static set of structures, outside human influence and historical change.

There's a moment for me in Noel Burch's work where the contradictions of his formalist position show forth in their inadequacy against the real complexity of the social phenomena he is trying to deal with. On page 59 of Theory of Film Practice (New York: Praeger Books, 1973), which is Burch's polemical introduction to the study of artistic form, Burch describes television as solely a formal activity. Against everything else that TV is, Burch understands TV solely as a system of technical codes which interact among themselves and whose only real reference outside the formal structure of a particular work would be to the tradition of previous uses of the same formal code:

"By breaking down the barrier between genres and in particular by quite naturally introducing a mixture of the 'lived' and the 'staged,' television has encouraged the creation of new forms and structures based on a deliberate mixing of genres and materials in them and has begun to explore the multiple dialectics that can result from such a mixture."

Any social understanding of television — its place as part of the productive means of a society (for example, the differing functions of television in the First and Third Worlds) — disappears as a formalist understanding. The history of an art as the replacement of aesthetic devices by new, and equally aesthetic, aesthetic devices comes to dominate analysis.

Burch's new book, To the Distant Observer: Form and Meaning in the Japanese Film forces even further a separation of films from history. Attempting the task of examining the meanings of films in living, historical contexts, Burch loses the concretely historical side of such a project by starting with the films themselves. He encloses history within the confines of rigid logic through the assumption that if a film has this-or-that form, it will of necessity have this-or-that meaning. The very distinction between the main title and subtitle of Burch's book signals the confusions which Burch's attempt to write history will fall prey to. There are two prepositions in the title, but whereas the "to" of the main title refers to an interaction, an open process between films and acts of observing, the subtitle loses this sense of process and attributes all the sense of films to qualities in the films and not in their interactions with people. Japanese films with a rabidly militaristic and propagandistic subject-matter turn out, in Burch's reading, to be politically progressive in the manner in which they play with forms specific to cinema (for example, how they handle off-screen and on-screen space, whether or not they adhere to rules of screen direction). Burch pits Japanese film against Western cinema, which he sees as tied to illusionism, a whole way of seeing which unjustifiably tries to use art as a way of expressing bourgeois values:

"In the West, since the eighteenth century, our major narrative arts — the novel, the theatre, and more recently, the cinema, have tended toward a kind of narrative saturation: every element is aimed at conveying, at expressing, a narrative essence" (p. 90).

Burch's critique of Western cinema is part of the attack on Realism which has characterized much of contemporary theory of the arts. According to such theory, several aspects of film can impel it toward Realism. For example, film's similarity to photography presents the possibility of a lifelike representation of the world (a representation made all the more lifelike by film's advantage over photography of adding motion to the image). Furthermore, editing allows film to bring different spaces into connection and so reduce their differences. Much of Burch's work analyses the eyeline match, in which an image of someone looking is directly connected to one of the objects looked at, with the result, Burch argues, that separate images come to share a common reality. Arguing that such a reality is created by the techniques of film and so is in no way a picture of a true reality, preexisting the film, Burch sees cinematic realism as a lie, an illusionism, which passes off the technical creation of film as an image of the world.

The value of Japanese film for Burch lies in its radical refusal to adopt the cinematic codes which would support Realism. For example, certain Japanese films will deliberately mismatch angles, lighting, etc., from shot to shot so that it becomes hard to find a shared reality between the shots. In Japanese films, Burch argues, the inability or refusal of techniques to uphold the illusions and illusionism of subject matter puts spectators outside any accepting involvement with the story or content of the film. The constant reminder by these films that they are films prohibits spectators from treating the films as images of a world outside of, prior to, the films. Representation gives way to presentation: the ways in which films present their contents become visible. The films become objects which spectators can take a distance from, examine, and criticize.

Burch claims that his practice as a critic — bringing into prominence a cinema which is in an oppositional relation to dominant cinematic practices — is Marxist. (The first page of the preface states that Burch's study is understandable "as part … of the modern search for a Marxist approach to art …") Moreover, he argues that this oppositional cinema is itself Marxist:

"The claim for the radicalism of Japanese practice is therefore not to be taken as the assertion of some metaphysical absolute but as an attempt to describe the situation which Japanese 'culture' effectively occupies today with regard both to the dominant ideological profile of Western Europe and the Americans, and to those practices, scientific, literary, and artistic, which instantiate the Marxist critique of that dominance" (p. 89).

Burch, then, presents his text as a perfect match of the object of study and the study of an object. It is his attention to form which allows his recognition of a formally complex cinema, and vice versa. But in what ways is this refusal of Realism a specifically Marxist critique?

Burch fundamentally derives his premises from the critique of the Western tradition in the early work of French philosopher Jacques Derrida. [2] Derrida's project has been to attack the certitudes of what he sees as the metaphysical underpinnings of Western thought. Such thought, Derrida argues, bases itself on a myth of presence, on an assumption (in many cases unconscious) that all phenomena have behind them a sense, a meaning, an ultimate truth (or set of truths) which exists as the guarantee of particular phenomena. For example, in this view, Christianity is a metaphysical system in which the particularities of any individual life are less important, less real, that the essence to which that life is ultimately accountable and from which that life gains whatever reality it has; the individual life is a superficial veil over the ultimate and final presence of God. Human endeavor is an attempt to raise oneself out of the human condition, which is a fallen condition. We have been banished from the fullness of Eden and from closeness to God's presence — back to transcendental truth: what religion will refer to as God or the Divine word, or what Derrida will call "the transcendental signified," that first cause to which all phenomena refer. In the metaphysical schema, material phenomena (for example, our seeming rootedness in a here-and-now) are mere stand-ins, signifiers, for the ultimate truth they embody in a derived and inferior fashion.

Derrida's Of Grammatology is a study of writing, a specific and in many ways exemplary example of the act of representation in which forms (material phenomena) — in this case words — exist only to point to a supposedly real, more fundamental existence elsewhere. The underlying suppositions of Western thought, Derrida argues, are based on the false Realism of such writing, which denies its own properties as a material so that it can all the better serve as a transparent window, a conduit, for truth to present itself, for human words to express the Word. Against this, Derrida argues that the only "thing" which one can say is really prior to all other activities is language as such — that is, as a set of forms which combine to create meaning rather than reflect meaning. A way to understand Derrida's point here is to think of the nature of a dictionary. To define a word, a dictionary refers to another word, but that other word also has a word standing for it which also has a word standing for it which … and so on. There is no final word which explains all others, and the only way this endless chain can come to an end is through an illusion — for example, the illusion (usually specially based) that one particular definition is the truest.

Influenced by the French structuralist linguist Ferdinand de Saussure, Derrida suggests that what we declare to be reality, what we represent to ourselves as reality, is finally, really, no more than an arbitrary construction, a fixing of the infinite field of possible permutations of meaning. For Derrida, reality is fundamentally what he calls a differance. This created word combines the idea of differing (every meaning is different from every other) and deferring (every meaning must base itself on another meaning, the result being an endless chain). Any claim to certitude in meaning, any attempt to fix reality, is for Derrida a lie, a desire to deny differance, to deny the real open-endedness of existence as a game (what Derrida calls "le jeu"). Derrida thus reads two possible attitudes and involvements in human activity. First is an irresponsible position which offers final meanings which are in fact arbitrary and without any real, ultimate foundation. Second, a more responsible position realizes the playfulness of language, its refusal to adapt itself naturally to any order of signification, to search instead for ambiguities, formal complexity, where previously there had only seemed to be representations.

In his reading of Japanese versus Western cinema, Burch repeats this split. Western cinema's representational drive — exemplified in the Hollywood film but "operative for many spectators of Michael Snow's WAVELENGTH or Godard's DEUX OU TROIS CHOSES JE SAIS D'ELLE" (p. 19) is the drive to present as essences a set of values which are really historically bound as the values of the bourgeoisie. Referring to diegesis — that is, the imaginary world a Realistic film constructs through its techniques — which is embodied in the diegetic effect "whereby spectators. experience the diegetic world as environment" (p. 19), Burch argues that

"the stages of formation and the ongoing elaboration of the basic mode of representation in the cinema of the West point to an all but universal tendency within the dominant cinema (as well as theatre and literature) to maximize and generalize the diegetic effect" (ibid.).

Burch sees Western cinema as a cinema of presence, effacing its film form to involve its spectators in a universe supposedly beyond the fact of the film itself. For Burch, there is nothing inherent in the cinema that should have propelled it toward representation. In work on early (pre-1915) Western cinema, Burch has claimed to discover a cinema that obeyed few of the rules of what cinema should be.[3] As one example, Burch cites the case of the early filmmaker, Edwin S. Porter, many of whose films deny any narrative links between separate scenes (to such a degree that Porter's adaptation of Uncle Tom's Cabin is incomprehensible unless one already knows the story. Instead of adding up to one total narrative effect, the scenes in many Porter films refuse to express a story. It was this kind of cinema, one which Burch argues gave full rein to arbitariness and a playful lack of fixed techniques and purposes, that was repressed, reduced, fixed into the diegetic cinema that Burch sees dominating Western film practice.

For Burch, Japanese cinema is, like primitive Western cinema, a practice which resists codes of representation; it is the differance which Western cinema does not admit. Like primitive cinema which was inspired by vaudeville and other playful arts, Japanese cinema derives from popular practices (for example, the formal experimentation of Kabuki theatre) which refused to surrender their own awareness of form to the needs of realism and diegesis. Indeed, through a survey of Japanese arts, Burch argues that at a fundamental level, the Japanese way of life is opposed to the fixity of meanings. This appears most directly in the orthography of the Japanese language, which combines phonetic and non-phonetic systems without subordinating one to the other and prevents any sense of the rightness, the naturalness of any one system. Japanese cinema, Burch argues, is not one which holds its signifiers to a function of communicating a signified. Instead, there is a separation of elements in which the priority of the signified is exposed to criticism, is exposed in its fragility. For example, one central practice of the Japanese cinema before the pressure of certain "Westernizing" tendencies was the use of Benshi, a male narrator who stood next to the screen and provided his commentary, often a critical one, on the proceedings of the film. The Benshi, Burch suggests, reasserted the film's quality as film.

Now in certain cases, the presence of the Benshi might have done just that. But the effects (psychological, emotional, ideological) of such a technique cannot be decided in advance just on the basis of that technique. Any technique has a different force, indeed a different identity, in each individual situation in which it finds itself. As he criticizes the myth of presence, Burch unwittingly admits and indeed starts from a central premise of that myth: that there are some techniques which inherently move toward Realism and others which inherently move away from it. Burch reverses the Realist argument (for Burch, it is unrealistic technique which is valuable; for the Realist, it is the contrary). But he stays within the same framework with his acceptance of the myths of inevitability and universality.

Against such a myth (whether Realist or anti-Realist) of the inevitable meaning of certain forms, I would argue that any technique, any practice, is only one practice among others in the social field. Any practice is interrelated in a complex way with other practices — for example, class relations, ideological relations, sexual relations — influencing them (and influenced by them) but not reducible to any other practice. Language, for example, is only a component of the social fabric. This is not to argue, though, that it has no force, that acts of language have no social power. Quite the contrary, the structure of social practices is such that each practice must have an effect on other practices. But this force is variable, so that it cannot be assumed that language or any other practice will always work in the system in the same way. For example, in a class society, the effect and force of language will vary across class lines.

For example, the petit-bourgeoisie often equates linguistic practice (for example, the language-transforming practice of the avant-garde) and revolutionary practice, thus arguing that a radical use of language will lead to a radical kind of action. One of the reasons for such a stance is related to that class's power and possession in the realm of words and its relative vulnerability in political realms (such as state power). For example, as Walter Benjamin suggests in his Charles Baudelaire: a Lyric Poet in the Era of High Capitalism (trans. Harry Zohn, London: New Left Books, 1973), the aggressive stance of Charles Baudelaire's poetry toward the public life of Paris care from Baudelaire's recognition that the marketplace mentality of this public had made the role of the lyric poet, like Baudelaire, a marginal one. But precisely because of this marginality, Baudelaire could strike back only in the limited realm of poetry, trying to turn the bourgeoisie's language against itself.

That a practice has (or could have) a particular force in a particular social situation is not to grant it a total or enduring power. More important, it is not so much anything in the practice itself that has force but its relation with other elements, which can make it effective in a particular way. For example, whatever force a Baudelaire's writing might have depends not only on meanings and forms in the poetry but also on a whole array of equally necessary factors. Among them are a system of distribution (for example, publishers interested in printing poetry) and appropriate forms of reading (including the very existence of readers with the time and training to read lyric poetry). At the very least, claims about this or that effect for a particular practice need to account for the social world in which that act occurs. It is this sense of the historical situation (for example, who is the audience for what art and why) which is regrettably absent in much investigation of artistic practices. Instead, most theories tend to be object-oriented, assuming that if the artwork has a particular quality, it will have a necessary effect. In valorizations of the avant-garde, especially, any attention to the variability of possible spectator responses drops out as the theory creates a "formal spectator," a spectator outside of history, outside class. Yet in another way that spectator is inside class, since the generalized spectator which the theory invokes is usually a projection of the theorist's own values onto the spectating act. [4]

The problem of formalism is twofold. First, it grants force to a practice on the basis of internal qualities of that practice. The potential effect of the practice is pulled out of a context and given an unequivocal and universal definition. Second, formalism assumes that the force which it has thus imputed to a form will tend to be the same for all spectators. Enumerating, describing materials in art forms, formalism assumes that the effect of practices derives foremost from those materials themselves. All that's required of the human observer is a point of view appropriate to the discovery of that form and its singular meaning. In other words, a given form (for example, the invisible cut) has in formalist understanding a meaning in general, an effect which, as in all formalisms, is a consequence of the nature of the form.

While critics in the Derridean tradition inflect this formalism by an attention to practices (such as Primitive cinema) which they see as escaping the impositions, the fixities, of set forms, this promotion of oppositional forms is no more than a modification in which the original formalism — generalities about the effects of narrative, about the tie between ideology and formal practice —continues to hold sway. The valorization of avant-garde practices in formalist theory is a valorization of those practices against: against narrative, illusionism, and representation as inevitably bound to certain effects as a consequence of their forms. The problem is that the effects which formalism imputes to practices are effects projected onto forms by the theory of those forms. In other words, it is the theory which creates meaning, and not the objects of the theory.

Now there is nothing necessarily wrong with this aspect of the theory. Every theory — indeed, every act of understanding — in some way constructs its objects rather than simply understanding objects separate from the observer's point of view. But this suggests that any act of theorizing necessitates a self-critical attitude, an attention to what knowledge one is coming up with and why, an attention to the ways procedures of investigation influence the kinds of discoveries one makes.

In the case of film theory, this could mean an attention to differences or similarities between a theorist's ways of looking at a film and other situations of looking. For example, one could ask if certain social and class privileges allow the theorist to look at films in a unique way (such as the recourse by Burch and other theorists interested in moment-by-moment inflections in form to motion-analyzing equipment, which allows them to slow down or reverse or stop films when they want). It is a way of looking which makes the film a different one than it would be when projected nonstop in a theatre. Burch's practice is precisely an individualistic one in which a solitary critic ("I was able to screen rare films in my hotel room when I unwittingly arrived in Kyoto on the eve of a holiday" — p. 8) creates meaning by his particular mode of analysis, but then he passes this meaning off as meaning contained in the film and not in the interpretation. This is not to say that Burch's way of viewing is wrong while some kind of "average" or "ordinary" viewing (as if there is such a thing) would be right. Rather, I'm suggesting that every viewing takes its specific identity from its context (who's viewing, why, how), a context which is determinant for meaning. This context cannot be washed away; it can, however, be studied, reflected upon.

The precise function of formalism in its attitude toward viewing is to deny the interpretive place, the context, of the theorist; it is to move from meanings between text and viewer to meanings supposedly objectively in the text. If, as formalism assumes, "there is a generality of ideology in the institution, 'before' the production of a particular ideological position," as Stephen Heath and Gillian Skirrow contend — in other words, that forms already tend toward a particular meaning before they exist in a specific historical situation — then the critic's role is to deny the specific ways in which he or she makes the meaning of the film and instead to move toward some sort of ideal ideological meaning which the text supposedly has. [5] Such an approach has tended to create an image of the theorist as an observer outside history, outside class. It is an image of criticism which, in the least self-conscious theory, assumes that particular films will always mean in set ways to its audiences.

In Burch, this inability to think out his own place as theorist in relation to the objects he is studying shows up most clearly as a contradiction and incompatibility of the goals he sets for the study of Japanese film. Burch echoes French cultural critic Roland Barthes' use of Japan as a fictional construct in his book, L'Empire des signes (Paris: Editions d'Art, 1970; reprinted 1980): Burch approvingly quotes Barthes' declaration that one can

"without claiming to represent or analyze any reality whatsoever … gather somewhere out in the world … a number of features … and with these features form a system. It is this system which I [Barthes] will call Japan."

If, as I've suggested, any attempt to know a system "somewhere out in the world" is necessarily filtered through and altered by the knower's own perspective, then Burch's borrowing of an East as a system of codes useful for intervention in the practices of the West has nothing inherently wrong about it. [6]

But two hesitations about such a project are necessary. First, given that there is a real, living Japan, there may be a kind of cultural imperialism in an attempt to create a fictive Japan — the real one denied or even repressed in those moments when it doesn't fit the model. [7] An example of this occurs on page 173, when Burch interprets the oft-noted (and incorrectly designated) use of "low-angle camera position which has come to be identified with the films of Ozu." (There are few low-angle shots but many scenes shot from a low camera height.) Burch notes that

"many 'expressive' explanations have been offered, of which the most pertinent is Sato Tadao's suggestion that this position endows all pro-filmic activity with a 'ceremonial quality.'"

But Burch goes on to counter this interpretation:

"This peculiarly Japanese reaction is perhaps difficult for us to appreciate fully except insofar as religious and other ceremonies in Japan do take place close to the floor and involve low bows and prostration. However, the key to Ozu's specific, radical use of this procedure lies in more theoretical considerations …" 

Here is a clear case of Burch's refusal to admit that his reading is precisely that: a specific interpretation. Instead, against the Japanese critic's imputed inability to know what Japanese cinema is up to, Burch has the "key." The phrasing by which Burch assigns "radical use" to Ozu's films and not to Burch's reading is precisely that procedure which attempts to objectify a subjective position. Although Burch's title suggests that he is trying to assume the position of outside observer and not participant, this position is constantly belied by his attempt to speak for the Japanese, to explain what their texts mean. This in a writer who admits he doesn't know Japanese very well and relies on "authority" texts for much of his historical information.

Burch even goes so far as to see his inability to understand spoken Japanese as an advantage since it supposedly allows him to concentrate on formal aspects of "nodes of representation" and not be distracted by the expressive functions of sound. Yet sound is not just an aspect of content but also part of the formal structure. Even the most seemingly representational, expressive Hollywood film would gain in formal complexity and would seem less expressive and more experimental if one didn't understand English. From his assumed position of superiority Burch declares that Japanese thought is lacking in theoretical insight or rigor — he talks of "the Japanese disdain of theoretical practice" (p. 12). And so he steps in to provide that theory:

"One of the principal assumptions of this study is that the critical framework developed in France over the past decade … provides elements toward an understanding of the far-reaching implications of le texte japonais (the Japanese text]" (p. 13).

Even if Burch's text reveals an inevitable shaping of object by theory, it is still necessary to evaluate the uses of that shaping. If all theory is selection, reduction, an structuring, we can nonetheless ask the value of what has been selected, how it's been reduced, and what effect that structuring will have. In asking this of To the Distant Observer, the second problem of Burch's historical method presents itself.

Burch's claim to construct les texts japonais enters into direct contradiction with his other goal: to write a real, objectively true history of Japanese cinema. On the first page of text, Burch writes, "My approach is, of course, historical in every sense." But history, in most senses, can't be written by a distant observer, especially when that distant observer filters the object of his/her history through different cultural structures (in Burch's case, his place in a particular tradition of critical theory interested in questions of formal innovation). Burch superimposes his concerns — the value of formal innovation as against representation and expression — onto the whole value system of Japanese culture without examining how or if these might be relevant or even operative values in another system.

In this, it seems to me that he ignores a fundamental point about art's integration into a society's systems of representations, a point most clearly elucidated by the art historian Ernst Gombrich, in his Art and Illusion (Princeton University Press, 1960). Namely, what a society considers as fitting or subverting its systems of representations is not something in, or objectively definable in, the art itself but rather is a definition by the society of the place of that art. To take one of Gombrich's examples, the first drawings of rhinoceroses by European travelers to Africa depicted the rhinos as wearing a coat of armor. The artists were assimilating a new experience to one they already know (coats of armor for humans). However, because this was how rhinos looked to them, the Europeans overlooked, did not even see, the disparities between the drawings and the actual look of rhinos. The forms the artists used took their meanings from within a social context; it was this context which defined what was real and what wasn't.

Formal innovation, Gombrich suggests, is not some inherent value nor some eternally progressive attribute of all art at all places, at all times. Indeed, Gombrich goes so far as to argue that a particular way of seeing, upheld by a group's whole system of values, may lead to a kind of overlooking, a not-seeing of elements which don't fit the way of seeing of that particular group. The early drawings of rhinos may seem unreal in the twentieth century but it is our own value system, different from an earlier one, which determines this sense of "unreality."

Burch's text is precisely am act of interpretation, a point of view, which won't admit that that's what it is. Revealingly, the essay, "Positions," which Burch co-wrote with Jorge Dana, demonstrates the kinds of sleight-of-hand which Burch has to engage in to avoid confronting the question of just what reading and interpreting are. [8] "Positions" argues that even though moments in Fritz Lang's SECRET BEYOND THE DOOR (1948) unveil the project of illusion in a film committed overall to the tenets of illusionism, these moments are overlooked by Western spectators whose habit of seeing films as expressions of life causes them to ignore, or minimize, non-expressive or even anti-expressive elements:

"the upward tilted master shot shows them [the newlyweds of the story] against a back-drop onto which the shadows of some stylized foliage is projected. It is a completely 'unreal' image which, at the sane time, could never be perceived as such by the 'average' cinema-goer accustomed as he [sic] is to accepting landscapes painted, back projections, and the like, as image of reality" ("Positions," p. 65-66, my emphasis).

Burch and Dana acknowledge that SECRET BEYOND THE DOOR has a moment which is not immediately a moment of Realism. But because the film is a Hollywood film and not a Japanese one, Burch and Dana have to explain this moment away, to suggest that the audiences wouldn't have noticed its unreality. While Burch admits here that force of habit can change, can reduce, the potential effect of formal innovation, he refuses in To the Distant Observer to admit that the same kind of practice could happen in the watching of Japanese film, that spectators there could also learn to read the films as "images of reality." Significantly, that such films often appear expressive has been precisely the experience of many viewers (see, for example, Sato Tadao's aforementioned understanding of Ozu, an understanding which Burch has to dismiss as "peculiarly Japanese").

What formalism institutes is a clear-cut split between two practices — representational art and representational ways of seeing on the one hand and innovative art and innovative ways of seeing on the other — and it irrevocably decides which side certain films belong on. History takes the form of two monoliths facing each other. It is this starting point which consigns formalism to its limitations. Most prominent among these limitations is its inability to deal with contradiction in anything but a mechanical way. The monoliths are in place, complete and whole, and the only possible contradiction that can exist is the eternal one between them. Burch's theory is a replay of the assumptions of manipulation theory, of an elitist theory of art which assumes that a mass art like that of Hollywood can never be anything but repressive and manipulative of its audience. As the case of SECRET BEYOND THE DOOR shows, Burch sees a whole order of cinema as tied, without any internal contradictions, to a repressive function, and another cinema tied, without any internal contradiction, to a critique of that repression.

It is certainly true that there is a regularity to the presence of certain kinds of forms in certain kinds of cinema. For example, as the quantitative research of Barry Salt demonstrates, there is a statistically verifiable preponderance of certain shot lengths, camera distances, etc., in Hollywood films of the 1940s. [9] But that doesn't necessarily imply that any particular effect inherently exists in those techniques; all it literally tells us is that those techniques recur to such and such a degree. Any assignation of a particular effect to the forms will have to come either from the projections of interpretations onto form or from historical research on the ways of seeing which adhere in particular viewing situations.

More than that, generalizations about a form can only work at the expense of denying the ways particular films can employ those forms in the whole fabric of the film. Formal analysis is not irrelevant to political criticism — the deployment of techniques is indeed a source of the meaning-effect of a film. But formal qualities can only be studied in terms of their particular viewing situations and in particular films. In an extreme lapse into formalism, Christian Metz suggests that off-screen characters automatically invite audience identification since the audience is in a similar off-screen state. [10] But this is to ignore the ways particular kinds of off-screen characters (for example, the monster lurking off-screen as the target of an on-screen character's look) might deflect any generalizable quality of identification, if indeed such a general condition exists. Metz's position here is a return to that "generality in the institution" which Stephen Heath and Gillian Skirrow refer to and which encloses the open-endedness of artistic practice within the bounds of categorical thinking.

In a useful rejoinder to the argument that there is a political inevitability to the meaning of film techniques. Philip Rosen has shown how one cut in SEVENTH HEAVEN (Frank Borzage, 1927) undermines the previous coherence of the film's narrative. [11] A classic Hollywood film, SEVENTH HEAVEN deals with two lovers separated by war but who deny any danger to their situation because they believe that their love is higher and stronger than all other forces. The hero is gravely wounded at the front, and we see a priest administer the last rites. In the next scene, the same priest goes to the girlfriend back in the States to tell her of the boyfriend's death, and now that she is on her own, the girlfriend begins to accept the courtship of another man. Cut to: the hero struggling through a crowd to get home, not really dead after all.

Rosen reads this ultimate denial of the laws of death as a consequence of Borzage's contradictory beliefs as both a materialist and a spiritualist. Although perhaps too reliant on biographical information and analysis of authorial psychology, Rosen's reading of contradiction here is an important one because it emphasizes the contradictory insertion of any text into a social moment and the contradictory place of any particular technique in the particular practice of a film. The cut here gains its meanings from its place here, in this film, in relation to Borzage's politics, in relation to Rosen's own perspective as critic.

In contrast to this open-ended understanding of the politics of technique, of the possible meanings of the institution of cinema, formalism tries to limit contradiction by deciding in advance what is contradictory and what isn't. It treats certain practices as always fixed in meaning and so ends up as a more theoretically refined version of the complex-art/mass-art split which has plagued theory of film from its inception. Mass art in this view is the world of representational texts on the one side and passive spectators on the other side, bound by the manipulating powers of the text. Rosen's point, against this, is that texts don't express or practice an ideology. They exist as part of a complex and contradictory interweaving of ideological practices, a plurality of effects which differs for each different insertion of the text into social practices. Any text is internally contradictory because its social field is contradictory.

Formalism, however, can only recognize external contradiction; the cinema is a given, a fixed institution working to set ends, a collection of meanings which is always predictable in its workings. Contradiction, in this formalism, is either something that the text generates and then, represses (as is the case for Burch with SECRET BEYOND THE DOOR) or something that can only assail the institution of cinema from a position of marginality (as is the case for Burch with the Japanese cinema — a cinema most of whose texts are not available in the West).

In concentrating on narrative and representation as the forms which it deems most linked to bourgeois ideology, contemporary formalist theory gives an exaggerated importance to those forms. It assumes that the ideological force of a social practice, as either supportive or critical, lies solely in the way that practice upholds or refuses illusionism. Political struggle becomes a battle of the coded against that which escapes the repressions of the code. The existing political system becomes a place of homogeneity, coherence, organization, while subversion becomes the realm of the alogical, the trans-rational, the avant-garde. This sets up a dichotomy which is problematic both as a dichotomy and in the values or qualities it assigns to respective terms.

What happens to the notion of progressive cinema in such a schema is revealing. Coming upon a preexisting, monolithic form (the so-called "Hollywood paradigm"), oppositional cinema can only be understood in formalist theory as a play with codes, an excess, a surface disturbance of the monolith's form. The diversity of forms of political struggle can find no place in such an either/or approach. The idea that political struggle in art might take any other form than that of formal innovation gains no recognition from formalism. The result can only be an uncritical endorsement of avant-garde art. (I am not suggesting that there is no justification possible for avant-garde political art. I am merely suggesting that formalism answers the questions of such art before they're even asked.)

The historical fate of many avant-garde movements, their recuperation, suggests that irrationality, trans-logic, or whatever other qualities are attributed a subversive function in art can as well be potentially socially functional and supportive qualities. In some way, what has happened in formalism as a politics of art is similar to what French sociologist Jean Baudrillard sees as often, happening in reductive applications of the Marxist critique of political economy. Out of the multitude of practices which constitute the political identity of a society, one practice (for example, the practice of economic relations, according to a reductive Marxism is isolated from the complicated field of the interaction of practices and made to serve as the tool which explains everything else in an unequivocal way. [12] In contrast, Baudrillard suggests that politics and political control may not be so much a matter of control by one particularly dominating form or code (such as a code of Realism) but by a fluidity, a free-floating exchange and production of signs, whose political force lies in the fact that they have no one meaning, that, instead of expressing one value, they express a number of meanings at the same time. By this, the world can seem richer, it actually is.

Baudrillard goes on to suggest that this proliferation of signs (as in the different brands of the same product) has to be seen as a strategy in the game of social distinction and the maintenance of social hierarchies. Some groups actually possess values and power while other groups can only mimic that possession in the substitute realm of signs, not realizing that their ability to manipulate signs, to generate new signs, to play with forms, is no real power at all but only its alibi. For example, the person who can choose from an array of cars has the choice of different forms (a blue car, say, rather than a red one). But this sort of choice is a limited one, one which hides the fact of all the things that one may not have the power to choose (for example, one may be able to choose between a red and green Volkswagon but still not have the power to choose a more expensive car or to live where one does not have to rely on private automobiles).

What formalist theory will valorize as formal play, as innovation in excess of narrative and representation, Baudrillard therefore sees as a potentially reinforcing component of social position. The play of form makes a nice fit with the way capitalism attempts to maintain its values simply by a change of names, by a recourse to the new, the different. Bound to a real order of social relations, people often accept the compromise position of power at the level of signs, not realizing that a play with signs, an attempt to change value through signs, may be a detour. Like Derrida's understanding of language as a play of differances which connects to no reality beyond it, formal play, as Baudrillard explains it, is an endless game, a continuous substitution of sign for sign — a play which cannot break out of infinite circulation in the realm of signs to engage in any kind of effective praxis. Signification becomes independent of the real social position of its players who can have a freedom with words and forms that they can't have elsewhere. Moreover, it is this separation which allows the social hierarchy to continue:

"This false dynamic is in fact completely overrun by the inertia of an unchangeable social system and its discrimination in the realm of real power" (Baudrillard, p. 55, my emphasis).

This is to suggest that much of the formal experimentation that a formalist critic like Burch values and encourages serves the precise function of making people see freedom where there is actually constraint of a very different sort. In his essay, "Upon Leaving the Movie Theater," Roland Barthes epitomizes the limitations of such a position when he argues that the central repressive function of film is its fixing of the spectator to an order of illusion, a mirror-like activity:

"the image captivates me, captures me, I adhere to the representation, and it is this adherence which founds the naturalism (the pseudo-nature) of the filmed scene." [13]

Barthes is then able to suggest that a different attitude in the spectator, an attention to other aspects of the film experience, would cut against this:

"I have two bodies at the same time: a narcissistic body which looks, lost in the close mirror, and a perverse body, ready to fetishize, not the image, but precisely that which exceeds it: the grain of sound, the auditorium, the blackness, the obscure mass of other bodies, the rays of light, the coming, the going…" (ibid.).

But this excess has always been part of the film experience, one of the effects we go to movies for. The fascination of going to the movies is often that fascination for story plus a something more: a play of signs, a liberation of forms, a breakdown of illusion. This richness of experiences, this burst of signs which is the seeing of a movie, may exceed a narrative or the expression of a value system, but, as Baudrillard's analysis of commodities suggests, this excess is not necessarily socially excessive.

Ultimately, what a valorization of form and what most avant-garde theory lack is an understanding of the relation between artistic practice and other practices, especially the practices of social interraction. Granting an autonomy and effect to art that it doesn't necessarily have, the adherents of an avant-garde aesthetic diminish political struggle by generalizing from their own situation and rewriting history from a particular and limited perspective.

However, at the same time, the real complexity and struggles of history exceed and show up the limitations of the formalist position. Significantly, the course of Burch's own book gives in to the force of a kind of "political unconscious," a reminder of the political struggles beyond formalism. As if he in some unconscious way realized the limitations of an aesthetic so committed to form alone, the later chapter's of Burch's book deal with directors such as Nagisa Oshima who are explicitly radical in their assault on the meanings — cinematic but also non-cinematic — of Japanese culture. In these final 35 pages (of a 363-page book), there is a glimpse of a cinema which has change as its specific project, but it is no more than a glimpse. Despite his own claims to a Marxist aesthetic, Burch's attention points elsewhere, and in this lies the problem of his work.

Notes

1. For a useful introduction to some of the political issues which Burch's work raises, see the debate between Martin Walsh and Chuck Kleinhans in Jump Cut, 10/11.

2. Burch's explicit reference is to Derrida's De la grammatologie (1966), which is available in English translation as Of Grammatology (trans. Gayatri Chakravorty Spivak, Baltimore: Johns Hopkins University Press, 1976). A useful introductory piece by Derrida is "Structure, Sign, and Play in the Discourse of the Human Sciences" in The Struturalist Controversy: the Language of Criticism and the Sciences of Man, ed. and trams. Richard Macksey and Eugenio tionato (Baltimore: Johns Hopkins University Press. 1972).

3. This material is forthcoming as a book, Burch has so far published two essays on Primitive Cinema in English: "Porter, or Ambivalence" in Screen 19, no. 4 (Winter 1978-1979): 91-105; and "Film's Institutional Mode of Representation and the Soviet Response" in October 11 (1979): 77-96.

4. I owe the term "formal spectator" and the concept it refers to to Robert Vianello, Film Studies, UCLA.

5. Stephen Heath and Gillian Skirrow, "Television: a World in Action," Screen 18, no. 2 (Summer 1977): 7-59.

6. "The people who most need to study this [Japanese] cinema in its most 'radically' Japanese form are those committed to constructing a thorough-going critique of the dominant modes of Western cinema" (p. 17).

7. As recent feminist critiques of the use of the category of "Woman" as Other have suggested, there is also something fundamentally oppressive about any appeal to a particular group (for Burch, the Japanese) as inherently critical by the fact of a supposed otherness. The consequence of such a procedure is to turn groups into essences, to give repressed groups no power other than a virtually magical (and unconscious) one as outsider. This is to rob social forces of the real power of their possible interventions in history.

8. Noel Burch and Jorge Dana, "Positions," Afterimage 5 (Spring 1974): 40-65.

9. Barry Salt, "Statistical Style Analysis of Motion Pictures," Film Quarterly 27, no. 1 (Fall 1974): 13-22.

10. See "The Cinematic Apparatus as Social Institution — An Interview with Christian Metz," Discourse 1 (Fall 1979): 7-37.

11. Philip Rosen, "Difference and Displacement in SEVENTH HEAVEN," Screen 18, no. 2 (Sunnier 1977): 89-104.

12. I base my contents here on Baudrillard's Pour une critique de l'economie politique du signe (Paris: Gallimard, 1972). In subsequent texts, Baudrillard abandons any affiliation with Marxism.

13. Roland Barthes, "En sortant du cinéma," Communications 23 (1975) (translated by Susan White and Bertrand Augst as "Upon Leaving the Movie Theater," University Publishing 6, Winter, 1979, p. 3).