The Poststritcturalist Mutation vocation of the cinema" as a kind of "writing machine." Through exemplary readings of films like .Mand India Song, Ropars explodes and disperses filmic signification rather than "taming" it. For Ropars film texts potentially put into play active, unsynthesizable "structural conflicts." The disjunctive capacity of montage, especially, can dismantle the sign by playing on the "differences" between material signifiers. Films are fissured by two forces, éeriture and counter-éeriture, one tending toward a disseminating scriptural energy, and the other toward sense and representation. (Bakhtin would have said between "centrifugal" and "centripetal" forces.) Brunette and Wills (1989) deploy Derridean categories in order to interrogate various totalizing notions which they see as surreptitiously informing film theory and analysis: the notions of narrative film; Hollywood as a self-identical coherent system; the primacy of the visual, seen as analogous to the primacy of speech over writing in the logocentric tradition. (Many of these notions had already been questioned without recourse to Derrida, of course.) Calling for a move beyond totalizations, the authors invoke the possibilities of an "anagrammatical" reading practice that sees cinema as writing, text, "an interplay of presence and absence, of the seen and not seen, in relations not reducible either to totalization or transcendence" (ibid., p. 58). Deconstruction has also influenced film theory and analysis indirectly, through certain writers themselves influenced by deconstruction, who rarely write about film but who are frequently cited by film theorists. Thus Judith Butler's work on gender becomes a key reference for queer film theory, while Gayatri Spivak's work on the subaltern subject and Homi Bhabha's work on "hybridity" and "nation and narration" become frequent (although often ornamental) references in film-theoretical writing. In political terms deconstruction has been seen as progressive, in that it systematically undermines certain binary hierarchies - male/female, West/East, black/white - that have historically buttressed oppression. Derrida has aligned himself with the feminist critique of Lacanian phallocentrism, for example. On the other hand, critics of deconstruction have complained that it is easily co-optable by academic 184 Textual Analysts elitists (as, for example, in the Yale School of literary deconstruction) and that its grand claims of "subversion" tend to be merely rhetorical. At times, in hyper-deconstructionist discourse, essentialism becomes the equivalent of original sin, resulting in a less-essentialist-than-thou sweepstake. Deconstruction also shifts political valence depending on who or what is the object of its critique. Progressive when it interrogates historically rooted social hierarchies (Man over Woman, West over East), deconstruction becomes regressive when it runs after the chimera of a completely de-essentialized thinking, handing over to language and discourse the collective agency rightfully belonging to human subjects. Textual Analysis The question of the text was at the very heart of Derrida's work, and j Derrida himself performed textual analyses (of Rousseau, Saussure jff and others). Deconstruction was on one level a form of textual ex-í egesis, an "unpacking" of texts, a way of interrogating their unspo-"\ ken premises while being alert to their discursive heterogeneity. And although textual analysis traces its long-term antecedents to biblical : exegesis, nineteenth-century hermeneutics and philology;_theFrench ; pedagogical method of close reading^Í£x^Uc0w?kÄ?.JMtSl>-3^A, -American New Criu^ism'sJ^imn^ ate antecedents include Léyl^to-^sj^s^ kstudy ortEe""open work,"J!lolandJB^rtJ^ ^woric^jmd^'tex^ asymptomatic reading" and "structuring absences,^and Derrida's workon^^ ,,. The emergence of the "film text" was thus rooted in multiple problematics and intertexts. The term transferred from literature to film the respect traditionally accorded the sacred word (first religious and then literary) and thus served to garner prestige for a maligned medium. In religious terms, film, too, has its quantum of revelation." When films are-texts rather than movies they become 185 Textual Analysis funnily of the same serious attention normally given to literátu Uvtiial analysis is also a logical corollary to auteurism: What would Miuhors" write, after all, if not texts? At the same time, the film text T\ a rünction of semiology's focus on film as the site of systematical! up; mized discourse rather than as a random "slice of life." The pre sumption of textual analysis that film as a medium deserves sériou ;,MuJ t distinguishes it not only from literary elitist writing but also lioni a journalistic criticism which sees film as mere entertainment II ľ uline Kael's boast'that she would never see a film more than oik. before writing about it had been made by a literary critic with itjiiid to Hamlet or Ulysses it would have been taken as a sign of 11 -i less or incompetence. Literary criticism, as Barthes pointed out tt.»s ilways a matter of "rereading." '; The concept of "text" - etymologically "tissue" or "weave" - con- í ttpiualizes film not as an imitation of reality but rather as an artifact a ,.. -i struct. In "From Work to Text" Barthes made two distinctions! '^■'ŕkľ._ľ^ Jlfincd as fo-e- PhenonLcn^Lsurrace of the object, for tWipledie. book one hoidsjn_oneVhand, i.e. a complered_product i.o.1 Jyíng~äh intenděcT and_pre-eristent* meaning. "Text" was de-*JiV ^^J^I^^^0^c^~^ÍSf^S^^ a producüonabsorbing -"'-LCt 1^. reader together!'"We now know," Barthes wrote, "that "rfic t.xt is npTáTi^ 'theological' meaning -UIk 'message' of an Author-God) but a multi-dimensional space in ;«In». I a variety of writings, none of them original, blend and clash" '1 . nes, 1977, p. 146). In S/Z Barthes fiirther distinguished be-'^,| the '"readerly" and the "writerly" text or, better, between rui'c -ly and writerly approacbesto texts. The readerly approach privily those values sought and assumed in the classic text - organic ii ml, , linear sequence, stylistic transparency, conventional realism. It >wits authorial mastery and readerly passivity, turning the author "ito a god and the critic into "the priest whose task is to decipher the U riting of the god" (Barthes, 1974, p. 174). The writerly approach, "' contrast, fashions an active reader, sensitive to contradiction and '-terogeneity, aware of the work of the text. It turns its consumer l»"o a producer, foregrounding the process of its own construction 11 d piomoting the infinite play of signification. 186 Textual Analysis Building on his own background in literary theory, Raymond m jjejlojaťatidressed some of the difficulties in extending HťeTärymod-"cíTíolíIm in his essay "The Unattainable Text." Whereas literary criticism emerges from centuries of reflexion, film analysis is of recent date. More important, the filmtext, unlike theliterary text, is jTorJ^cjuotable" (Bellour was writing prior ro_theexjgtěnce oŕ'Tľlne7" Jicans, VCRs, laser-disks, andxahkLteÍ£jdsÍQn^a^irne_when the very scarcity contributed to the mystique of film analysis). Whereas lit-"eřatuřTánd literary criticism share the same medium of words, film and film analysis do not. While the film medium deploys Metz's five tracks (image, dialogue, noise, music, written materials), the analysis of film consists of words. Critical language is therefore inadequate to its object; the film always escapes the language that attempts to constitute it. Bellour then compares film to other artistic texts in terms of their coefficient of "quotability." The painterly text is quot-ř: able, and can be taken in at a glance. The theatrical text can be £ rendered as written text, but with a loss of "accent." Bellour then * analyzes the uneven susceptibility of the five tracks of cinematic ex-- pression to verbal rendering. Dialogue can be quoted, but with a : loss in tone, intensity, timbre, and the simultaneity of bodily and ; fecial expression. In the case of noise, a verbal account is always a translation, a distortion. The image, finally, cannot possibly be ren-t dered in words. Individual frames can be reproduced arid quoted, " but in stopping the film, one loses what is specific to it - movement .tsdr" The text escapes at the very moment one tries to seize it. Given this obstacle, the analyst can only try, in "principled despair," to compete with the object he or she is attempting to understand. ^Metz^ilistinguished between two complementary tasks, a kind of jhot-rgyerse shot dialogue, as it were, between (1) film theory (the study of film language perse) arid (2J^hn^alyjsisJJ^ile_dnematic language is the object of cine-semiological theory^ the text is the object of filmolinguíštic ^a^wS^prTprictice, as we shall see, the distinctiorTis not always so clear). In Language and Cinema Metz developed the notion of the textual system, i.e. the undergirding structure or network of meaning around which the text coheres, even in cases, such as Un Chien Andalou, where the structure is one 187 Textual Analysis of willed incoherence. The structure is a configuration arising fr0m the choices made from the diverse codes available to the filmmaker >(. The textual system does not inhere in the text; it is constructed bv the analyst, in Language aná cinema Metz was not concerned with providing a "how-to" book for textual analysis, but rather with determining its theoretical "place." Textual analysis, for Metz, explores the mesh of cinematic codes (camera movement, off-screen sound) and extra-cinematic codes (ideological binarisms of nature-culture male-female), either across a number of texts or within a single text All films, for Metz, are mixed sites; they all deploy cinematic and non-cinematic codes. No film is constructed uniquely out of cin-.' ematic codes; films always speak of something, even if, as in the case of many avant-garde films, they speak only about the apparatus itself, or about the film experience, or about our conventional expectations concerning that experience. Metz's formulations had the advantage of socializing, as it were, the artistic process of creation. By foregrounding éeriture as the re-lelaboration of codes, Metz envisions film as a signifying practice not .dependent on obscure romantic forces like inspiration and genius but rather as a reworking of socially available discourses. However, in some respects Metz's socialization did not go far enough. In this sense, the Bakhtin-Medvedev critique of the Formalists in The Formal Method in Literary Scholarship can be extrapolated so as to apply to the Metzian view of textual systems. The Formalists described [ textual contradiction in íer.-p.s recióiení .of social-struggle, in metaphors evoking combat, struggle, and conflict. SňlCovsky, for exam pie, compared the advent of a new school of literature to a revolution, "something like the emergence of a new class."1 However, even the Formalists retreated from the implications of their own metaphor -it was "only an analogy" - and literary contradiction remained in a hermetically sealed world of pure textuality. Bakhtin and Medvedev, in contrast, took the Formalist metaphors seriously, especially those terms evocative of class struggle and insurrection: revolt, conflict, struggle, destruction, and even "the dominant" - but made them apply equally to the text and to the social itself (Pechey, 1986) The Metzian and the Formalist views of the text might be usefully 188 Textual Analysis complemented, then, by the Bakhtinian concept of heteroglossia j e. a notion of competing languages and discourses as they operate^ within both "text" and "context." The role of the artistic text, within a Bakhtinian perspective, is not to represent real-life "existents" but rather to stage the conflicts inherent in heteroglossia, the coincidences and competitions of languages and discourses. A social semi-otic of the cinema would retain the Formalist and the Metzian notion of textual contradiction,""but rethink it through heteroglossia. The languages of heteroglossia, Bakhtin argues (in words that echo Metz's affirmations about mutually displacing filmic codes), may be "juxtaposed to one another, mutually supplement one another, contradict one another and be interrelated dialogically."2 James Naremore's essay on Cabin in the Sky, for example, takes this "discursive" approach, seeing the film as relaying distinct discourses (mral-folk-lorist, Afro-urban-modernist, etc.). The publication of Metz's Language and Cinema was followed by an international deluge of textual analyses of films in journals such as Screen and Framework in Britain, Iris, Vertigo, and Ca in France, Camera Obscura, Wide Angle, and Cinema Journal m the United States, Contracampo in Spain, and Cadernos de Critica in Brazil. (Roger Odin focused on essays in French and found 50 analyses of this type by 1977.) Such analyses investigated the formal configurations of textual systems, isolating a small number of codes and then tracing their interweavings across a film. Among the more ambitious textual analyes were Kari H^ncr's analysis r'fShock Cr-rri-c' :•, Stephen Heatlrs analysis of Touch oJĹvil, Pien ^ Laudn/s analysis of Intolerance, Thierry Kuntzeľs analysis of The Most Dangerous Game, and Cabier/. analysis of Toung Mr Lincoln. What, then, was new in the semiotic approach tojextual analysis? Fim,~thlfalw~m£Öiöll^ sensitivity to the. filmic sjgnifier^an^jQjsg&óĚ^ as op- posedto the traditional emphasis on character and píotTSecond, analyses tended to be methodologically self-aware; they were at once about their subject - the film in question - and about their own methodology. Each analysis became an exemplum of a possible approach. In contrast to journalistic criticism, the analysts cited their 189 Textual Analysis wn theoretical presuppositions and critical intertcxt (many "analy-•s began with quasi-ritual invocations of the names of Metz, Barthes j-isteva, or Heath). Third, these analyses also presupposed a radi-illy different emotional stance toward films, one characterized by a ind of Brechtian distantiation, an oscillation between passionate )ve and critical distance. The analyst was supposed to adopt a schizo-hrenic attitude, both loving and not loving the film. Rather than a ngle screening, the analyst scrutinized the film shot-by-shot (the ěvelopment of VCRs has since that time democratized the practice f close analysis). Analysts such as Marie-Claire Ropars and Michel larie developed elaborate schemata for notation, registering such odes as angle, camera movement, movement in the shot, off-screen >und, and so forth. , Given the closeness of attention of such analyses it became impos-ble to try to say everything about a film. As a result, many analyses' )cused on synecdochic fragments of films. Thus Marie-Claire Ropars evoted 40 pägélľto the initial shots of Eisenstein's October and .ocha's Antonio das Mortes, while Thierry Kuntzel dedicated long nalyses to the ouvertuře sequences of films such as M, King Kong, id The Most Dangerous Gamey seen as condensed matrices of mean-lg. The dedication of many pages of critical writing to a brief seg-lent also indirectly demonstrated to high-art elitists that the same medium despised by others was actually the scene of veritable cornucopias of meaning. The analyses also varied widely in scale. The limits oťtne text might.be defined by a single image (for.example, 3 Ronald Levaco's and Fred Glass's analysis of the MGM logo), by a ji single segment (Bellour on The Birds), by an entire film (Heath on Touch of 'Evil), by the entire oeuvre of a filmmaker as examples of a "plurifilmic textual system" (René Gardies on Glauber Rocha), or even by a vast corpus of films (Michele Lagny, Marie-Claire Ropars, and Pierre Sorlin on the French cinema of the 1930s, and Bordwell, Staiger, ajidJľhampSOTjr^ classical-H-oliywoxiCL cinema). Textual analyses rejected the traditional evaluative terms of film criticism in favor of a new vocabulary drawn from structural Unguis-. tics, narratology, psychoanalysis, Prague School aesthetics, and lit- 190 Textual Analysis erary deconstruction. In what was perhaps an over-reaction against i " traditional film criticism, textual analysts often completely ignored issues traditionally central to film analysis: elements like character, acting, performance. Although most of the analyses generated by this wave belonged, broadly speaking, to the general semiotic cur-rent,"not all of them were rigorously based on Metzian categories. Marie-Claire Ropars Wuilleumier's extremely intricate analyses of such films as India Song and October synthesized semiotic insights with a more personal project inspired by Derridean grammatology. Many textual analyses were influenced by literary textual analyses, for example Julia Lesage's extrapolation of Barthes's "five codes" to Renoir's Rules of the Game (in Nichols, 1985). Some textual analyses were inspired-by Pröppian narratological methods (e.g." Peter Wollen on North by-Northwest)^ by Lacan's "return to Freud (e.g. Bellour on North by Northwest)^ or by other theoretical currents. While some textual analyses sought to construct the system of a single text, others studied specific films as instances of a general code informing cinematic practice. Here, too, the distinction is not airways clear, however; Raymond Bellour's analysis of The Birds offers both a microcosmic textual analysis of the Bodega Bay sequence of the Hitchcock film and an interrogation of broader narrative codes shared by a larger body of films; to wit, the constitution of the couple as the telos of Hollywood narrative. In two books, Kristin Thompson (1981; 1988) offered a programmatic alternative neo-' Formalist- me'tricd of textual ahalýsis/pérfóřS "the gräin^ôf^éniiotícs^Alfred Guzzetti (1981), meanwhile, offers a blow-by-blow account of the Godard film in terms of sound, image, and intertextual reference. The theoretical discourse concerning the cinema that developed in France in the 1960s was taken up in the 1970s by the British journal Screen and subsequently migrated to the United States and to many other countries with the growth of cinema studies programs, many of them with a strong Parisian link. (The Centre Americain d'Études Cinématographiques, which sent American students to Paris to study with leading French semioticians, was crucial in this regard.) Left-leaning versions of semiotics favored a 191 r k:> . I: ,. • Ku Qrrutinizing social and artis-subversive work oťdenaturahzanonby^ tic productions in order to discern the c ^ ^ operative in them. Film theory ^c"^tíona, disciplines, not course to the left of many other more ua _ ^^ only because ofa strong «French connecuo^^ ^ ^^ ^ quently moved dramatically «^ ^itural disciplines simultaneous emergencealongside su'n culture studies. M a as women's studies, ethnic studles,;^ P p way by the «mouldy result, film studies was never plague d ini tn namotá fields figs," entrenched conservatives who dominated like literature and history. industry also had insti- The emergence of film theory as a growth ^ ^ ^ ^.^ tutional causes: the inauguration^ ^^ Australh) ^ in major universities in France, Bntain' 0f theory testified to Brazil, and elsewhere. SOP^^^i thus indirecdy pro-the intellectual seriousness of mm ^ » ^ departments, vided a rationale for the creation of cínem ^ ^.^ ^ ^ Just as film had to legitimate itsdf as ^tuUOnal home base in legitimate itself as a discipline. With its in ^ acquircd con. the academy and the publishingindust^\ bbjsm of traditional siderable prestige and dissemination. ^ cukure and for film literary academics, with their scorn tor p.p^ ^ ^ ütcratuK| as a medium supposedly inferior to co ^^ to demon. painting, and music, inadvertently proaa ompcnsate by ^ strate its own seriousness, and at times ro tuoso displays of theoretical prowess. . , _____3 Interpretation and its Discontents . • ^reived bv filmiemiptics„came In the 1980s textual analysis as^C^nf On the one hand, l^ter-utt^^ textual analy- poststructuralist currents ^^^^^^^sjmtiajk:. :sTsTsll-akmTeWsem finitivdf captures film's meamngbyem^-----3' 192 Interpretation and its Discontents codes. On the other hand? the emerging field of cultural studies was not terribly invested in textual analysis. Its attitude was summeďup later in the words of Cary Nelson, Paula Treichler, and Lawrence Grossberg in their introduction to Cultural Studies (1992), where they state that "although there is no prohibition against close textual readings in cultural studies, they are also not required, [since] textual analysis in literary studies carries a history of convictions that texts are properly understood as wholly self-determined and independent objects." "~ Jacques Aumont and Michel Marie (1989) outline four possible critiques of textual analysis: 1 Its relevance is limited to narrative cinema. 2 It "murders to dissect," ignoring the organic unity of the text. 3 It reductively "mummifies" film by reducing it to its systemic skeleton. 4 It elides film's context, its conditions for production and recep-^ tion. i The first of these critiques misfires (since textual analysis is applicable to any object), while the second seems rooted in hostility to analysis per se, especially when performed in relation to an "unworthy" medium. But the last two have some force, and are in fact interrelated. When textual analyses are reductive, it is precisely because they are ahistorical and therefore fail to take production and reception into account. And the charge of ahistoricism is not answered satisfactorily by Aumont and Marie's suggestion that analysts "also" do history. The roots of the "decontextualization" of some textual analysis lie in the ahistoricism of two of the source movements of semiotics: Saussurean linguistics - particularly its tendency to cut off language from history - and Russian Formalism, with its preference Tor a purely intrinsic analysis. When analysts within the filmolinguistic tradition recommend that film scholars should also study - within "a kind of amicable division of labor - history, economics, sociology, and so forth, they recapitulate the approach taken by the Formalists themselves, who also recommended first the immanent study of the 193