Dance and Fetish: Phenomenology and Metz's Epistemological Shift

Dance and Fetish: Phenomenology and Metz’s Epistemological Shift DOMINIQUE CHATEAU AND MARTIN LEFEBVRE There is . . . something unique in the cinema, it combines two things that no one had combined up to now: the brute presence of the world and the subtleties and refinements of human speech. The cinema is the world having finally acquired speech [le cinéma c’est le monde devenu enfin parlant]. —Christian Metz1 Christian Metz is remembered today as having almost single-handedly transformed the culture of film studies. This widely held view was summarized by one commentator, who wrote that “with Metz a new research paradigm is born, as well as a new generation of scholars. The ontological theories are followed by methodological theories.”2 According to another, “Metz exemplified a new kind of film theorist, one who came to the field already ‘armed’ with the analytic instruments of a specific discipline, who was unapologetically academic and unconnected to the world of film criticism.”3 Of course, Metz didn’t just surge like a meteor on the scene of film studies. His arrival was “prepared” by the filmologie movement spearheaded in Paris by Gilbert Cohen-Séat and by two early film semiology essays published by Roland Barthes in La Revue internationale de filmologie.4 Yet it is Metz who is rightly remembered as the figurehead of film semiology. 1. From an undated manuscript (most likely from the mid- or late 1960s) entitled “Exposé-conférence de caractère très général sur le cinéma comme moyen d’expression.” Ms. CM1412. All manuscripts cited in this article come from the Fonds Christian Metz of the Bibliothèque du film in Paris. Reference to them will use the current manuscript number. English translations are ours. The Fonds has not been catalogued and was not available to the public at the time of writing this essay. Martin Lefebvre wishes to thank the estate of Christian Metz for granting him access to archival materials and to acknowledge the support of the Social Sciences and Humanities Research Council of Canada, which made it possible to consult the Fonds. 2. Francesco Casetti, Theories of Cinema, trans. Francesca Chiostri and Elizabeth BartoliniSalimbeni, with Thomas Kelso (Austin: University of Texas, 1999), p. 91. 3. Robert Stam, Film Theory. An Introduction (Oxford: Wiley-Blackwell, 2000), p. 108. 4. Roland Barthes, “Le problème de la signification au cinéma,” Revue internationale de filmologie 32–33 (1960); and “Les ‘unités traumatiques’ au cinéma. Principes de recherche,” Revue internationale de filmologie 34 (1960). On the filmologie movement, see the special issue of CINéMAS, “La filmologie de nouveau,” ed. François Albera and Martin Lefebvre, 19, nos. 2–3 (2009). OCTOBER 148, Spring 2014, pp. 103–132. © 2014 October Magazine, Ltd. and Massachusetts Institute of Technology. 104 OCTOBER In examining the role played by phenomenology in Metz’s work it is important to begin by noting that he did not train formally as a philosopher. Enrolled at the École normale supérieure (rue d’Ulm) from 1951 to 1955, Metz elected to pass examinations for a licence in German studies (in 1953) and for an agrégation in lettres classiques (1955) at the Sorbonne. These were followed by a diplôme d’études supérieures in ancient Greek (1954) and, in 1971, by a thèse d’état in general linguistics, defended at the newly created Université Paris V—René-Descartes, on the grounds of prior publications.5 And yet it was difficult in Paris in the mid-’50s and early ’60s not to come into contact with phenomenology. Sartre and MerleauPonty had been giants dominating the intellectual stage in the aftermath of the war until the arrival of Lévi-Strauss and the rise of structuralism. Moreover, whoever was interested in the cinema would equally face strands of phenomenology in the writings of critics and theorists as diverse as André Bazin, Marcel Martin, Amédée Ayfre, Albert Laffay, Jean Mitry, and several filmologues, including Etienne Souriau, René and Bianka Zazzo, Albert Michotte Van den Berck, and Henri Wallon. Indeed, inasmuch as film critics and theorists as far back as Hugo Münsterberg and Rudolf Arnheim had concerned themselves with problems such as the psychology of film perception, cinema’s “impression of reality,” or its ability to “reveal” reality, they would inevitably come into contact with phenomenology or phenomenologically inclined psychology. Metz, of course, championed structuralism and semiology. His early contributions are altogether closer to linguistics than to philosophy, and his “filmolinguistic” work, as he sometimes referred to it, is for the most part devoid of overt philosophizing. Regarding his thèse d’état in 1961, for which he filed a project entitled “Cinéma et langage,” Metz explained in an annual report to his initial rapporteur, Georges Blin, that he sought to provide a juncture between filmologie and Saussurean linguistics.6 Now, neither structural linguistics nor semiology can be said to be natural allies of phenomenology. In fact, for many, the rise of structuralism in the 1960s marked a rejection of any type of “philosophy of the subject” and thus explicitly signaled the waning of phenomenology and of its dominion over much of the humanities. Both discourses have been associated with different—some would say unbreachable—epistemic “attitudes,” semiology being primarily positivistic and relying on a descriptivist scientific attitude, while phenomenology—in several of its incarnations—showing itself to be resolutely speculative and non-positivistic, more concerned with “comprehension” or “understanding” than “explanation,” and more interested in diachrony and the role of the subject (as conscious subject of knowledge) than in synchronic structures and immanent objects. But as different as structuralism and phenomenology could be in theory, in practice there nonetheless 5. The subject for his diplôme d’études supérieures (or master’s thesis) was “Isocrates and Pan-Hellenic Politics in Fifth and Fourth Century B.C. Greece.” 6. Blin was a professor of literature at the Sorbonne. He joined the Collège de France in 1965 and held the chair in modern French literature until 1988. Metz dedicated the first volume of his Essais to him. By the time Metz defended his doctorat d’état in 1971, Blin had been replaced as rapporteur by the linguist André Martinet. Dance and Fetish 105 existed certain points of contact through the works of several individuals. Some phenomenologists, such as Merleau-Ponty or Mikel Dufrenne, carefully read Saussure and sought to incorporate aspects of structuralism into their work, while phenomenological themes sometimes surfaced in the work of structuralists like Jakobson or Lévi-Strauss even though the latter, for example, clearly rejected phenomenology’s involvement with individual experience and consciousness and with the intentionality of the subject.7 In the case of Metz, phenomenology and structuralist semiology openly coexisted in his first publications, and, as we shall see, this coexistence continued—though in a more muted and paradoxical (or inverted) form—well into his later, psychoanalytic period. This paper proceeds through three chronological sections, examining in turn Metz’s early “filmolinguistic” period (1964–1967), his middle or pan-semiological period (1967–1975), and his late psychoanalytic period (1975–1985). In all three of these periods phenomenology plays an important and sometimes pivotal role. I. Early works: Semiology as Phenomenology or Phenomenology as Semiology At least three of Metz’s early key essays—“Le cinéma: langue ou langage?,” a programmatic piece, initially published in 1964 in Communications 4, announcing the project of filmolinguistics, or film semiology, which would lead to the development of the “grande syntagmatique” and culminate in the publication of Langage et cinéma in 1971; “A propos de l’impression de réalité au cinéma” from 1965; and “Remarques pour une phénoménologie du Narratif,” published the following year—are important for grasping the initial phenomenological strand in his work. These last two essays were in fact compiled in the opening section of the first volume of Essais sur la signification au cinéma (1968) under the heading “Approches phénoménologiques du film.” Interestingly, this first volume of Metz’s writings was published at Klincksieck in a collection edited by Mikel Dufrenne, an influential phenomenologist whose own work centered on aesthetics. In fact, just a few years after the release of that first volume, Dufrenne presided over the jury at Metz’s doctoral defense, alongside linguist André Martinet and literary scholars Georges Blin and Roland Barthes. Dufrenne’s influence on Metz’s early articles can best be seen in the copious manuscr ipt notes Met z took while reading Dufrenne’s t wo -volume Phénoménologie de l’expérience esthétique, first published in 1953.8 Indeed, several passages from these notes found their way into “Le cinéma: langue ou langage?” These passages, as we shall see, concern the problem of “cinematic expressivity.” 7. See especially Claude Lévi-Strauss, La Pensée sauvage (Paris: Plon, 1962). This was an all-out attack on Sartre, phenomenology, and the cogito—though the book was dedicated to the memory of Merleau-Ponty. For an excellent account of debates in French philosophy since the 1930s and of how Merleau-Ponty managed to ally himself with those who would, after his death, become the enemies of phenomenology, see Vincent Descombes, Le Même et l’autre (Paris: Minuit, 1979). 8. Mikel Dufrenne, Phénoménologie de l’expérience esthétique, 2 vols. (Paris: PUF, 1953). 106 OCTOBER “Le cinéma: langue ou langage?” is a rather long and somewhat sprawling essay that moves uneasily at times between aesthetics, linguistics, and phenomenology, seeking nonetheless to set forth the project of a semiology of cinema. Metz begins by synthesizing the existing literature on the theme of “language and cinema”—a literature he had been reading since 1961, when he began work on his doctorate—and then moves on to consider various issues that a Saussure-inspired structuralist investigation of cinema would have to face. One of these issues concerns the distinction between expression as the phenomenological order of meaning and signification as the linguistic or semiological order of meaning. Metz, as we shall see, carefully considered this distinction in his handwritten notes on Dufrenne’s book. For the moment, however, we may simply note its first occurrence at the outset of “Le cinéma: langue ou langage?,” where he rehearses the Bazinian opposition between Eisenstein and Rossellini. Metz writes: From the point of view of expressive mechanisms we can distinguish between the “natural” meaning of things and of beings (continuous, global, without any distinct signifier: thus the joy that can be read on the face of a child)—and deliberate signification. The latter would be inconceivable if we didn’t already live in a world of meaning, yet it is conceivable only as a distinct organizational act by which meaning is redistributed: signification loves to sharply delineate discontinuous signifieds that belong to an equal number of signifiers. Its task, by definition, is to give form to an amorphous semanticism.9 But while Bazin uses the two filmmakers for the purpose of aesthetic criticism, Metz dispels the prescriptive aesthetic opposition and redeploys the terms of the debate with regard to what he conceives to be the distinct domains of concern of phenomenology (especially the phenomenology of art) and semiology with respect to meaning. This leads him to define “cinematic specificity” on two levels: that of filmic discourse—which is the phenomenological level of artistic expressivity—and that of image discourse—which, in principle at least, corresponds to the semiological level. It is this distinction that is first mapped out in Metz’s notes on Dufrenne. As a phenomenologist investigating the nature of aesthetic experience (which includes the experience of art but is not limited to it), Dufrenne developed a whole section of his argument by distinguishing between expression on the one hand and signification or representation on the other. Dufrenne’s point is simple enough, though: According to him, art—even verbal art—becomes aesthetic through expression, never through signification or representation. That is to say that an art object—a poem, a painting—may signify or represent something, but 9. Christian Metz, “Le cinéma: langue ou langage?,” Essais sur la signification au cinéma, vol. 1 (Paris: Klincksieck, 1968), p. 45; our translation. (It is well known in the film-studies community that Metz’s essays, at least up to The Imaginary Signifier, are not always adequately translated with regards to technical terms. Therefore we have opted to use our own translations.) Dance and Fetish 107 not in its identity as aesthetic object or work of art. Whereas signification or representation may count as a facet of the work, only expression is commensurate with the work qua work of art (i.e., the work in its specifically aesthetic aspect). Dufrenne defines expressivity in art as the meaningful correlate of sensual form and subject matter, which transcends or exceeds any signification or representation—any use of codes.10 But as Metz indicates in his notes on Dufrenne, two forms of expression must be distinguished: natural expression and aesthetic expression. He writes: “Natural expression is the fact that things have meaning. Aesthetic expression is the fact that certain things (works of art) have a certain meaning (a style).”11 But such meaning is distinct from that of signification or representation. In his notes, Metz illustrates Dufrenne’s idea of expression as well as the division of natural and aesthetic expression with a shot taken from Eisenstein’s unfinished ¡Que viva Mexico!, an example that returns almost verbatim in “Le cinema: langue ou langage?”: Let’s consider Eisenstein’s famous image (¡Que viva Mexico!) representing in triangular form three peons being trampled by horses. Where does its natural expression lie? It is in the fact that the faces of the peons express horror (or terror, or suffering). . . . [B]ecause it is the hallmark of the real—here the human face—to have the power to express things, the analogical reproduction of this human face will express the same things. In this same “shot” of the film, where does the aesthetic expression lie? It is elsewhere, it is in this triangular construction—which has nothing natural about it—a construction that has a form and that expresses something like majesty, grandeur, which expresses not horror or terror, but the fact that this horror and this terror merge with the struggle of the Mexican people, with the beauty of the landscape, with the revolutionary grandeur of a history, etc. . . . The aesthetic expression isn’t: the face of the peon, but instead: the fact that three peons have been taken and arranged in a specific fashion. To put it differently, aesthetic expression assimilates but exceeds natural expression. At the level of the signifiers: the triangular form assimilates and exceeds the faces of the peons, the landscape, etc. . . . At the level of the signifieds: the painful grandeur of Mexican history assimilates and exceeds the pain of the peons. Both are nonetheless expressions since painful grandeur can be read onto the triangular form itself—and pain can be read onto the faces of the peons themselves. In both cases, the signified can be read on the totality of the signifier—or, better yet, in both cases there is no signifier and no signified but a thing-having-meaning. But it is not the same thing and it is not the same meaning.12 10. Of course, not all art forms have a subject matter: For Dufrenne, abstract painting, music, architecture, dance are often expressive without signifying or representing anything. 11. Christian Metz, undated manuscript notes on Phénoménologie de l’expérience esthétique, ms. CM872 B8, p. 27a. 12. Christian Metz, undated manuscript notes on Phénoménologie de l’expérience esthétique, ms. CM872 B8, p. 27a/b-28a/b. 108 OCTOBER It is precisely this train of thought that leads Metz in “Le cinéma: langue ou langage?” to declare the cinema—like verbal art—to be an art of connotation. More specifically, it is an art form whose aesthetic expressivity builds upon (i.e., assimilates and exceeds) what it designates (or represents) in great measure through analogy.13 We see Metz here grappling with the various ways cinema can come to stand for something and trying to account for both the phenomenological and semiological conceptions of meaning. In an attempt to clarify the problem at hand he offers the following diagram; the upper part pertains to semiological meaning and the lower part to phenomenological meaning: The early Metz adopted Dufrenne’s view according to which the primary difference between signification (now reordered as a subtype of designation by Metz) and expression is that, with the latter, “meaning is immanent to what is perceived.”14 Of course, this was an idea shared by phenomenological approaches to meaning, and to cinema in particular, from Merleau-Ponty to Bazin, all the way to Mitry. Yet, however useful the above diagram may have been for Metz in laying bare the basic opposition between semiological and phenomenological conceptions of meaning, it was obvious for him that it didn’t account for the fact that designation in cinema—which for the most part falls under the heading of representation—is also expressive by virtue of the world’s natural expressivity, which it reproduces photographically. The problem, in a sense, is that one could account for cinematic designation in two ways: On the one hand, the image and what it shows—the signifier and the signified—are distinct, as is the case with language (though unlike language they are analogous). On the other hand, because of photographic analogy, the expressive qualities of the world are carried over to the image that represents it. And when these expressive qualities are used with style (Metz here borrowing Dufrenne’s conception of style), a supplementary, aesthetic meaning emerges: aesthetic expressivity. The idea finds its final form in “Le cinema: langue ou langage?” when Metz asserts that “in the cinema connotation is homogeneous with denotation, and is, like it, expressive.”15 Metz’s concern for expressivity in this essay seems to originate from early versions of his project for a doctoral dissertation. In an outline submitted to Georges Blin at the beginning of 1963, he presented his project in two parts. The first section was said to be “historical” and centered on expressivity and verbal language in the cinema, looking at both the silent and sound periods and investigating the issue of “expressive” acting.16 The second section was presented as “philosophical, 13. Literary art, on the other hand, must develop expressively from what language signifies. 14. Christian Metz, “Le cinéma: langue ou langage?,” Essais sur la signification au cinéma, vol. 1, p. 85. 15. Ibid. 16. Expressive gesture and its relation with language was an important theme in Merleau-Ponty’s work starting with Phénoménologie de la perception (Paris: Gallimard, 1945). Dance and Fetish 109 filmological and linguistic” and concentrated on the relations between art, style, and language, as well as distinguishing “image discourse” and verbal language. Considering his notes on Dufrenne and the subsequent essay of 1964, it is fair to infer that phenomenology likely played a substantial role in thinking through this would-be thesis. However, as we shall see, the phenomenological idea of some meaning wholly immanent to perception is not one that would sit well with Metz’s later work, though, somewhat paradoxically, he was nonetheless careful not to refute phenomenology entirely. After 1964, the bulk of Metz’s efforts were devoted to investigating cinematic “designation” (or denotation), and he mostly shied away from aesthetic matters and phenomenological considerations.17 On the rare occasions when aesthetics came into the orbit of an argument, he never gave in to the temptation of some critics to call on phenomenological principles as a basis for a prescriptive poetics. Thus, in “Le cinéma moderne et la narrativité,” first published in Cahiers du cinéma in 1966, he severely criticizes those who would celebrate the “new cinemas” of the ’60s on the grounds that these films offered a more fundamental realism or a new form of objectivity better suited to fulfill the cinema’s revelatory “essence.” He writes: If [by these ideas] one implies some sort of cosmophanic power, some sort of aptitude for all revelations that the cinema carried inside itself from birth, but for which it would have taken a long time to become aware, then one is reverting to a mythology that was justly criticized by Jean Mitry, a mythology that hides behind the frock of phenomenology an essentialist realism whose only accomplishment is to salvage, at the level of the “natural meaning of things,” the terrorist univocity of signification which is otherwise being fought against in the name of ambiguity; these are the most disputable aspects of André Bazin’s and Roger Munier’s theories. . . . [T]he real lesson of phenomenology . . . isn’t by any means in the unilateral triumph of cosmophany, but in the obstinate affirmation of a definitive to-and-fro between the things that are and the person for whom they are, in this insurmountable “adverse spectacle” Valéry speaks of, and which can be found in its entirety in the assertion “there is,” the elementary form of existence and equally that of cinema: the “there is,” because it implies that there is something and that there is 17. There are at least two exceptions to this. In May 1971, Metz gave two talks at a symposium held at the Château de la Brangélie for the Centre régional de documentation pédagogique de l’Académie de Bordeaux: “Existe-t-il une approche sémiologique de l’esthétique?” and “Didactique et esthétique.” The first one was recently published in 1895, Revue d’histoire du cinéma, no. 70, 2014, pp.154–67, with an introduction by Martin Lefebvre. Moreover, though from a slightly different perspective, it could be argued that all of Metz’s work was centrally concerned with aesthetic matters, as argued by Martin Lefebvre in “Metz and Aesthetics,” The Semiotic Paradigm. The Thought of Christian Metz, ed. Margrit Tröhler, Julia Zutavern, and Guido Kirsten (Amsterdam: Amsterdam University Press, forthcoming). 110 OCTOBER someone for whom there is something, summons both the filmed object and the process of filming it.18 Such “to and fro” between subject and object is obviously one of the axiomatic ideas of the tradition inaugurated by Husserl and further developed in various guises by the likes of Heidegger, Sartre, and Merleau-Ponty. In Merleau-Ponty’s philosophy it is what gives both direction (sens) and meaning (sens) to embodied experience (or embodied intentionality). And it is precisely to such directionality and meaning that Metz turns his attention in his two most overtly phenomenological essays: “A propos de l’impression de réalité au cinéma” and “Remarques pour une phénoménologie du Narratif.” The first piece comes on the heels of research conducted by filmologues such as van den Berck, Edgar Morin, and Henri Wallon. Other thinkers associated with phenomenology are also cited, namely, Bazin, Laffay, and Mitry. Metz here is interested in what he considers the specific form of “empire” that films have on their viewers and their perception. Another way to put it would be to ask how the cinema present s it self to consciousness through percept ion, i.e., what is it s “meaning”? It lies, Metz claims, in the “feeling of directly witnessing a spectacle that is almost real,” “a filmic mode of presence that is largely credible,” and whose effect “rallies the masses and fills movie theaters.”19 Indeed, even non-realist, fantastic films find the source of their effectiveness in the fact that, in the cinema, even “the unreal appears realized and offers itself to the gaze under the guise of a factual efflux (surgissement événementiel) and not the plausible illustration of some purely conceived extraordinary process.”20 According to Metz, two factors are responsible for the cinema’s impression of reality: There is first what cinema as a vehicle for images (and sounds) presents to perception and, secondly, the unreal or imaginary nature of its fictions or diegesis. As a perceptual vehicle cinema offers a rather intricate configuration of moving photographs, and though he is not explicitly cited, Merleau-Ponty’s statement about the cinema “as an extremely complex form inside of which a very great number of actions and reactions are taking place at every moment” comes to mind in reading Metz’s efforts to describe the gestalt—he speaks of “fixed structures and figures”— commensurate with the impression he seeks to explain.21 Metz begins by noting that film images, inasmuch as they are “photographic,” partake of the photographic impression of reality described by Roland Barthes in his “Rhétorique de l’image,” an impression grounded in the rendering present for contemplation a reality from the past, a “having-been-there” distinct from the here and now of contemplation. As a result, photographic consciousness is restrained from full immersion in the present that characterizes the perception of the actual, for what the photograph depicts— 18. Christian Metz, “Le cinéma moderne et la narrativité,” Essais sur la signification au cinéma, vol. 1, pp. 193–94. 19. Christian Metz, “A propos de l’impression de réalité au cinéma,” Essais sur la signification au cinéma, vol. 1, pp. 14–15. 20. Ibid., p. 15. 21. Maurice Merleau-Ponty, “Le cinéma et la nouvelle psychologie,” Sens et non-sens (Paris: Nagel, 1948), p. 112. Dance and Fetish 111 although highly indexical of reality—stays resolutely in the past. It follows that photography’s intentionality—Metz uses the term visée, which can be translated as aiming22—is purely spectatorial and translates into “an attitude of exteriorized contemplation rather than a magical or fictional consciousness.”23 Therefore, although they share a parentage, photographic intentionality must be distinguished from filmic intentionality and its surgissement événementiel, which supersedes the purely external contemplation of photography. But it needs also to be distinguished, explains Metz, from another parentage, namely, theatrical consciousness, whose vehicle is the real, though what it represents is fictional. If neither photography nor theater is capable of providing the impression of reality to the same degree as cinema, it is important to consider what it is about the latter that characterizes our consciousness of it. Of course, the key discriminating factor with regard to photography is the advent of motion. Here, Metz reprises some of the arguments made earlier by the filmologues: Movement adds the mobility of the world to photographic analogy and therefore adds an extra layer of reality to what it shows us; it also gives corporeality to objects, which are now more likely to appear as detached figures on a ground.24 Both aspects indirectly bring more reality to film images, i.e., they increase the traces of the real into films. A third aspect, however, one not sufficiently analyzed by the filmologues, according to Metz, concerns movement’s direct contribution to the impression of reality inasmuch as it always appears on the screen as real and actualized, unlike photographic objects, which are perceived as absent and belonging to the past. So that even though the movement seen on the screen has technically occurred in the past (as the actors or the camera were moving), “the spectator,” writes Metz, “always perceives the movement as being present.”25 It is a presence, not a representation or a “copy.”26 As a result, “the impression of reality in cinema is also the reality of the impression, the real presence of movement.”27 Yet this conclusion seemingly introduces a paradox, or at least poses the question: If the impression of reality brought about by film is due to real presence (that of movement), why is it not greater still in the theater, where the entire vehicle is real? Any confusion, explains Metz, comes from not having sufficiently distinguished between the two factors mentioned earlier, that is to say between 22. Etymologically, intentionality is connected to the idea of pointing at something. Indeed, the Latin intendo means to “point at,” to “aim at,” or to “stretch toward” something. 23. Christian Metz, “A propos de l’impression de réalité au cinéma,” Essais sur la signification au cinéma, vol. 1, p. 16. 24. Especially Edgar Morin in Le cinéma ou l’homme imaginaire (Paris: Minuit, 1956) and Albert Michotte van den Berck in “Le caractère de ‘réalité’ des projections cinématographiques,” Revue internationale de filmologie, tome I, nos. 2–3 (1948). 25. Christian Metz, “A propos de l’impression de réalité au cinéma,” Essais sur la signification au cinéma, vol. 1, p. 18. 26. Metz explains why: “Movement is ‘immaterial,’ it offers itself to sight but never to touch. This is why it will not admit two degrees of phenomenal reality, the ‘real’ and the copy . . . to reproduce the sight of it is to reproduce its reality. In truth, one cannot even ‘reproduce’ a movement, one can only re-produce it by way of a second production that, for the spectator, belongs to the same order of reality as the first.” Christian Metz, “A propos de l’impression de réalité au cinéma,” Essais sur la signification au cinéma, vol. 1, pp. 18–19. 27. Ibid., p. 19. 112 OCTOBER what is represented by each medium—the content, the diegesis—and the vehicle of representation. In the theater the entire vehicle of representation is real; however, and precisely for that reason—as has been shown by writers such as Jean Leirens and Henri Wallon—what is represented, the imaginary diegesis or fiction, does not appear real or believable. In the theater, the perception of reality on the side of the vehicle thus outweighs the impression of reality on the side of the diegesis. With photography, however, it is the other way around: What is shown, although analogous to reality, is too removed from the perception of reality hic et nunc to give us the impression that it is really taking place now or to sustain a believable diegetic universe. The cinema, Metz argues, finds a rare point of equilibrium between the two: Its analogous and somewhat corporeal images, as well as the presence of movement, endow the imaginary, absent world of the diegesis with an impression of reality the likes of which have never before been experienced. Phenomenologically, then, cinema’s impression of reality is a quality of its fiction, that which the film, as semiological system, denotes. It is the phenomenal form that narrative—which otherwise has its own gestalt—expresses in the cinema. It is toward this other gestalt, that of narrative in general, that Metz turns for the last overtly phenomenological piece of his early period, “Pour une phénoménologie du Narratif.” Here, explains Metz, Narratif, with a capital “N,” stands for an expressive category akin to those studied by the phenomenological aesthetics of Dufrenne or Etienne Souriau, categories such as “gracefulness,” “sublimity,” or the “elegiac.” Capitalized, Narrative stands for the quality of narrative—for “narrativeness”; it offers a substantive for an adjective, in the same way that “gracefulness” stands for whatever possesses “grace.” According to both Dufrenne and Souriau, such categories are existent ial—and therefore presumably ant i-idealist— equivalent s to Kant’s transcendental categories that are active in the to and fro between subject and object, i.e., they make possible experience, perception, and affect and are thus subjective and yet they also objectively belong to the structure of a work. Metz himself describes this constant in-betweenness when he writes that these categories can be seen as both “universal categories of the world as it appears or of man as he apprehends the world.”28 Narrative, that is to say: narrativeness or the impression of narrative, thus becomes in Metz’s words one of the “great forms of the human imaginary,” one that can be described as such: “a closed discourse that proceeds by unrealizing a temporal sequence of events.”29 Both narrativity and the cinema thus appear as “anthropological facts” with the former acquiring an additional quality, namely, the impression of reality, upon meeting the latter. It is clear to see why “A propos de l’impression de réalité au cinéma” and “Remarques pour une phénoménologie du Narratif” combine to make up a separate section in the first volume of the Essais. Less clear perhaps is the overall division of labor that affects phenomenology and semiology in the work of the early Metz. The opening paragraphs of “Remarques pour une phénoménologie du Narratif” are 28. Christian Metz, “Pour une phénoménologie du Narratif,” Essais sur la signification au cinéma, vol. 1, p. 35. 29. Ibid., p. 28. Dance and Fetish 113 especially revealing in this regard. Here Metz addresses himself to his fellow structuralists, who by 1966 had begun the project of the structural analysis of narratives, or structuralist narratology, on the basis of the earlier work done by the Russian Formalists and Vladimir Propp (Morphology of the Folktale) and the more recent studies by Lévi-Strauss on the structure of myths. That same year a group of scholars—most of them associated with CECMAS30 and l’École pratique des hautes études—had published issue No. 8 of the journal Communications, which was entirely devoted to structuralist narratology. Included in that group, alongside Barthes, Umberto Eco, A. J. Greimas, Tzvetan Todorov, and former filmologue Claude Bremond, was Metz himself, who first published his “Grande syntagmatique du film narratif” in that same issue. The introductory paragraphs of “Remarques pour une phénoménologie du Narratif” are meant to clarify the position of phenomenology with regard to structuralism and semiology. It is summarized very succinctly by Metz: “Structural analysis must always assume, at a prior stage that may be either explicit or implicit, something like a phenomenology of its object; or, again, that signification (which is constructed and discontinuous) always makes explicit what could only have first been lived (‘vécu’) as a meaning (‘sens’) (which is perceived and global).”31 In other words, structuralism and semiology’s task is to rigorously (or scientifically) analyze and explain what must first strike consciousness—through perception—as a global quality (an indecomposable impression, form, or gestalt) emerging in the constant back-andforth between object and subject. The goal of structural analysis, then, is to lay bare the object’s system, to decompose what was “already there” for consciousness—but not to decompose the impression associated with the object, its gestalt. Now, inasmuch as Metz turned his attention to various problems of film denotation (especially the signifieds of large chunks of signifier such as the synt agms of the grande synt agmat ique, which—unlike the single photographic shot—do not represent analogically but clearly signify), he was able to avoid any further phenomenological investigation. Yet this initial division of labor, between discontinuous and constructed signifieds and indecomposable or global meaning, would soon become stressed under the growing empire of semiology. By the time Metz republished his early articles in the first volume of Essais, he was already critical of the idea of some indecomposable layer of meaning given directly to perception without codification, without any distinct or special signifier of its own. II. Middle Works: The Empire of the Code Metz’s new attitude toward phenomenology was largely influenced by developments in semiology—especially the works of Umberto Eco (whom he met in 30. The Centre d’études des communications de masse, connected to the École pratique des hautes études in Paris. Metz had been invited to join CECMAS by Roland Barthes for the 1963–64 academic year. The Center was founded by philosopher Georges Friedman in 1960. By 1973 it was run jointly under the directorship of Friedman, Edgar Morin and Barthes and became the Centre d’études transdisciplinaires. Sociologie, anthropologie, sémiologie (CETSAS). Since then it has changed its name a few more times and in 2008 became the Centre Edgar Morin. 31. Christian Metz, “Pour une phénoménologie du Narratif,” Essais sur la signification au cinéma, vol. 1, p. 25. 114 OCTOBER Pesaro in June 1967) and Emilio Garroni32—around the themes of analogy (as a key principle for cinematic denotation) and connotation. Thus, in the 1968 version of “Problèmes de dénotation dans le film de fiction,” as published in the first volume of Essais, Metz writes: “Contrary to what I believed four years ago, it doesn’t seem at all impossible today to surmise that analogy is itself coded without, however, ceasing to function authentically as analogy in relation to codes of a higher level, which are brought into play only on the basis of this assumption.”33 He goes on to distinguish between two vast areas of “signifying organizations”: “cultural codes” that define the culture of a social group and “specialized codes,” which require learning (such as la langue, the codes of politeness, the codes of road signs, etc.). Cultural codes, writes Metz, “are so ubiquitous and so well ‘assimilated’ that their users generally consider them ‘natural’ and even an intrinsic component of what it is to be human (even though they are clearly products since they vary in time and space).”34 And in a long footnote added the same year to “Le cinéma: langue ou langage?,” Metz distances himself from his former phenomenological assumptions, divesting himself from phenomenological idealism and transcendentalism as they might pertain to the viewer’s consciousness: If the sum of the effects of meaning we call expressive, motivated or symbolic, etc., appear “natural”—and are indeed so, in a certain way, for example for a phenomenology or a psychology of meaning—it is in great measure because they are very deeply rooted in cultures and because they emerge in these cultures at a level that lies far below (en deçà) that of the various explicit, specialized and informative codes.35 Reprising the example from ¡Que viva Mexico!, he adds: [In “Le cinéma: langue ou langage?”] I gave, among other examples of expressivity, what is known, quite suitably, as “facial expression.” However, it is certainly not due to the effects of “cinematic language” (or any other explicitly informative code) that the film spectator manages to decipher the expressions he reads on the face of the film’s hero. However, it isn’t through the effects of pure nature either, for facial expressions have meanings that vary from one civilization to the other (witness: the extreme difficulty we experience in trying to 32. Both authors published works in 1968 that influenced Metz, Eco’s La struttura assente (Milano: Bompiani) and Garroni’s Semiotica ed estetica (Bari: Laterza). Metz’s meeting with Eco in Pesaro was decisive in this regard, as he himself recognized in a footnote added in 1968 to “Problèmes de dénotation dans le film de fiction” (p. 115, n. 2, Essais sur la signification au cinéma, vol. 1). Eco’s contribution to the Pesaro conference was published in 1967 under the title Appunti per una semiologia delle comunicazioni visive (Milano: Bompiani) and soon afterward integrated into La struttura assente. 33. Christian Metz, “Problèmes de dénotation dans le film de fiction,” Essais sur la signification au cinéma, vol. 1, p. 114. 34. Ibid. 35. Christian Metz, “Le cinéma: langue ou langage?,” Essais sur la signification au cinéma, vol. 1, p. 82, n. 2. Dance and Fetish 115 “understand” facial expressions in Japanese films). Nevertheless, it remains true that in French films, we understand them quite “naturally,” that is to say, thanks to the effect of a very ancient and very deep knowledge that resides in us, that functions “all by itself,” and that—for us—henceforth merges with perception itself.36 Several passages in this footnote, including the final sentence, signal that Metz’s break with phenomenology was not absolute. And yet, June 1967 was a turning point for him in this regard, as he embraced what can only be considered a form of “pan-semiologism”—understood as a radical critique of naturalism and expressivism—that, as we shall see, literally turned upside-down the initial relation he had assumed prevailed between phenomenological and semiological meanings. According to the early Metz, “designative” works of art—works of literature or cinema, for instance—must do with three strata of meaning. Take the case of verbal art. Composed of language, it is both expressive and designative, though linguistic signification is indebted to expressive meaning—which acts as the lower stratum supporting the entire edifice. Reflecting on this in his notes on Dufrenne, Metz writes: [F]or language to code significations, the very fact of signification must exist; that it can say things that signify something to the mind, or that it is a distinctive attribute of the mind to grasp significations amounts to the same thing, for it is one of the aspects of the perpetual reversibility of the phenomenological “there is.” . . . It is because we soak in a world where everything is naturally signifying (expression) that language, by way of codification and de-naturalization, both ulterior and secondary to this fundamental expressivity, to this primary signification, has been able to form itself. . . . In short, there are two things: on the one hand, there is the moment of arbitrary codification, the moment that truthfully establishes language—and, on the other hand, there is what lies underneath this moment. What words signify is arbitrary, but the fact that they signify is “natural.” . . . In the world things have meaning. On that ground man creates words, i.e., artificial things. The word is an object, a tool, a fabricated object. We fabricate it arbitrarily. But once fabricated it becomes a thing in the world, like other things in the world. . . . And like all things in the world the word takes on a natural meaning, which, in a way, is a sort of naturalization after the fact—through habit—of its artificial meaning.37 The third stratum, as seen earlier, is the purely aesthetic one, that which concerns 36. Ibid. 37. Christian Metz, undated manuscript notes on Phénoménologie de l’expérience esthétique, ms. CM872 B8, pp. 35a/b-36a. 116 OCTOBER aesthetic expression or connotation and whose meaning is style, genre, or poetic atmosphere and the various impressions that works of art are likely to elicit as we experience them. Now, we know that in his earliest works, Metz claimed cinematic denotation to be expressive, directly reproducing the expressivity of the world by way of analogy. However, since expression is the domain of phenomenological inquiry into the meaning of the world, there seems to exist an uneasy juxtaposition between phenomenology and semiology at the most “semiological” of levels, namely, that of denotation. In other words, as long as semiology conceives of cinematic denotation as expressive, it runs the risk of either merging with or becoming a subspecies of phenomenology (its linguistic-influenced methodology notwithstanding). Undoubtedly this issue must have had a role to play in the move that led Metz away from the “image” and from the conceptual category of the “sign” and instead toward the study of “large signifying units”—such as those of the grande syntagmatique—units that had little if anything to do with analogy in their designative function and seemed therefore unconnected to the phenomenological tradition in film studies. In fact, Metz comes close to saying so himself in 1968 when he writes that: The notion of analogy must be handled with care. It is true that for a genuinely cinematic semiology analogy represents a kind of stumbling block: at those precise points where it takes charge of filmic signification (notably the meaning of each “motif” taken separately), all specifically cinematic codification fails. This is why, to my mind, filmic codes must be sought on other levels.38 By then, of course, Metz had sided with Eco in the debate over analogy (or iconism39) and adopted the view that analogy is coded through-and-through and therefore not expressive—i.e., not natural nor direct—except in a sort of naive, though no less useful (more on this below), phenomenological way.40 As a result, the initial distinction posited by Metz between film art and literary art according to which the former is connotatively homogeneous (expressive connotation building on expressive denotation) while the latter is connotatively heterogeneous (expressive connotation over inexpressive denotation) also dissolves itself at the theoretical 38. Christian Metz, “Problèmes de dénotation dans le film de fiction,” Essais sur la signification au cinéma, vol. 1, p. 113. This essay predates Langage et cinéma and Metz’s formal distinction between “filmic” and “cinematic,” whence the use here of the less appropriate term. 39. It took over thirty years for Eco to have a change of heart on the matter and, under the influence of Peirce’s semiotic theory, retract his early views on iconicity. See “Iconism and Hypoicon” in Kant and the Platypus. Essays on Language and Cognition (New York: Harcourt Brace, 2000). 40. Interestingly enough, it could be argued that Merleau-Ponty’s own conception of consciousness as embodied is indirectly responsible for the decline of phenomenology at the hands of a pan-semiologism for which (incommensurable) cultural differences only exhibit arbitrariness and, therefore, codification. This seems to be Emmanuel Lévinas’s point in his critique of Merleau-Ponty in “La signification et le sens,” Revue de métaphysique et de morale 69, no. 2 (1964). See especially pp. 149–50. Dance and Fetish 117 level—though again the idea might be acceptable at a naive, impressionist, or pretheoretical level. In short, what appears to be expressive—including the analogical recognition of visual objects—is merely so through acculturation (or cultural assimilation) and blindness, including in some cases ideological blindness, through force of habit. If, for the early Metz, phenomenological meaning grounds semiological signification—to the point where, with regard to filmic denotation, they might even seem to overlap—the situation is now clearly inverted: Coded semiological meaning now grounds what appears naively as the directly perceived natural gestalt of phenomenological meaning. Clearly, one implication of the new pan-semiological outlook is the removal of the world from Metzian film theory, thus fulfilling the spirit of Saussureanism, which requires that substance be expelled from the structural (and formal) study of language. Through analog y, therefore, film denotation doesn’t directly express or deliver the world to perception anymore; rather, it encodes it. Of course, the logical upshot of this theory is the rejection of perception itself as being natural and consequently the exclusion of the world not just from visual images but from perception as well. Hence, in 1971, in Langage et cinéma, Metz goes one step further in adopting Eco’s pan-semiologism by linking the so-called analogical code of figurative images (images that enforce an “impression of resemblance” to the world) to natural perception as being itself a coded process. He writes: It isn’t unwarranted to recall the partial resemblances between filmic perception and everyday perception (sometimes called “real perception”), resemblances that certain authors (myself included) have sometimes misinterpreted. They are not due to the fact that the first is natural, but to the fact that the second isn’t. The first is coded, but its codes are partly the same as those of the second. Analogy, as Umberto Eco has clearly shown, isn’t between the effigy and its model, but exists—while remaining partial—between two perceptual situations, between the modes of decipherment which lead to the recognition of an object in a real situation and those that lead to its recognition in an iconic situation, in a highly figurative image such as that of the film.41 Metz’s distance from phenomenology couldn’t be more pronounced. Indeed, whereas the worldly “there is”—the idea that the world is always already there— forms t he basic presupposit ion of eit her t ranscendent al-idealist ic or existential-embodied phenomenology, it is now understood as the mere naturalized result of a deeper process of codification.42 This process functions as a 41. Christian Metz, Langage et cinéma (Paris: Larousse, 1971), pp. 208–09. 42. See, for instance, Merleau-Ponty’s insistence on the irreducibility of the world as given, as “being there before any possible analysis I could make of it,” Phénoménologie de la perception, p. iv. 118 OCTOBER cultural and empirical equivalent to the Kantian a priori, and its effect is to theoretically shut off both language (i.e., all coded systems) and the subject from an independently given world: Nothing now enters perceptual consciousness that is not always already coded.43 These developments in Metz’s thinking also led him to revise his initial conception of connotation. To be sure, connotation was an especially vague term in early structuralism, comprising all second-order forms of meaning not accounted for by denotation. As we saw earlier, Metz first identified it with a phenomenological theory of aesthetic meaning in film. According to this theory, aesthetic qualities or aesthetic meaning compose a third stratum of meaning, one that lies “above” the natural expressivity of the world and “above” denotative meaning in representational or figurative art. In Dufrenne’s theory, aesthetic meaning is the outcome of the perception of artworks as aesthetic objects. It is what their sensible form reveals to consciousness through aesthetic perception; what Dufrenne also calls style and which for him manifests a quasi-subject, namely, an aesthetic world that has an expressivity of its own constituted by the consciousness and seasoned skill of the artist. By the early 1970s, however, Metz’s conception of connotation had changed focus—it became unhinged from the outlook provided by Dufrenne’s aesthetic—and was as equally denaturalized as denotation.44 For one thing, as Metz himself admitted, the new purview opened up by Eco and the recognition that a vast array of codes came together in something like a film—and not just those that pertained to the recognition of objects onscreen—created a context where connotation seemed a less urgent problem. Secondly, Metz questioned the idea that connotation was a form of meaning that ought to be “immediately read onto the object” of perception as a whole (as Dufrenne would have it). In an essay devoted to it written in 1971, he claims that one problem with his previous understanding of connotation was that he had failed to consider the existence of a specific signifier of connotation, an entity conceptually or theoretically distinct from the entire denotative code (the conjunction of the signifier and signified of denotation) and capable of marking variations that acquire meaning (become pertinent) only at the connotative level. Here, Metz returned to the work of linguist Louis Hjelmslev, who explained that texts belonging to a given semiotic system were not always homogeneous. For instance, a single text written in a given language (say, a novel written in English) could belong to several styles or use different idioms in different sections. The problem then is to distinguish them semiotically without losing sight of the otherwise homogeneous quality of the language (English) as a first order or denotative plane of meaning. On the basis of 43. Paul Ricoeur once characterized structuralism as “Kantianism without a transcendental subject,” in Le Conflit des interprétations (Paris: Seuil, 1969), p. 50. 44. Dufrenne recognized the importance of culture for the perception of aesthetic expression, but saw it as an enabling condition, not an a priori condition for it. Expression, he wrote, “ought to and even must be mediated by culture,” but this doesn’t mean that it is created by it or that it exists only relative to it. See “L’art est-il langage?,” Revue d’Esthétique 19, no. 1 (1966); repr. in Esthétique et philosophie, vol. I (Paris: Klincksieck, 1980), p. 107. 119 Dance and Fetish that initial plane, a second-order plane of signification is considered that takes into consideration the various stylistic or idiomatic categories to which the variations belong. Thus, what isn’t pertinent as a variation at the denotative level becomes pertinent at the connotative level, and connotation acquires a signifier that is conceptually and theoretically distinct from the whole denotative code even though, as Metz explains, they may be materially indistinguishable (connotation still rests on denotation). Concluding his essay on connotation, Metz writes: Strictly speaking, it wasn’t a falsehood, after all, to assert that connotation is nothing else but the choosing of different ways of establishing denotation. Yet this formulation is awfully incomplete since the very possibility of the process it describes is connected to a condition it fails to notice, namely, that the various ways of filming something (all of which denote, it is true) must be, on the other hand—and in relation to a different cultural code—especially imbued with connotative values. Otherwise, the various ways of filming will only produce synonymous denotations.45 Finally, by the mid-1970s, Metz’s concerns for connotation—which he also referred to as “non-literal” or “symbolic” significations that emerge after analogy (i.e., after denotation)—pulled further away from Dufrenne’s phenomenological conception of expressivity and centered instead on its production at the intersection of rhetoric and the unconscious’s primary and secondary processes through the figures of metaphor and metonymy.46 However, Metz’s turn to psychoanalysis during this period signaled a return to preoccupations from his early work, namely, the impression of reality. III. Late Work: Psychoanalysis and Phenomenology or Negative Complementarity Le Signifiant imaginaire, whose essays were written between 1973 and 1976, signaled an important change in Metz’s work by introducing psychoanalysis into it. In this regard, Metz was following the lead of Jacques Lacan, who, more than anyone else at the time, was responsible for combining psychoanalysis with structural linguistics, Freud with Saussure and Jakobson—albeit with a few adjustments, especially on the side of linguistics.47 Certainly, though, it was the absence of the world, the absence of substance, that, correlated with the concept of value as differ45. Christian Metz, “La connotation, de nouveau,” in Essais sur la signification au cinéma, vol. 2 (Paris: Klincksieck, 1972), p. 172. 46. For instance, Metz discusses Dr. Smirnov’s famous dangling pince-nez in Battleship Potemkin as both connotation and metaphor: “The pince-nez connotes the aristocracy: metaphor,” in Le Signifiant imaginaire (Paris: UGE, coll. 10/18, 1977), p. 241. 47. What immediately comes to mind, of course, is the “inversion” of the relation of signifier to signified as initially diagrammed by Saussure and the emphasis given by Lacan to the signifier as a result of Saussure’s insistence on the separability and arbitrariness of both planes (that of the signifier and that of the signified). 120 OCTOBER ential, made Saussurean linguistics appealing to Lacan as a model for the unconscious. Once translated into the terms of psychoanalysis and radicalized by Lacan, it’s also what caused much incompatibility between Lacan and phenomenologists such as Merleau-Ponty—the latter finding more to agree with in the theories of Freud and Melanie Klein. Given the deep epistemological divergences between Lacan and Merleau-Ponty up to the latter’s death in 1961, one might expect phenomenology to disappear completely from Metz’s work.48 This, however, was not quite the case, as references to phenomenology and a qualified endorsement of it are still found in Metz’s writings from the mid-1970s. One such endorsement comes up in sections III and IV of the book’s eponymous essay, “Le Signifiant imaginaire,” which are concerned with problems of identification and the function of the scopic drive. Section III begins by questioning the specificity of the cinematic signifier, which is said to rest on audiovisual perception. Compared with the signifier of other media, such as literature, painting, or music, writes Metz, “the cinema is more perceptual” in that it “mobilizes a larger number of the axes of perception,” and yet is “less perceptual” than the theater or the opera once we envisage “the status of these perceptions . . . rather than their number or their diversity.”49 One immediately recognizes here a key phenomenological theme, though its treatment will be largely psychoanalytic. Interestingly, it is also partly a reprise of the phenomenological argument developed a decade earlier in “A propos de l’impression de réalité au cinéma.” As was the case in that essay, the specificity of cinema is associated with its perceptual regime, with the fact that the cinematic signifier gives us images of “uncommon perceptual richness” while “registering at an uncommonly deep level, in its very principle,” that what is seen is not “really the object” but “its shadow, its double, its replica” appearing in a “new sort of mirror.”50 What follows is one of the most famous and influential pieces of film theorizing ever, with Metz comparing the situation of the film viewer to Lacan’s description of the child looking at itself in the mirror, all the while, in an alienating fashion, (mis)identifying itself with its reflection as it moves towards becoming the subject of the unconscious. Especially significant for our purpose is the way Metz characterizes the spectator’s identificat ion wit h t he camera in terms t hat recall Husserl’s concept ion of t he transcendental ego of pure consciousness (as distinct from the empirical ego). 48. Lacan, for instance, considered that Merleau-Ponty’s openness toward psychoanalysis never managed conceptually to move beyond Freud’s idea of the preconscious. See Jacques Lacan, “MerleauPonty: In Memoriam,” Les temps modernes 17, nos. 185–86 (1961). See also James Phillips, “Lacan and Merleau-Ponty: The Confrontation of Psychoanalysis and Phenomenology,” in Disseminating Lacan, ed. David Pettigrew and François Raffoul (Albany: SUNY Press, 1996). It is noteworthy, however, that Metz distanced himself from Lacan and the strict Lacanian school: “You know,” he said during an interview, “I am not a Lacanian. There is a misunderstanding about my position, because I borrow some concepts from Lacan’s work. I use three or four words taken from Lacan, and I am considered in some places as a Lacanian, but I am not.” See “The Cinematic Apparatus as Social Institution—An Interview with Christian Metz,” Discourse 1 (Fall 1979), p. 8. 49. Christian Metz, Le Signifiant imaginaire, pp. 62, 64. 50. Ibid., pp. 65, 64. Dance and Fetish 121 “The spectator,” writes Metz, “identifies with himself, with himself as pure act of perception (as wakefulness, as alertness): as the condition of possibility of what is perceived and therefore as a sort of transcendental subject, prior to any there is.”51 In short, what we have here—in the cinematic mirror, if you will—is the spectator misrecognizing himself as the transcendental ego of Husserlian phenomenology, of pure perception as apperception. For Metz, in a way, the phenomenological model of spectatorial consciousness becomes the alienating inverted image, the false consciousness or camera obscura of the spectatorial self. And it is on these grounds that Metz proceeds to critique Bazin and the idealist-phenomenological tradition in film theory for being blind to the deception it falls prey to, failing as it does to recognize the alienated nature of the spectatorial self as subject of pure perception. And yet, it follows nonetheless that film phenomenologists are not entirely wrong. After all, the mirror stage is a necessary passage towards the formation of the subject, and the ego cannot escape being deluded in front of the mirror. In like fashion, Metz assures us that primary cinematic identification with the camera’s gaze (as I/eye)—and therefore alienation—is essential to any film’s intelligibility. 52 Accordingly, phenomenological film theory has some legitimacy, if only negatively: Cosmophanic conceptions impart rather well the “feeling” experienced by the deluded self of the spectator; they provide us with excellent descriptions of this feeling and, to this extent, have something scientific about them and have advanced our knowledge of the cinema. . . . For it is true that the topographical apparatus of the cinema resembles the conceptual apparatus of phenomenology, with the result that the latter can cast a light on the former. . . . The “there is” of phenomenology proper (philosophical phenomenology), as an ontic revelation referring to a perceiving-subject (= “perceptual cogito”), to a subject for whom alone there can be anything, has close and definite affinities with the establishment of the self in the cinema signifier. . . . In this regard, cinema truly is a “phenomenological art,” as it was so often claimed, and as Merleau-Ponty himself claimed.53 There is therefore a certain degree of complementarity, in Metz’s mind, between psychoanalytic semiology and phenomenology, much like, for Engels, the dialectic always requires the unity and conflict of opposites. In Metz’s own work, this complementarity can be witnessed in the way “Le Signifiant imaginaire” returns to the issues initially raised by his earlier essay on the impression of reality, with the new work still relying on the same basic (phe51. Ibid., p. 69. 52. Metz writes: “But with what does the spectator identify during the projection of a film? For he certainly has to identify: identification in its primal form [the mirror stage of infancy] has ceased being an actual necessity for him, but in the cinema—under the threat of the film becoming incomprehensible, considerably more so than the most incomprehensible films—he continues to depend on a permanent play of identification without which there would be no social life.” Le Signifiant imaginaire, p. 66. 53. Ibid., pp. 74–75. 122 OCTOBER nomenological) premises, the same fundamental conception of cinema (or cinematicity, the quality of cinema). Indeed, recalling the piece from 1965, Metz writes: “My approach, at that time, was purely phenomenological and didn’t owe much to psychoanalysis. However, the latter confirms me in my first opinion.”54 This opinion can be summarized by a few ideas: Cinema as a medium holds a specific power over its spectators; this power results from a peculiar regime of presence and absence that differentiates it from other media and art forms while enabling spectatorial identification and heightening adhesion to fiction.55 In the 1965 essay, these ideas all merge in the cinema’s impression of reality. We saw earlier how, for the young Metz, there is, within the dominion cinema holds over us (“l’emprise filmique”), “the secret of a presence and of a proximity that rallies the masses and fills movie theaters,” and how this secret rests on an identification with fiction made possible by the uncanny experience of presence/absence on the screen.56 It is to this very same dominion of the cinema that Metz returns in “The Imaginary Signifier,” reconfiguring it as the outcome of the scopic and invocating drives that sustain perception and desire—and likewise sustain the institutions of the cinema, including the filling of movie theaters. In explaining how, according to Freud, these drives distinguish themselves from instincts, Metz insists on how they thrive on the absence of their goal, of the objects elected to fulfill them, by the very distance their originating organs (the eye, the ear) must maintain to them. Yet, claims Metz, what characterizes film when compared to other visual or aural media is a doubling of this absence, what he calls “a supplementary and specific turn of the screw riveting desire to lack.”57 Two arguments are presented to support this claim, both of which appeared in the earlier essay on the impression of reality. The first one concerns the amount of detail found in cinema’s audiovisual representation. Here, the greater ratio of absence manifests itself superficially in the fact that film, which provides us with an indefinite amount of sights and sounds, lets one perceive more of the world than most other media, and therefore “absents [more] from our grasp.”58 54. Ibid., p. 92. 55. In an undated handwritten manuscript (likely produced in the mid-1960s) for a talk on the impression of reality in the cinema, Metz wrote: “The fundamental fact that characterizes the psychological attitude of the spectator in front of the screen is that the spectator doesn’t see images (he didn’t come to the movies for that reason), he sees a story; which is to say that he follows and reads a narrative as the images flow [tout au long des images]. The cinema has given us the habit of longitudinal reading, oriented toward what comes next (to the detriment of a transversal reading that would stop to consider each image). The great idea of Jean Mitry, that cinema is made from the image, not for the image, is absolutely fundamental.” He adds: “The impression of reality is not only a perceptual problem, it is also an affective and aesthetic problem. A good film gives a greater impression of reality than a bad film (where the characters are caricatures, outrageous), even if both are essentially produced according to the same perceptual techniques. That is because we don’t believe only in the images, but also in the story; and it is often the story that lends its reality to the images in return. There are therefore two impressions of reality in cinema, and they are in constant dialectical reaction with one another. Emotion counts as much as perception.” “Exposé sur l’impression de réalité au cinéma,” ms. CM1430. 56. Christian Metz, “A propos de l’impression de réalité au cinéma,” Essais sur la signification au cinéma, vol. 1, pp. 14–15. 57. Christian Metz, Le Signifiant imaginaire, p. 85. 58. Ibid., p. 86. Dance and Fetish 123 Secondly, there is the fact that, unlike the theater, film gives to perception objects that, notwithstanding the details they possess, are nonetheless absent, unreal, imaginary. On the one hand this implies that voyeurism in the cinema is different, closer in spirit to the “unauthorized scopophilia” of the Freudian primal scene, since what is absent cannot truly acknowledge the spectator’s gaze.59 On the other hand, and here we find the phenomenological argument of 1965 yet again, such absence leads to heightened believability and identification or adhesion to the fictional, imaginary world of the film: In the cinema as in the theater, the represented is by definition imaginary; this is what characterizes the fiction as such, independently of the signifiers that are in charge of it. But in the theater, representation is fully real, whereas in the cinema it is imaginary, the materials being already a reflection.60 In other words, if absence operates doubly in the cinema—at the level of the signified and that of the signifier—so does fiction, according to Metz.61 Which is why, for him, fiction belongs to a phenomenological account of cinema, or, to put it differently, why the cinematic includes the fictional. Therefore, on the negative side of phenomenology—the side of alienated spectator consciousness —“every film,” writes Metz, “is a fiction film”; whereas on the positive side of psychoanalysis every film is doubly imaginary. What psychoanalysis brings to Metz’s phenomenology, moreover, is a more positive explanation of identification (and its alienation) through the analogy of the mirror. Indeed, in the essay on the impression of reality Metz criticizes the 59. Ibid., p. 89. 60. Ibid., pp. 92–93. 61. Interestingly, in “Le Signifiant imaginaire,” Metz omits to mention filmic movement, which, in “L’impression de réalité au cinéma,” was argued to be a form of presence. One has to wait until 1985 to see movement find its way into the psychoanalytic argument. In an essay initially published in English, “Photography and Fetish” (October 34 [Autumn 1985], pp. 81–90), Metz returns to the idea first developed in “Le cinéma: langue ou langage?” (and elsewhere in the Essais), that cinema is a machine that suppresses film frames (in 1964, this is expressed in the idea that it is the shot—and therefore not the photogram—that is the basic semiological unit of film). But he returns to this idea in a different guise, one better formulated, in a shorter and slightly different French version of this essay that he presented in 1986 at a conference held at the Collège international de philosophie on “Photographie et philosophie”: In cinema, writes Metz, “unfolding [“le déroulement”] mobilizes the viewer’s attention to the detriment of the link that unites each image [each frame] to its referent, at the detriment of indexicality, which, although still intact, becomes less perceptible” (Christian Metz, “Photo, fétiche,” in Pour la photographie, vol. 3, ed. Ciro Bruni [Sammeron: GERS, 1990], p. 118). For Metz, then, this felt waning of indexicality produced by movement, by the film’s unfolding, detaches the film image from its anchor in the real world at the phenomenal level, which explains in part why it is perceived as so “intimately fictional” (“la narrativité bien chevillée au corps,” Essais sur la signification au cinéma, tome I, p. 52). The upshot of this conception is a different place for fetishism in the experience of film and that of photography. Though other factors also contribute, in the end Metz concludes that film is a “powerful activator of fetishism” (“Photo, fétiche,” p. 123), disseminating disbelief as belief on large quantities of fleeting perceptions, whereas photography, because of its fixity, is “more apt in becoming itself a fetish” (ibid.). 124 OCTOBER accounts given by Wallon and Leirens as being negative.62 Leirens, for one, claims that presence in the theater hinders identification, while absence makes it possible in the cinema.63 In manuscript notes on Leirens’s book—likely written before completing “L’impression de réalité au cinéma”—Metz sketches his criticism in clear terms: There is something truthful about this theory: It is because, among other things, it is too carnal that theater gives an impression of artifice. But it isn’t for that reason alone; there is also the artifice of theatrical works (their tone, the style, the situations, the actor’s diction, etc. . . . ), in short, a diegetic artifice often absent in the cinema (this is one of the causes of the [cinema’s] impression of reality). By contrast, to claim that we identify in the cinema because it is immaterial doesn’t really hold. Such reasoning only considers negative processes (it’s true that in the cinema the absence of flesh-and-blood humans leads, for the spectator, to the absence of any risks regarding anti-identification. Because the actor is unlikely to start sneezing, there is little risk that the spectator will start laughing. But it isn’t sufficient that the obstacles to identification disappear for identification to effectively come about. One must still explain the positive process of identification. For this, the cinema’s immateriality won’t work. On the contrary, we must look among the positive powers of the cinema: How is it that the cinema so much resembles life, [or] that we forget that these are images, etc. . . . [ ?] In short, the complete break between the universe of the diegesis and that of the spectator (a break achieved in the cinema, but not in the theater)—or, if one prefers, the principle that “in order to seem real it must be entirely false”—in short, then, this phenomenon only has a role as an enabling condition of identification (it neutralizes some possible anti-identificatory processes), but it does not constitute the positive cause, certainly not the sufficient cause, for identification.64 And yet, in “Le Signifiant imaginaire,” it is precisely absence—what Metz calls here “immateriality”—that proves, after all, to be the positive factor in cinematic identification. For it is the presence of absence (as absence of presence), in Metz’s scheme, that nourishes the scopic and invocative drives, ensuring the imaginary functions of 62. A well-known child psychologist (and filmologue in the last years of his life), Henri Wallon carried out experimental work on mirror-image recognition in animals and children that was influential on Lacan’s formulation of the mirror stage. See Henri Wallon, “Comment se développe chez l’enfant la notion de corps propre,” Journal de psychologie normale et pathologique (1931), pp. 705–48. 63. Jean Leirens, Le cinéma et le temps (Paris: Editions du Cerf, 1954). 64. Christian Metz, undated manuscript notes on Jean Leirens’s Le Cinéma et le temps, ms.CM959. Dance and Fetish 125 both the signified and the signifier. It is absence that—along with photography and movement—actively (or positively) transforms the spectator’s perceptual consciousness into a “false consciousness” and therefore ensures (mis)identification with the camera and adhesion to the fiction (what translates phenomenologically into spectatorial consciousness as cinema’s impression of reality), without which film would be incomprehensible.65 Metz himself, of course, recognized this complementarity between phenomenology and the psychoanalytic semiology of cinema. For instance, in a later section of Le signifiant imaginaire, he states that: [T]he theoretical contribution of psychology (the study of perception, of consciousness), and its extensions into classical filmologie, must therefore be complemented by that of metapsychology, which, like linguistics— which it doesn’t supplant—can contribute to renew the study of film. To the extent that the impression of reality is linked to the perceptual traits of the signifier, it characterizes all films, diegetic or not; but inasmuch as it participates in the fiction-effect, it is specific to fiction films and to them alone. It is before such films that the spectator’s consciousness adopts a very particular sort of intentionality [“visée de conscience bien particulière”] that merges neither with that of the dream, nor with that of the daydream, nor with that of real perception, but partakes of all of them and inserts itself at the center of the triangular figure they draw: a type of gaze whose status is at once hybrid and precise, and that establishes itself 65. See especially Metz’s rewording of this phenomenon at the end of the section entitled: “Film/rêve: la perception et l’hallucination” (part II of “Le Film de fiction et son spectateur [Étude métapsychologique]” in Le Signifiant imaginaire): “[O]ther than the filmic state, there are few situations in which the subject receives particularly dense and organized external perceptions while being internally predisposed by his immobility to over-perceive (“sur-recevoir”) them. The classical film plays on this pincer movement, for which it has itself set up both claws. It is the double reinforcement that makes possible the impression of reality, it is thanks to it that the spectator, starting off from the material presented on the screen, the only thing given to him at the outset (spots of light that move upon a rectangle, sounds and speech that come from nowhere), will become capable of achieving a certain degree of belief in the reality of an imaginary the signs of which he is supplied with, capable of fiction, in short. Now, the capacity for fiction, as we too often forget, isn’t exclusively (or primarily) the capacity—unequally shared and therefore a true source of bliss for aesthetes—to invent fictions, it is above all the existence of a historically constituted and much more widespread socially regulated regime of psychical functioning, called fiction. Before being an art, fiction is a fact (a fact of which certain art forms can take possession). The relation between this fictional capacity and the film of narrative representation is close and bidirectional. [On the one hand] diegetic cinema could not function as an institution . . . if the spectator . . . was incapable of steadfastly adopting and renewing at will the special regime of perception that I am trying to analyze here in Freudian terms. [On the other hand], film, as I have tried to demonstrate elsewhere in a more phenomenological cast, produces an impression of reality much more vivid than that of a novel or a painting, since the intrinsic nature of the cinematic signifier, with the exceptional ‘likenesses’ of its photographic images, with the real presence of movement and sound, etc., has the effect of inflecting the fiction-phenomenon, albeit so ancient, toward historically more recent and socially specific forms,” Le Signifiant imaginaire, p. 144–45. 126 OCTOBER as the strict correlative of a certain kind of looked-at object. . . . Faced with this cultural object, which the fiction film is, the impression of reality, the impression of dream and the impression of daydream, cease to be contradictory and mutually exclusive, as they usually are, in order to enter into a new set of relations wherein the gap that normally lies between them, while not exactly annulling itself, admits an unprecedented configuration that makes room for overlapping, alternating balance, partial agreement, shuffling, permanent circulation between the three: authorizing, in short, a kind of central moving zone of intersections where all three can “encounter” each other on a singular territory, a confused territory that is common to all three and yet doesn’t abolish their distinctness. . . . The fiction film thus holds the strange power of momentarily reconciling three very different regimes of consciousness. . . . 66 This quote is taken from the book’s third part, “Le Film de fiction et son spectateur (Étude métapsychologique),” which is nothing short of an attempt to describe the being of the film (at least in its dominant institutional mode in the West) through the split consciousness of the spectator. Which is to say: to use psychoanalysis to do the work of phenomenology. Thus, Metz speaks of the film spectator’s visée de conscience [intentionality] whose specificity is to merge the impression of reality with the “impression of dream” and the “impression of daydream.” Be that as it may, complementarity between phenomenology and psychoanalysis (or semiology) doesn’t translate into epistemic equality, however. Indeed, we mentioned earlier that, next to either psychoanalysis or semiology, phenomenology provided a naive, even pre-theoretical form of description. We conclude this essay, then, with what, for Metz, came to constitute an epistemological distinction between these different types of approaches. At the time he was writing the essays that make up Le Signifiant imaginaire, Metz—who never disavowed linguistic-inspired semiology—was asked to contribute a paper for a Festschrift in honor of Mikel Dufrenne. The book, Vers une esthétique sans entrave. Mélanges offerts à Mikel Dufrenne, was published in 1975 by Union Générale d’Éditions and part of a collection edited by Dufrenne himself. Metz’s essay, however, is a somewhat strange contribution in homage to a famous phenomenologist. Indeed, it adopts a very strict semiological approach in dealing with traditional phenomenological themes, in a way, moreover, that might even seem anathema to phenomenology. This essay, entitled “Le perçu et le nommé,” investigates the relation between perception (of both world and images of the world—as well as sound perception) and naming.67 Refusing to settle the debate as to which, perception 66. Ibid., p. 174. 67. Christian Metz, “Le perçu et le nommé,” Vers une esthétique sans entrave. Mélanges offerts à Mikel Dufrenne (Paris: UGE, coll. 10/18, 1975), pp. 345–377. The section on sound was translated into English and published as “Aural Objects” in Film Sound: Theory and Practice, ed. Elisabeth Weis and John Belton (New York: Columbia University, 1985). Dance and Fetish 127 or language, first informs the other—a debate he characterizes as “pointless”— Metz dispenses with the phenomenological account of perception, replacing it with a semiological model borrowed from Eco, according to which perception like language, is arbitrarily shaped by society (and therefore not natural).68 For Eco, visual perception functions by way of schemas, with consciousness identifying not so much the entire manifold of sense but only certain traits of the iconic signifier (shapes, contours, shading, texture, size, orientation, etc.) that social subjects come to deem pertinent in the context of their culture. These traits belong to what Eco, and Metz after him, calls “recognition codes.” Thus, for example, with regard to snow, Inuits are said to “perceive ten distinct objects,” whereas Frenchmen in 1975 “see a unique object with variable determinations” because their recognition codes are different.69 The implication is that neither the perceived object nor its qualities belong to the natural world in itself, but rather to a socially constructed world—however unaware of its social status or of its categories we may be. What reveals itself to perception, therefore, isn’t the world but society or culture. Recognition codes are said to function equally in both everyday perception and in the perception of figurative images with a high degree of likeness to the world (such as film images). Naming what is perceived through such codes, however, requires that what is recognized or identified from within the manifold—the visual signified—be correlated to a linguistic unit. Metz offers a detailed account of this process that shows how this correlation—or intercoding—is achieved. He begins by noting that a likely candidate for a verbal account of the perceived object is the word. However, words are already signs that arbitrarily combine a phonic (or written) signifier with a signified, and there is no code for joining the linguistic signifier with the iconic signified. Instead, the naming of what is recognized must translate the iconic signified of the recognition code into a linguistic signified (conceptual component) of the word, what linguists call a sememe. Thus, writes Metz, [T]hanks to the pertinent traits of the iconic signifier the [perceiving] subject identifies the object (= he establishes the visual signified); from there, he moves to the corresponding sememe in his mother tongue (= linguistic signified): this is the precise moment of naming, of crossing the inter-codic bridge; once in possession of the sememe he can pronounce the (phonic) signifier of the linguistic code. The cycle is complete.70 There is, furthermore, a second level of articulation between the visual and the linguistic. Here, Metz calls on the work of A. J. Greimas, for whom correlations 68. Christian Metz, “Le perçu et le nommé,” Vers une esthétique sans entrave. Mélanges offerts à Mikel Dufrenne, p. 355. 69. Ibid. 70. Ibid., p. 360. 128 OCTOBER exist between Eco’s visual traits (the iconic signifier) and those smaller semantic units that bind together to make up a sememe and are called semes. Metz gives the following example: [I]f, in an ethnographic film, we see an object that is unknown to us (say, a hunting weapon or a musical instrument), and if the anterior extremity of this object possesses a distinctly shaped rounded part, we will not hesitate to perceive the latter as the “head” of this utensil which is otherwise impossible for us to further identify.71 The point here is that those visual traits—“extremity,” “distinct or separated extremity,” “superior or anterior extremity,” “spherical extremity”—are also the semantic makeup (the semes), the pertinent semantic traits, of the sememe “head.” Thus, thanks to that relation between semes and visual traits, the linguistic signifier (“head”) can come to stand for both of them. According to this theory, then, the perceptual code of recognition parses the otherwise meaningless (or not yet meaningful) sensual field into pertinent traits (iconic signifiers) that are said to correspond (but why?) to the smallest semantic units of language (semes, linguistic signifieds), thus enabling a verbal description of the world even when objects aren’t recognized (and can’t be named). As one can see, little in this account of either perception or language can be deemed truly phenomenological. Rather it is a purely semiological account from which the world—including our lived or embodied experience of it—is carefully kept at bay from both.72 Recognition codes are considered to offer a more sophisticated construal and a deeper understanding of the world and of perception than phenomenological description.73 Thus, in the case where objects are actually 71. Ibid., p. 366. 72. As mentioned at the outset of this paper, Merleau-Ponty read Saussure (as early as 1947) and became one of the first exponents of Saussurean linguistics in France. His understanding of Saussure was nevertheless greatly idiosyncratic, to say the least. Though Merleau-Ponty subscribed to Saussure’s attempts to both divorce the study of language from the ontological problem of reference and avoid the idealistic conception of language as representation of ideas, he nonetheless considered it incapable of conceiving the relation of language to the world (which, for him, requires consideration of the embodiment of speech) and of attending to the expressivity of language itself (see Merleau-Ponty’s later conception of parole parlante or “speaking word,” developed in contradistinction to what he called parole parlée or “spoken word”). Directly addressing the issue of naming in Phénoménologie de la perception, Merleau-Ponty refused to distinguish between recognition and verbal designation: “The denomination of objects doesn’t follow their recognition, it is the recognition itself” (see Phénoménologie de la perception, p. 207). 73. In an entirely revised edition of La struttura assente which Eco published in English in 1976 under the title A Theory of Semiotics, we find him—unlike Metz—somewhat hesitant as to possible relations between semiotics and phenomenology: “According to Husserl the dynamic act of knowing implies an operation of ‘filling up’ which is simply an attribution of sense to the object of perception. He says that to name an object as /red/ and to recognize it as red are the same process, or at least that the manifestation of the name and the intuition of the named are not clearly distinguishable. It would be worth ascertaining to what extent the idea of ‘meaning’ found in the phenomenology of perception agrees with the semiotic notion of a cultural unit. A rereading in this light of Husserl’s discussions might induce us to state that semiotic meaning is simply the socialized codification of a perceptual experience, which the phenomenological epoché should restore to us in its original form. And the sig- Dance and Fetish 129 recognized and named—like recognizing a “tree” in a photograph and naming it—one must keep in mind that what is named isn’t something that belongs to nature (some inalienable presence that is “already there”) with which we interact. The correlation, rather, is between something that exists exclusively in visual culture (the visual signified as “cultural unit”) and something that exists exclusively in language (the linguistic signified). But how is this translation possible without referring to the world, to an actual—directly experienced—tree, one free not so much from perception as from linguistic and perceptual codes? Metz’s answer is that both visual and linguistic signifieds qua signifieds necessarily belong to what Hjelmslev called content purport, that is, the entire domain of unformed meaning or semanticism in general, the amorphous field of whatever may be signified. For that reason, content purport is not something the can be perceived, but it connect s all signifying systems and manifest s it self in all signifieds. It is the imperceptible horizon or reservoir of meaning, of all that can be signified to human consciousness—but only through semiotic systems and therefore not through perception. The question, then, is: What place, if any, is there for phenomenology in this epistemological field? The answer comes in the short concluding section of “Le perçu et le nommé.” Here Metz depicts phenomenology as both anterior and posterior to semiological (and psychoanalytic) description. This is nothing short of a hermeneutic circle, with phenomenology (or perception) offering a naive, but required, starting point for a transcendent scientific description that seeks to explain it—thus inverting the early Metz’s initial epistemological outlook. “We are all phenomenologists when the need arises,” claims Metz.74 Thus he regards semiological and psychoanalytic concern for the perceptible signifier of cinema as offering a continuation of phenomenology on the grounds that both can only begin their work with regard to what is first perceived, and without which they would have nothing to go on. We saw earlier how Metz made what seems to be a similar claim in “Pour une phénoménologie du Narratif,” namely, that structuralism must begin with a phenomenological account of meaning. And, in the final words to his unpublished book manuscript on witz, the unconscious, and language—a manuscript completed in 1986—he clearly indicates that he thought of nificance of daily perception (before the epoché intervenes to refresh it) is simply the attribution of a cultural unit to the field of perceptual stimuli. . . . Phenomenology undertakes to rebuild from the beginning the conditions necessary for the formation of cultural units which semiotics instead accepts as data because communication functions on the basis of them. The phenomenological epoché would therefore refer perception back to a stage where referents are no longer confronted as explicit messages but as extremely ambiguous texts akin to aesthetic ones” (A Theory of Semiotics [Bloomington: Indiana University, 1976], p. 167). It should be said, however, that what is here perhaps conceivable in Husserlian terms would not be so for Merleau-Ponty, as epoché’s movement away from the external world and toward the “immanent” content of inner consciousness is contrary to his phenomenological description of lived perception; a description for which naming, or the attribution of a “cultural unit,” to an object of the world doesn’t exhaust the variegated—and sometimes even contradictory—meanings that make up our embodied experience of it. 74. Christian Metz, “Le perçu et le nommé,” p. 376. 130 OCTOBER Le Signifiant imaginaire as a work of psychoanalytic phenomenology.75 There is indeed a sentence to that effect in the lead essay of the latter, where he notes—as an aside, in a parenthesis—that “in any domain . . . we must begin with a phenomenology of the object which we seek to understand, with a ‘receptive’ description of its appearances; only afterward can critique go forward; psychoanalysts, it should be remembered, have their own ‘phenomenology.’”76 He is more specific still in “Le perçu et le nommé”: When I think of my own field of work, that of cinematographic analysis, how could I hide from myself—and what good would that serve?—that an entire domain of cultural knowledge, one without which the first viewing of a film wouldn’t even be a viewing, nor would the subsequent ones which are more analytical, less descriptive (or at least in a different sense of the word), more “semiological,” if we must use the word—that an entire domain of knowledge already present in immediate perception is necessarily mobilized in order that I may begin working? And this knowledge, how could we fail to understand that it is—is and is not—the “perceptual cogito” of phenomenology? The content is the same, though the status we give to it isn’t.77 And yet, inasmuch as Metz adheres to phenomenology as offering a necessary point of departure for analysis, he parts from it—and from his call to adopt a phenomenological outlook in “Pour une phénoménologie du Narratif”—as soon as he seeks to go behind perception, as it were, by way of semiological codes or through the workings of the unconscious in order to “unravel” the “conditions of possibility,” the “production structures” and the “objective determinations” of subjective experience. Thus he replaces one form of idealism—that of classical phenomenology—with another form of idealism, one without a (conscious or whole, Cartesian or Husserlian) subject—that of structuralism. In the end, if phenomenology comes “first” for the later Metz, it is for its experiential value. However, unlike semiology or psychoanalysis, it ultimately fails at delivering anything like the positive, objective truth achieved by theory. 75. “There are (at least) two sorts of applied psychoanalysis. That which bears on people (writers, etc.) or bears on those imaginary people that are the characters of fictional works; and that which concerns itself with processes, mechanisms or apparatuses that have their own (affective) logic, one relat ively independent from individuals. In a sense, the latter is a psychoanalysis of anonymit y. Psychoanalysis or, most times, psychoanalytically inspired phenomenology.” In L’Esprit et ses mots. Essai sur le Witz, unpublished book manuscript, CM1512 (1986), pp. 328–29. 76. Christian Metz, Le Signifiant imaginaire, p. 74. We have opted to keep the French “Critique” instead of the term criticism used by Ben Brewster in his translation of the essay in order to underline the Kantian flavor of the relation Metz is developing between phenomenological description (as explicandum) and the “scientific” knowledge of semiology and psychoanalysis (as a priori explicans). 77. Christian Metz, “Le perçu et le nommé,” p. 376. 131 Dance and Fetish IV. Conclusion The movement examined here—for this is what it truly is in the end, namely, a dance of sorts around the problems of experience and perception—shows Metz trying to reconcile two ideas: an idea of what cinema is, what it does, and what it is to experience it—as art, as narrative, as impression of reality, as impression of dream and daydream, as language—and an idea of theory and of its epistemic promises. It is the latter idea that will distinguish him and with which he left his imprint on film studies. This is the Metz who we said at the beginning is now remembered for transforming the culture of film studies, opening the way to “methodological” film studies, and paving the way for “theory.”78 That this is a dance is perhaps best illustrated by the way Le Signifiant imaginaire repeats the same steps as those of the early Metz, though this time the dancer has done an about-face. Aesthetic expressivity first understood as connotations resurfaces through the study of metaphors and metonymies in their connection to the work of an unconscious “structured like a language”—thereby repressing aesthetic issues. The impression of reality and narrativity return as the negative points of origin for a study of spectatorial desire, etc. The image of a turning point with Metz’s meeting with Eco also belongs to this dance, as it reverses the direction of the work without changing its basic components, what we called earlier Metz’s “opinion” of the cinema, which seems “theory free” and yet is that which Metz unwearyingly seeks to justify through theory, through a discours de la connaissance. At the end of his essay on photography, “Photo, fétiche,” Metz claims to have used theory—namely, psychoanalysis—as a fetish, which in this case would mean using it as something that “hides away” a traumatic sight or idea of loss, death or absence, one that the subject nonetheless wants to preserve—hence Octave Mannoni’s famous “Je sais bien, mais quand même . . . ”79 Now, it is clear in reading Metz, especially Le Signifiant imaginaire, that what theory hides away as fetish is the film, the object of experience itself, as that through which one experiences the cinema. Let us recall a famous passage from Le Signifiant imaginaire where Metz writes that: To be a theorist of the cinema, one should ideally stop loving the cinema, and yet still love it: to have loved it a lot and to only have detached oneself from it by the other end, by taking it for the target of the very same scopic drive that made one love it. To have broken up with it, as we break up relationships, not to move on to something else, but to find it again as the spiral turns. . . . [To not] lose sight [of the cinephile that we once were], but to keep him in custody. Finally, to be and not 78. See D. N. Rodowick’s recent book, Elegy for Theory (Cambridge, MA: Harvard University, 2014), esp. pp. 152–200. See also the comment ar y by Warren Buckland in his Film Theor y: Rat ional Reconstructions (New York: Routledge, 2012), esp. pp. 73–92. 79. Octave Mannoni, “Je sais bien, mais quand même . . . ,” in Clefs pour l’imaginaire: ou l’Autre scène (Paris: Seuil, 1969), pp. 9–33. 132 OCTOBER to be him, since these are, in the end, the two conditions according to which one can speak of him.80 This balancing act, he adds, may seem “acrobatic.” Indeed it is. It is precisely the acrobatic dance we have been discussing and around which (or through which) moves theory: phenomenology, semiology, psychoanalysis—though phenomenology is quickly discredited for not being theoretical enough, or perhaps for unveiling itself too readily, if we recall the image of the striptease Metz uses repeatedly throughout Le Signifiant imaginaire. Indeed, for anyone charting the progression of phenomenology in Metz’s work, before and after the “epistemological shift” produced by his meeting with Eco in 1967, it seems clear that phenomenology and structuralism conflict with each other epistemologically. They conflict, namely, in the way they emphasize what stands as a source of knowledge—whether, for instance, one goes “behind” perception or not. Thus Metz’s final subsumption of phenomenology under semiology and psychoanalysis as a naive or direct form of knowledge, a pre-theoretical form of knowledge. Yet, at the same time, Metz did not feel this change of perspective to be a radical one, one that would actually alter or discredit his earlier work. And he was not ready to dispel phenomenology altogether. Quite the opposite, in fact; in “Le perçu et le nommé,” he claims that his current approach “is not a complete rupture of horizon” from phenomenology since it makes experience the starting point of inquiry.81 In a way, however, the reason for this compatibility is simple enough, and it lies in the fact that both classical phenomenology and structuralism, as divergent (or even incompatible) as they may be with regard to the role of subjective consciousness, emerge from the same great Aristotelian-Thomistic tradition as inflected toward transcendental idealism by the work of Kant. Thus emerges Husserl, on the one hand, privileging the subject and intentionality, and structuralists, on the other hand, who displace idealism onto consciousness-free structures, believing that perception needs to be subsumed by positivistic-objective laws that aren’t given to direct observation though they must be rationally reconstructed to account for it. This is the true fetish, of course. And it is precisely these two forms of idealism—the second one being really a “mock” idealism—that Merleau-Ponty sought to sublate (in the sense of Hegel’s Aufhebung) in his late and uncompleted works where, unwittingly, he edged toward a different epistemology, one whose strands, we can say in closing, can be felt in philosophers such as the later Wittgenstein (especially in On Certainty) and Charles S. Peirce (in his development of an abductive logic of vagueness) though it is yet to be felt decisively in film studies. 80. 81. Christian Metz, Le Signifiant imaginaire, p. 24. Christian Metz, “Le perçu et le nommé,” p. 376.