- False Movements: Or, What Counts as Cinema for Deleuze?
In Cinema 1: The Movement-Image, Gilles Deleuze argues that the movement of cinema is real movement and not an illusion or an effect. “Cinema,” he says, “does not give us an image to which movement is added, it immediately gives us a movement-image.”1 This can be taken as a foundational axiom of Deleuze’s diptych on cinema. At first glance, the claim may not seem particularly surprising or controversial. But by proposing in 1983 that the movement-images of cinema are real indivisible movement, not imaginary or illusory movement, Deleuze placed his conception of cinema in opposition to the dominant currents of film theory at the time.
Here, for example, Deleuze’s quarrel with Christian Metz’s semiological paradigm is already implicit. Metz, as Deleuze explains, takes film to be something that consists of utterances and “at the very point that the image is replaced by an utterance, the image is given a false appearance, and its most authentically visible characteristic, movement, is taken away from it.”2 Similarly, any attempt to read cinema as being premised on certain absences or effaced mechanisms—as in Jean-Louis Baudry’s essays on “the [End Page 261] apparatus” or Jean-Louis Comolli and Jean Narboni’s influential statement of purpose on the goals of ideological criticism—can be seen as further attempts to immobilize the image by seeing cinema’s movement and time as effects produced by means of the apparatus.3 By affirming the identification of image and movement and offering cinema as the actualization of this identification, Deleuze treats as a false problem the problematic of representation that much film theory in the 1970s revolved around.4
So, does Deleuze merely return us to the phenomenological perspective that had been the target of the semiotic, psychoanalytic, and Marxist paradigms that dominated the era? One might think so, to read some of the current appropriations of Deleuze that assimilate his work with a return to a phenomenological emphasis on embodied spectators in contrast to an emphasis on codes and significations. But Deleuze’s conception of the movement-image is equally a critique of phenomenology. As he explains, phenomenology must grasp images as images of objects and therefore as representation, locating movement not in the image itself but as a “Gestalt which organizes the perceptive field as a function of consciousness.”5 A movement-image, on the other hand, according to Deleuze—and here is where he locates the radical novelty of cinema—accomplishes in itself what modern science and philosophy have sought to demonstrate: that there is no opposition between the psychological and the physical. Due to its automatism, it produces a self-moving image. Physical movement and mental image are one.6
As anyone who has read Cinema 1 knows, the origin of this claim that movement and image are equivalent is attributed to Henri Bergson’s “discovery” in Matter and Memory (1896).7 But to make the claim that cinema consists of movement-images, Deleuze must also take up an argument with Bergson himself. In Creative Evolution (1907), published eleven years after Matter and Memory, Bergson described cinema as the very model for false movement, the commonsensical if mistaken notion that movement can be recomposed from immobile sections in time.8 According to Deleuze, Bergson’s mistake, his failure to recognize cinema as an ally, could be explained by the fact that in the first decade of cinema when he was writing, cinema was not yet cinema, or rather, it concealed its essence that would only emerge with the development of montage and the mobile camera. This idea of a primitive cinema that precedes the discovery of editing is by no means unfamiliar—indeed, it can be found in most traditional histories of cinema—but for Deleuze it has a specific meaning that he derives from Bergson. According to Deleuze, movement in the earliest [End Page 262] films does not inhere in the image as such. The origins of cinema disguised its true novelty by aligning the screen with a view. As he puts it, “We can therefore define a primitive state of...