Mediated Associations: Cinematic Dimensions of Social Theory 9780773570351

Table of contents :
Contents
Acknowledgments
Introduction
1 The Affective Associations
2 Three Social Apparatuses – and a Fourth?
3 Cinema's Optics
4 The Close-up: Circuits of Communication
5 The Out-of-field: Sociability and Sociality
6 The Flashback: Cultural Dislocations and Global Actions
7 Cinematic Interaction: Spatial Displacements and Global Scenes
8 Symbols and Secrets
Conclusion
Notes
References
Index
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
R
S
T
U
V
W
Z

Citation preview

Mediated Associations Cinematic Dimensions of Social Theory

Mediated Associations builds on current debates over the relationship between society and the cinema, and extends the critical dialogue that has been emerging between cinematic concepts and methods of social analysis. Drawing from a broad range of philosophical, sociological, cultural, media, and cinema theorists, Daniel O'Connor develops a unique conception of the power of cinematic apparatuses. He expands our understanding of how cinema effectively resonates with its viewers and draws our attention to the constitution and control of aestheticcinematic communities. Rather than focusing on the abstract and individualizing character of cinema, Mediated Associations elucidates the collective character of cinematic objects. O'Connor argues that social theory must come to terms with the new mobilities and speed of cinema, and the various ways in which the affect - as a virtual moment of collective experience - is inserted into the flow of movement and structures cinematic events. In considering the primacy of the affect to cinematic forms of power, he examines the way in which cinema controls our associations, reconstituting our manners and habits of sociality and sociability in subtle and complex ways. DANIEL O'CONNOR is assistant professor in the Department of Sociology and Anthropology, University of Windsor.

This page intentionally left blank

Mediated Associations Cinematic Dimensions of Social Theory DANIEL O'CONNOR

McGill-Queen's University Press Montreal & Kingston • London • Ithaca

McGill-Queen's University Press 2002 ISBN 0-7735-2397-9

Legal deposit fourth quarter 2002 Bibliotheque nationale du Quebec Printed in Canada on acid-free paper. This book has been published with the help of a grant from the Humanities and Social Sciences Federation of Canada, using funds provided by the Social Sciences and Humanities Research Council of Canada. Funding has also been provided by the Associate VicePresident Research, the Executive Dean of the Faculty of the Arts and Social Sciences, and the Department of Sociology and Anthropology of the University of Windsor. McGill-Queen's University Press acknowledges the support of the Canada Council for the Arts for our publishing program. We also acknowledge the financial support of the Government of Canada through the Book Publishing Industry Development Program (BPIDP) for our publishing activities.

National Library of Canada Cataloguing in Publication O'Connor, Daniel M. (Daniel Michael), 1960— Mediated associations : cinematic dimensions of social theory / Daniel O'Connor Includes bibliographical references and index. ISBN 0-7735-2397-9

i. Motion pictures—Social aspects. PN1995.O36 2002

3o6.4'85

I. Title.

C2002-901869-2

Typeset in 10/12 Baskerville by True to Type

Contents Acknowledgments vii Introduction

3

1 The Affective Associations 15 2 Three Social Apparatuses - and a Fourth? 33 3 Cinema's Optics

73

4 The Close-up: Circuits of Communication

94

5 The Out-of-field: Sociability and Sociality 107 6 The Flashback: Cultural Dislocations and Global Actions 127 7 Cinematic Interaction: Spatial Displacements and Global Scenes 147 8 Symbols and Secrets 162 Conclusion Notes

180

191

References Index 203

195

This page intentionally left blank

Acknowledgments I am particularly indebted for his endorsement of this project, to Rob Shields, whose work has deeply influenced my own. Barry Rutland and Charles Gordon offered insightful comments on one or more chapters. In admiration of his work, his insights into Deleuze's philosophy, and for setting the example of the kind of scholarship I wanted to produce, I would like to thank Constantin V. Boundas. Aurele Parisien at McGill-Queen's University Press assisted and promoted this project. Anonymous reviewers for the Aid to Scholarly Publications Programme and McGill-Queen's University Press provided a host of valuable and constructive suggestions that have helped to make this a better book. I appreciate the careful copyediting of the manuscript by John Parry and the support of the University of Windsor. For their unwavering encouragement and support throughout the writing of this book, I owe a debt of inestimable gratitude to my parents, Dolores and Bud O'Connor. Mostly, I would like to reserve my deepest, heartfelt thanks for Suzan Ilcan. Without her meticulous reflections and knowledgeable counsel, this work would not be possible.

This page intentionally left blank

Mediated Associations Cinematic Dimensions of Social Theory

This page intentionally left blank

A theory of cinema is not 'about' cinema, but about the concepts that cinema gives rise to and which are themselves related to other concepts corresponding to other practices, the practice of concepts in general having no privilege over others, any more than one object has over others. It is at the level of interference of many practices that things happen, beings, images, concepts, all the kinds of events. The theory of cinema does not bear on the cinema, but on the concepts of cinema, which are no less practical, effective, or existent than cinema itself. (Gilles Deleuze, 1989: 280)

Introduction

In this book I examine various ways in which social apparatuses such as sovereign displays, legislative sign systems, disciplinary arrangements, and cinematic technologies exercise power. Social apparatuses exercise power through spectacular displays that are designed to control, regulate, and govern the associations of those who witness them. The fact that they can accomplish this is a consequence occupying strategic social sites that serve as focal points of public attention and forums for associating with others. Social apparatuses deploy various intermediary structures and interfaces to exercise control. Their ability to govern or control associations (both material and symbolic), through indirect mechanisms and via social interfaces, means that ruling no longer implies a direct face-to-face confrontation between the structures of ruling and those whom they rule, nor does ruling require the consent of the ruled. They are also not repressive agencies that work behind our backs or structures that work deep behind the scenes. The social apparatuses that constitute the subject matter of this work occupy a grey area between direct confrontation and consent. They are productive structures that work in the open to construct what we read, hear, and see, and that in doing so shape our associations. What is at stake in the analysis of these social apparatuses is the risk of overlooking the more mundane exercises of power simply because they are commonplace. Perhaps the most insidious forms of power are those that control everyday associations, that operate right before our eyes, in the open, and in full public view. In terms of

4 Mediated Associations social theory and the advancement of our understanding of indirect social relations, it is important to understand the role that media and intermediaries play in the formation and perpetuation of these relations. This work addresses these issues by developing a conception of power that is exercised through social interfaces and mediated associations. I develop this conception through an examination of the cinema apparatus and the ways in which it mediates associations by means of its specialized social interface. I build on the critical dialogical relation that has been emerging between cinema and social theory in the works of Deleuze, Denzin, Jameson, Kuhn, Stacey, and Virilio and to draw on other social theorists' whose conceptions are not of cinema per se but are suited to its analysis. Building this dialogue requires developing cinematic concepts and methods of analysis that form conjunctions between social theory and cinema. These conjunctions are invaluable for those who work in either or both fields. Theorists of society, culture, and cinema always run the risk of either immobilizing their subject matter or adopting a fixed standpoint relative to it. Unlike other cultural productions, cinema's products are neither inert nor still. They do not take the form of finished products and so do not lend themselves to the kind of analysis typical of other cultural artifacts. Objects of cinema are cinematic movements. They are consumed as aspects of motion and objects of circulation that are on a line of flight and so disappear as fast as they appear. As such, they are very different from the kind of objects we are used to analysing. Engaging in a social theory of cinematic processes with cinematic concepts requires us to think about the nature of images and the power of the social imaginary. In many ways, it also forces us to confront the problem of movement and change outside the conventions of inert, ready-made structures and closed analytical frameworks. The problem is to develop analytical tools and concepts that are up to the speed of cinema's complex audiovisual apparatus and that are fleeting enough to grasp the qualities of this highly mobile medium and enable us to understand its power dynamics and power relations. This means entering into a relation with cinema in much the same way that Deleuze works on the margins of philosophy and cinema — namely, to conceptualize the production of images in terms of their power to form associations through the displacement of time and space. As a social apparatus, cinema is both a discursive and a non-discursive social formation. Cinema not only produces an order of signs, it also constitutes fields of visibility and directs the formation of associations in these fields. Conceived in this way, it is comparable to other audiovisual apparatuses that have served both as a focus of public attention and as a medium through which power relations are publi-

5 Introduction

cized and articulated. In this regard, I follow a Foucaultian trajectory in the analysis of power-knowledge relations and the associations that they govern. In particular, I focus on what Foucault calls the dispositif, or social apparatus, as the primary social mechanism by which power is articulated (see Deleuze, 1992a: 160; Shields 1991: 43). More than just regimes that communicate signs, images, or ideologies, social apparatuses are material assemblages that constitute space and coordinate movements through time. All communicative media operate through a material medium. As I conceive it, power is always articulated in both discursive and non-discursive dimensions. By means of a comparative analysis of social apparatuses I seek to shed light on the similarities and differences between cinema and other, more or less spectacular, social apparatuses such as sovereignty and discipline that aim to exercise power over the masses and articulate their associations. The instances of power that I have in mind are not the conventional social institutions, such as bureaucracies and other unofficial structures, whose associations are both consensual and co-ordinated by authoritative imperatives. Such institutions form, as Weber would have it, following Hobbes and Rousseau, by voluntary acts of surrender to the authorities, to their sayings, and to their direction (1978: 2 1 2 ) . Understood in this way, these acts of surrender put an end to resistance. They are signed for security reasons - that is, for protection from others or for the protection of property from others. Surrender is also conceded for the defence of a way of life, such as to ensure the continuity of tradition, perhaps because of an irrational belief in the sanctity of other sayings and the promise of salivation, or perchance for a position in a hierarchy of domination where everyone, including superiors, is subordinate to the same formal sayings and directives. Because there is something returned for submission, the basis of the authoritative association is thought to be a symmetrical exchange. Paradoxically, the outcome of the symmetrical exchange is a dissymmetrical structure, where the authorities rule absolutely. But I approach this paradox in a different way and ask a different question. Instead of inquiring why people consent to be ruled, I look for the mechanisms, structures, and strategies that produce relations of domination and subordination. Approaching the problem this way is reminiscent of Marx's analysis of exchange relations and the processes by which relations of equivalence are transformed into hierarchical relations (1978). Specifically, I am reminded of how symmetrical exchanges between independent producers are transformed into dissymmetrical capitalist relations. As we know, this transformation occurs only when one, very special commodity - namely, labour power - becomes an agent in the circuits of exchange.

6

Mediated Associations

As I show in chapter 2, there are also three other social agencies that have, through their circulation, brought about comparable dissymmetries and evoked other forms of power relations - namely, public executions, legislative signs, and discipline. Cinema represents an evolved form of disciplinary apparatus, which I call "surveillance." Social apparatuses, as distinct from authority structures, neither repress opposition nor are absolute. As I conceive of it, power is always exercised in relation to bodies that resist, that have the potential to resist, or have the ability to act otherwise. In its discursive and nondiscursive formations, power does not simply see or say. As Rodowick reminds us, power organizes and controls the conditions of seeing and saying (1997: 197). It exercises these strategic functions in public, in the theatrical forums of the social apparatus. Social apparatuses are defined by their regimes of publicity, by their styles and strategies of producing, disseminating, and collecting audiovisual cues and indications. They are planned for the public. Through various media forms, they are designed to be seen and heard, or to be read; they produce bodies of knowledge and form associations rather than simply repressing information. Through various schemes, techniques, devices, and methods of control, social apparatuses constitute an archive of the articulation of power-knowledge relations. The tradition of public executions of criminals remains a good starting point for the analysis of the social apparatus. This apparatus involves a public gathering in a common locale such as the public square and features spectacular displays of "seeing and saying." The spectacle has to be seen to be effective if it is to produce and reproduce a dyssymmetrical, hierarchical relation between an all-powerful sovereign and the undifferentiated mass of spectators that are its subjects. The mediators of the sovereign spectacle include the body of the condemned, the scaffold, or raised stage, and the presence of border guards. Through its strategic display, the body of the condemned is transformed into a mechanism to affirm the edicts of sovereignty and to convey the marks that establish or re-establish the limit of its jurisdiction. For all to see and hear, and in a language that is both unambiguous and accessible to all its subjects, sovereignty produces a bodily discourse of indelible marks that serve to stake out its domain. The other intermediaries of the spectacle - namely, the scaffold and border guards - serve as safeguards against ambiguity. By limiting public participation, they prevent the spectacle from spilling over its boundaries and becoming a forum of popular discontent. Given the proclivity to overturning sovereign meanings through participation, the popular forum is comparable to Bakhtin's conception of the carnival (1984: 265). When it could no longer contain the spectacle or the meaningful associations that it produced - that is, when the spectacle became a carnival - sovereignty

7 Introduction

divested its responsibility for maintaining order within its territory and promptly sold off jurisdiction to other interested agencies. The images of the soveregn were replaced when the image of a reasoning and 'readerly' public formed the foundation of a new social order. Through the articulation of legislative signs, the advertising regime constituted a new kind of social apparatus. In this regime, simple, everyday associations and the pleasures associated with resistance and unlawful acts that are celebrated in popular culture and circulate in popular discourse become threats. They are threats not to the orders of sovereignty but to the image of social order generally. The concern is that the small, everyday offence will become a contagion and will spread and multiply. The strategy of this regime involves, first, the arrest of the everyday offences, no matter how trivial. As Barthes (1973: 132) reminds us, any ready-made association in circulation can serve as a vehicle for the advertising regime. Second, it seeks to find and to demonstrate in a public forum the chain of associations linked to the offence. And, third, it aims to release offenders back into circulation so that they may articulate new, more forceful associations to the public. Instead of eliminating offending bodies and effacing their offences, the advertising regime puts the recalcitrant association to work as a vehicle to create semiotic associations and circulate myths of social order to the public. At every crossroad and public transit way, or at any place where people gather to work, learn, or play, this regime planned innumerable public-works theatres designed both to punish and to educate. The advertising regime includes resistance as a condition of its schemes. It uses resistance as a medium of dissemination. Where the advertising regime aims to punish and deploy only the more reasonable associations characterizing offending actions, disciplinary regimes set as their target all the unruly associations and habits of association, or the way of life, behind the act. To break all these unruly habits and forge new ones, they require longer-term removal from circulation and long-term exposure to another, more controlled circuit of bodies, information, and light. To control the flows in these circuits, discipline creates an audiovisual field such as the panopticon and deploys in it a structure known as "the gaze." The gaze structure establishes a dissymetrical relation in its audiovisual field between seeing and being seen (or hearing and being heard). Through the deployment of optic (and sonorous) contrivances such as shutter screens and other specialized interfaces designed to control the flow of bodies, information, and light, the disciplinary apparatus creates a hierarchy of ocular positions or a spectatorial division of labour. Within this division, discipline evokes a power relation between an anonymous viewership and those made into its image. For the latter, for those caught in its long-term exposures, the apparatus aims to produce self-conscious associations.

8

Mediated Associations

With an unmistakable audiovisual presence and a mass appeal, the cinema apparatus should be considered among the panoply of audiovisual apparatuses that occupy the public and the public square. As a further elaboration of discipline, the cinema apparatus deploys a comparable gaze structure, one that assembles various classes of spectatorship and structures spectatorial relations. I call this apparatus "disciplinary cinema." In this apparatus, spectators can, on the one hand, identify with the camera, as Jameson (1990: 15) argues. Here, the camera acts as a control surface to establish a one-way flow of images and information. The class of spectators occupied with the camera constitutes, as Denzin argues, an anonymous viewership that acts as surveillance agents (1995: 102). On the other hand, spectators can identify with the image in various ways, as Stacey indicates. According to Kuhn, these spectator positions are constituted by the to-be-lookedat-ness of the image. In the relation of camera-viewer and image, disciplinary cinema produces and reproduces a hierarchical set of associations based on a spectatorial division of labour. In this differential structure, women become the over-exposed images of an anonymous masculine gaze as the feminine positions are defined by their to-belooked-at-ness. To be preoccupied with this habitus (I use this term in the broadest scene to denote the body-image and all manners of being, posing, moving, looking, and dressing that are associated with it - what Foucault calls a "lifestyle") is to actualize discipline in surveillance-conscious acts. As cameras (and other mnemonic devices) migrate or "swarm" from disciplinary enclosures such as prisons and schools to malls and parking lots, disciplinary self-consciousness gives way to surveillance-consciousness. As an extension of disciplinary vigilance, disciplinary cinema structures the visual field so as to produce and reproduce the opposition of viewer and image. The mechanisms and processes that support this dyssymmetry, that produce the distinction between seeing and seen, and reproduce this kind of power relation should be an aspect of central concern. As I conceive of it, the aspect of the cinema apparatus that supports the opposition of viewer and image must occupy both a central position in its scheme and the interval between viewer and image in order to maintain or mediate their relation. Disciplinary architectures have always deployed control surfaces to regulate flows of light and information. Like the shutter screen in Bentham's plans for the panopticon prison, these surfaces are pivotal to the disciplinary plan. They are, quite literally, inserted into the relationship where they act as mediating terms. Originally, disciplinary control surfaces were very simple and limited in what they could do, in terms of the kind of meditations they could exact and of associations they could produce. Developments in the

9 Introduction

cinematic apparatus have, however, enhanced both the characteristics and potential of these control surfaces. Now they serve as elaborate, mobile interfaces with the capacity to articulate complex space-time associations. In chapter 3, I use the term "montage" to characterize the method or technique for deploying this interface in cinema apparatuses. Montage encompasses a set of strategies and plans for controlling associations in cinematic societies. Through the production and circulation of intermediary surface structures, montage works in the intervals of discipline, in the gaps within and between its relative enclosures. Here, it controls the flow of light and information, modulates change and transition, and regulates the movements in and out. It does so through the insertion of a moving interface. The interface can appear and take on a variety of forms. While this interface works to maintain the anonymity characteristic of viewership, it also opens up a space of communication between viewer and image. Instead of maintaining distinctions, the interface permits the controlled movement of viewership into the image, in order to surround it with sense or anticipation, to give it a temporal value or motivate its movements. By means of the mediating agency of montage, a variety of hybrid image-forms are produced, such as cinematic perceptions and utterances. Montage establishes these formations and associations dynamically. They are articulated in time and therefore noticeable only in passing. They can be easily overlooked if we assume a closed vantage point or an inert interface. The cinematic interface is a structure occupying time rather than inhabiting space. It is a transitional and transitory moment in the interval between seeing and seen. Film theorists conceive the cinematic interface as the face in closeup (often referred to as an "insert"; see chapter 4) or various other equivalent structures of faciality, such as the unidentified sound source (chapter 5) or the flash back (chapter 6). While the closeup has been identified as excessive and pornographic by Baudrillard (1988: 42-3) and others, the kind of excess that I have in mind is not the long-term exposure of the disciplinary gaze, nor does it signal the end of knowing and seeing or the end of distance and difference. I say this because cinematic faciality, as initially developed in chapter 4, resists capture by the frame. Deterritorialized, or radically displaced, from its mute cellular existence in the disciplinary apparatus, this interface sheds its identity and escapes all orders of meaning. It comprises an excess of expressiveness that escapes or spills over the frame. It displays an expressiveness that cannot be explained by preceding events, nor can its expressive potential be exhausted in those that follow. Faciality serves as the expressed of the event (see Patton 1996a). In other words, it is the way in which the

I o Mediated Associations event is lived at the level of the affect. The expressiveness of the face signals a potentiality or a potential movement, which can become actual only by changing qualitatively. Faciality is a potential moment of qualitative change between one state and another. We often experience it as anticipation. While the disciplinary interface constitutes a gaze structure to oversee the relative positions of bodies in space and to make them its virtual image, montage mobilizes the power of the interface to surround the image with what an anonymous gaze could potentially see. This cinematic gaze no longer functions to objectify a visual field. Instead, it adds a supplemental dimension to the image and to the field of perception. Following Deleuze, I call this dimension the "out-of-field." The out-of-field is that which we do not see but which could be seen by what Boundas terms "an otherwise other" (1993: 36). A singularity rather than an identifiable one, this other adds a social dimension to perception. It brings with it an awareness of a world outside, a world behind our backs, to the side, or elsewhere. It opens up the potential of seeing or doing otherwise. Yet it is only a potential. It is not in itself directed to anything in particular. In terms of social theory, these potential associations resemble Simmel's conception of sociability and its characteristic aimlessness. They parallel Maffesoli's conception of unicity (1996: 105) and Bauman's conceptions of undirected being together (1995) and aesthetic communities (2001: 71). They are also reminiscent of the not-so-"docile" bodies that Foucault (1979) calls variously mixtures, interminglings, and mysterious associations. These potentially unruly bodies, or what I call affective associations, can form anywhere people gather. They constitute the corpus of power relations. Power captivates these potentially recalcitrant relations: it shapes and makes use of them; it puts them to work; it mobilizes and steers their potential. Faciality is the primary cinematic image in the sense that it displays qualities of firstness, in accordance with Peirce's definition, and produces signs that are icons. Because they form the nexus of affective associations, these cinema icons occupy a central place in the cinematic apparatus. By deploying the iconic qualities of faciality in strategic ways, montage mobilizes the affect and its potential. Through various montage strategies, these aspects are inserted into circulation where they form circuits and relations with other images. As I conceptualize them in chapter 5, these mediations produce a variety of hybrid image forms and indirect associations. By linking these aspects of sociability with sensory states, montage creates cinematic perceptions that change the value of sensory experiences. By linking them with activities, montage produces cinematic actions where movements are given their impetus and their sense of direction. Moreover, by

11

Introduction

inserting the affect-image in the interval between different sensory states or between different movements montage creates various transitional schemes ("sensory-motor schemes" as Deleuze [1991: 155] calls them). Here, montage functions as an agency, in the sense that its interventions and mediations differentiate and make a difference. The problem is that the difference they make involves the transformation of affective associations into differential relations. With the power to qualify perceptions and motivate actions, cinema's mediating agencies intervene between seeing and seen in a manner different from discipline. By structuring expectations of what is to come or what can return, they create new dyssymmetries in the audiovisual field. According to Virilio, these dyssymmetries involve relations of seeing and foreseeing (1989: 2-3). More than the relative placement of bodies in space, relations of seeing and foreseeing are associations of difference constituted through movement - that is, through the qualitative transformation of spaces and the qualities of transformative actions. The interventions of montage transform the affect and its sociable associations into differential relations. In this process they mobilize faciality traits and transform them into what Bourdieu (1977: 72) has termed "the habitus." For Bourdieu, the habitus is a structure with the capacity to make a difference. However, in Bourdieu's (1993) conception, the habitus is limited by its institutional settings or fields. I extend this conception by placing the habitus in relation to the out-of-field. Here, the habitus becomes a structure of transformation and a differentiating agent, capable of provoking difference and forming differential associations. I conceive of the habitus as faciality transformed into a mediating agency. In this conception, the habitus occupies the marginal space between fields of vision or action. Here, it acts like a boundary marker to divide and differentiate. At the same time, however, it also works as an intermediary to form conjunctions between its differentials. These secondary links are formed by mobile and itinerant structures in the circuits of montage and by the circulation of icons. The habitus - a secondary formation of these circuits - produces distinctively cinematic associations. These associations are established before they are perceived and grasped before they are understood. They are anticipatory associations. In accordance with Peirce's conception of secondness, these secondary cinematic formations produce signs that are indices. The indexical, anticipatory associations formed in the cinema apparatus have a close parallel in Simmel's social theory and in his conceptions of sociality and sociability (1950: 43-4). While sociabilities can be seen as undirected or disinterested forms of being together, socialities are interested and directed forms of association.

12 Mediated Associations Socialities comprise the cultural manners of association in polite society. For Simmel, sociability comes from sociality, it transcends sociality as it rises above everyday interest. I approach this relation with different priorities. From my perspective, sociability forms an undercurrent of social life, one that threads through it, that rises up from time to time, and that has the potential to unsettle everyday associations. Socialities, in contrast, constitute secondary formations that transform sociable associations and control them. The cinema apparatus and its methods of montage are pivotal in this process of transforming the affect and its potentials into socialities and anticipatory associations. In this regard, and throughout chapter 6, I elaborate two related forms of socialities that the Hollywood cinema produces and reproduce - namely, the duel and the difference of opinion. The cinematic duel and the difference of opinion are forms of sociality constituted in the circuits of montage. Both involve anticipatory associations that revolve around an oblique interface or mediating term. In cinema's various renditions, the duel is different from its classical counterpart and the polite, face-to-face exchange of firepower that this form of sociality entails. The cinematic duel involves a rivalry or difference between two states, or states of affairs, that are set infinitely far apart in terms of their qualities. The interval between these states is occupied by movements that actualize its potential through various transitional formations and transitory structures. These transitions involve relations of seeing and foreseeing. The associations that are produced in passing from state to state are both differential and anticipatory. The weapons of the cinematic duel are particular kinds of signs identified as indices. They are deployed to establish indirect, indexical relations such as feints, traps, and parries. These forms of association constitute the basic strategies of the duel as a kind of anticipatory encounter. Taken together, the ensemble of movements, mobile associations, and anticipatory signs constitutes the habitus of cinematic duels. Counter to the duel, the difference of opinion involves two actions, or equivalent gestures, that may be markedly different in terms of their qualities, infinitely close, or only slightly different. These differential manners disclose an oblique intermediary state, one that could serve as a potential account of conspicuously different manners. When manner differentials are only slight, however, the intermediary can divide into two states that are infinitely different in terms of their qualities or become a single state that has become ambiguous. In either configuration, the intermediary state is not given but, rather, constitutes a matter of opinion. Opinions, unlike other knowledge formations, are always open to dispute, in terms either of the quali-

13 Introduction ties of the states associated with them or of the qualities associated with those who advance them (for example, their credibility or powers of persuasion). The social character of opinion consists of the group that one associates with its expression, the anticipatory membership that we acquire through its expression, and the kinds of differential associations that can form with other, rival groups expressing contrary opinions. As an ensemble of montage associations, the difference of opinion (as analysed in chapter 7) is another form of the habitus of cinematic societies. The difference of opinion is cinema's mediated version of the form of sociality that we know as "table talk." In addition to the various secondary formations of montage and the indexical relations they produce, montage also produces symbolic encounters and secret associations (chapter 8). Following Peirce (1955), I see these associations as characterized by thirdness and as producing forms of sociality that I call "secret associations." One of the characteristics of the secret and of the related practice of discretion is the potential presence of an excluded third party. If the secret is kept, it is kept from another. While this third party may know of these mysterious associations and be aware that such groupings exist, he or she is excluded from its plans and kept outside its circuits. While third parties are implicated in the constitution of the secret association, they also form an independent aspect or term in it. Peirce would call them "interpretants." In cinema, this kind of association is produced by the strategic insertion of an aspect that relays the potential of the out-of-field and its faciality traits to establish differentia] relations with fields of visibility or action (chapter 8). Here, faciality takes the form of an everyday object, but with a difference. The difference that it entails is its refusal to settle into expected patterns. Like a key that does not fit any lock, these aspects are reminiscent of Baudrillard's (1990) non-functional objects. Yet they are more than just objects of collection. They form part of a system of mediated communication. As symbolic aspects they exert their independence in order to interrupt or suspend other practical associations and break up their routines. They form an interval in the practical association that opens up another order of possibilities. Instead of extending into practice, symbolic aspects constitute short circuits that return repeatedly to the same place or the same problem. Each return discloses a different possibility, and each possibility disclosed is equally co-possible. No return can exhaust the symbolic potential. Through its mediation, and by constantly modulating its character, interpretants perpetuate the mysterious qualities of the association. It does so by keeping members of the association from identifying the

14 Mediated Associations subject-matter of their sociability. Instead of relations formed in opposition to a system of power-knowledge, the focus of opposition diffuses and becomes problematic. What forms instead are socialities that are taken up with and preoccupied by activities that circulate around a subject-matter that neither party has experienced - namely, the habits of gossip, rumour, and story-telling.

There seems to be a growing feeling within media, literary and art theory that the affect is central to an understanding of our information and image-based late capitalist culture, in which the so-called master narratives are perceived to have foundered. Fredric Jameson notwithstanding, belief has waned for many, but not the affect. If anything our condition is characterized by a surfeit of it. The problem is that there is no culturaltheoretical vocabulary specific to the affect. Our entire vocabulary has been derived from theories of signification that are still wedded to structure even across irreconcilable differences. (Massumi, 1996: 221)

1 The Affective Associations

In assessing the power of mass media - that is, those forms of mediated exchanges whose products are designed for mass consumption - many scholars analyse these forms of cultural production as they would any other form of cultural artifact produced in quantity; they are fetish objects (see Bhabha, 1994; Denzin, 1995). Fetish objects are forms of social activities, associations, and social relations that are treated as things, as objects of contemplation, or, as Baudrillard claims, as objects of fascination (1983b) or collection (1988). Barthes (1973, 1977) analyses the fetish object as captured popular associations that are returned to circulation. These approaches treat cultural productions as if they were things, as finished products or finished works that circulate, are disposed of, and are consumed as such. Reduced to inert figures and stills, cultural artifacts wait for movement to be added to them, for power to motivate them, and for forces to set them in motion. Like the passive surface of a mirror (or the distortions of a crystal), cultural artifacts do not produce anything; they can only reflect external images or record or echo the movements of other, more productive agents, bodies, and forces. From this standpoint, cultural productions are images of external movements. As 'mirrors of production' or 'mechanical reproductions,' they reproduce social estrangement or reflect back a fractured image of life in fragments. The quality of these (re)productions depends on the true-

16 Mediated Associations

ness of the surface, the smoothness of the reflective plane, and the qualities of the medium itself. A good medium, like a perfectly subservient scribe or inscription machine, neither adds nor subtracts anything, thus making it impossible for observers to tell the difference between the original and its reproductions. Why, then, would we need to see the original when we already have an image of it, a slavish imitation or copy that is made to correspond to the original point for point? The perfect copy can make the original obsolete, superfluous, or even redundant. It can replace it, stand in its place (like a representative), or take over for it like a double. Can the image speak with the same authority as the original that it replaces (or displaces)? Indeed, the problem of the image is a problem of knowledge, but it is also a problem related to power and resistance. The problem of the image redoubles when we approach that form of popular cultural production known as cinema. Cinema is an apparatus that produces images, but of what? Some commentators, such as Christian Metz (1982; 1974), would have it be a kind of transcription machine whose images copy external movements by means of coded communiques. The resemblance between model and copy is, in this case, indirect and not point for point. If these copies are decipherable, they are so only for those in possession of the code book. In this particular semiotic approach, the organization of images and the organization of things are brought into correspondence through their code structure - where the code is the reasonable content, or what is understandable. While images and things may differ in appearance, they are identical in underlying structure. It is a cunning logic that establishes these correspondences and makes differences comparable (homologous) . Whether we locate these structures of identity in the psychology of mind, in its laws (reason), in its deep grammatical structures, or in the nature of things generally, these codes are legislative mechanisms designed to order the world and rule over its associations. If media produce copies, these copies can also be amusing, like the distorted reflections of the fun-house. Does laughter undermine the authority or representatives, as Arendt claims (1970: 45), or do even the most distorted images still retain the power to affect people? Clearly, some reproductions are better than others and make manifest the difference between what they express and what they are supposed to represent. But what if the medium were not so subservient and could act or move on its own account? If there is some distortion, refraction, or resistance owed to the medium itself, then not only does the idea of the media as faithful servants of reproduction become problematic, but the whole nature of representation itself comes into

17 The Affective Associations

question. Such is the character of the simulacra discussed in Baudrillard's work. Baudrillard's simulacra are displaced signs that no longer refer to an original; they refer only to other signs and form circuits with them. But the problems of simulacra become even more complex as we enter the domain of cinema and its moving images. Instead of seeing cinema as yet another mechanism for capturing or reflecting images of movement that are external to it, I argue that cinema is a social apparatus that produces cinematic movements through processes of montage. Montage is a process that puts special mediating agencies into circulation, where they serve as social interfaces. While these interfaces can be seen as fetish objects, they are not passive reflections of past relations and associations. They are active producers of all kinds of sociations, past, present, and future. While these mediating agencies work by displacement, they are not necessarily displaced signs. The objects that circulate in the cinema are moving images that produce signs. It is through their capacity to displace space and time that these mediating agencies have the power to create associations and to create displaced, imaginary locations for these associations to form. In the absence of a public square to serve as a focal point for common gatherings and public events, cinema, though its images, produces new qualitative localities and qualitative focal points for the formation of montage associations. Various writers suggest that the forces of globalization - that is, the movements of capital, of commodities and people, combined with changes in production - have virtually eliminated local culture (see Giddens, 1994; Jameson, 1990). These forces not only displace and disperse what used to be called, the "people," who once shared an attachment to a place, region, or nation state. These forces are also implicated in the conquest of space, which uproots local cultural markers and regional sign-posts and strews them about as simulacra in their wake. Paradoxically, mass media participate in this spatial conquest at the same time as they seem passive reflections of it. Consequently, contemporary media culture becomes a fractured ideological mirror for a homogeneous mass of isolated, disaffected, and dissociated individuals, who consume displaced signs and massproduced cultural artifacts. In contrast, however, other writers suggest that numerous sites within this mass society are responsible for the appearance of new cultural localities - new forms of local communities with new forms of social relations. Unlike status groups, whose ties are based on social positions or class locations with a delimited field and who are identified by their productive or consumptive practices, these mysterious, transient, and heterogeneous assemblages are

18 Mediated Associations

said to be bound only by affective ties, which cut across existing social boundaries. Moving away from the view that modern culture is homogeneous and global, Michel Maffesoli (1996) points out that new kinds of superficial communities of taste and distinction have emerged out of these modern forms of social displacement. He calls these community formations "neo-tribes." These bodies are imaginary, or virtual communities because they are not defined by principles of territory or practices of territorialization. Their habits, beliefs, and practices are not bound up with maintenance of borders or the staking out of class distinctions in the way that Bourdieu (1977) describes. As Maffesoli characterizes them, these associations form across conventional social distinctions and mix them up. They exceed the explanatory power of conventional demographic and structural analyses of the exogenous factors of group formation and affiliation. In contrast to more rigid social organizations, these group formations are transient and nomadic. Maffesoli, following Simmel's (1950) work on sociability, suggests that these mysterious associations are spontaneous group formations that appear above the rational interests and contents that delimit conventional social relations. They form around so-called sacred sites, or public squares, or sites that serve as modern-day equivalents (Maffesoli, 1996). These sites can appear anywhere and serve as cultural localities and places of assembly for the "tribus" or "little masses." These assemblages are not organized on rational lines, nor can they be characterized as unities. Instead, Maffesoli views these mobile and transitory neo-localities as potential sites of "unicity," where the bonds of association are affective ties and opinions rather than rational orders or orders of truth. Clearly, Maffesoli aims to go beyond monolithic explanations of contemporary society and the reduction of its masses to a mass of disaffected and estranged individuals. His analysis of unicity and the formation of affective associations takes us part way to this goal. This framework, like earlier ones (see Simmel, 1950, for example), presupposes a sociability drive or instinct as the motivations for these kinds of social mixtures. Evocation of drives or instincts often stands in place of an analysis of the processes of group formation and of the social conditions that create possibilities for these kinds of associations. The purpose of this chapter and those that follow is to address these issues. I show how power relations, especially those that work through various media and media forums, condition these associations. From my perspective, affective associations such as those characterized by unicity are mediated associations. The sacred sites and public squares that they form around are already occupied by media icons and

19 The Affective Associations

complex social apparatuses that helped form these assemblages. Furthermore, because of the way these poststructural groupings and associations are grounded, rival groups and competing associations may emerge, with different affections and opinions. Sites that exhibit this character of unicity are also potential sites of opposition. It is therefore important to show the tensions and the active and reactive forces that animate and mobilize these "little masses" and their popular cultural associations. In many ways, this study of cinema and its masses parallels Foucault's investigation of power relations in public forums, through architectural forms and discursive formations. Like its predecessors - that is, the other social apparatuses that have occupied the public and the public square - the cinema apparatus is an assemblage and interassemblage of frames and planes, openings and closures, and manoeuvres addressed to the public. My perspective emphasizes the structures and processes that mediate power relations and focuses on indirect power-knowledge associations produced by these mechanisms of mediation. If cinema is to be considered in relation to the masses, it is not because it is a passive mirror that slavishly reflects or reproduces external movements in the way that cameras or other surveillance devices do. It is because it deploys a mobile social interface that is not identical with the camera, one that serves both as a focal point of the affect and as a mechanism to control the flow of affective associations. If box-office and movie rental receipts are any indication, cinema has mass cultural appeal that greatly exceeds that of so-called high-cultural productions such as theatre, opera, symphony, dance, and ballet. As Eisenstein (1949) once said, cinema is the art of the masses. It has preoccupied the public for more than 100 years now. When we think about cinema as a particular forum of popular culture, we should remember that cinema is not only the art of the masses, it is an art in motion. What we call "masses" are more than simply serially linked individuals sharing a singular sentimental attachment to fetishized, partial objects or simulacra in the conventional sense. Different forms of associations and transitory associations have coalesced out of this undifferentiated mass. It is important therefore to determine the nature of the associations within and between these social groupings, to understand the character of the places and spaces in which they bide their time, and to conceptualize the images that form their art. With ever-increasing speed and volume, information circulating in the mass media has an unsettling effect on local cultures. This is especially the case when local is taken to mean a relatively closed, fixed, and territorially bounded space for idiosyncratic events and activities

2O

Mediated Associations

that are peculiar to its inhabitants. The very idea of a local culture, with its folk-ways and folk art, has been rendered largely obsolete by the global flow of bodies and information, not to mention the cornmodification, packaging, and circulation of culture as well as the proliferation of tourist sites. Jameson, for example, distinguishes contemporary cultural forms from "folk art" practices. While he sees folk art as both "popular" and "organic expressions of distinct communities and castes" (1990: 19), he suggests that "[t]he historically unique tendential effect of late capitalism on all such groups has been to dissolve and to fragment or atomize them into agglomerations of isolated and equivalent private individuals ... [T]hus the 'popular' as such no longer exists, except under very specific and marginalised conditions (1990: 15). Globalizing tendencies not only shatter the boundaries of local culture, they also break its bonds and transform its cultural expressions. In their wake they leave only an estranged and fragmented mass of individuals. Jameson views the mobility of the mass media in many ways simply as a metaphor for production under late capitalism, since it is the repetitive motion of mechanical reproduction itself that serves as the source of public entertainment and enjoyment. This repetition is as much a matter of creating always the same old thing using stereotyped plots or genres as it is a matter of repeating, in the sense of duplicating or imitating an original. Jameson argues that "the atomized or serial public of mass culture wants to see the same thing over and over again ... When you watch a cop show or detective series, you do so in the expectation of the stereotypical format and would be annoyed to find the video narrative making high cultural demands on you" (1990: 19)Similarly, Eco emphasizes that the pleasure of contemporary massmedia forms (films, television, novels) derives from "narrative repetition" and from the many "devices" used to make the expected, the stereotypical, or the cliche appear unexpected. The popularity of modern remakes, retakes, and series is a function of the mass of "naive readers" who get "to recover, point by point, what they already know, and what they want to know again" (Eco, 1985: 164). We are satisfied either because we find what we expect or, as in the case of parody, because we appreciate the clever ways in which our expectations are tricked or frustrated - that is, the many ways in which "our cultural competence is tested" (Eco, 1985: 171). Baudrillard takes up and elaborates this idea of testing. He suggests that the mass media subjects everything, audiences included, to a perpetual optical examination (1983b: 117-30). The displacement of locality and the desire for its familiarity are popular themes in media analysis.

21 The Affective Associations With the ability of information and images to cross great territorial expanses and spatial boundaries instantly, and with more and more sites linked to communication and information networks, various forms of mass media participate in this regime of spatial conquest. Just as the internet was initiated for strategic purposes, other vanguard media technologies, Virilio argues, are historically correlated with machines deployed in military conquests of space (1991: 23). For example, the development of the machine-gun correlates with the emergence of cinematography machines, nitrocellulose with celluloid film, 1 and radar with video. Further, alongside the war machine, "there has always existed an ocular (and later optical and electro-optical) watching machine capable of providing soldiers, and particularly commanders, with a visual perspective on the military action underway. From the original watch-tower through the anchored balloon to the reconnaissance aircraft and remote sensing satellites, one and the same function has been indefinitely repeated, the eye's function being the function of a weapon" (Virilio, 1989: 3). As early as the First World War, film sequences were used in aerial reconnaissance. At that time, "the general staff had no other means of regularly updating their picture of reality, as artillery constantly turned the terrain upside down and removed the topographical references crucial to the organization of battle" (Virilio, 1989: I). From this perspective, the arsenal of the war apparatus comprises both machines for projection (firepower) and reception (its seeing machines). Virilio also supposes that a day will come when ubiquitous seeing machines will supplant projectile delivery systems as the primary weapons of war, when strategies of deterrence (firepower symmetries/dissymmetries) will give way to strategies based on differentials of "seeing and foreseeing," and when winning will be equivalent to "keeping the target in constant sight." In short, the war of projectiles and missiles will give way to "a war of pictures and sounds" (1989: 2-3). As the war machine is directed to ends other than the capture or defence of territory and property, conflict will become virtual. Not only has there always been a watching machine alongside the war machine, so there has been an amusement machine in which the visual technologies of representation and surveillance - comprising both projective and receptive aspects - have served as a medium for diversions, material advertisement, and mass cultural forms of entertainment, such as the movies. Designed primarily for war, in their local deployment these optical machines have also had a profound affect on culture. We know, for example, that the enhanced motion/motoricity of modern vision machines has altered conceptions of space (see Murdock, 1993: 535), as it tends to displace

2 2 Mediated Associations former ideas of exteriority and distance with a kind of virtual presence, or a proximity that brings everything close up. This technomobility has resulted in a net loss of spatial dimension through overexposure and the progressive disappearance of local "cultural markers" (Virilio, 1991: 23). For Baudrillard, the increasingly "pornographic" (i.e., the excessive, over-exposed, overly visible) nature of information has led to the elimination of "stage distances" and "social horizons" (1987: 42-3). This visual presence has been linked to the development of what Kellner calls "media culture" (1995: 35), which increasingly dispslaces where heterogeneous, localized community customs and practices in the homogeneous totality of a global media village. Once considered structural barriers for maintaining the distance of inside and outside and the difference of insiders and outsiders, as well as for preserving the "reflective distance" between self and other (see Falk, 1993), spatial and temporal boundaries have been virtually shot full of holes by these optical machines. As a consequence, boundaries now function more like "osmotic membranes" (Virilio, 1991: 17) or "limen" (Shields, 1992b: 195), whereby differences enter into frequent and familiar contact. With the image of society reduced to an aggregation of particles, the forms of mass media, such as cinema, are not organic cultural expressions. Instead they seem to be celebrations of simulacra. For Baudrillard, simulacra are stereotypical or analytical models, abstracted of use or functionality. Without origin or referent (1983b: 120), they are also technical- or pseudo-objects (1990: 55-75). While the "real" objects reveal as much as they hide, simulacra hide nothing. Lacking a secret or a referent, they have nothing to disclose other than what is already there (see also Butler, 1987: 108). They are ready-made and are given all at once as objects of fascination, preoccupation, and obsession. I envision simulacra as uprooted sign-posts in the wake of a violent storm. Deplaced from their place and free to move with the next strong gust or to float along with the prevailing winds, these nomadic signifiers have nothing to say and, as Rodowick (1999: 47) argues, have nothing to represent. They reveal nothing and advertise nothing (except themselves). Lacking a sense of direction, they neither indicate a place nor mark the route to one. They do not warn of conditions or tell of events to come. They have no content or context and are completely devoid of sense. Be they words, images, or things, simulacra are displaced objects cut out of their place in the scheme of things and cut loose from their moorings. In this dizzying whirlwind, signs make sense only in relation to other uprooted signs. They are not so much consumed as collected and assembled. As Baudrillard argues,

33 The Affective Associations

"the single object is not enough: there always has to be a succession of objects, with the [consumer's] ultimate aim of having a complete set" (1990: 44). Just as one might try to re-collect the pieces of a "life in fragments" (Bauman's 1995 expression), the aim of collection is to assemble something resembling a whole or to form from the puzzle of fragments some image of it. The paradox of simulacra is this: while they are singularly complete, collectively they constitute an unfinished project; while they hide nothing, they motivate collecting projects that are organized around, and obsessed with, the next in the series of simulacra and that aim at finding the final term necessary to produce an image of the whole an image, Baudrillard argues, that is nothing but the "person of the collector" himself or herself (1990: 47). But because the final term is elusive, the project of recollecting this modern persona remains unfinished and unfinalizable. Consequently, simulacra stimulate an everrenewed search for the missing artifact that will complete the narcissistic mirror and reflect back a totalized image of self. I agree that modern media machines are powerful mechanisms of cultural displacement and that they are implicated in modern schemes to conquer space and local culture. But if they are indeed powerful, they cannot be passive mirrors copying and reproducing images of movements external to them, nor can they be simply reflections of the atomized mass that they are supposed to produce. The one-sided conception of simulacra as fetishized partial-objects only adds to the confusion. Fetishism, as I understand it, refers to products independent of processes, to dead labour, rather than to the sensuous activity of production, change, and transformation. Dead labour is monumental and timeless, a passive object of contemplation, while activities and processes are conceivable only in the moment of their passing. This distinction is especially pertinent to cinema, for the images of cinema cannot be reduced to stills or mere objects of contemplation without separating them from the motion and processes that make them uniquely cinematic. Cinematic simulacra are qualitatively different because they are in motion and in process. While they are displaced terms, they are also replaced as inserts in cinematic assemblages. As inserts they are images of the outside, images of alien bodies, and thoughts from beyond that interface with interiorities and transform them. They are images of the past or future that have unsettling effects on the present as they mobilize or motivate it and so anticipate its change. They are images that are beyond the acts of framing, but at the same time they are aspects of an assembling process that brings them into contact with what goes on inside the frame. If these images of cinema are signs of the times,

24 Mediated Associations it is not because they are just passive objects for individual collections or reflections of mechanical reproduction, but because they form a kind of cultural equivalent of locality for the dislocated masses who collect and assemble around them. If local culture still exists anywhere, perhaps it is a neo-local culture that bides its time in relations with cinematic simulacra and the imaginary dimensions of locations that they produce. The way that cinematic assemblages effect spatio-temporal dimensions can be understood in two ways. On one hand, the cinematic dimension could be quantitative and relative. In this sense, cinematic images are either microscopic or telescopic magnifications of detail that derive from a form of cutting out that isolates fragments and produces partial objects. These cinematic structures are abstracted from their place, from their everydayness and everyday association, and therefore forfeit their social identities. On the other hand, when these structures are stitched into cinematic assemblages they can also produce changes of dimension and form new images of the whole. Understood as abstract elements in new assemblages, these temporally articulated structures open up dimensions of space that are qualitative and absolute. I say this because our experience of space is conditioned not only by what is in it or by what we see. Often deriving from unidentifiable sources, places and spaces also have an ambience that invokes feelings or expectations. The ambience, or mood, that we associate with a place can be fleeting and can change or shift abruptly. This is the experience of moving from one place to another without going anywhere. It is a kind of qualitative movement. The question is, how do we understand locality? Can we understand it as a quality of time or a temporal displacement rather than as a quantum of space? If so, what are the implications for sharing these transitory moments and for forming spatio-temporal associations? To the mass society thesis, and its reliance on serial recognition (repetition of the overdone and overexposed) by atomized individuals, we must add a paradoxical corollary: from the growing massification of society there have crystallized a variety of heterogeneous, unstable micro-groups, known in the literature as "affectual tribes" (Maffesoli 1996: 6) or "mass-folk hybrids" (Collins 1995: 44-5). The paradox, as Maffesoli sees it, is that the very processes involved in the massification of culture are both cause and effect of such tribalism. Bourdieu (1993) also aims to counter the tendency to homogenize mass culture with a conception of the habitus. But this conception serves as a means of entrenching cultural distinctions between different groups in

25 The Affective Associations society. The varieties of taste and manner of the habitus serve as metaphors for objective social arrangements, including social class, and naturalize ready-made structural and hierarchical differences. Not to be confused with Bourdieu's distinction grouping, these tribal groupings tend to cross class, intellectual, moral, political, and local boundaries. Instead of forming distinctions, these groupings coalesce around imaginary, mythical, and sacred forms such as heroes, saints, emblematic figures, and icons, which serve as matrices (empty forms) for their collective sentiments and the experience of communing with others. They are comparable to an aspect of Weber's (1978) analysis of the temporal orientations of social organizations. Here, Weber identifies a special form of transient community structure with an orientation to the present (or a presence). These charismatic communities, as he calls them, are temporary organizations that are largely undifferentiated in terms of social roles and have no permanent positions of hierarchy (1978: 243). Such communities form by repudiating involvement with the everyday world, its rules, routine, or ordinary activities (Weber, 1978: 243-5). They exhibit, more directly, a form of solidarity that centres on non-rational, affective ties to a locus of extraordinary qualities. The extraordinary and its qualities constitute a kind of power that coalesces in the community. So long as the extraordinary remains as such and continually renews or exhibits its extraordinary qualities, the community will persist. When the extraordinary loses its charm, the basis and focal point of community organization disappears. Consequently, the community loses its raison d'etre and will disband, unless it is able to establish another basis of association by, for example, transforming its group activities into ordinary, everyday routines and habits. Maffesoli employs the term "socialities" to distinguish the habits of the affective association from the more rationalized, individualized, and disciplined social bodies that tend to arrive only after the event of sociality (1993b: 12-14). However, and as Amirou defines it, socialities are manners of relating that derive from forms of sociability. He suggests that socialities are the "result of sociabilities crystallized in different types that historians can easily describe (courtesy, politeness, seduction, conflicts between people, pilgrimages, hospitality, and many other social habits of hypothetical origin and meaning)" (Amirou, 1989: 118). It was Simmel who initiated the distinction between these two concepts. From his perspective, socialities are the socializing manners of polite society, whereas sociabilities are the play forms of association that emerge spontaneously in and from these social encounters. Amirou therefore turns Simmel's distinction on its

2 6 Mediated Associations

head by giving primacy to sociability and making it the basis for other, more ritualized habits of association. From my perspective, it makes sense not only to distinguish these two forms of association but also to give primacy to sociability over sociality in the analysis of group associations. The inversion of perspectives is especially important for the analysis of mediated associations and of the ways in which power interfaces with them. Sociability characterizes those forms of encounter that occur on the margins, or at those nodal points where territorial planes and orbits overlap and intersect. Simmel calls these nodal points "sociability thresholds" (1950: 47), referring to those sites where functionality and pragmatic interests are suspended or set aside and the spell of normal, everyday interaction is broken. Sociability takes place in a space beyond the frames that enclose everyday interactions. Normal interactions are those forms of communication motivated largely by the practical demands, aims, and interests of the "contents," or the "objective aspects of situations" in which they are caught, and to which they are responses. In contrast, sociable encounters (often characterized by the circulation of rumours, small talk, or gossip, rather than simply by the transmission of information from one setting to another) are not motivated by the demands of a determinate milieu. They lack a sense of co-ordination with structured and bounded (or striated) social places. The sociable occupies (or carries with it) its own indeterminate, qualitative space-time, which lacks motivation (extension or direction) but is none the less "moving" (intentional) or "pathetic" (in the sense of pathos; see Deleuze, 1991: 96). The sociable therefore moves along a line of flight that "escapes" the motivated "laws of the milieu" (Maffesoli, 1996: 16), its logos, and its habits, though not simply in order to pass into another milieu or territory, but in order to surpass territorialization and occupy the open space between. So long as the sociable encounter retains its marginal or deterritorialized dis-position, and therefore refrains from expressing, acting out, or responding to, or even enjoying the logos or truth of readymade situations, it retains its potential (or puissance) to constitute an event that is the basis of the affective association, its temporal qualities, and its esprit de temps. At its most minimal, sociability inheres in an alternative logic of associations that threads through the coded norms of formal, even ritualized, situations and transforms them. Thus when Princess Diana spontaneously extends a caring hand in the midst of a formal hospital tour or otherwise breaks the formal protocol of a situation, we can see this sort of pathos - what is often referred to as the human element - which cuts right through the for-

27

The Affective Associations

malized plot of roles and assigned places. This remains the case even when such gestures are later appropriated by the media and other institutions to become expected gestures or staged photographs. At its maximal, sociability amounts to a swelling of affection, rather than a welling up of emotion, that occurs between people in the face of a common event, even a media event such as the funeral of the Princess of Wales. Without ever completely abandoning the mass society thesis, Maffesoli (1996) suggests paradoxically that, in many ways and in many different sites, neo-tribal, affective associations have coalesced out of this mass of individuals. While this perspective forms an interesting counterpoint to theories of massification, individualization, and alienation under modernity, there is little research, except in the way of sporadic claims, on what role, if any, mass-media forms play in the formation or facilitation of affective associations. This question may be all the more poignant when applied to cinema, since cinema, above all other art forms, has proven to be the "art of the masses" both in the sense of evoking their patronage and in the sense that they, like capital, have become increasingly mobile, whether for work, travel, or leisure. Is this simply because it celebrates the unfinished project of the modern individuum of mass culture, or is there an undercurrent of sociability that draws people together in affective associations? Does not cinema allow people to abandon the objective aspects of their situation and their pragmatic interests in it, at least for a time? It has long been noted that the suspension of disbelief also plays a significant part in forging affiliations with what goes on in the image and in establishing a feeling of co-mingling or mixing with a whole that is in the process of formation. This process of suspending disbelief unsettles the so-called natural faculty of dubious judgments that characterizes the abstract Cartesian individuum, its supposed duality in relation to its objects of cognition, and its effort to overcome this duality by mental (cognitive) efforts. To counter this characterization, Boundas (1993) argues that acts of perception need to be conceived differently (see also Rella, 1994).2 More than simply reiterating ready-made formulas, cinema fosters a sense of participation in the working up of a whole. A carnival of sorts, perhaps parallelling the way television treats its audience as participants in the so-called semiotic democracy (see Fiske, 1989b: 95) or establishes a site of affective living in the so-called affective democracy (see Grossberg, 1997: 143), cinema invites its audience into a kind of sociable encounter with otherness and forms a space within which this encounter can take place beyond everyday conventions and interests.

2 8 Mediated Associations To address the issue of spaces of sociability, it is necessary to see how the cinema apparatus works. To do so, I draw on Deleuze's cinematic theory. Deleuze, following Eisenstein's (1942; 1949) early work on the subject, focuses on a conception of montage as the basis of cinematography. Early, single-point-of-view cinematography comprised fixed, spatial shots, allowing movements to remain the property of the figures that it framed. Characters and bodies changed and exchanged their relative positions while the camera simply recorded these changes and exchanges. The fixed, spatial shot soon gave way to montage, which selected and assembled separate shots, most of which could remain fixed and spatial. Montage liberated movement from the bodies framed by the camera and became an element of filmic composition. Montage effectively displaces the mise-en-scene as the unit of analysis. The territorialized movements that occur on the set or in the frame are not as important as what comes next or as what lies beyond the frame. Montage assemblages form osmotic relations between frames and shots. They open up the frame of the image to its outside and place it in relation to what lies beyond its horizon of visibility. Montage establishes a form of mediation between the image and its potential exteriority. Correspondingly, montage also exposes an interval space or gap in successive framings and shots and in so doing creates a space of potential transitions or transitional events. As Trinh claims, "like montage, and like every single (other) film component, cinematography is in itself a multiplicity of cinema intervals. As the image arises, it vanishes, doomed to disappear for the film to be" (1999: xii). The analysis of montage begins outside the frame and the set, in the deterritorialized any-place that lies between or beyond these structured enclosures. This is where a distinct kind of cinematic movement occurs. Through a variety of technical means and image forms, montage constitutes what is qualitatively moving about the cinematic image. Cinematic movements are distinct from the quantitative movements of characters or objects in the delimited or striated space of the frame. As Eisenstein claims, "[T] he juxtaposition of two separate shots by splicing them together resembles not so much a simple sum of one shot plus another shot - as it does creation ... [T]he result is qualitatively distinguishable from each component element viewed separately" (1942: 7). But this cinematic movement has a dual character which defines a fundamental cinematic tension. Movement appears, on one hand, as a velocity or potential that moves virtually, without going anywhere, and, on the other, as directed or motivated movements that always go from one point to another. In the former sense, montage is responsible for defining the space of the affect, an essential filmic dis-

29 The Affective Associations continuity or interval of movement cut from its situational moorings. In the latter sense, montage defines various forms of movement that are actualized - that is to say, are placed or stitched into a setting or milieu - in order to constitute an overall sense of coordinated motion and filmic continuity. This dual quality introduces a tension or "duel" (a pun that will be more significant in later chapters), into cinema. Deleuze claims that this tension is the basis of cinematic motion - a movement that is specific to cinematic form and quite separate from the mechanical passage of frames before a projector lens or of bodies before the camera. Methodologically, montage serves as the object of investigation for moving pictures. I say this because it is the method that moves this audiovisual apparatus by co-ordinating entire sets of scenes, frames, and actions. Montage works because it captures (albeit indirectly and by means of likeness) the interval of motion or that which passes between shots. Deleuze contends that the primary image of inbetweenness in cinema is the affect-image. The affect-image is defined by the close-up, also known as an affective insert. The substance of the close-up is the face, typically the human face, although Deleuze will suggest that any object can be "facified." Some scholars might argue that because the close-up involves a cut, or a cutting out of an image from the profilmic material of the set (see Aumont, 1987: 35), it offers a cinematic metaphor for psychoanalytic structures of the unconscious (for example, castration complexes; see Kaite, 1991: 176). Others - Baudrillard, for example - associate this process with the fetishization of the partial object (simulacrum) corresponding to commodity relations under capital (see also Kroker and Levin, 1991, and Debord's discussion of the partial object-image, 1994: 12, 43). Barthes, in contrast, prefers to see the cinematic fragment as a still and as an act of quotation that form the basis of cinematic parodies (1977: 66—7). For Ingmar Bergman, however, the possibility of drawing near to the human face constitutes the "primary originality and distinctive quality" of cinema (cited in Deleuze, 1991: 99). As Rutland (forthcoming) suggests, the face, our relation to it, and the responses that it evokes constitute a primordial, pre-discursive potentiality (an otherness) that haunts all discursive formations. The face expresses qualities and emanates power. It is both capable of being moved - that is, of reflecting movements or actions that are external to it - and of motivating movement; through the pores of the eyes it exudes a sovereign gaze (in Foucault's sense) and through the mouth, orders and commands (sign regimes). In relation to the face and its emanations, there is also

30 Mediated Associations

the face of the subject, who, caught in a sovereign gaze, gathers up what is given (or said) and emits a response or expression, an action or a speech act. By the same token, we can read off (indexically) from a subject's expression or action a sense of the situation in which the subject finds himself or herself, which motivates his or her actions or speech acts, and to which these actions or speech acts are responses (for example, a look of terror). The face is verbose and sociable; it emits and receives, releases as well as captures, signifying signs. Signifiers always reterritorialize (take place) on the surface of the face - the signifier is always facified (Deleuze and Guattari, 1987: 115). The face can be both moved and mover, it influences and is influenced, it directs and is directed, it touches and is touched, but none of these links to the exterior accounts for its mediating potential. For the face is not simply a passive mirror reflecting movements that are external to it. It is a mobile surface that selects, distorts, and transforms the excitations that it receives into expressions. Expressions are movements (changes, transformations) that are peculiar to the face. The movements of the face are intensities or affects. They are not exactly passive (like emotions), because they are filled with motion, and not yet active, because their movements are not of the kind that are directed toward practical aims, objects, or ends. The affect, or its likeness in the expression, is an undirected, non-linear form of qualitative motion with the potential to make contact, to move, to touch, or to extend into action. These movements are potentially contagious. The face occupies the space between the poles of received and executed movements. This space is like a threshold between moving and being moved or an interval between that which is reflecting and that which is reflected. The close-up is typically inserted in cinematic assemblages to stake out such thresholds, points of change, turning points, or critical instants. But by itself, it is indeterminate or unfinalizable, as Bakhtin might say (1981). It expresses a non-discursive potency that can be controlled only by strategies that suture it to its surroundings or link it to given states of affairs. This is no ordinary face. It is not typically perceived or understood in this way. As a rule, faces stand in relation to settings in which people act and perceive. Here they serve to identify individuals or to mark their place or post within the co-ordinated spaces of the social structures (see O'Connor, forthcoming). Framing the face in close-up suppresses both its individuating and its socializing functions. It does this by dissolving its spatio-temporal co-ordinates as well as its relationship to a particular body or specific individual. The close-up isolates qualities in order to call forth pure affects, such as shock or surprise. Different from the face situated within a co-ordinated state of affairs, there is also an expressed state of affairs. An expression might direct us to a given state

31 The Affective Associations of affairs or an event that has occurred. It might even anticipate an event that could happen or a state of affairs that might be possible. While each of these anticipatory states or events could serve as an affirmation or an actualisation of the expressed, each actualization nevertheless differs in kind from the qualitites of the expressed by this extraordinary motor-surface. A quality such as decisiveness is different from any actual decision, and redness is different from any specific instance of red. Similarly, the expressed is a qualitative potential or a potential movement and remains so, whether or not it is actualized by a content or in a practical context. The expressed, like sociability, is indifferent to contents. From Simmel's perspective, the face with its qualitative traits is like a "pure form." I, in contrast, conceive of the face as an affective content that can take on a multiplicity of forms, that can be actualized in many different ways, and where no one form is able to exhaust its potential. As a consequence, the face often expresses the qualities that are suspenseful. By itself, captured in close-up, this affective image is a pure potentiality that is both impersonal and distinct from every individuated state. As Balazs states, "the expression of a face and the signification of this expression have no relation or connection to space. Our sensation of [place] is abolished. A dimension of another order is open to us" (in Deleuze, 1989: 96). Because of this displaced character, Deleuze claims that affect is a nomadic singularity (1991: 98), one that can freely enter into combinations and conjunctions with other images. The overall effect is the formation of new, complex entities. The complex entity can also be the conjunction of many faces, faces that begin to resemble one another, not because they merge into a unity of the one, but because they have lost all their individuation. The undirected potential motion that occupies this deterritorialized close-up space constitutes an affective ambience, irreducible to the practical demands of moving on to the ordinary business of everyday life. This any-space-whatever of the affective association is neither a sedentary place of habitual or instrumental responses nor simply a functional or utilitarian site to be moved through in order to reach some other transcendent goal. Rather, it is a space of biding time that is outside the more serious business of everyday affairs. Sociologically, this affective space and the icons that it produces may hypothetically constitute the matrices around which the neo-tribes coalesce in celebration of their sociability, their puissance, their unicity, or their "undirected being-together" (Maffesoli, 1996: 81). In many ways, this media interval could be seen as the site for a kind of collective association that expresses something akin to the "vitality" of the marketplace (see Bakhtin, 1981), the disinterestedness of the sociable encounter (see Simmel, 1950), or the "ambience" of the public square and its "public

32

Mediated Associations

ways" (see Maffesoli, 1996: 157). Like Bauman's media "idols," they may also accomplish the small miracle of conjuring up "the 'experience of community' without real community, the joy of belonging without the discomfort of being bound" (Bauman, 2001: 69). The characteristics expressed in these places of displacement parallel what postmodern theorists see as a lack of urgency or a lack of goaldirectedness and a general lack of faith in modern progress and science. These features of the postmodern condition are claimed to be associated with the failure of modernity's grands recits (Lyotard, 1984: xxiii). Hebdige (1988: 186) identifies three negations that found the postmodern condition. They are as follows: first, the negation of totalizing discourses (or socio-fictions) that prescribe human goals or define an essential human nature, such as the idea of a unitary subject, capable of intentional, transparent communication and unmediated action on the world; second, the negation of teleology or a scepticism regarding the ideas of decidable origins/causes, authorial purposes, or historical destiny; and third, the negation of Utopia. Postmodernity accords little faith to those master narratives that purport to understand the human condition or the global situation as well as the grand plans and modern strategies that authorize the responses necessary to solve its problems and bring about change. These schemes may have foundered in their attempts to direct the masses and to define their goals, but alongside these grand schemes, there are a range of subtle, minor strategies aimed at policing crowds, at controlling or sanitizing the public ways, and at breaking up its mixtures and associations. Even former sovereign powers exercised these strategic functions in small-scale manoeuvres, in additional to global wars. They produced spectacles designed to attract the little masses that were their subjects and to rule their everyday associations. Parallel strategies also appear in cinema. They too converge on a kind of spectacle, but in this case the spectacle involves moving images that produce, direct, and motivate associations. Such manoeuvres are characteristic of power relations. Power does not simply repress or eliminate its opposing forces but acts on them, analysing, dividing, coordinating, and orchestrating their movements to enhance their effectiveness or economize their operations. Power operates on puissance. It is responsible for both constituting and variously actualizing the potentiality of associations so that they become reconciled with or seamlessly sutured into the continuity of a social order. It is in this sense that I discuss the power of social apparatuses to transform "little masses" (Maffesoli, 1996) into more orderly associations. As I see it, cinema is a social apparatus that has emerged from the shadow of other social apparatuses that have shared much the same aim.

If we forget the theory of visibilities we distort Foucault's conception of history, but equally we distort his thought and his conception of thought in general... Foucault continued to be fascinated by what he saw as much as by what he heard or read, and the archaeology he conceived of is an audiovisual archive.

(Deleuze, 1986: 50)

2 Three Social Apparatuses and a Fourth? In this chapter I introduce the concept of social apparatus and survey some of its more common formations and deployments - the scaffold, the legislative sign-post, the panopticon prison, and cinema. I focus on the audiovisual features of these social formations and the characteristics of the mediating structures and agencies (aspects) that they embody and the various ways in which these aspects are deployed in public domains. Social apparatuses work in the open and through the creation of strategic openings. They assemble and address public forums. They create social localities, serve as focal points of association, and are involved in activities aimed at directing these associations. I show how social apparatuses assemble social localities and use mediating aspects to constitute power-knowledge relations. This discussion provides a comparative basis for introducing the cinematic case as a distinct social apparatus. Through its audiovisual formations, cinema puts a new face on display in its cinematic locations and gives power a new look. While the elements of the social apparatus are comparable, the specificity of each apparatus derives from the assembly and deployment of its elements and aspects and from the kind of associations that they produce. I therefore suggest that simply applying the ready-made frameworks of other apparatuses to cinema is insufficient to generate a full account of the effects of this audiovisual machine. Crucial to my analysis of the social apparatus is the emphasis that I place on mediation and on the concept of mediation. While all the

34 Mediated Associations social apparatuses that I analyse exhibit forms of mediation, each operationalizes it differently, through different mediation structures and processes. The kind of power-knowledge relations that are the subject matter of this chapter, and of those that follow, work through indirect means and by strategies of mediation. Not all systems of power are engaged in direct face-to-face confrontations. Even the most obvious forms of direct physical confrontation are not secure from mediation. They are often appropriated en bloc as, mediating structures in other systems of power-knowledge, whose aim is not to engage in a direct assault. That is to say, some power relations require a middle or mediating term. Power-knowledge relations are relations that persist, even if only for short periods. They are processes that cover time. Because they do so, the relations themselves cannot be immediate, nor can they be the property of terms in isolation or outside the relation. Relations are differential in that they occur only between unlike terms. There is no difference between one and one. To simply repeat the same (to match blow for blow) is to deny a relation, just as to return or respond with the same (an eye for an eye) is to refuse the relation that we call gift-giving. The gift relation is a form of repetitiondifference. The gift is different each time that it returns and must be so if the gift relation is to endure. It is the difference that makes a relation possible, but the mere appearance of difference is insufficient to explain the relation itself. For a relation to occur, each must partake in a common quality, a likeness that neither term can possess in isolation as a property or an attribute. The qualities of a relation must be external to the terms involved in it and different from their differences. It is only through the difference of differences that relations endure. The difference of difference that makes relations endure is not an identity. Identity itself is a framed concept. It is a reflection of confinement or an enclosed universe. It is an image of the settled and of the stability of those who settle, and it is the denial of difference that comes with exhaustion at the end of a struggle - that is, when further recalcitrance is made impossible, or when there are no other foreseeable possibilities, or when all lines of flight have been foreclosed. Identity is the product of capture or surrender. For example, Weber's (1978: 212—54) theory of authority and of the orders of domination (i.e., the bureaucracy, charismatic community, and traditional office of dominance) revolves round an identity politic. From this perspective, these structures of ruling are based on accounts of the motivations (rational or otherwise) that lead people to surrender, subject themselves to, or identify with the will of another.

35 Three Social Apparatuses - and a Fourth? POWER RELATIONS

These legitimate forms of rule, like their opposite, brute force, are totalitarian forms of governance, since they reign only "ideally" or in the absence of resistance. Weber's reflections on these settled mechanisms are therefore very one-sided (admittedly so: "one-sided accentuations" form the basis of his method; Weber, 1949: 90). Whether as force or as authority, these structures of domination are variations of the apparatuses of capture. They are iron cages, as Weber used to call them, that come about when power (macht) is monopolized (see Weber, 1978). These structures are at the extreme limits of the exercise of power - that is, where its relations have stopped or when the difference that motivates the relation has been surrendered. However, between the polarities of capture and consensus lies an interval space where settlements are not secure and where the limits of closure have not yet been achieved. This unsettled margin is a terrain of open contest; it is a place of challenge and response, where everything is not already solved or worked out in advance. This terrain is the place of the difference of difference. It is the locus of power relations as I conceive them. Michel Foucault (1983) claims that power is activity in a double sense and cannot be reduced to the ability to act on the still, or on stable or inert matter, or to the ability simply to dispose of finished products. It is action on an actual or potential action and thus does not work through capture or consent. The principle of power lies somewhere in the interval between these extremes in the conduct of conduct. Power presupposes resistance as the condition of its exercise. Incapacitation always does violence to objects or things. Consensus reigns over those unable or unwilling to resist. By dispossessing die other of the will to respond or of the means of flight, these mechanisms eliminate the very conditions necessary for the exercise of power. Violence and submission are power's absolute limits, not the basis of its exercise. The real, material basis of power is the potential to move or act (puissance or virtual power). Power works on this potential and aims ultimately not to eliminate it, but rather to control its flow. Its domain includes the processes and flows of work (labour processes) and its products (commodities), but also other kinds of activity. All types of movement (of bodies, knowledge, light, even images) are potential subjects of power. The mechanisms that process and control the movements and circuits of power-knowledge are specific kinds of social formations. In order to understand these social formations, we must consider the elements and components necessary to their analysis. I argue that these

36 Mediated Associations formations are more than simply discursive regimes or systems of knowing or saying. Foucault's conception of the dispositif signalled a shift in the analytical emphasis of social formations. In addition to discursive formations, other spatiotemporal and optical dimensions come under scrutiny in terms of their power-knowledge effects. As Shields points out: "Dispositif is distinguished from episteme primarily because it encompasses the non-discursive as well as the discursive" (1991: 43). This dimension changes the character of the analysis of social formations. Deleuze calls these social formations "machines which make one see and speak" (1992b: 160). Foucault conceives of the dispositif or social apparatus (a term of equivalence used by Shields, 1991; Deleuze, 1992a; Bhabha, 1994; Patton, 1994; Rella, 1994; and Escobar, 1995), as "heterogeneous ensembles that consist of: discourses, institutions, architectural forms, regulatory decisions, laws, administrative measures, scientific statements, philosophical, moral and philanthropic propositions - in short, the said as much as the unsaid. Such are the elements of the apparatus. The apparatus itself is the system of relations that can be established between these elements ... whether discursive or non-discursive" (1972: 194-5). "Social apparatus" is a general term that encompasses the structures and processes involved in any specific regime of power. This conception also fosters a comparative approach to different modalities of power (such as those identified by Foucault as sovereign, sign, and disciplinary regimes) in terms of their transformation spaces and mechanisms designed to control the flow and movement of bodies and information. In this conception, there is no apparent analyses of the relationship - and possible overlap - of the sociological concept of "institution" and the Foucauldian concept of "apparatus." I submit that, unlike institutions and the a priori bodies whose nature they are charged with constraining or socializing and whose social function is primarily one of control by imposing limits, the investigation of social apparatuses attempts to enter into the domain of the strategic situation where these essential bodies and their associations are produced as objects of knowledge and subjects of power prior to their subsequent marshalling in institutions. The structures and processes of social apparatuses are not identical with the hierarchy of organized positions and role constraints that characterize formal institutions and bureaucratic settings, such as prisons, schools, factories, and hospitals. Social institutions are characterized by the equilibrium that they offer, by the constraints and consistency of their moral rules and roles (Durkheim, 1951; 1933), by their formal positions and offices (Weber, 1978: 217-54), and by their commonly held, though variable, patterns of temporal value orienta-

37 Three Social Apparatuses - and a Fourth?

tion (Weber, 1949: 112, Parsons: 1951: 169). There are two characteristics that to me seem common to institutional analyses. They are patterns of relations that persist over long periods of time, and they follow a master historical trend of social change - that is, a process of division of labour whereby multi-functional institutions and roles invariably give way to specialized ones (see Durkheim, 1933). Aiming at greater historical specificity (arguably) and at gaining a more supple framework, contemporary theorists have introduced notions such as uneven differentiation to take into account differential speeds of institutional change and have introduced the concept of dedifferentiation to designate those reactionary syndromes that reject institutional complexity in favour of less differentiated forms of social organization (see Colomy, 1986). Whatever their complexion, institutions work continuously to exercise their key adaptive, controlling, goal-directed, and integrative social functions despite undergoing radical structural changes. They are systems of enclosure that work to establish and maintain firm boundaries, such as the boundaries that are formed by what Durkheim (1982: 50) terms "social facts." They emphasize consensus or submission to formal rules and roles and therefore attempt to rule out opposition. Some social institutions - prisons for example - are conceived as homogenizing systems of interiority, or what Goffman (1961) calls total institutions. As a reprieve from the tyranny of such total institutions, Weber (1978) supposed that the most exceptional figure might occasionally come along to break the chains of despotic unreason or to guide the enthralled from their bureaucratic internment. This opening in the institutional framework of society would be only temporary, however, because the saving grace that served as its motivating force was still subject to human frailties and could be only short-lived. After their momentary reprieve, the liberated would invariably find a reason or a strategy of succession to return to their traditional chains or bureaucratic cages. Momentary reprieves notwithstanding, if social institutions are synonymous with systems of constraint or systems of interiority, then social apparatuses are not institutions. Social apparatuses are drawn from "diagrams" (abstract machines) or "maps of the social field" (Deleuze, 1986: 34-5). For example, discipline - one of Foucault's most noteworthy social apparatuses - is not identical to its form of actualization in the concrete mechanism of the panoptic prison or to its realization in schools, factories, and barracks. Each actualization operates with a different social cartography and produces a different set of structures, architectures, processes, and expected associations ('You are in school now, not at home"; 'You are in prison now, not at school"; 'You are in the army now, not at work";

38 Mediated Associations

... ). Discipline is a discontinuous "carceral archipelago" (Foucault, 1979: 301). It began with a diagram of an alternative cartography (see also Rajchman, 1999: 47-9) of the social world that followed the movements of the great plagues that swept through Europe and the movements of people that constantly changed the social terrain. The so-called strategies of confinement that followed in its wake produced not so much systems of interiority as structure movements between that which was designated as interior and the forces of the outside. Disciplinary structures form at marshalling points, at city gates, docksides, or other equivalent spaces of passage, movement, or change. Despite de Cauter's (1993) claim that social apparatuses are closed systems, if we follow Foucault's analysis closely, we see that even the most conspicuous forms of confinement are not at all synonymous with systems of (en)closure. They are also systems of exteriority that are shot through with calculated openings or lines of (f) light. Discipline, as we see below, is a mobile and itinerant set of strategies that can, so to speak, slough off their attachment to heavy institutional structures in order to become light-weight mechanisms that can be transported to almost any site. While it may be a strategy of long-term exposure, discipline also occupies a site of continual transit and transformation. We should also bear in mind that discipline is only one form of the social apparatus. There are also other sites that are much more short-lived and sporadic. This is an important distinction to make because, in addition to the great spaces of places or interiority that the institutional authorities are supposed to police, there are also innumerable spaces of flow or passage that form on the margins of a place. These are sites of movement and change. They are spaces of transit and transformation from one place to another, from one state to another, or from one phase of training to another. These spaces are what I call "strategic sites." It is there that we find the social apparatuses of which I speak. Social apparatuses, such as discipline, are strategies of movement that work in the frameworks of social institutions and between them to shape movements and associations of movement that form or can form wherever differences come into proximity or contact. It is because the spaces of places of the social institution are only relatively closed that they are subject to flows of all kinds that cross their boundaries. Talcott Parsons (1951), an early sociological theorist of the media, already understood that social systems are engaged in a process of continual exchange with their outside. However, his proposed cybernetic model of system exchanges presupposes that exchanges of information and energy are balanced and symmetrical and therefore help maintain the equilibria of power and knowledge systems. What

39 Three Social Apparatuses - and a Fourth?

this perspective misses is an analysis of the structures and processes that form on the margins, that shape the mixtures that form in these sites, and that produce dissymmetrical flows and unequal exchanges. Firmly bounded systems and system boundaries - often considered as rigid territorial frames in need of constant protection and maintenance - are being revisited and reconceptualized. They are understood variously as permeable or osmotic membranes, marginal sites and liminal spaces, intervals of movement and change, places of passage, interfaces between moments of closure, and surfaces where the inside and outside meet. The once-exclusive and -exclusionary boundary has become a surface, or interface, between adjacent places, regions, bodies, substances, phases (of training), or stages (of development). These are potential sites of research into the innumerable points where supposedly independent, differentiated, often-polarized substances, bodies, ideas, images, individuals, groups, and systems make contact, interact, mix, and intermingle. These intervals and intermediaries are potential sites of research into the dynamics of power-knowledge relations. Understood as active, productive relations rather than as acts of seizure, arrest (Barthes, 1973), fixation (Bhabha, 1994), or repression, power is related to movement and to the intervals of movement - to the spaces of flow between the spaces of places that once delimited the analysis of social institutions. The analysis of the social apparatuses of power and knowledge that I undertake does not seek to uncover the secret, methodical plottings of an ordered, arranged, or systematic history that develops ideologically or unfolds gradually in a variety of (sublatable) forms or modes. The pretensions of historical method aside, my perspective is primarily methodological, insofar as it seeks a comparative understanding of the different methods of power and knowledge that have been deployed in strategic sites as well as a reflexive understanding of the objects, subjects, and bodies produced by them. Rather than constituting a hierarchy of forms, or a history of the past in terms of the present, I aim simply to come to terms with, and develop comparative conceptions of, the multiple social apparatuses that aim to govern and control strategic sites and the social bodies that form around them and that regulate their capacities for movement, their spatial and temporal displacements, their deterritorializations and reterritorializations, and their mixtures and intermingling. Social apparatuses are media of control that operate in the spaces between institutions, but also within and through them. Institutions are spaces of places that are shot through with openings, with places of entry and exit, as well as with spaces of possibility and points of resistance. These other sites provide the fertile ground for the social apparatuses that exercise their

40

Mediated Associations

power everywhere and only on a body's puissance - that is, on its capacity or potential to act and move. All social apparatuses, from sovereignty to discipline and beyond, feature regimes of light, regimes of enunciation, as well as lines of force that cross between the visible and the utterable and constitute their power dimension. My argument is that each apparatus therefore comprises, first, an optical machine. An optical machine consists of lines or planes (plans) of light that structure fields of visibility and invisibility, illuminating some objects and causing others to disappear. Second, each apparatus includes a sonorous machine, comprising lines of enunciation, or that which can be enunciated in discourse or uttered in a system of signs. These "modes of symbolic expression" (Patton, 1994: 163) often serve to authenticate or authorize presences and absences in the visual field. Each social apparatus is an audiovisual machine whose specificity lies (a) in its particular regime of light or visibility, (b) in its style or form of enunciation (i.e., discursive regularities), (c) in the lines of force that forge associations between the seen and the said, between situations and responses, verbal or non-verbal, and between the visible and its representations, and (d) in the mediating structures or agencies that serve as interfaces between spaces or states and determine both quantitative and qualitative dimensions and directions of flow. The social apparatus itself is the conjunction of these aspects. The specificity of any social apparatus therefore derives from the way in which these aspects are articulated or brought into relation, the strategic conjunctions that they form, and the forms of association over which they preside. I begin my analysis of social apparatuses by emphasizing aspects of visibility. As Escobar (1995) claims, the idea of visibility and its specific deployments (as a panoptic gaze) have become synonymous with the apparatuses of social control and the emergence of what Bogard (1996) terms the "surveillance society." Escobar also suggests that the role of vision extends far beyond these new technologies: "The birth of science itself was marked by an alliance that almost two centuries ago 'was forged between words and things, enabling one to see and to say' (Foucault, 1975, xii). This alliance was enacted by the empirical clinician upon opening the corpse for the first time 'to really see' what was inside. The spatialization and verbalization of the pathological inaugurated regimes of visuality that are still with us. From the analysis of tissues in nineteenth-century medicine through the microscope and the camera to satellite surveillance, sonography, and space photography the importance of vision has only grown" (Escobar 1995:155-6).

41 Three Social Apparatuses - and a Fourth? While previous physiological and anatomical experts were largely resigned to the analysis and interpretation of the extrusions (for example, in phrenology) and humours that characterized a body's disposition, the flaying and slicing clinician opened new pores in the body's already porous surface that would allow the light of the truth to penetrate and illuminate, chart and organize, its darkest regions. The early social physicians (Saint-Simon, Comte, Spencer, and Durkheim) would apply the same strategy to the social body by identifying its primary and necessary organs (institutional patterns), tending to its needs, and deciphering its disposition or state of health. This kind of analysis of the social body required a reassessment of its foundations. This reassessment focused not on the so-called founding conventions, contracts, or accords, and it would not seek to evaluate these agreements in terms of their rationality. The new analysts would focus instead on observable regularities, the persistence of their patterns, and the visible organization of things generally. These observable regularities would stand as the limit of what could be legitimately said or claimed about the social body. These social facts, as Durkheim called them, exerted a force that was beyond doubt, and they did so whenever there was an attempt to move against them. While you cannot see them directly, you can feel their force because they occupy space in much the same way as walls do. As physical barriers, constraints, or limits, they cannot simply be wished or reasoned away; they can be overcome only by the application of a superior force. Like overriding laws of nature, these facts are external to the will and consciousness. They impose on the latter, on bodies, and on the observation (see Durkheim, 1982: 69) in an undeniable way. They determine the conditions of vision. The externality and reality of social facts are visible in collective patterns of action - that is, in the ways that collective bodies are forced to deviate, or move around, them and in the way they are channelled by them. These social facts are called "collective representations." They are manifestations of the power of a collective conscience. Like planets that cannot be seen directly, the power and presence of this collective force are made known and verifiable in the representations that they produce - that is, in the aberrant motions that they produce among other bodies. While the social fact is an absent presence, whose presence is attested to by its representations, other aspects of representation in the visual field are not accorded the same seriousness. These are images proper. They are considered internal representations. As such, they do not occupy space, and they offer no resistence to the will that creates them or dreams them up. If such an image were able to affect a body's way of moving, either the image would have to be taken

42 Mediated Associations seriously and considered as an external, physical presence, or the effect that it produces would have to be considered erroneous, mistaken, a mere deviation based on fantasy or opinion rather than on the solid reality of the fact. Is deviation in and of itself enough to confirm (affirm) or deny the solidarity of the social fact or to differentiate the factual from the imaginary? The difference between the fact (as collective representation) and the image (the fictitious, no less than individual, representation) is one of degree. The difference is only quantitative and therefore affirmed only via statistical enumeration. The social fact is a statistical regularity or statistical artifact. What counts as a social fact is an opinion, belief, or practice with numbers on its side. It is a form of orthodoxy that has assumed the status of reality, one that is said to enclose and circumscribe the basic institutional patterns of society and to determine which practices are necessary for its continued functioning. What Durkheim misses, however, is an analysis of the mediating forces that manage the fields of visibility that he describes, the various ways in which the visible field is structured by power, and the ways in which visibility is deployed to bring some objectivities, types of behaviour, and practices to light, and to hide others. What are the power relations involved in determining the dominant conditions of vision? Durkheim first describes institutional patterns and then articulates their relevance and necessity in a discourse of morality. I, in contrast, analyse social apparatuses as regimes of visibility, power, and knowledge that form on the margins between the internal and external and that mediate between facts and images. Before these distinctions are galvanized and their boundaries are affirmed, in this grey area of potential opposition, there is a space of other possibilities where a multiplicity of seemingly contradictory conjunctions and potentially contradictory associations can form. All social apparatuses involve the encounter of bodies, the effects of one body on another, and the kinds of associations that these bodies compose as a consequence of their intermingling, or, conversely, the power relations that organize and co-ordinate their effects. Discourses are not simply imposed on bodies, nor do they order their relations. Discourses, narratives, and texts do not move or circulate freely, nor do they act on their own accord. Their momentum is derived externally. That which incites, seduces, and constrains discourse is an ensemble of power relations. Discourse, especially moral discourse, is but the archive of these relations. In itself, discourse is insufficient to account for the dimensions of power-knowledge. Strategies of spatialization in the non-discursive field are also important: "Anyone envisioning the analysis of discourses solely in terms of temporal continu-

43 Three Social Apparatuses - and a Fourth? ity [i.e., narrativity] would inevitably be lead to approach and analyse it like the internal transformation of an individual consciousness. Which would lead his erecting a great collective consciousness as the scene of events ... Endeavouring on the other hand to decipher discourse through the use of spatial, strategic metaphors enables one to grasp precisely the points at which discourses are transformed in, through and on the basis of relations of power" (Foucault, 1980b: 70). Power relations are constituted at the interstices of the discursive and the non-discursive, where the non-discursive is the space (territory, terrain, body, domain, soil, horizon, region, site, landscape) for the production and deployment of the discursive regularities (marks, signs, representations, symbols, images). The formation of discursive-non-discursive conjunctions requires structures and agencies to form interfaces, to mediate the contact of these aspects, and to constitute their associations. The relative stability of discursive enunciations, as truths or certainties of representation, signals the relative stabilization of an ensemble of power relations by embedding them in the spatial and ensuring their persistent visibility. Similarly, changes in the discursive regimes of enunciation signal changes in the exercise of power and new forms of domination. But no system of domination is ever absolute. Total victory means eliminating the opposing force, eliminating the power relation, the subjects and objects of knowledge, and the means of producing truth(s). As Foucault reminds us, power is exercised only over individuals or collective subjects faced with possibilities (1983: 225). So, when he describes torture as the art of maintaining life in pain (1977: 33), he is articulating a form of power relation that delays or defers death and thereby produces calculated openings for the development of knowledge - that is, the extraction of truth from a body in the form of confession. Strategies of power that enjoin regimes of light and enunciation form mobile systems of power that cut through institutionalized frameworks and transform them. This conception of the social apparatus is fruitful for comparing strategic relations across a number of dimensions and directions, and it is especially useful in exploring the four types of associations that systems of power and knowledge have established with and within the audiovisual field. SOVEREIGN ASSOCIATIONS AND PUBLIC EXECUTIONS

The social apparatus of sovereignty included a visual machine, the scaffold: a raised platform designed to be seen while maintaining the

44 Mediated Associations integrity of the mise-en-scene (or staged spectacle) by keeping spectators at a safe distance. This scaffold is a mechanism of framing. It works to divide and differentiate seeing and scene and so to limit participation. The difference and distance that it evokes, often reinforced by the presence of border guards, established a kind of spectatorial division of labour, or a "division of spectatorial tasks," as Debord (1994: 37) defines it. Primarily an apparatus of projection, sovereignty emanates images and signs, casting its radiant, stellar force in all directions, for all to see. It is nothing, however, without an impressionable body, a white screen so to speak, on which to project its images. The screen is only a means of reproduction and representation; it is the collective that is assembled only to reflect or celebrate the images of sovereign omnipotence. Sovereign power has to be seen to be effective - that is, if it is to carry out its strategic function. As Foucault states, "an execution that was known to be taking place, but which did so in secret, would scarcely have any meaning" (1979: 57-8). The technological function of any viewing screen is to be, according to Nelson, "blank, a surface of nothingness, upon which can be projected images that originate from a different source, the projector." To equate a spectatorial crowd, a population, or a country with a screen is "to recognize that it plays the role not of originating any images, for after all, a screen cannot do that, but only of receiving the projected image" (1992: 64-5). The "good screen" is therefore the one whose surface is free from all imperfections or protuberances - that is, of anything that might interfere with the flow of representations or the spectacle's "totalitarian monologue." As Minh-ha argues, the totalitarian monologue is based on the blind denial of the spectator as "reading subject and meaning maker-contributor" (1991: 93). These spectacles of sovereignty and the practices of the public execution are not, as Thompson (1995: 125) claims, direct face-to-face, or dialogic, relations between a sovereign and a crowd. As I see it, these spectacles are mediated processes and forms of indirect encounter. The middle or mediating term in the sovereign-subject relation is the body of the condemned. What is on display is not the sovereign, it is the exercise of sovereign power. This power is exemplified in the spectacular marking of the body. It is by means of the spectacular mark that sovereignty displays its exceptional capacity to make meaning that is, by directly inscribing its mark on a body's surface. Like the fantastic inscription-machine described in Kafka's The Penal Colony, sovereign marks indelibly associate a signified with a material body. The process is often accompanied by establishing associations between sonorous and visible elements in the testimonials that the accused is

45 Three Social Apparatuses - and a Fourth? forced to provide. Each act of inscribing (or adorning) the body's surface not only affirms a dissymmetrical relation, it also establishes that the direction of power is one way and irreversible. While it acts directly on the body, sovereignty also has a more strategic function and a necessary connection to the public square, which, as Foucault claims, can be instantiated anywhere, at the scene of the crime, at crossroads or crossing points, or any other place of meeting (1979: 58). Here, in view of the public, sovereignty deploys the body of the condemned as a vector of the impression, to impress on the public its meaningful associations, and to affirm that the power to form these associations is solely the jurisdiction of the sovereign. Through its marks, sovereignty stakes out its territory and the limits of its jurisdiction. The sovereign spectacle is an audiovisual machine and apparatus of power and knowledge. While it acts directly on bodies, it indirectly targets the undifferentiated mass of witnesses to its display. In this sense, it is indiscriminate in its application. In its planned excesses, sovereign power needed to be spectacular - that is, to be both visible and verifiable to the homogeneous social body that was its subject. From this point of view, the power of the spectacle is limited to producing and containing the meaning of the event. Its framing mechanisms are strategically placed1 to prevent this staged order of meaning from being flooded and becoming ambiguous. I say this because the same social body that served as a relatively effective projectile mass when launched at enemies was also potentially unruly: it could also reverse its field and turn against that which incited it or set it in motion. Resistance to sovereignty and to its spectacle of violence came in the form of carnivalesque plays of meaning whereby the crowd would reflect back inverted and deformed images. Parallelling Bakhtin's (1984a) discussion of popular festive, parodic forms of resistance, turning criminals into heroes helped to break authoritarian chains of associations, erased the dividing lines that limited popular participation, and travestied the entrenched spectatorial division of labour. The inability to maintain staged boundaries and to contain the meaning of the spectacle were the crises that both hastened the collapse of the sovereign project and constituted a new diagram of the social field. At its extreme point of application, the power of the sovereign theatre lies in its capacity to segment and distribute the body that it targets. At this point sovereignty also ran up against another jurisdictional limit where its power was limited to the power of life and death: "a body destroyed piece by piece by the infinite power of the sovereign constituted not only the ideal, but the real limit of punishment" (Foucault, 1977: 50). This form of cutting and dispersion

46 Mediated Associations of the body is an act ultimately ending in death. As Foucault notes here, sovereign power, in its grand theatrical design, also reached its limits over the social body in its ability only to cut out and eliminate any excrescent element and to cut off any line of flight that might exceed the integrity of its boundaries. By eliminating its mediating term, sovereignty eliminated its mechanism for creating lasting impressions. From its modified social terrain left in the wake of the failures of the sovereign projects, another regime would seize on the opportunity to create another social apparatus, one that would take advantage of the resistance that formed in opposition to sovereignty. ADVERTISING REGIMES AND LEGISLATIVE SIGNS

For some time now, the analysis of the audiovisual medium has been dominated by formal semiology and meta-linguistics (see, for example, the work of Metz, 1974; Kaplan, 1987; Fiske, 1989a, 1989b, 1978; Stam et al., 1992). Here, the dominant tendency has been to treat images as signs. On the one hand, this has resulted in attempts to show a mimetic relationship between images and reality by articulating the synchronous analogical codes that link images to a reality (everyday experience) to which they are supposed to refer (Metz, 1974). In this sense, images are placed within the classical referential semiotic framework comprising signifiers, signifieds, and their arbitrary relations. On the other hand, contemporary non-referential semiology treats only sign fragments and their digital codes. From this point of view, signs have achieved a high degree of deterritorialization, for they are no longer placed in a relation of designating bodies and things or even of indicating states of affairs. Instead, these deterritorialized signs refer only to other signs in syntagmatic signifying chains or circuits (as in Fiske, 1978: 54). When signs refer only to other signs, every sign becomes a signifier and its signified, another sign, which in its turn becomes a signifier, and so on. This makes the ultimate signified the Signifier itself and establishes the redundancy or circuitousness of the sign process. Because the signifying chain has effectively displaced the signified (along with any sense of paradigmatic support), the result is an infinite postponement or deferral of meaning. Refusing to be tied down, the signifier flees its body as much as its embodiment. Like the name that survives its owner, or the statement, its author, the restless signifier glides along a slippery, smooth surface, easily jumping from one signifying circuit to another.

47 Three Social Apparatuses - and a Fourth?

Levi-Strauss (1969), solves the problem of the relations of signifying chains by inventing totemic messages, which make the (binary) opposition of one set of signifiers homologous or analogous to another by the logic of a code. For example, the statement "We are wolves" is rendered reasonable (i.e., analogical) by translating the statement into the template A is to B as C is to D, or "We are to others as wolves are to sheep." Here the problem or tension between One and an Other is explained or solved elsewhere - that is, by the natural relation of wolves and sheep. To say that clan A is descended from wolves, and clan B, from sheep "is nothing more than a concrete and abbreviated way of stating the relationship between A and B as analogous to a relationship between species" (Levi-Strauss, 1969: 100) - that is by imposing on this relation a set of ready-made, structural contrasts (dualisms) to which a culture has tacitly agreed (such as raw/cooked, inedible/edible, and nature/culture). This mythic system (named "bricolage") makes possible a system of reference that equates one signifying chain to another or perhaps to many others on many different registers, planes, or circuits. Similarly in Barthes (1973), the language sign becomes impoverished, is emptied into a form, and is reduced to a mere signifier as soon as it is captured by myth (a second-order system of signification). For example, Barthes illustrates this by reading an image, an advertisement for Panzani. In this case, "the bringing together of the tomato, the pepper and the tricoloured hues" - that is, of the yellow, green, and red of the background poster - constitutes a signifying chain. At the same time, however, this signifying chain, the linguistic message of the first order, is emptied of its significant content, in order to become a pure signifier, or form, of another second-order signified, namely a concept (see Figure 2.1). By means of this articulation, a word, image, picture, or gesture, even a whole book, can serve as a signifier for a single concept (Barthes 1973: 120). This impoverished, empty signifier becomes a means or medium, a stolen vehicle, that is filled with ideological import and used to promote and circulate a concept - Italianicity, in this example (Barthes 1977: 34). It does so by drawing on ready-made familiarities that are already in circulation as dominant cultural stereotypes. This second-order sign system (i.e, the relation of empty form and concept) "stands in a relation of redundancy" (1977: 34) with the firstorder visual or linguistic message. It takes over the substance or meaning of the first-order message and uses it as an alibi in the perpetration of a sign-crime. Meaning, defined by Sartre as "the natural quality of things situated outside a semiological system" (in Barthes 1973: 133, n. 11), is myth's alibi (Barthes, 1973: 121-3). Like a virus,

48

Mediated Associations Sign (Signification)

Second-order (myth)

Signifier (form) First-order (language)

Signified (concept)

Sign (meaning) Signifier

Signified

Figure 2.1 Stratigraphy of the sign

myth uses frameworks of meaning to house and disseminate its concept. The process is colonial. Myth turns the impoverished, hollowed-out sign-body and its meaningful associations and expressions of meaning into a signifier in a foreign-language system. Nothing, it would appear, is safe from the larcenous scheming of myth and its capacity to formalize the expression. As Collins argues, "One of the great truisms regarding cultural production ... is that all forms of art and entertainment have become one form of appropriation or another, whether it be called pastiche, parody, revivalism or just plain retro" (1995: 92). While myth systems typically prefer to work with already impoverished, incomplete images, "where meaning is already relieved of its fat, and ready for signification," even the most fiercely guarded and firmly bounded language system is vulnerable to mythological interpretations. For around any finished, completed, or finalized meaning, there always remains "a halo of virtualities where other possible meanings are floating" (Barthes, 1973: 127). These virtualities can either infect meaning (robbery by colonization) or encompass it and direct it off course (by abduction or press-ganging2): "When meaning is too full for myth to be able to invade it, myth goes around it, and carries it away bodily. This is what happens to mathematical language. In itself, it cannot be distorted, it has taken all precautions against interpretation: no parasitical signification can worm itself into it. And this is why, precisely, myth takes it away en bloc; it takes a certain mathematical formula (E = me 2 ), and makes of this unalterable meaning the pure signifier of mathematicity" (Barthes, 1973: !32). Even the resistance that modern poetry offers - that is, its apparent lack of order of signs - is captured by myth and transformed into an empty signifier "which will serve to signify poetry" (Barthes, 1973: 134). In the end, resistance is futile: "myth can always, as a last resort, signify the resistance which is brought to bear against it." Even the resistance that silence offers can be surrounded by myth and made to signify repression. Myth is a form of power that uses resistance as the basis of its exercise and its means of dissemi-

49 Three Social Apparatuses - and a Fourth? nation. In essence, resistance is transformed into a mechanism of power. The means of inf(l)ection notwithstanding, any (apparently) closed or disordered language system, as well as an open system such as the expressiveness of articulated language and the expressive image, is merely an empty gesture, for the play of mythic transformations, and a vector for the dissemination of its concepts. As Deleuze and Guattari argue, the "pure formal redundancy of the signifier could not even be conceptualized if it did not have its own substance of expression" (1987: 115). Any expression or expressive entity can become a formalized substrate to which generic or deterritorialized sign-concepts affix themselves and can therefore be enlisted in the service of the signifying regime. "[N]ot only written discourse, but also photography, cinema, reporting, sports, shows, publicity, all these can serve as support to mythical speech ... (I)t is because all the materials of myth (whether pictorial or written) presuppose a signifying consciousness, that one can reason about them while discounting their substance ... We shall therefore take language, discourse, speech, etc., to mean any significant unit or synthesis, whether verbal or visual: a photograph will be a kind of speech for us in the same way as a newspaper article; even objects will become speech, if they mean something" (Barthes, 1973: 110-11). Because of the generic way in which Barthes applies the concept of language, everything is already a potential signifier of languagicity. It is, after all, the "analogy of meaning and form" that establishes the crossing and mixing of strata, of first-order associations with those of the second-order, and turns the signifier into something duplicitous; with one facet turned towards the materiality and particularity of the meaningful association as the other turns away and assumes "the look of generality" while seeming "neutral and innocent" (1973: 125—6). Sign-Crimes at the Cross-mads

The analogical model analysed by Barthes and Levi-Strauss (among others) was originally pursued by the legislative-sign-regimes of the eighteenth and nineteenth centuries. These regimes emerged in the face of a badly distributed system of social justice. For example, sovereign power, in an effort to cover its excesses, including its punitive excesses, had appropriated the right to sell legal offices. The resulting multiplication and dispersion of the number of judicial authorities created a discontinuous, overlapping, and contradictory legal apparatus. From this new diagram of the social field, subsequent reform

50 Mediated Associations movements planned a new economy of justice that would "assure its better distribution, so that it should neither be concentrated at certain privileged points, nor too divided between opposing authorities; so that it should be distributed in homogeneous circuits capable of operating everywhere, in a continuous way, down to the finest grain of the social body" (Foucault, 1979: 80). Central to the development of this economy was a new sensibility that would no longer perpetuate the idea of an unbridgeable gulf between the lawmaker's arbitrariness and those subject to it (nor the idea of a lawmaker's reasonableness opposed to the criminal as a monster vomited by nature - an idea that would come much later). Rather, this new general economy was to set itself against the very reasons (interests, motivations, values) that led the criminal to commit the crime. The emphasis on reasoned and reasonable motivations also corresponds to the development of what Thompson calls a "reading public" that had access to public events by means of the printed word (1995: 126). In this new economy it is the social body as a whole that is attacked by the crime, rather than the sovereign. Punishment therefore had to shift from the vengeance of the sovereign to the defence of society. Crime was now injurious to the social body because of the scandal that it produced, the whole "series of disorders that it is capable of initiating," and the danger that "everyday offences might multiply" (Foucault, 1979: 90). Punitive strategies should not therefore target the crime per se. Crime is dangerous in its pedagogical function, in the examples that it sets, in the models that it constructs, and therefore because of the contagion, repetition, dangerous copies, or representations that it is capable of engendering. In Foucault's words, "one must punish exactly enough to prevent repetition," yet the "penalty must have its most intense effects on those who have not committed the crime" (1979: 93, 95). Therefore, in addition to its specific point of application, this new punitive economy should also produce generic effects. In Canada, for example, Charles Buncombe's Report on Prisons and Penitentiaries (1836) suggested that the goals of the new punitive strategy should be threefold - to deter others from crime, to prevent the repetition of offences by offenders, and to reform those who break the law (see Ekstedt and Griffiths, 1988). This strategy of linking the specific and the general, or what could be called the two faces of deterrence,3 required what Foucault calls a semio-technique and a new apparatus of re-presentation. The semio-technique (as we would call it) proposed by the penal reformers rests on the hypothesis that the motivation or reason that incites the criminal action is the same as that for any other normal or everyday action; it is motivated by the advantage, pleasure, reward,

51 Three Social Apparatuses - and a Fourth?

convenience, value, and so on that one associates with it or expects to gain by it. Put in the terms of Barthes's model, the signifier action has an associated signifier - pleasure - which constitutes a sign of the first order, or pleasurified action (pleasurified action is distinct from the crime passionel. Because the crime of passion does not signify a rational motive it became an excuse rather than an offence.) The habit of meaning formed by this primary chain of associations establishes the path or circuit that carries the action (crime included) along a specific line, in a specific direction. In criminal matters this directed flight must also be subject to verification and the signifying chain reproduced for all to see. It is therefore necessary "that the legal procedures should not remain a secret, that the reason the defendant is condemned or acquitted should be known to all" (Foucault, 1979: 96). While there is always a risk of glorifying and celebrating the actions of a defendant by this public representation (as was the case under sovereignty) , the ultimate aim of this strategy is to set up a counter-sign system that would essentially steal the momentum of these everyday associations and the popular tales that celebrated disorder - namely, to find a punishment for the crime whose idea is such that it robs forever the crime of its pleasurable associations. To counter the trajectory of crime—pleasure associations requires establishment of a crime-punishment association. In order for the latter to be effective, it should, first, have at its disposal slightly more force than the former - that is, there should be a quantitative difference between the opposing forces, so that one would be led to the conclusion, by simple penal arithmetic, that the displeasure of possible penalty was more weighty (by degree) than the possible pleasure to be derived from the crime.4 Second, if crime was to be receded in the popular imagination that is, emptied of its meaning and supplied with another significance and direction - the counter-sign should also be disseminated or circulated as widely as possible. The act of publicizing this punishment-sign should arrest5 any further representations or copies. As Foucault claims, long before he or she was regarded as an object of science, the criminal was imagined as a figure of instruction: "Let us conceive of places of punishment as a Garden of Laws that families would visit on Sundays (Brissot, 1781). Punishments must be a school... where everyone may learn the significations ... This, then, is how one must imagine the punitive city. At the crossroads, in the gardens, at the side of the road being repaired or bridges built, in workshops open to all, in the depths of the mines that may be visited, will be hundreds of tiny theatres of punishment" (Foucault, 1979: 113)It was the business of these instructional mediations and mediated forms to arrest and invert the popular discourses and tales that

52

Mediated Associations

celebrated criminality. In order to do so, captured signs had to be returned to circulation. To make them most visible and achieve the most rapid and widest broadcast, the reformers favoured public works as one of the best possible penalties: "Public works meant two things: the collective interest in the punishment of the condemned man and the visible, verifiable character of the punishment. Thus the convict pays twice; by the labour he provides and by the signs that he produces. At the heart of society, on the public squares or highways, the convict is the focus of profit and signification. Visibly, he is serving everyone; but, at the same time, he lets slip into the minds of all the crime-punishment sign: a secondary, purely moral, but much more real utility" (Foucault, 1979: 109). By these mechanisms a large mass of first-order signs were conscripted for service as mediating forms of a mythico-legal discourse. Third, for the crime-punishment association to be effective, it would have to background6 the force relation that produced it. The links between crime and punishment should not appear as an arbitrary (despotic) chain or a political proposition. The links should be immediate, and the sutures, invisible. It should appear as if the circulating signifier-body naturally called up its own punishment. If punishment followed crime as certainly and inevitably as night followed day, then the application of the penalty would be simply a natural consequence of pursuing a prohibited line of action. No one could be held responsible for an association that derives from the very nature of things. As Barthes asserts, myth systems do not aim to hide anything, so there is no need of an unconscious to explain how it works. Myth works by exposure: "its function is to distort, not to make disappear" (1973: 121). It does not efface the crime or eliminate the offending body; it deforms its meaning while retaining its remaining framework (the hollowed-out sign-body) to control and promote alternative associations through mythical images. I use the term "image" here to designate an absent presence or a way of presentifying absence. The mythical signifier plays on this difference; its form is empty but present, its meaning absent but full (Barthes, 1973: 124). The mythical signifier can therefore serve to summon up, by means of association, the presence of a generic concept like punishment or imperiality. Forcefully rerouting the signifier and then dissimulating this act of transgression remove any trace of intervention. It is as if something took over the sign-body of the criminal, arrested its movements, turned it away from its associations and intended course, pointed it in another direction, along another path, towards another goal, and then removed the directional force, lest it get bound up or confused with the direction indicated. The overall effect is to make the discrete, par-

53 Three Social Apparatuses - and a Fourth?

allel signifying chains reach a natural junction or crossroad (typically in the public square). This meeting place would be populated by innocent looking warning signs indicating the presence of future obstacles (punishments) and the necessity of taking detours to avoid them. Any definitive mode of linking across the gap separating the signifying chain constitutes a narrative scheme. While such "schemings of language" (Chevalier, 1990) are acts of transgression - constituting a "leap," as Barthes (1974) puts it, "over the wall of antithesis" - the aim is also to naturalize flows across this juncture. The logic of the legislative-sign regime is analogy. Its formula is designed to establish one plan of transgression as the preferred mode of movement and transit. By dissimulating the work of transgression, the act of encoding relations appears seamless, as if governed by natural, ready-made formulae. Combining schemes of transgression and naturalization reduces language to the deontological function of what Deleuze and Guattari (1987) call "order-words." These order-words are mythological imperatives designed to govern and master relations, associations, and movements across a gap in semiotic systems or between one system of associations and another. However, narrative mastery does not imply simply that there is one preferred manner of linking sign fragments; rather, it commands or authorizes certain performances (readings) at the expense of others and thereby tells us what we must think or how we must act. As an exercise of power, the aim of the master narrative is to constitute reading habits and to overrule our everyday associations through their circulation and by dint of repetition in innumerable forms. The metaphors used to describe such language schemes centre on the tactical deployments of space and movement. Their manoeuvres go from place to place and set up potential obstacles in the interval constituting the space of transgression. To normalize transgressive space is to restrict or prohibit the unwarranted (carnivalesque) associations that might occur there. This is accomplished by steering the flow away from such aims and directing it to others through various representations of signs of control. Our everyday movements are subject to an immense variety of such signs. Often the punishment is literally inscribed on the sign. The sign of the crime also warns of the fine, or the degree or duration of punishment, for violating the legislated order-word. It is a simple penal arithmetic of advantages and costs where, all else being equal, the superior weight of the embodied punishment association should be just enough to impede the momentum of the offending event or prevent its generalized repetition. Stop signs, no-parking signs, and other no-go signs are ubiquitous in our fields of visibility, each calling

54 Mediated Associations

up ideas of punishment, presenting obstacles, and forcing detours. These are not spectacles in the sovereign sense. They are more like spectres of power that haunt the places and spaces where meaning used to reign supreme. While the semiological apparatus of power instantiates ideas of control in almost every public venue conceivable, and while it may appear that no communication, no matter how meaningful, is invulnerable to its schemes, there is another apparatus of information, knowledge, and mediation that is equally powerful, if not more pervasive. In a world of myth there is no longer anything to interpret that is not already itself interpretation, where the ultimate signified is the great empty signifier itself, which goes on copying itself to infinity. Perhaps we should simply forgo the expressive altogether and no longer count on reaching something like a first-order association. Barthes's solution, drawing on Flaubert, is to constitute a third-order system that would rob myth in its turn. The first term in this thirdorder semiological chain is the second-order myth-sign. Consequently impoverished in its mythical significance, it is turned into mere ornamentation - the term that Barthes employs to designate the empty form of a third-order process called demystification. The problem of interpretation (verstehen) took a related turn with the advent of movable type and the mechanism of textual reproduction, i.e., the printing press (the basis of McLuhan's Gutenberg revolution). This device made it economical to copy and disseminate the root text of a tradition. Words that had once been the closely guarded secret of an elite became accessible to the masses, and a religion that was once "scarcely perceptible in practice" (Weber, 1958) earned a new, practical significance. Instead of relying on the medium of priests, others would have direct access to the order-words and interpretations would spread like a disease. New interpretive mediums such as Martin Luther and John Calvin were formidable decoders. Their respective reiterations of the calling and fate would later combine both to celebrate common labour (this-worldly activity) and reassess the value of its products (material wealth). Just as labour becomes associated with prayer, its products take on ethereal values (surrounded by "metaphysical niceties," as Karl Marx put it), which estrange them as objects of utility, consumption, or subsistence. Wealth becomes a sign of election, an image of salvation, a marker of status in the other-world. Securing one's future, by producing ever more signs of it, becomes the object of relentless activity. Idleness and its associated consumption habits (the enjoyment of wealth, greed, gluttony, and so on) became the deadliest of sins. Later, Weber (1958) argues, this process turned irrational as the great signifier (wealth),

55 Three Social Apparatuses - and a Fourth? formerly a sign of grace, was transformed into an end in itself, as it was emptied by the forces of another tradition that inf(l)ected it and took it over. This final catastrophe of meaning, as Baudrillard (igSga: 102-3) might call it, was brought about by another media revolution and another apparatus of power. The apparatus in question emerged out of modern architectural arrangements that were designed to do more than simply arrest movements, frame up signifier-bodies, and distort the rational elements of everyday associations. Under the cloak of leniency and humanitarianism - or partly as an "unintended consequence" of authentic benevolent and religious sensibilities or of the cultivated sensibilities of travelling persons, promoting these principles (see Garland, 1990: 159, 227-8) - a more extensive modern apparatus of control took hold of the social body. This apparatus, which has come to be known as discipline, comprised a set of social strategies that encompassed a much broader ensemble of associations. The targets of disciplinary interventions are all the mysterious associations, rational or otherwise, that are beyond its system of knowledge and do not accord with its system of power. Discipline targets ways of life rather than disorderly action. Here, the total array of associations, including all the ordinary unruly and wayward ones, that constitute a way of life now stand at the centre of social disorder. As a consequence of this new diagram of the social terrain, Bentham, perhaps the first modern city planner, imaged a different kind of punitive city. Using the art of the frame in novel ways, he redesigned the cityscape as a complex audiovisual machine that would create, for all those who entered it, an all-encompassing, artificial social and moral milieu of discipline. DISCIPLINARY ASSOCIATIONS AND PANOPTICON PRISONS

"Up to this point, it requires manual labour, but from now on, the apparatus works automatically."7 Disciplinary apparatuses also have optical machines. These give a selective spatial presence. The school has an optical machine for creating systems of presence and absence; the prison, a dissymmetrical-looking machine. These panoptic machines involve "dissociating the see/being seen dyad; in the peripheric ring, one is totally seen, without ever seeing. ... It is an important mechanism, for it automatizes and disindividualizes power. Power has its principle not so much in a person as in a certain concerted distribution of bodies, surfaces, lights, gazes. ... Any individual, taken almost at random, can operate the machine" (Foucault, 1979: 201-2). It is in

56 Mediated Associations this sense that Virilio defined the "vision machine" as realizing a selfsufficient, spectatorless gaze (1988: 5). The panopticon is the exemplary disciplinary apparatus. Its optical machinery reverses the principles of the dungeon (to enclose, deprive of light and hide), as well as those of the palace (designed to be seen) and the fortress (designed to survey an exterior). Premised on the notion that light was more powerful than (despotic or signifying) chains, the encompassing space of Bentham's enlightened, ideal prison city8 was doubly functional. On one hand, the panopticon served to trace out a territory, to control movements in and out of the territory, and to structure lines of flight. The panopticon defined an image of a relatively closed universe in which space curved in on itself, with a focal point of illumination (point of convergence of the lines of light) occupying the city centre or public square. The focus of attention was a non-luminous, dark body. Akin to a black hole - a dark object formed by the collapse of radiant stellar matter - this object neither emits nor reflects but rather captures all other forms of radiant energy. Like an independent variable, the power of the dark object is detectable only in its effects - that is, in the distortions of movement that it produces. Bentham's dark body was a central watchtower that could potentially capture all the images of movement (like the eye of the camera) and from which no movement could escape. The rest of the panoptic machine was composed of many partial frames that were placed strategically on the horizon of the enclosure to form back-lit cells. These constituted spaces of subjective habitation. These frames that surrounded them were designed to control lateral associations and flows of communication between denizens, while one frame opened to the centre, establishing a line of communication with it. As a strategy of normalization, the aim of the panoptic regimen was to constitute an organized body where all partial, confined, and limited subjective perceptions would be linked in perpetual communication with another, privileged, objective perception that encompassed them. Canada's first experiment in this new scientific management of criminals, Kingston Prison, built in 1835, reflected these latest penological reforms. Prisoners were not allowed to communicate with one another in any way. In the dining-hall they sat in rows with their backs to each other, and in the chapel the prisoners were arranged so that they could see the preacher but no one else. Prison rules forbade the inmates to "exchange looks, winks, to laugh, nod or otherwise gesticulate to each other." The success of the system was dependent on the "absolute prevention of intercourse among the convicts." If communication was necessary - in the workplace, for example - it could take

57 Three Social Apparatuses - and a Fourth? place only by means of signs or under direct monitoring by the guards. Any infraction was punished by a public flogging, six to twelve lashes public only in the sense that it was to be seen by the rest of the inmates. With so many silence rules, there were also many infractions, and the frequency of punishments escalated from the time the prison opened; there were 770 punishments in 1843, 2,102 in 1845, 3,445 in 1846, and 6,063 m l^47- One prisoner, during an eight-year stint in Kingston, had been whipped thirty-five times, receiving a total of i, 182 lashes (see Carrigan, 1991). The economy of operation of this surveillance apparatus rests on one-way flows of light and information; the visibility that flowed from the centre to the frames should retain only a virtual presence, one that would be invariant to the subject's point of view. "Bentham laid down the principle that power should be visible and unverifiable. Visible: the inmate will constantly have before his eyes the tall outline of the central tower from which he is spied upon. Unverifiable: the inmate must never know whether he is being looked at, at any one moment" (Foucault, 1979: 201). Unlike a theatrical space where there are as many scenes as points of view, and everyone sees a different play, the panoptic space was designed such that any point of view was subject to the same vision, much like the all-seeing eyes in the picture that appear to follow the observer about the room. The design was intended to provoke a permanent feeling of visibility and its accompanying effects, quite independent of any actual (and verifiable) seeing or spectacular application of power. These design mechanisms, incorporated to achieve one-way flows of light, were specialized shutter-screens in the frames of the watch-tower. These stylized interfaces permitted the guards in the tower to observe the inmates at any time, while the inmates were unable to see whether or not they were being seen at any one moment. For discipline to work, it has to see without being seen and hear without being heard.9 Bentham also planned to incorporate listening devices in the enclosure - namely plumbing them by means of tin tubes that would link centre and cell. But this plan was scrapped, since there was no available technology to establish telemetry and allow for the one-way flow of information necessary for the anonymous application of power. In the absence of the necessary sound interface, lateral communication had to be curbed by other, legislative or punitive means - flogging, in the Kingston case, or in Philadelphia's silent system, isolation for those who violated its hierarchical principle of speaking only when spoken to. The panoptic apparatus relies on images of movement as a means of inducing or producing self-consciousness (although perhaps a more

and 6,063 in 1847. One prisoner, during an eight-year stint in

58

Mediated Associations

apt term is camera- or surveillance-consciousness - that is, an awareness of the third eye and its anonymous gaze). In this case a virtual record-of-movement (like an omnipresent bug or listening device) acts on the actual and potential movements of those encompassed by the sphere and its one-way flows. The strategy of the panoptic machine is to turn the virtual image of movement into a force that will affect the actual and potential movements of bodies in its space (and beyond). Far from simply containing, capturing, or immobilizing bodies, the panoptic machine, as a form of disciplinary power, deploys a body's own capacities to move and to act both as a means and as an instrument of domination and subjectification. It transforms the social body into a body of knowledge. In this aspect the panoptic regime is a rough analogue of the sign regime. However, in the latter case excessive or redundant associations are stored in a body that is then used as a medium of circulation, while in the former, excessive or redundant information radiates and escapes the body. It is stored elsewhere, in virtual memory. Confronted with a mute, non-responsive overseer instead of an analyst who interprets, the subject does all the interpreting (Deleuze and Guattari, 1987: 138). Under the auspices of the panoptic regime, the body image - the image of a body capable of acting (an image of what a body can do) - is the infinitely renewable, material condition of this exercise of power. In this regard, power is "a set of actions on other actions" and so presupposes that "the other (the one over whom power is exercised) be thoroughly recognized and maintained to the very end as a person who acts" (Foucault, 1983: 220). There is, in all instances of power, an effect of one body on another. This mixture or intermingling of bodies produces bodily effects; some of these assemblages enhance strength or potential. Other bodily mixtures produce effects that result in decomposition. For example, the introduction of poison mixes with our bodies to produce the effect of disintegration. Likewise, discipline produces an effect of one body on another. It is not a direct, forceful encounter; rather, it is a form of controlled optical encounter between an anonymous any-body and another pre-assembled social body. Its affections, or mixtures of bodily effects, act on the potential actions of the assembled to create other orderly assemblages. Discipline is designed to produce deferred effects. It acts on future actions and on future states of the body - that is, on the potential actions of the body that it assembles. Discipline's affective associations are designed to affect what the body can do by transforming its own capacities and potentials into bodies of knowledge. By now, at this phase of our training, we are all aware of what these bodies can do.

59 Three Social Apparatuses - and a Fourth? THE ART OF A S S E M B L A G E

Discipline is not a grand scheme that has as its target the undifferentiated mass of the social body as a whole. Discipline works through division, by dividing and further subdividing space, time, and motion. Its optical machinery is micro/telescopic. It divides up space and movement into smaller and smaller fragments, subjecting each to intense and extensive scrutiny - that is, to the long exposure of the disciplinary gaze. It makes things that were not previously visible observable and measurable by dividing constellations and assemblages into innumerable points of illumination. These actions specify surveillance and make it functional. As Foucault puts it, discipline works on the little things and the small details. As a micro-strategy, discipline operates on a segmented social body, decomposing it into smaller and smaller fragments (a practice that also serves to multiply and disperse the number of sites for its implementation and exercise). But this mastery of the art of distribution is only part of the disciplinary strategy. While sovereign power (and the political economy of the mark) reached its limit in the spectacular theatrics of decomposition and dispersion, an act ending ultimately in death, disciplinary regimes aim to master the art of (re)assembly and rank, of arranging, ordering, and linking fragments of space, movement, and time, in a reconstituted image of the whole. This reconstituted whole parallels Durkheim's image of organic society and the specialized institutions that he describes - each with its own specialized functions to perform as determined by the presumed needs of the whole, each with its own highly specialized legislative and judicial apparatus, and each acting as a repository of the social body's collective force. To explain the necessity of these diffuse institutional practices, Durkheim borrowed a discourse of systemic needs that ended up serving as a form of teleological determinism instead of an explanation (see Turner and Maryanski, 1988: 112). Under the guise of an organic morality, dilettantes capable of entering into various assemblages would no longer be venerated, nor would those (undisciplined ones) who refused specialization. Discipline is a strategy of power-knowledge, programmed to constitute, illuminate, interrogate, and police the smallest movements in the spaces that it frames. It works by division and the ordering of divisions. Like sovereignty, discipline decomposes bodies, but its target is not the body of the condemned. Its target is the social body in its various associations and compositions, and its aim is to break up any inorganic and unauthorized assemblage (i.e., any dangerous mixture, contagion, mysterious association, or compact social grouping that remains

6o Mediated Associations outside the disciplinary field), to terminate any grouping that is not supervised by authority or arranged according to the dissymmetries of hierarchy (Foucault, 1979: 239), and to dissipate any such grouping into discrete and elementary singularities (individuals). Unlike the indiscriminate character of sovereign power, the discriminatory apparatus of the prison tailors punishments to the singularities constituted in its spaces. Rather than focusing on the body of the criminal or the act that constitutes the crime, disciplinary power is applied to the image of delinquency and to the mysterious ensemble of "lifestyle" associations that culminate in the delinquent act. As Foucault (1977: 252-5) reminds us, discipline punishes lifestyle practices and unauthorized associations, not crimes. More than just a building or an institution, discipline is a strategy of traces. Traces are the visible inscriptions or imprints left by the passage of a body. They are records of its style of operating or habits of moving and its possible directions. Traces are after-images of a body's "way." Since every body (human or social) has its own unique modus operandi, ambulatory style, and sense of direction, its traces are its signature the inscriptions that identify its individuality (see Deleuze, I9g2a: 5). Unlike the sovereign line offeree, which directly emblazons the body's surface, discipline's soft touch establishes a connection with a body image rather than a body-sign. It does this by inscribing (at least potentially) an image of a body's acts on a receptive surface. If every action or decisive movement is potentially inscribed and stored somewhere as an objectified trace or memory, and therefore potentially punishable at some point, by an anonymous someone, then it is the possible objective consequences, or rather the anticipation of these consequences projected on the screen of the imagination, that influence decisions to act in the present. Because punishment is only potential or imagined, no spectacular display of force has to reach the body. It does not act directly on the body or its meaning, but acts on its past and future actions or on what it has done and what it is capable of doing. By simply recording these signature traces, the disciplinary apparatus affirms the assignment of motion to these figures. It holds them accountable for their movements, both past and future, and evaluates their ability to hold these movements - that is, to display appropriate postures, to hold a pose, or to maintain their composure. Disciplinary figures are distinguished in terms of their form or composition - that is, comparatively, as a system of rank-ordered deviations from a pure form or an ideal state (i.e., a model or template10), either of body type (as in early criminological theories") or of bodily postures and attitudes that illustrate either good or bad form - that is, as good or bad copies of the ideal. Good copies are rewarded by changes in rank. Bad

61 Three Social Apparatuses - and a Fourth?

copies - those that deviate from the ideal - are set back to begin the process of training over again. In current practice, even the idea of enforced training or of imposing a reformed way of life on inmates has become largely obsolete. In Canada, under the "new opportunities" model (Ekstedt and Griffiths, 1988: 329), it is now up to the inmate to change his or her ways. A case-management team (guards, living-unit officers, parole officers, parole tribunals, and probation officials) subjects offenders to a battery of tests as it observes and assesses the way of life of offenders in

and outside the institutions to see whether the new self-inflicted pro-

grams of treatment have modified the offender's disposition. Failure at any point can lead to the offender's being sent back a level to begin again the process of classification and to work through the graduated hierarchy of punishments. Moreover, and more generally, the disciplinary system is discontinuous by the very nature of its exercise. Each disciplinary internment has its own plan, habits to be acquired, and forms of knowledge. From one system of internment to the next, from one apparatus to the next (from family, to school, to factory, barracks, or prison), from one rank in the internal hierarchy to the next, one begins again, from a zero-point, and ends with an examination designed to evaluate the body's disposition. Disciplinary power is predicated on what bodies can do, not on what they are or what they mean. If its other (its subjects) were predetermined in such a way as to be incapable of action or reaction - i.e., inhabiting a closed system and lacking a field of possibilities from which to actualize one or several possible reactions - then the conditions favourable to this power-relation would not be present. Power is exercised only over individuals or collective subjects faced with possibilities. The dungeon is a simple scheme of enclosure. Its thick, heavy walls are designed to hide or conceal as well as to prevent entering or leaving (primarily to secure bodies for trial and for punishment, not as a means of punishment). In contrast, Bentham's original design for back-lit cells shows conversely that disciplinary enclosures are full of "calculated openings." While each space of internment is a closed set, and while each framed space is potentially divisible into an infinite number of smaller subsets (a function of micro/telescopically "zooming in"), the system is not closed. Disciplinary enclosures, as with all systems, are never completely closed. In other words, "there is no relationship of power without the means of escape or possible flight" (Foucault, 1983: 225). These relative openings, or lines of (f) light, allow light to pass to the interior and converge on a central point that specifies the position of the eye-subject and the location of

6 2 Mediated Associations

the gaze (see Baudry, 1980: 27-8). It is an illusion to suppose that the disciplinary gaze emanates from someone within the enclosure. Rather, it is a centre where lines of light converge, where the actions or signature-traces of a body are (potentially) recorded. This receptive apparatus stands as the virtual repository of perceptions, observations, and information that any body can occupy. It is virtual because it is screened off from investigation: one should not know whether one is being seen (or recorded) at any given moment. The disciplinary gaze is anonymous. It is as nameless as the face behind the surveillance camera or its equivalent, the one-way mirrored glass, either of which functions as a repository of movement. In many ways, the functioning of the disciplinary apparatus surveyed to this point could be seen to parallel the functioning of the early cinematographe machines. For example, Edison's first machines were bulky, immobile, and confined to the studio. Like today's ubiquitous systems of surveillance (for example, video cameras positioned in shopping malls, parking lots, gaming houses, and corner stores) early cinematography comprised fixed, single-point-of-view, spatial shots, which allowed movements to remain the property of the figures in its frame (a central feature of the disciplinary strategy of acting on actions). Surveillance, as Bogard (1996: 77) asserts, "is a central, constitutive feature of all modern organizations. More individuals are targets of surveillance than in any other point in history." In what I would call a single-point-of-view cinema, characters and bodies change and exchange their relative positions, while surveillance cameras simply record their movements and spatial displacements. Consequently, cinema itself is seen to be but another extension of the disciplinary apparatus, reproducing its division of labour and redeploying its strategic aspects. It is also seen as a means of producing and training an army of surveillance satellites, thereby multiplying and further dispersing discipline's surveillance function throughout society (and fostered by policing programs such as Crime-Stoppers and Neighbourhood Watch). POST-DISCIPLINARY CINEMA

In the early days of Kingston Prison, the majority of sentences were short - from one to three years - during which time prisoners were prohibited from having visitors or contact with the outside. The only visitors to the institution were those curious individuals who, having paid an admission fee, came to take in the sights. A means of defraying some of the operating expenses (see Carrigan, 1991), this sport was open to the public six days a week from ten a.m. to noon, with a

63 Three Social Apparatuses - and a Fourth?

matinee from one to three. To the light side of the disciplinary apparatus and the self-consciousness of those all-too-public bodies caught in its orbit we must add an analysis of the dark side - that is, those random, anonymous any-bodies who operate the surveillance machine and populate its black hole (i.e., those in camera versus those perpetually on camera). By including the curious spectator and the spectator-effect in the analysis, the disciplinary apparatus not only produces delinquency, it is also said to produce the guardians of the panoptic order, its voyeurs. Presumably, like the delinquent, the shadowy figure of the voyeur is the spawn neither of nature nor of upbringing but is produced by an apparatus of power-knowledge and is therefore subject to it. Through its specialized interfaces, disciplinary apparatuses produce a hierarchy of vision and support a highly differentiated spectatorial division of labour. On one of its facets, there is the receptive surface that functions as a mnemonic recording device; on the other, the expressive surface that has nowhere to hide. In the case of Kingston Prison, these are the prisoners and the guards and voyeurs, respectively. Between the poles of reception and expression the interface controls the flow of light while producing and reproducing the dissymmetry, difference, and distance between the dyads seeing-seen, darkness-light, private-public, investigatorinvestigated, and voyeur-delinquent. One could make the argument that these divisions of labour remain in force in cinema, despite the technological changes that discharged the camera machines from the studio (the heavy houses of confinement) and allowed their inscribed information to be projected and rescreened in other houses of darkness and light. In what I call "disciplinary cinema," the emphasis turns, or rather returns, to the anonymous voyeur-anybodies who come to take in its sights - that is, the would-be guardians of the disciplinary city, or, as Denzin prefers to call them, "surveillance agents of the state." "This voyeur would gaze into the sacred, hidden places of society. In so doing, the voyeur would reproduce the concepts of public and private life which were central to the cultural logics of capitalism; that is capitalism and the liberal, democratic theory of the state required that a division between the public and the private be maintained" (Denzin, 1995: 102). This is perhaps one explanation for the spread of panoptic regimes: they are strategies that strike an accord with liberal discourses and supporter reproduce its optical distinctions. Denzin (1995), following Metz's (1982) history of literary figures, argues that the voyeur, cinema, and psychoanalysis all emerged at about the same time between 1900 and 1913 - and that cinema was made for private

64 Mediated Associations individuals (1982: 113). But by assuming that the twentieth-century cinema created this new social type (i.e., the voyeur, or Peeping Tom), Denzin misses both the history of the voyeurism and the apparatus that produces the voyeur-individual pair. In addition, and because of the technological interventions that subsequently set things in motion, the voyeur can no longer be considered in camera. As subjects of new regimes of power-knowledge, these "life-style consumers" (see Shields, iggaa) are no longer secure in their anonymity and place. Much like their delinquent counterparts, they too are surveyed, prodded, and wired by an array of surveillance techniques and machines in order to test and measure their receptivity. They have also been carved up, ordered, and ranked into communities of taste and preference by an army of marketeers and analysts (see Shields, iggaa: 14-17; Denzin, 1995: 117). Is the distinction between the poles not as clear as it was once thought to be? Who is the object-subject of the gaze? Is it possible to reach such a point of identification, or is everybody simply caught in the chimaerical, circular, Benthamite light trap (another form of carceral chain)? There are after all, if only hypothetically, some points in space where gravity is so intense that it forces rays of light to form a circle (according to relativity theory). Anyone placed within this closed circuit, and who would look straight ahead, would see another, who is actually himself or herself from behind. While the illusion of the other remains, the actual object of investigation disappears as something other, and one ends up only observing, and therefore policing, one's own self/image. But to collapse visibility, as well as the concepts of spectator and voyeur, into one body not only eliminates the power relation as a social relation, it also homogenizes the divisions and distinctions that discipline works so hard to produce and sustain. ENGENDERING THE

DISCIPLINARY

D I V I S I O N OF LABOUR

If discipline is about policing unsupervised mixtures and producing new ones, through division, hierarchy, and cellular organization, it not only evokes the seeing-being seen distinction, in a parallel way it also produces and reproduces other social distinctions as well. Mulvey's (1989: 19) analysis of dissymmetries in visibility shows the gaze structure to be a monolithic apparatus saturated with patriarchal values. In Feminism and Film, Humm (1997: 69) has reiterated this claim, suggesting that the panoptic gaze is also a masculine gaze, regulated by the division of the visual field into active subject positions that see without being seen and passive object positions constituted as other.

65 Three Social Apparatuses - and a Fourth?

Because the subject position is masculine, the gaze structure produces and reproduces gender differences in a hierarchical system of unequal exchanges where one party has the privilege of invisibility in relation to the image or spectacle (Stacey, 1994: 21). This difference/distance is the source of voyeuristic pleasure in relation to the image - that is, the pleasure of power that accompanies the controlling look (Kuhn, 1985: 28). Relative to the masculine gaze, the object of the gaze, the image, is positioned both as feminine and as an object of masculine viewing pleasure. It is possible to distinguish two approaches or levels of analysis here - one that deals with images of women and another that deals with women as image (Stacey, 1991: 142), although it is perhaps more fruitful to integrate the perspectives around issues of spectatorship. Distinguishing between the gaze as a structure that objectifies and subordinates the feminine and the look of the feminine object on the screen or in the photo, Kuhn (1985) offers the following synthesis. If the gaze is designed primarily to produce a feeling of self-consciousness - that is, a perpetual awareness of one's to-be-looked-at-ness (Kuhn, 1985: 31) - then a corollary is found in the images of women in "pornographic" photos (i.e., images with a gendered gaze structure). Pornographic images of women typically show them with an averted gaze, indicating an apparent unawareness of their looked-at-ness, or, conversely, the gaze is returned in the form of a "come on" look that indicates a kind of pleasure in being seen. In both cases, the face and its look are key signifiers. For the woman spectator, such images show either that there are no places or moments where women can experience private pleasure without at the same being aware of their potential for being seen (adding a new dimension to privacy) or that women take pleasure in their subordination and identify with their self-conscious positions as exhibitionists. Humm suggests that because the gaze is masculine and women are positioned as objects, both pornographic and cinematic images assume the spectator to be masculine and encourage his voyeurism (1997: 39). She also shows how the female body is refigured in contemporary cultures and draws our attention to the ways in which women's bodies are feared, pathologized, or re-created in films such as Alien, Carrie, and Dead Ringers, respectively (1997: 58). In the relation of masculine seers and feminine objects, it is difficult to conceive of feminine subject positions in relation to these structured conditions of vision. Consequently, Smelik (1998: 29) suggests that female spectators have to adopt masculine-voyeuristic, masochistic-exhibitionist, marginal, or non-subject positions. Doane, in contrast, contends that women cannot assume masculine-voyeuristic positions, for they can

66 Mediated Associations

never achieve the requisite distance from the image required for voyerism (1982: 78-80). Because women are the image (Doane, 1982: 80) and never own the gaze, they must adopt the masochistic positions that the gaze structure demands of them. Constituted and overdetermined by their subordinate to-be-Iooked-at-ness, female spectators' only avenue is identification with the image of women. This is not to suggest that female spectators constitute a homogeneous mass, however. While they may not reach the distance and detachment to address the image from the voyeuristic point of view, female spectators can, according to Stacey (1994), attain various degrees of distance from the image and therefore produce various forms of identification. These forms of identification range from devotion/worship (a distance-produced fascination or appreciation from afar that reinforces the otherness of the image and affirms an immutable boundary between self and the image ideal while nevertheless allowing for the recognition of certain admirable qualities and powers), through escapism (away of merging with the image-ideal to countermand temporarily the "drudgery of domesticity and motherhood"), to extracinematic practices (imitating or copying the behaviour, attributes, or qualities of images of women through partial recognition of common properties or qualities and their fragmented replication in imaginary games or through actual appearance and practice). On one pole, identification is the pleasure of passionate detachment, on the other, a closeness or proximity to the image that overcomes its distance by various means. This spectator relation is a power relationship because the apparatus keeps open these spaces of possibility and maintains the choice of doing or seeing things differently. The potentiality of choice is what power presupposes as the material basis of its exercise. In this case, the otherness of the other in the image (its qualities and powers) and the possibility of moving (empathically or practically) towards them and potentially becoming other constitute calculated openings and lines of flight that seduce the look and trap it. Variable identificatory practices maintain spectators as subjects who act and move while subjecting them to a disciplinary structure that enforces a gendered spectatorial division of labour that subordinates the feminine positions to a masculine gaze. If the gaze is taken as a metaphor for the affect - that is, for the effect that a body has on the state and actions of another - then the effects of these composition and optico-spatial associations are indeed deleterious. They can instill a perpetual state of self-consciousness and can embarrass, humiliate, and shame the object; they produce fear, desire for, or hate in, difference; they are sadistic, as they create a fetish-object; they intrude into the object's privacy, body, or social

67 Three Social Apparatuses - and a Fourth?

body; they assault the object in the dark, stab it in the back, brand it as delinquent, define it as mentally ill, hysterical, or exhibitionistic, and punish it for its sexuality. In the end one might wish for a world without mixtures, where there were no others, or perhaps this too is part of the disciplinary strategy. The power of the apparatus, whether as prison, pornography, or disciplinary cinema, is predicated on the dialectical interplay of distinctions between inside and outside, self and other. Between the poles of voyeur and object there is always a specialized interface designed to control the flow. The prison has its shutter screens, although now, as generalized surveillance, a panoply of other sensory devices serves much the same purposes, pornography has its photo-de-pose, and the cinema has its viewing gallery. But discipline involves more than just reifying differences and maintaining distinctions; as a process it always maintains the possibility of graduating or of moving between the poles of being and becoming other. The interface is not an immutable barrier and is much more than a simply reflecting surface, silverscreen, or mirror that does little more than reflect back an image of self. Interfaces are permeable membranes and media of passage. Between and beyond the "spaces of places" of disciplinary enclosures, there are also "spaces of flow," to borrow Arrighi's (1994: 23) distinction. Between any series of enclosures, no matter how close together, there will be gaps, intervals, marginal sites, places of both lateral and dyssymmetrical encounters and mixtures. It is here that discipline places its interfaces to structure the character and qualitites of these associations. Through its framing mechanisms, discipline not only establishes zones of visibility and visible bodies, it also produces zones of invisibility and anonymous bodies and establishes relations between that which it has parted. The intermediary zone and its control mechanisms modulate the kind of mixtures and associations diat can form. Discipline transforms the primary affect and its mysterious associations into a second-order association of power and knowledge. By dividing the visual field into aspects across this mediated dimension, it produces an intermediary effect - namely, leaving traces (rather than marks) of a punitive power in its object bodies so as to affect their future states and their potential actions and to predispose them to think about the possible objective consequences of any act. Through occupation of the interstices, discipline is able to exercise a kind of foreign policy to govern the mixtures and associations of bodies as they move between states. If discipline is successful, every disposition should be accompanied by myriad possible objective consequences - that is, by things that could happen or things that could happen again. In essence, every act becomes an act of potential

68

Mediated Associations

association with an anonymous anybody. It brings with it an awareness, a kind of foreknowledge, of the power of this other and what this other can do. The production of this anticipatory effect in disciplinary practice is comparable to the kind of power-knowledge system at work in the cinematic apparatus. While the analysis of disciplinary cinema takes us part way in establishing the terms of comparison between these two systems, to understand fully how power is exercised in the cinema apparatus we must address the distinctive characteristic and qualities of the cinematic interface and the mechanisms of its deployment - namely, montage. It is my contention that the cinematic constitutes an interface as a space of flow that is not reducible to the prison or pornography, nor does it produce the same kind of gaze structure inherent in the latter. The prison and pornography work through the deployment of an anonymous gaze, through the camera eye and its point-of-view (p.o.v.) ,12 whose relatively inert structure is designed to objectify and produce self-conscious subjects. Like discipline, the cinema apparatus deploys its specialized interface and gaze structure by designing the visual field to produce power relations. But cinema's structural design and method of assembling images produce a different kind of gaze structure, as they open up other spaces of possibility and other spaces of control. Kuhn (1985) has already identified the key structure in the power of this image - the face and its look. When the face is deployed strategically in the ensemble of moving images that define cinema, it takes on other qualities and powers. The face defines both the cinematic interface and another, distinctive, cinematic gaze. In social apparatuses, light pours through openings, is projected across gaps, and is distributed across surfaces. Light encounters various impediments, obstacles, and traps in the field of visibility, which produces perceptions, reflections, and images. Fields of visibility are shot through with lines of light. Signs are also fashioned. They are emblazoned or projected on surfaces. They are captured, ordered into formation, and redistributed in networks and circuits. The conditions of the sayable or expressible are structured by sign regimes of enunciation. Social apparatuses are also characterized bylines of force that pass between statements and visibilities, between moving bodies and bodies of words, establishing correspondences here, producing dissonance there. Social apparatuses are dynamic systems, not inert structures. They are systems of mobility and have various deployments. They are regimes of light, enunciation, and force, but they are also regimes of flow. Flow is perhaps the most cinematic of the analytical dimension of the social apparatus.

6g Three Social Apparatuses - and a Fourth?

Flow is a dimension of movement that passes between interior and exterior spaces, sometimes following the normal pathways and migration routes, sometimes going beyond the norm and the convention. Disciplinary apparatuses, as relative enclosures, are full of holes that allow lines of light from the exterior to pass and at the same time to fall strategically on certain points. These lines of light are also potential lines of flight, a means of escaping the enclosure — the condition that also makes possible the exercise of disciplinary power (O'Connor, 1997). Calculated openings are designed to structure the flow and distribution of bodies and light, while its panoptic centre serves as an apparatus for potentially capturing and recording these flows. This double strategy of calculated openings and mechanisms of capture ensures that the movements of bodies, light, and information are directed and channelled in one way. Whether we speak of the movements of images or signs, information or light, bodies or energy, or even energic bodies (agents collective or singular) with a rich capacity for movement and potential for change (or making a difference), all these flows of matter are caught up in the complex circuits of social apparatuses designed to control them, presumably without diminishing their capacity. What I propose in the following chapters is an analysis of cinema that extends disciplinary structures and goes beyond its spectatorial division of labour. Mobility is the key to the analysis of post-disciplinary cinema, as I prefer to call it. These new forms of mobility change again the conditions of spectatorship, the image, and the power relation entailed therein. As Virilio has pointed out on numerous occasions, ours is the society of motoricity, of acceleration and change, and of accelerated change. But it is also a visual culture, and what moves faster than light? What we need to do is to bring our analysis of the apparatus up to (its) speed. This movement towards mobility and its conceptualization begins, though does not end, with the Lumiere brothers' invention, unveiled on 22 March 1895. Unlike Edison's stationary camera, the Lumieres' cinematographe machine was lightweight and mobile. It liberated them from the studio and changed the functioning of the visual apparatus. It allowed them to take to the open space of the Parisian streets and to follow, rather than simply capture, its mundane movements and flows. By this movement of deterritorialization - that is, by following a flow or a line of flight - they broke the entire panoptic mechanism and transformed it. They quickly replaced fixed, spatial shots that allowed movement to remain a property of the figures in the frame - first, by moving the camera, by panning, zooming, and altering the depth of

yo Mediated Associations

field; and second, by stealing motion from other bodies, by mounting cameras on transportation devices or other vehicular contrivances. Not only had the Lumieres found an effective way to transport the volatile plastic film past the shutter screen (i.e., Lumieres' claws), they also made the whole mechanism transportable, licensing the camera to become a roving eye. After the initial public success of their invention and its display at the Paris Exhibition of 1900, the Lumieres turned their attention to the more practical demands of its manufacture and sale. It was left to others to exploit its potential for mobilizing bodies and their relations. Mobility recast the problem of spectatorship. In addition to the various modes of identification that are made possible through association with the image (as discussed above), the moving camera opens the possibility of other, distant forms of identification. Benjamin explains: The performance of the movie actor is transmitted to the public by means of an array of technical instruments, with a twofold consequence. The camera that presents the performance of the film actor to the public need not respect the performance as an integral whole. Guided by the camera [person], the camera continually changes its position with respect to the performance. The sequence of positional views which the editor composes from the material supplied [him or her] constitutes the completed film ... Also, the film actor lacks the opportunity of the stage actor to adjust to the audience during the performance, since [he or she] does not present [his or her] performance to the audience in person. This permits the audience to take the position of the critic, without experiencing any personal contact with the actor. The audience's identification with the actor is really an identification with the camera (1968: 228).

Comparable to Denzin's (1995) discussion of the voyeur gaze as a mode of identification with the movements of the camera, this perspective aims to take account of other forms of movement in the cinematic apparatus. These movements are further differentiated from the movements that take place in the frame. Moving camera positions create different points of view and change the frame of reference. Complied as an assemblage of camera movements and different points of view, the processes of cinematic production foster the further differentiation of seer and scene by creating positions of displacement or detachment. The objectivity of the film critic, the detachment of the analyst, and the voyeur who hides in the shadows are disciplinary figures that occupy strategic positions in discipline's optical division of labour.13 There is, however, another aspect of movement that is indi-

71 Three Social Apparatuses - and a Fourth? cated, but not thoroughly addressed, in Jameson's perspective on cinema. The aspect of movement that I refer to here is a product of the processes of cinematic production and presentation. It is a form of motion and an aspect of change produced within, and by means of, the moving assemblage of shots that we call "cinema." The capacity of cinema to record the motion of bodies and produce images of motion that are external to it is one aspect of cinematic production. Another aspect is the capacity to move the camera, to change its position, to steal the motion of other bodies, as well as its capacity to pan, zoom, alter the depths of the visual field. But I contend that the primary and distinctive quality of cinematic motion is its capacity to move the point of view, to direct the movements of change and transition from one point of view to another, to create and occupy the intervals of motion by means of cinematic interfaces and thereby to control the flow. The primary method of creating a uniquely cinematic movement is assemblage, or montage - the selection and conjunction of separate shots, most of which can be fixed and spatial (i.e., with very little camera movement). This cinematic method liberates movement from the frames that used to contain the image or the images of motion, and at the same time it liberates perception from the privileged vantage points that confined and naturalized it. Montage gives us glimpses of time freed from the fixity of place and space, as a process of passing and passage that moves through intervals and between places, spaces, and vantage points. These moments express duration as a qualitative aspect of motion and change. They help us understand the processes of change and transformation, both within and beyond cinema. It is this perspective on cinematic movement that I think fruitful for explaining transitory and mobile forms of cinematic associations. One of the key, often under-theorized, aspects of cinema is that it does not require additional motion or motivation for its images to make sense. The material of cinematography is neither textual nor narrative but consists of mobile sections or shots that already express relations of change and transformation. These shots are not stable surfaces (like the painting or the still) for reflection or interpretation. To reduce the shot to the still is to eliminate what is specifically cinemagraphic. A method for studying cinema must go beyond the analysis of captured figures (and their redeployment as linguistic figures and tropes) and the idea of frames with impermeable boundaries. The work of cinematography is not narration but rather montage, which selects and assembles mobile sections to constitute, albeit indirectly, an image of time that passes through each shot and each frame. To conceive cinematic motion it is necessary to consider boundaries as a fluid medium, or interface, rather than as empty spaces between fixed states, and

72 Mediated Associations therefore to view the processes of movement and change as more than simply the shock of successive incarcerations. The cinematic apparatus came into existence as a different kind of public spectacle with a new face and a new look. It is an apparatus with a distinctive gaze, qualitatively different from discipline's. This gaze opens up new dissymmetries in the field of visibility and produces a different spectatorial distinction. Its specificity lies in its ability to incorporate the distinction between seeing and foreseeing in a continuous lateral flow of moving images stitched together to form the assemblages of montage.

It was partly thanks to information provided by the Entrepreneur, the first battlefield observation balloon, that General Jourdan won the victory of Fleurus in 1794. In 1858 Nadar took his first pictures from a balloon. During the American Civil War, the Union forces equipped balloons with an aerial-mapping telegraph. Soon the army was rigging together the most varied combinations: camera-kites, camera-pigeons and camera-balloons predated the intensive use of chronophotography and cinematography on board small reconnaissance aircraft (several million prints were made during the First World War). By 1967 the US Air Force had the whole of South-East Asia covered, and pilotless aircrafts would fly over Laos and send their data back to IBM centres in Thailand or South Vietnam. Direct vision was now a tiling of the past: in the space of a hundred and fifty years, the target area had become a cinema location, the battlefield a film set. (Virilio, 1989: 11)

3 Cinema's Optics

There are various ways of looking at cinema. We can see it as a kind of sovereign spectacle for framing and limiting meaning, perhaps a signregime to deter everyday associations, or even a disciplinary apparatus for producing a relation of voyeurs and self-conscious bodies. Sovereignty fixes the frame of reference and smooths out the screen, the sign regime adds motivation to (overcodes) captured motion and redirects it, while the disciplinary apparatus breaks up motion into elementary units and, in the silent observatories, subjects them to a permanent optical test. Baudrillard, perhaps inclined to see discipline as the dominant form of control in modern society, concludes that "everywhere the test [referendum] functions as the fundamental form of control, by means of the infinite divisibility of practices and responses" (igSgb: 116). Cinema has a whole arsenal of projective, receptive, and reflective mechanisms at its disposal. It has incorporated not only the techniques of the frame and screen. It also has incorporated strategies for deframing, for refraining, and for mobilizing the frame of reference. But reducing cinema to what goes on in the frame tends to arrest its motion and, in so doing, to reduce its images to ready-made posters or pictures, figures or stills, even empty forms constituting the captured material of mythological interpretation. Dividing up the motion into its cells or frames produces a similar problem - that of partial-fetishobjects torn from their place in the whole (like uprooted signs or

74 Mediated Associations non-functional antique-objects) with the necessity of positing an external collection agent responsible for reassembling the pieces (the collector of antiquities). These perspectives reduce cinematic images to analytical fragments (textual or otherwise) that frame the content of what can be analysed or become arrested forms for other interpretations. In a medium such as film the process that underscores the selection and assemblage of visual sections is montage, as opposed to analogy (the favoured analytic of mythologists, as we saw in Barthes's analysis). Montage is a process that is neither identical to the semiotics of the script nor reducible to it. If there was an overarching semiotic governing both the narrative and montage, forging their identity, then the overall composition would have to be read; it would have to have movement added to it. Such an analysis forgets that the cinematographic image is already in motion and that this is its distinguishing feature. Alternatively, one could imagine a double semiotic, one coding the montage associations and another coding the semiotics of the narrative. This hypothesis would underscore the possibility of the audio and visual entering into a variety of forms of contrapuntal, polyphonic relations within the same composition and would preserve their independence. But the hypothesis explored in this chapter is that the relationship between montage and the script is not equivalence within one semiotic regime or a relation among many regimes. Instead, I argue that the image-sign, montage-narrative relation is equivalent to the apparatuses in which relations between non-discursive and discursive fields are either forged or found lacking. Under this hypothesis, we are no longer in the semiotic domain of the image-sign equivalence established by Metz (1974). We enter instead into the domain where this dependence is reversed, where images and their relations are treated like mixtures that form by the relations and associations of bodies and where discourses and signs are the products of these substantive encounters. As Jameson aptly points out: "movies are physical experiences, and are remembered as such ... film is an addiction that leaves its traces in the body self (1990: 1-2). It may seem odd to equate images and bodies, given that conventions of phenomenology and existentialism have determined images to be simply figures of consciousness. Because these figures lack extension, they are unable to inhabit or occupy space the way bodies do. As merely intentional phenomena, images also suffer from a fundamental poverty of relations. Although one can, for example, recall (presentify) a subjective image of the Parthenon, it is not possible to count the number of columns in the image as one could when in its pres-

75 Cinema's Optics

ence. Bodies, such as the Parthenon, are characterized by an infinite number of relations with other things and therefore constantly overflow the frames and positions of consciousness and go beyond its horizons of awareness. Owing to the positional or selective character of consciousness (its intentional poses), its figures possess only a finite number of relations with other things that act on it and to which it reacts. For Sartre, images are identical with our subjective consciousness of them (1991: 20). The poverty of images is a consequence of the fact that they are always partial and selective. They are restricted to confined points of view or limited frames of reference and to the particular subjective interests, aims, goals, and purposes that correspond to these points of view. But in relation to selective perception and its point of view, there are always virtual perceptions - that is, other poses, or points of view, that one could adopt and from which vantage point one could see other aspects of the phenomena under investigation (for example, reverse sides and hidden columns). These other aspects parallel the halo of virtualities (possible concepts) that surround the sign in Barthes's analysis or what is seen by the virtual observers hiding in Bentham's panoptic machine. What renders this extended perception distinct from the mere subjective image is its structure of alterity - that is, the possibility of others (other bodies, seeing other aspects) populating the horizon of the seen, beyond the frame of a given experience. According to Boundas (1993), this structure of alterity (not simply the presence of empirical others) is already implicit in the phenomena of perception. The potential for movement, for changing positions, and for the difference that this makes for our perceptions is already an aspect of perception. This is perception's social character. The potential to see, from another's vantage point, what lies beyond or outside our own is a dimension open to us as a consequence of living with others, but it is also a structure of our perceptual field. The awareness of depth, or the awareness that there might be other columns to count, is itself a function of this structure. Alterity, the other side of the image, and the possibility of joining and mixing with the others already there, either in reality or in the imagination, turn our partial-subjective images into cinematic perceptions, or perception-images, as Deleuze (1991) calls them. Without this structure of alterity, Boundas argues, consciousness would attain only partial, fragmentary, and incomplete images. Alterity adds depth and dimension (or extension) to the image (even on a flat screen). The ability to link up with these virtual others helps smooth the transition from one point of view to another. It establishes both the sense

76 Mediated Associations of perception and the potential for images that are more than partial. This potential is an image of the whole. Like the view from a prison cell, the difference between the image and perception (part and potential whole) is dependent on a conception of confinement that fixes the frame of reference and encloses the point of view. The fixed frame of reference is also responsible for capturing the interest: the partial, subjective, and dubious image (of Descartes or Husserl) is essentially a reflection of confinement. Movement, in contrast, produces an image in extension. Perceptions of depth and dimension are dependent on mobility, on the possibility of changing one's position (vis-a-vis an immobile figure), or on moving from one vantage point to another, passing through thresholds and assembling these changing frames. Conceptions of space are intricately intertwined with movement. With acceleration, the spatial distinctions of proximity, of near and far, and of present and absent become unsettled. The same can be said of the categories of the actual and of the imaginary, since these differing conceptions of being-inthe-presence-of-something also depend on movement or its lack. The other way to constitute a perception would be to have the image itself move and constantly change its frame. To move the image is to create a kind of virtual perception that undoes the image-perception opposition. Mobility is the key to establishing this virtual whole and for forming the basis of a kind of experience that is de-subjectified, dis-positional, and disinterested. "Fascination," the term that Baudrillard (igSgb: 62) uses to describe such experiences, is unfortunate, since it implies capturing, freezing, or deterring motion (64), as in the state of being spellbound. "Abandonment" is perhaps more accurate, for it connotes fleeing one's position, the nomadic movements of deterritorialization, vagabondage, and itinerancy. The image of the whole, as a potential, is identical with movement, with possibility of mobility, and with the openings that permit us to take flight. As such, the whole can maintain only a virtual existence. It cannot be given in the immediate or monumental sense. It cannot be contained in a frame. To attempt to do so is to change it qualitatively. To take the whole into fragments, to immobilize, isolate, or otherwise fix it in place is to produce a fetishized partial object (or an equivalent linguistic figure - a synecdoche). Since the whole lacks the particular goals, intentions, or purposes (direction) associated with the one, it can never be found in the practical objects of utility or even in a collection of objects (on this point Marx is instructive; the whole is a social relation, not a thing). While confinement establishes figures and poses that capture and frame up the imagination, the mobilized image opens up the potential to

77 Cinema's Optics explore the regions beyond the territorial limits, frames, boundaries, and divisions that constitute subjectivity. As Grosz writes, "the boundary between inside and outside, just as much between self and other and subject and object, must not be regarded as a limit to be transgressed so much as a boundary to be traversed. ... Boundaries do not so much define routes of passage; it is movement that defines and constitutes boundaries. These boundaries, consequently, are more porous and less fixed and rigid than is commonly understood, for there is already an infection by one side of the border of the other, there is a becoming otherwise of each term thus bounded" (1995: 131)Perception is essentially a form of encounter with Otherness and therefore contains a germ of difference in itself. Perception is not subjective and partial without at the same time possessing a virtual (nonsubjective) element that creates our awareness and conception of the whole. The subjective and objective do not constitute a duality of privileged, fixed centres of determination; they are polar elements necessary for perception. The domain of cinematic perception lies somewhere between the rigid subjective and the objective poles of perception, in the interval or the interface across which the poles communicate and by means of which a number of possible forms of association can occur. According to Deleuze (1991), it is this conception of cinematic perception (as a relation between actual and virtual, part and whole) that forms the basis of montage as it operates in cinema. To perceive cinematically we have to change the way we understand the frame. STRUCTURED ENCLOSURES AND CALCULATED OPENINGS

Architecture was the first art of the frame and of the inter-assemblage of frames and planes, designed to carve out territories and constitute places of dwelling and habitats. Art begins "with the house. That is why architecture is the first of all arts. ... It can be defined by the frame, by the interlocking of differently oriented frames, which will be imposed on all arts, from painting to the cinema" (Deleuze and Guattari, 1994: 186). The frame of the picture, of the pose or the ready-made figure, was instituted to delimit the identity of a given experience. The picture frame functions as a boundary, not only to "exclude all that surrounds it" (Simmel, 1994^) and to avoid intermingling or "spillage" between the inside and outside (Bauman, 1988) - much like the impermeable framework of the early jails, whose function was to code and secure the binary opposition of inside-outside - but also to provide viewers with a frame of reference,

78

Mediated Associations

to help them maintain their distance, and to prevent excessive excitation, which might extend beyond the frame into stimulus-evoking (bodily or sexual) responses. But architectures also include other kinds of frames, such as windows and doors. These frames are more like thresholds - points of interchange between inside and out or interfaces between the limits of enclosure and what lies beyond. As Simmel suggests, "life flows forth out of the door from the limitation of isolated separate existence into the limitless of all possible directions" (igg4a: 8). While the physical barriers of walls portion out a mute cellular or insular existence, "the door speaks" (Simmel, igg4a: 7). The door is a kind of frame whose significance lies not only somewhere between the poles of limitation and freedom but at the "threshold of representation and communication" (Minh-ha, iggg: in). Whether as architecture, art, or cinema, boundaries and structures are shot through with a variety of openings, thresholds, and gaps, through which the inside extends outward and by which the outside or alterity enters. The interval or margin is an open space of mixture, a place of hybrids, composites, and assemblages, where new bodies are created and new wholes are constituted. In cinema, the frame is a space that produces a variety of cinematic composites (cinematic perceptions, affections, and actions). As in modern architecture, the idea of calculated openings is an invaluable analytical device in understanding modern exercises of power. These are openings designed to regulate mixtures and control flows, to legislate and co-ordinate the movement of bodies, to create dissymmetry in the flows of light and information, to instill habits, to mark out the stages in the hierarchical evolution of moral subjectivity, and to produce organized bodies. But since power is a relation and is exercised only against recalcitrance, there is an inherent contradiction between spatial and temporal relations - that is, between relations that work to regulate movement and control flows across openings and those that work to disturb the flow and to interrupt the normal motion of bodies and things by occupying these openings. While institutional strategies concerned themselves with containment, with the mise-en-scene (movement in the set or what goes on inside the frame) and with exercising power over that which can be interiorized, internalized, or appropriated locally, cinema strategies deal with montage, or controlling what comes next - that is, with the osmotic relation between the frame, the shot, and what is beyond it (its outside), or with what takes place between successive framings and shots (in between in the interval or gap). Even if one were to stop the cinema apparatus, and analyse the still thus isolated, one would invari-

79 Cinema's Optics ably come across certain remarkable instances to contemplate - for example, "when the horse has one hoof on the ground, then three, two, one" (Deleuze, 1991: 5). But these remarkable singularities are not forms actualizing or embodying an eternal standard or model, nor do they await a narrative to order their arrangement or distribution (cf. Metz, 1974). The only really remarkable thing about them is that they are equally close to, and therefore equally distant from, other equally remarkable instances that are properties of motion. The analysis of montage begins with acts of deframing that constitutes cinema's extraordinary interface. This act of deframing unsettles the distinctions of close and distant as it breaks the system of co-ordinates typically used to pinpoint placements and determine precise locations. In so doing, montage sets up the cinematic interface in an indeterminate field. This field is a qualitative location. It is a space of potentials and qualities, of qualitative changes. In this way the interface produces a particular kind of cinematic activity - namely, qualitative movements. The singularities of cinema have little in common with the longexposure optics of discipline or with its spaces of confinement. Under systems of discipline, dissaggregated and excommunicated singularities relate to a fixed centre of determination. Discipline's cellular plan provides a clear-cut mechanism for determining the exact placement, position, or location of each part in the disciplinary array. This opticospatial arrangement forms a co-ordinate system (a co-ordinated space of visibilities), which in turn supports the interminable body-count and attendance-taking rituals for establishing presences and absences in the visual field. Cinema singularities are singular points of a movement that belong to movement and are immanent in it. For no matter how remarkable, unique, interesting, or ordinary from the point of view of their immobilization, each instant is simply an any-instant-whatever, whose only distinguishing feature is its equidistance from all the other instants in the moving duration of the shot. And every shot is a qualitative multiplicity of these any-instants-whatever. When approached as a mobile section, each frame is transformed from an enclosure that captures movement into a moment in passing. To reduce the shot to the still would indeed be pornographic (or photographic) and would miss movement at every point. For every frame communicates with an outside, just as every closed set refers to an out-of-field, or larger, unseen (virtual) set that encompasses or modifies it. For example, sound in the out-of-field can extend the scene beyond the visible frame. The sound of traffic or sirens emanating from beyond the frame can invoke the feeling of a larger urban setting, effectively displacing, qualifying, and reterritorializing the scene instantly. The

8o Mediated Associations out-of-field has transformative powers over the visual field. Consider how the haunting musical score in Jaws changes the intensity of what would otherwise be banal scenes. This is also why content analysis can never reach completion; the idea of a completely closed set with an impermeable boundary that could provide consciousness with a stable frame of reference is erroneous. The whole is never simply given all at once. There are always breaks in the presumed continuity of the enclosure; the outside penetrates and the inside escapes through its porous openings in a process of continual change and transformation. From my perspective, motion in the cinema apparatus comes in three discrete forms, with each form reflecting a different level of analysis. First, there are images of motion, or what goes on inside the frame (or shot), as bodies and things change and exchange their relative positions in the frame or set. There is no cutting at this level of analysis. Second, there are movements of the camera that change perceptions of the set and continually modify the frame of reference. There is no analysis of transitional moments at this level of analysis. And third, there are cinematic movements that are a consequence of relating movements of bodies or the camera to the intervals of motion that pass through frames and between the shots. Instead of simply positing montage as the shock of one image following another, or as a collection of images requiring external links added by the script or narrative, the special case of cinematic movement presents images acting on and reacting to one another independent of the script or narrative. If an image were to extend into the space beyond its enclosure, to escape the containment of its frame, and to begin to act on other images (or bodies in space), or if the mirror image of movement were suddenly to begin to move on its own account, then we would have to reconsider a whole set of taken-for-granted distinctions between the interiority, poverty, and immobility of images and the externality, richness, and motive force of the real. DIMENSIONS OF ANTICIPATION

Cinema is an example of what Virilio calls an "open system" (1995: 125), where the whole is never fixed or given all at once, but is itself variable. In an open system a part can belong to various relative sets simultaneously. By refusing settlement in any one place (set, setting, or situation), such nomadic singularities always transform their places of spaces into in-between or threshold spaces (see Grossberg, 1996: 180). By subjecting each place or locale to a relentless process of opening or deterritorialization, space no longer functions as a fixed

8i

Cinema's Optics

system of co-ordinates to define, position, or delimit movements. Instead, space itself becomes a variable of movement whose "boundaries are only produced and set in the process of passage" (Massumi, in Grosz, 1995: 131). The dualities of space, or inside-outside, close-distant, and present-absent, and the related categories of past and future, are "categories of absence" (see Shields, iggsb: 187). One could include desire among the categories of absence, since desire is typically defined as lack, by a lack of being or a lack of presence, which implies distance. These dualities are derived from the relative safety of the enclosure and the point of view of confinement. Mobility changes all that. Media mobility, as Virilio argues, has the effect of changing old notions of proximity and distance, along with the classic distinction between inside and outside (1995: 106). Such notions bring us very close to Baudrillard's image of a collapsed universe, to the implosion of social and spatial dimensions, and to the field of pornographic visibility that it defines. In this field (of simulacra) everything appears close up, all too visible, or given all at once. It is a "state of pure presence ... without even the faintest glimmer of a possible absence," where there is no longer even anything to see because everything is already there. For Baudrillard, the overall effect of "traversing all horizons," limits, and boundaries is that "others have virtually ceased to exist" (1988:32, 41). From this perspective, simulacra are ominous signs for the social; they are symptoms of its collapse. "The social void is scattered with interstitial objects and crystalline clusters which spin around and coalesce in a cerebral chiaroscuro. So is the mass, an in vacua aggregation of individual particles, refuse of the social and of media impulses; an opaque nebula whose growing density absorbs all the surrounding energy and light rays to collapse finally under its own weight. A black hole which engulfs the social" (Baudrillard, ig83a: 3-4). Baudrillard's perspective is very visual. Like discipline, this theoretical system has a light trap at its centre. It presents an image of a closed universe where nothing escapes the displacement of exposure. It renders the social as a cluster of visual sign-posts and fetishized-partial-objects that swirl around in decaying orbit. Far from being simply a closed universe, or an apparatus for staging the visible, cinema keeps open the possibility of deterritorializing and reterritorializing the seen/scene. It is montage that mobilizes the movements that threaten to leave behind the territorial principle. Cinema is, after all, a moving scene/seen, not just a surveillance apparatus for capturing movements (or their reflections) in a pre-defined scene or ready-made enclosure. Because shots are

82

Mediated Associations

selected and assembled (independent of the fact that each shot is already a mobile section), montage introduces openings into filmed motion (in the interval between mobile sections) and invents new techniques and strategies for dealing with the movements that pass in the interval. The optical machinery of cinema therefore functions differently from both sovereignty and discipline. Sovereignty defined a spectacular regime of power that had to be seen to be effective. Discipline structured light to fall on an interior and illuminate its details, like a searchlight, or an intentional but non-subjective1 consciousness, which summoned objects out of their mysterious darkness. While discipline establishes a hierarchical system of non-reciprocal, one-way gazes that see without being seen, cinema operates by introducing signs of the unseen, or what could be seen by another, to the seen. Without reducing scene to seen, it makes us grasp what we do not know, but what is perceptible to another (Deleuze, in Boundas, 1993: 36). For example, in Urry's (1995) discussion of the tourist gaze, the images that we have of a place (the "place-image," as he calls it) could not be formed without the presence of others. He argues that it is the "presence of a large number of other people, as are found for example in the English seaside resorts ... [that] give atmosphere to a place" ... "It is the presence of other tourists, people just like oneself, that is actually necessary for the success of such places" (138). By way of extension, one could argue that the ambience of a place is also due in part to the others who are not figures in the landscape. One's place image and its social ambience - its intrigue, for example - can derive from travellers' tales (those who have seen and who convey its image), from rumours, gossip, folktales, and seductive advertisements, from the artifacts and memorabilia collected there, and from the pictures that we take to show the folks back home (see Shields 1991). In this sense, the other does not have to be an empirical object populating our field of vision, as in the face to face encounter of gazes doing reciprocal perspectives (interactionism) or engaging in a subject-object dialectic (existentialism). The other can also make its effects felt without being actualized or materialized in one's visual field. The disciplinary regime already established the fact that the other was most effective and influential when it was not seen. Boundas (1993) goes even further and suggests that the empirical other must be "foreclosed" or displaced in order for this other - what he calls an "otherwise-other" - to make its effects felt. This otherwise-other does more than reflect back a self (cf. Denzin, 1995: 112). It is a structure of alterity living on the margins and constituting the horizon of visibility beyond the frame. Like a spectre

83 Cinema's Optics

haunting the scene, it conveys a Zeitgeist, like Maffesoli's conception of unicity, an esprit de corps, or a form of togetherness or sociability, as in Simmel, as well as the ambience or sense of space and place, as in Lefebvre's (1991: 31) discussion of lived space and Maffesoli's (ig93b) sacred any-places. But it is also a poltergeist that makes its presence felt by stomping round, whispering, and rattling chains in the out-of-field. Film-makers have long been acquainted with this conception of the other and have deployed it in numerable ways (as we see below). Put simply, in a world where the structure of alterity ceased to function - that is, in a world without others - we would be constantly colliding with other bodies that would act on our actions with the force of projectiles. "The absence of the other is felt when we bang against things, and when the stupefying swiftness of our actions is revealed to us" (Deleuze, 1990: 306). The entire structure of anticipation - the expectation of what is to come, what is beyond, what we do not see and cannot be sure of (the horizon of possibilities) - is a function of this structure: "The part of the object that I do not see I posit as visible to Others, so that when I will have walked around the object to reach this hidden part, I will have joined the Others [already there] behind the object, and will have totalized it in the way that I already anticipated. As for objects behind my back, I sense them coming together and forming a world, precisely because they are visible to, and are seen by, Others" (Deleuze, 1990: 305). Similarly, the depth of our perception is a possible width for others, thus enabling us to relativize proximities, to distinguish foreground and background, and to know when objects are hidden behind others. This other illuminates the margins of the world that we do not see, warning of assaults from behind or the side, anticipating the effects of our actions, smoothing transitions from one perception to another, and filling the world with benevolent murmuring. For as Deleuze suggests, "When one complains about the meanness of Others, one forgets this other and even more frightening meanness - namely, the meanness of things were there no other" (1990: 307). Surely anyone who has seen a film is made aware of all these dimensions - of depth, of space and duration, and of course of anticipation. None of these effects, realism included, would be possible if the screen were simply flat and the frame fixed in its boundaries. For contemporaneous with all that is seen, there is a "virtual" unseen set, a beyond or out-of-field, that wards off (en)closure, prevents the set from closing in on itself, and prevents the mere (voyeuristic) identification of the scene with the seen or with what the camera sees. Like a line of flight, this virtual cinema-other passes through the porous membranes

84 Mediated Associations

of the seen and, at the same time, is that which allows the seen to pass into what comes next and warns of events to come. Here we might extend Eco's (1985: 166-71) thesis beyond the desire for repetition or its lack, to say that the pleasure of contemporary mass-media forms stems primarily from our relationship with this other, as it mobilizes our perception, liberates our limited and subjective point of view, renders a sense of place and the value of things, anticipates changes, and unlocks the secret potential of things to come. Whether things actualize as anticipated, and we get what we expect, always remains to be seen, for this other is the temporal structure of events and the means by which they flow, change, or endure. Rendering of meaning, typically undertaken from the point of view of the here and now or the fixed standpoint of civil society, as Marx (1978: 145) used to say, is all the more difficult when the interpretive subject-matter is porous and actively resists capture by constant flight, by changing its shape or form, refusing to settle, to be pinned down, captured, or naturalized. Since it is a structure, the question is not who this other is or means (because it is unstable, changeable), but what it can do. And the problem is not one of trying to contain or locate its source, since it refuses settlement, but rather one of assessing its affects. We can, however, give it a sur-name: faciality. As Deleuze and Guattari claim, "certain assemblages of power require the production of a face" (1987: 175). Perhaps nowhere is the difference between the regimes of en-closure and the open, cinematic system more apparent than in the place that faciality occupies therein. QUALITATIVE LOCALITIES

The face can serve as a means of distinguishing or characterizing a person. Synonymous with the signature, mug shot (photo de pose), or fingerprint, the face can be individuating and serve as a means of identifying individuals and distinguishing bodies. The face can be socializing, in the sense that it manifests a social role or a social location. Taken as a reflection of the settings in which people normally act and perceive, the face becomes habitus (see Bourdieu, 1993), where one's looks distinguish the place from which one originates, the place where one belongs, or the group to which one belongs (the ethnic face). The face can also be communicating, not only between characters or between roles (the subordinate face, or the face in the money-shot described by Linda Williams [ 1989]), but also as an expression of the internal consistency in the sense of auto-communication between a character and its role - when one looks the part, or its lack, as in the case of the automaton where the face fails to reflect or respond appro-

8 5 Cinema's Optics

priately to its scene. These are instances of the face subordinated to space or place. The face is also verbose. It emits and receives, releases and captures, gathers and expresses many signifying signs. Signifiers always reterritorialize on the surface of the face, so that signifiers are always facified (Deleuze and Guattari, 1987: 115). Signatures (a means of individual identification), the habitus (a means of group identification), and signs of communication (identification or the failure to identify with an individual or a group) are examples of the face captured and inscribed by a particular regime of enunciation. The face is made to bear these inscriptions, although no actual physical force ever touches or marks its surface. These signs should be understood in terms of the particular apparatuses, the regimes of enunciation, and the lines of force that make it signify or that "induce it to speak" (so to speak). But in the cinema apparatus, the face takes on a different complexion. As Bergman defines it, "the possibility of drawing near to the human face is the primary originality and the distinctive quality of the cinema" (in Deleuze, 1991: 99). It is typically the face that populates the cinematic close-up, but Eisenstein (1942) goes further. For him, the closeup is the face, suggesting that any close-up calls forth either a face or a facial equivalent. More than just a quantitative enhancement of detail, or a moving nearer, the close-up marks for Eisenstein a transition from quantity to quality. It is a "qualitative transformation" or "a leap beyond the limits of the possibilities of the stage - a leap beyond the limits of situation: a leap into the field of the montage image, montage understanding" (1949: 239). What Eisenstein advances is a new way of understanding the dimensionality of the cinematic close-up and the changes of dimension that it brings about. This change of dimension has two dimensions. First, this change can be quantitative and relative. It may consist of the magnification of a part of the set and serve as a means of optically enhancing its little details - in much the way that one might apply a microscope or a telescope to zoom in on, and isolate, an interesting bit from the rest, or the way a drover might isolate an individual from the herd for special treatment. In this case, the resulting close-up is an image of a partial object or fragmentary element that has been abstracted, cut or torn from its place in an overall constellation of elements. It is an image of an individuality, a unique but partial element separated from its place and its relations to other elements in that place. The effect is an image that is both small and large at the same time, which displaces elements and dislocates the field of visibility - as is the case in Baudrillard's conception of the abstracted signifier and its corollaries: the

86 Mediated Associations pornographic, fetishistic gaze of sign consumers, their taste for collecting, and their fascination with the series or the serial project of collecting a self. Second, this change of dimension may be qualitative and absolute. Like the proverbial straw that broke the camel's back, at some point an increase or diminution of quantity will pass a threshold and bring about a qualitative, irreversible change of state. This is how Eisenstein envisions the close-up, as a turning point, a threshold, a point of qualitative transformation. In as much as it dislocates, it also opens up a new kind of dimensionality and a new kind of space - namely, a transformation space: a space of possibilities between the state of the before and the state of the after. For this reason, I consider it a space of becoming and of movement rather than a space of being or settlement. It forms a conjunction between this and that or, in Simmel's terms, an opening or doorway between here and what lies beyond. From this perspective the close-up defines a qualitative locality and a location of qualities. Whatever occupies this interval or threshold space will not only have to forgo placement; it will have to surrender its identification or the identifying marks that determine where it belongs, its place of origin or the group to which it belongs, and its individuality. Probyn understands this dynamic image of in-betweenness as "the moment when the trapeze artist has let go of one ring but hasn'tyet grasped the other" (1996: 42). Here, the size of the gap, the space between one ring and another, between one state and another, may very well be measurable and quantifiable (sometimes large and sometimes small), but the space of the interval can also be assessed in other terms. It can be assessed in terms of its intensity, and the intensity of the scene/seen can be different, singular and unique, each time it is experienced. In cinema, the close-up does not simply isolate the part from the whole. Because of its location in the interval, the facified close-up determines the value of the whole scene in terms of its intensity and affective qualities. To illustrate this notion of the close-up as facification, we can turn to the example of the clock-striking-midnight in the work of Eisenstein (1942: 21) and Deleuze (1991: 87). The clock, in this example, is shown in several shots that gradually draw near to its sur-face. One of its fixed traits is to be reflective, to reflect back light and information - that is, to tell the time. Its other trait is dynamic and expressive. Because its hands form a series of micro-movements, even virtual, imperceptible ones, it marks an accent towards a critical instant, to the point where midnight becomes affective, intensive, not only reflecting

87 Cinema's Optics the hour and telling the time, but also expressing the quality of midnight as a fateful hour and determining the "value of what is seen" (Eisenstein, 1949: 238). Scenes such as a clock-face at midnight, a close-up, are typically employed to stake out such critical instants - for example, "those instants in which water becomes a new substance - steam, or ice-water, or pig-iron - steel" (Eisenstein, 1949: 173). It is also used to mark turning points. Pacification is preparatory, of the progress either towards a limit or of crossing a threshold. Moreover, it expresses these qualities virtually, without ever having to actualize them in the visual field. The face is a cinematic movement: a surface of capture and inscription and a surface of expression, transformation, and change. Whole buildings are often facified: they have a look, an expression all their own; their facades are illuminated, sometimes polished to a highly reflective gloss (corporate faces), sometimes worn and tired; their surfaces are adorned and decorated, painted and made up, inscribed, reinscribed, and over-scribed (with graffiti); they are populated with control surfaces, windows, doors (bars and cameras), places of entry and exit (for controlling flow); populating the cityscape they modify the "atmosphere" of a place or space (the ambience of the city). Whole landscapes can be facified. Eisenstein (1949: 237-8) calls them "large scale" objects or faces, primarily to distinguish them from those poorly framed, photo-graphic close-ups that only display or present. Another example of facification occurs in a scene from Clint Eastwood's film Unforgiven, made in 1992. The scene comprises a series of four shots (the cut "§*£" defines the shot): A shot of the rain at night. The hero, Munny and his sidekick the Schofield Kid, ride into the frame of the shot and stop. Munny hands the Kid a whiskey bottle and keeps one for himself. They exchange dialogue, then the Kid rides off. Munny continues to ride into the town (Big Whiskey), the bottle in hand. S»S The camera tracks slowly, pointed toward the ground made wet by the rain. A whiskey bottle is thrown into the frame. It hits the ground and splashes as the camera continues to track along the ground (with jerky movements as if mounted on a horse), then pans up to see the town. Two torches can be seen in the distance in front of the town's saloon. J