Closed Circuits

Screening Narrative Surveillance

Garrett Stewart

• Publisher: University of Chicago Press
• Year: 2015
• Print ISBN: 9780226201498, 9780226201214
• eBook ISBN: 9780226201351
• Topic: Media & Performing Arts
• Subtopic: Film & Video

The recent uproar over NSA dataveillance can obscure the fact that surveillance has been part of our lives for decades. And cinema has long been aware of its power—and potential for abuse.

In Closed Circuits, Garrett Stewart analyzes a broad spectrum of films, from M and Rear Window through The Conversation to Déjà Vu, Source Code, and The Bourne Legacy, in which cinema has articulated—and performed—the drama of inspection's unreturned look. While mainstays of the thriller, both the act and the technology of surveillance, Stewart argues, speak to something more foundational in the very work of cinema. The shared axis of montage and espionage—with editing designed to draw us in and make us forget the omnipresence of the narrative camera—extends to larger questions about the politics of an oversight regime that is increasingly remote and robotic. To such a global technopticon, one telltale response is a proliferating mode of digitally enhanced "surveillancinema."

1

THE PRYING “I” OF MONTAGE

Fritz Lang’s way of making visible in M (1931) the often fierce prying of the film’s own optical prowess typifies not only his cinematic temper but the long tradition of camerawork it influenced. Cinema, like surveillance, is a dialectic between the seen and its seeing—as well as, from the late 1920s on, between the uttered and the overheard. And surveillance, like cinema, is often the work of mediation. Surveillance had already taken paper form, often photographically imprinted paper form, in the plotted paranoias of Lang’s narratives (in wanted posters and forensic photos, for instance) long before the director introduced his own projective technology into given screen plots as an explicit instrument of moving-image capture and accusation: the heavenly filmic archive of Judgment Day in Liliom (1934), the incriminating courtroom footage in Fury (1936).1 Yet his filmmaking was always implicitly just that: a form of cornering and exposure.

In this sense, the broad place of cinema in the panoptic regimes of modernity, rather than its secondary narrative emplacement in particular storylines, had been an undertheme in his work from the first time Lang trained his monocular lens on the supposedly unsuspecting actor overseen in role. And with Lang in mind, surveillance needs also to be located on a conceptual axis with its opposite (or complement) in disguise, impersonation, eluded recognition. In the first scene of the 1922 seedbed film, Dr. Mabuse: the Gambler—beginning an auteurist cycle ending only four decades later with the eponymous Thousand Eyes of the same mastermind—the criminal overlord at first fans out and reshuffles photographs of himself (in various disguises) like playing cards in the game of deception, choosing which one to deal out next for his on-camera (rather than, for the actor, behind the scenes) makeup artist. One of these photos is soon brought to life in the second sequence of the film when Mabuse appears to full view, though no longer in propria persona, to prosecute his anarchist schemes. The transition is medial as well as dramatic: from photograph to the kinetic mirage known as cinema. Almost a decade later in M, one may sense the same material tension between still and moving images linked in more subterranean ways to the epistemology of the photographic index and the tracking operations of the moving-image apparatus.

In this discussion’s original setting as a conference paper under theumbrella notion of “Moving Modernism” (at Oxford University in thespring of 2011), I had trusted my title, “Frame-Advance Modernism:the Case of Fritz Lang’s M,” to ring a bell—or, better, throw a projector switch. Yet the case for the frame in this sense, as transparent photograph (or photogram) on the backlit spinning reel, is one that repeatedly needs advancing, rather than being taken as axiomatic—and all the more, I find, in the spreading intermedia landscape of new modernist studies, where theory is going to the movies as never before. Faced with the kinetic hypothesis of modernist mobility, certainly cinema comes to mind right away: moving pictures. Not so-called at first, though. Too candid, that transferred epithet, displaced from motor process to visible motion. What moves are indeed only the pictures, in order to picture movement. The actual moving of single transparencies, single backlit photograms, yields virtual movement. This is what people came for at first: the new magic, even before the specific attractions. In some of the earliest projection venues, in fact, the first image was held like a slide (made possible by water cooling) so as to highlight the subsequent wonder of its launch into action. Each screening thus served to bear forth the medium from its own genetic origin in the still.

Like moving pictures, the term motion pictures came later too: also true to the apparatus, at least in a roundabout way. The motion of anything on-screen is what the projector’s own motion pictures, pictures in the thrown beam of change itself. This transition per se between celluloid frames or increments is then transferred to spatial transit across the screen frame. The real advance (pun allowed) of the flickers was the frame advance. Which is what encourages Friedrich Kittler, in Optical Media, to go so far as to subordinate cinema to the digital. For him, the necessary intermittence of the projected image in its syncopated pulsation of frame/[bar]/frame is thus discrete, binary, and in itself, though photographically composited, ultimately nonanalog in its motion effects.2 This is exactly what the early modernist philosopher of time Henri Bergson disliked about film, its simulation rather than capture of durée, as did chronophotographer Étienne-Jules Marey, objecting to cinema’s nonanalytic conflation of poses, its squandering of the properly discrete graphing of difference in the illusory spectacle of procession.3 Assimilating a binary on/off to the internally traced difference (rather than sheer alternation) of cinema’s cellular increments, each retaining the afterimage of the last and invaded in protention by the next, we might call this not the zero/one but the zer/o/ne effect. No anachronism is necessary to see this. I am therefore less concerned than Kittler to rethink cinema’s material strip in the backward light of digital oscillation than to detect—in the narrative editing that exploits this original filmic microframe—one director’s inadvertent forecast of an ocular mediation that characterizes the surveillance ethos of a whole (and now electronically implemented) rhetoric in contemporary cinema. This is what we might call Fritz Lang’s perpetual modernism, the weird currency of his camerawork. In the grips of his shot logic, the future is now.

A Thousand Eyes

Tom Gunning’s magisterial The Films of Fritz Lang certainly earns its broad and definitive title. At the same time, within its sustained variorum-like commentary on the interleaved themes of Lang’s cinema is embedded, in effect, an entire stand-alone monograph on audiovisual surveillance as the destiny of the medium itself, a book-within-the book that might lay its own apt claim on the subtitle of the entire work: Allegories of Vision and Modernity.4 In Gunning’s auteurist scope almost no surveillance moment is overlooked, both in terms of POV camerawork and in regard to prosthetic technologies brought to bear on such a general epistemic framework. His commentary logs nearly every instance of this second-degree inflection of narrative optics (later audio) from as early as the silent urban thrillers whose theme is summed up by the 1928 title Spies: films that centrally reflect modernity’s “new systems of social control through a panoptic system of surveillance” (94). In topic and technique, such narratives register a counterplay of violence, bottom-up versus top-down, that is most schematic in 1933’s The Testament of Dr. Mabuse. Thus Gunning: “As the ordinary criminal threatens the state’s monopoly on violence, the master criminal threatens the state’s unique employment of the panopticon of surveillance and information-gathering” (95). What we’ll come to in the Postface below, via Deleuze’s sense of an overthrown panoptic model in “societies of control,” cannot deny what Gunning is out to notice in these early sound films.

And the threat of coercive oversight in Lang often takes machinic form. From the simple “interlocking technologies” of “the railway, the pocket-watch, the telephone system” (98) in the opening episode of Dr. Mabuse: The Gambler, we move (with Gunning) across Lang’s whole German and then American career—alongside nontechnological surveillance, for instance, in You and Me (1938), Woman in the Window (1944), and House by the River (1950)—from the incorporated function of cinema itself as a spying machine (in Liliom and Fury, again) through to the striking (almost literally arresting) use of television across the last half decade of Lang’s screen work. In While the City Sleeps (1955), a newscaster’s direct video address terrifies a criminal in hiding as if it were a private transmit rather than a broadcast dissemination (an effect quite possibly derived from Orwell’s telescreen in 1984, a device that might really be able to see the cornered subject during its invasive harangues). In Lang’s last film, The Thousand Eyes of Dr. Mabuse (1960), with its new iteration of the title figure’s posthumous power play, the ghost of the evil genius turns the criminal Cornelius into a master technician of closed circuit TV. This involves hidden automatic sound cameras spying in part on dialogue episodes whose intermediate transmission itself advances the plot. In a world that, for Lang, had come to seem increasingly “saturated with observation” (Gunning, 470), this new transmission technology moves—across these last two films—from intimidation to incrimination.

For Gunning, there is less media rivalry here than surveillance genealogy: “Television . . . simply realizes the role Lang had envisioned for it in his science-fiction scenarios” (471), as for instance with the closed-circuit surveillance screen communicating between factory floor and overlord in Metropolis (1936). The telos of all media optics: equally to show and to see. Mabuse’s original out-of-body, beyond-the-grave effect has evolved into the instrumental telepathy of mediation as telepresence. So that, here again, this last “allegory of vision and modernity” is constellated as nothing less than the regime of surveillance. It is fully in view of, and in debt to, Gunning’s broad grasp on this motif in Lang that I will be calling out certain cinematographic details—and even tacitly photogrammatic ones—that slip past notice in his account of M, even when in one case the shot in question is actually discussed in passing. This is to say that though Gunning’s omnibus scope leaves no stone unturned in revealing the thematics of surveillance in film after film, the optical substrate of a given film, unearthed thereby, may still invite—and certainly in the foundational case of M—further cinemato/graphic notation regarding the implicature of its montage in the systems of coercion it articulates.

Much depends on scale. With Lang’s M, the issue of fixity versus motion is not just, as Jacques Rancière sees it in Film Fables, a tension—in its own right quintessentially modernist—between mimesis and diegesis.5 While for Rancière this formative division of labor pits image against plot, each reciprocally “thwarting” the other, a similar though invisible tension, one level down, is at work between photogram and frame line. What did Lang seem to know and to show about this underlying aspect of the image file—and its prehistory—at the very moment when, in his first sound film, he was double-timing silent cinema with his first audial track? What did the intermittent audial synchronizations he had budget for, or interest in, underscore about the fundamental intermittence of the strip? And how did he commandeer this apprehension for the pacing of narrative impact?

Medial premonitions aside, these are the historical questions that underlie this chapter—and that will direct our attention to images in the film that borrow their composition and framing from adjacent graphic arts of the period. The reasons for this are more than thematic. It’s not just that Lang’s montage aligns a time-based medium with the alternative pictorial formats it subsumes in the process. It does so with immediate technical as well as narrative repercussions, thus rendering inter-art comparisons less tangential than they may sometimes seem. Film is like sculpture in motion, like mobile panoramas, like large-scale narrative painting, and, increasingly in modernism, like dance, like cubism, like war and its mechanized sightlines, etc.—where the likeness serves to keep the very distance it would bridge. This is, of course, as it should be, for the most part. With one medial exception. My work concerning films previous to the digital advent—studying as it did the relation of film to photomechanical imaging—could reasonably dispense with the framework of similes. Film isn’t like photography. Film is photography (or was, until computerized imaging), with the difference amounting mostly to internal differentials in high-speed sequencing. And, within narrative films, some photographs, printed rather than transparent, figure that fact—now genealogically, now elegiacally, now ironically.

But when one speaks about a certain photographic style more broadly, rather than simply the fact of photography, the axis of comparison widens once again. This is the case when contemplating an entire Weimar art movement, the Neue Sachlichkeit, as it inflects a contemporaneous film like M. How might this aesthetic practice, in painting as well as photography, be more than just a cultural ambience in Lang’s film? More than a tacit intertext? How, that is, might on-camera satiric portraiture reminiscent of the canvases of George Grosz, for instance, alongside shots evoking still-life photographic treatments of urban industrial Germany and its merchandizing displays in the interwar years—with their stringent geometries rejecting in every way the inward urgencies of expressionism—be seen as developing in M something, for want of a better term, more inframedial? This is to ask: without actual photographs on camera until the film’s climactic scene—where the serial child murderer, Beckert, is confronted with implausibly enlarged prints of his victims in the neutral photographic mode of studio portraits—how, building toward this, might the very different pictorial zeitgeist of the period find an impact in, rather than just on, Lang’s narrative? An impact, that is—in this first (and perforce, in this respect, most experimental) of his sound films—on the narrative’s own disclosures about the already “mixed medium” of speeding photo-transparencies (picture plus motorized projection) under the pressure of the new hybrid mediation of audio/visual synchronicity.

This is a synchronicity only fitfully secured in M—and often against the separate tread of image. Sound triggers an independent graphic pattern in the film’s first extended sequence. Accompanying the forlorn mother’s repeated shouting out of “Elsie” in this anguished missing-person episode (the first scream we hear in Lang’s cinema), we see a veritable Neue Sachlichkeit portfolio of photo allusions processed in cinematically static (rather than optically frozen) fixed frames that step off the various architectural echo chambers, absent Elsie, of the mother’s cry.6 Much later, as if bookending this effect, we come upon serial images of the deserted corporate offices from which the killer has been removed—this, in another quasi-photographic dossier of depopulated and hence unmoving images redolent of the period’s stripped-down, almost clinical aesthetic. These reveal the confines of an impersonal urban space, now ransacked, that is immediately recognized as typical of New Objectivist photography, including in the mix, this time, certain openly stop-action freeze-frames. In this optical episode, the less sound, in fact, the better. Thus isolated, these shots come forward as the recognizable photographic icons, and intrinsically silent film frames, that in fact they are. But their fixity italicizes more than that. For by this point Lang’s spectator is wholly entailed, rhetorically, in the ocular articulation of plot that these images slow to a sprocket-driven crawl. So that what ultimately distinguishes the narrative’s two phases of serial (photogrammatic) stasis, early and late—even while implicitly linking them at the level of allusive photographic composition—is the relation of the second series, by then, to a disembodied surveillance motif. More than just saturating Lang’s work before and after M, this emphasis on motivated but invisible viewing is everywhere implicit, as well, in a certain lineage of film theory (“the system of the suture”) that repeatedly queries the relation of optic frame to an invasive and disavowed spectatorial gaze.

The Implicated Spectator: Sous(ra)ture

A film like M can help remind us how that theoretical lineage has faded away in academic discourse without exhausting its full utility. In the related terms of literary deconstruction, the act of writing, like the ideology it sometimes explicitly inscribes, is the putting under erasure (sous rature)—in the name of meaning—of a text’s basic lexigraphic operations. For apparatus theory in film studies, concerned instead with explicitly cinematographic erasure—and speaking English as well as French this time—“suture” names a related level of collaboration between text and ideology in the grip of disavowal, masking narrative cinema’s essential gaps in service to diegetic coherence.7 Implicit in this theory is the overridden action of the track as well as of editing. For what happens from frame to frame, subliminally, is also denied in the shot plan and its spatial discontinuities. This sense of denial or suppression derives in passing from the deepest mythographic self-consciousness of apparatus theory: a primal anxiety over the ruined illusory immediacy of all screen experience.

It goes this way, in the earliest formulation of the theory. With obvious (even when occasionally unstated) shades of the Lacanian imaginary stage, the original screen trauma is the recognition of sight as mechanically directed. The task of subsequent cinematic narrative is to assuage this anxious notice in an almost surgical fashion—through a masked pastiche whose own functional invisibility as such, once acceded to, in turn sutures the viewer into the piecemeal shots of montage in order to anchor their credibility as world space. Resisting the initial disenchantment of the gaze, suture arises to address, whatever its immediate local function, a utopian nostalgia for the pure seen. This is an optic not even yet demarcated as an image—just the sheer manifestation of the visible: apparatus disappeared unacknowledged into apparition. On this account, it is almost as if every movie, once its spools are set in motion, replays in projection—in the first wonder of its animation, subsequently undone by the machinations of framing and plot—the mirage of immanent life. Editing arrives to sustain, against all odds, this sense of presence.

But as soon, of course, as the screen figment is recognized (or say reconceived) as a manufactured image, which may of course be with the speed of simultaneity—or in other words as soon as the sighted picture, as image, is taken in as organized “scene”—the act of viewing (rather than just looking) has set in. The result is not to occlude the prospect, of course, but to bracket it as selective, worked, intentional, in short framed, and to this extent made rather than given, tainted with labor and artifice alike. The world no longer comes forward partially disclosed to us; instead, its separate aspects have been deliberately chosen from, for subsequent display. Recognizing this marks the primal fall from pure visibility into viewing, long before this loss is reinvested by either moving camera or edited shot-changes into a new narrative economy.

Suture theory understands the manifest manipulation of camerawork as serving always to reframe and (again its surgical sense) stitch over such a lost jouissance, because a fractured integrity, of the visible (ever and already vivisected). This is especially so, and most obviously, in the shot/countershot pattern of later narrative developments, where the very idea of the looked-at, an idea focused on the framed face before us, is a discovered illusion. For this looking happens within—how soon we forget—a wholly constructed interspace or interface (never a real human conjuncture) between detached framed images. To phrase it in the heuristic hyperbole of the discourse on suture, the prenarrative dream of immanent visibility, fallen into framing and spectatorship, has become the managed panic of continuous alternating evacuation, where the spliced wound between image and represented world, as well as between us and image, can never be owned to—and remains, as such, a cognitive prototype for bad faith.

One pattern above all is quintessential. In suture...

Table of contents

• Cover
• Title Page
• Copyright Page
• Dedication
• Contents
• PREFACE Returns of Theory
• INTRODUCTION Narrative Spycams—A Foreshortened View
• 1 The Prying “I” of Montage
• 2 Telescreen Prose
• 3 Feedback Loops of the Technopticon
• 4 In Plane Sight
• 5 The Othering of Lives
• 6 Digital Reconnaissance and Wired War
• 7 Retrospecular Eyes
• 8 Parallel World Editing
• POSTFACE On Mediation as Interface
• NOTES
• INDEX