Introduction: “I am Highway Watch”

In 1998,¹ The American Trucking Association (ATA) founded Highway Watch, a programme designed to teach truck drivers “what to look for when witnessing traffic accidents and how to assist emergency vehicles to find crash sites” (Boston Herald, 2004). In the wake of 9/11, this guidance on what to look for, how to look, how to use the windscreen as a screening device for times of emergency, became something of a gift to the US Department of Homeland Security (DHS). By 2004, the DHS’s Transportation Security Administration (TSA) had entered into an agreement with the ATA, providing $40 million to extend the programme to “school bus drivers, highway maintenance crews, bridge and tunnel toll collectors and others”, so that they would be able to “recognize and report suspicious activity” (DHS, 2004a). Against a backdrop of the burning twin towers, the promotional video represents truck drivers as first responders, displaying “I am Highway Watch” badges alongside the US flag on their uniforms. “All of those eyes out there on our nation’s highways”, proclaims the voiceover, “they see things that maybe don’t fit into the picture somehow. We want them to watch out for unusual activity and call the Highway Watch call centre”.
Highway Watch is but one example of a vigilant mode of visuality that is deployed routinely on the “homefront” of the so-called war on terror. Of course, there is nothing substantively new about engaging the eyes of people in their everyday routines and journeys in order to identify suspicious behaviour or the apparently “out of the ordinary”. In many ways, this is a rearticulation of the “behind the blinds” surveillance of 1950s suburban neighbourhood watch. Yet, contemporary modes of watchful politics are particularly geared to the anticipation of events, deploying a kind of precautionary principle that governs through the suspicion of a possible future threat (Ewald, 2002). As Highway Watch moves from the reporting of accidents, crashes and incidents during or after the event (where sight provides “eyewitness” evidence) to the post-9/11 preemption of terrorist attack, before the event (where sight becomes foresight), the emerging watchful politics is vigilant: it “looks out” with an anticipatory gaze.
Set in post-9/11 Los Angeles, Paul Haggis’s film Crash reveals something quite different about the watchful politics of a fearful life lived predominantly on the highway. The watchful look is present in its many (dis)guises – through windows and windshields, via glances at passers-by in hoodies or hijab, in a police officer’s decision to stop and search. Yet, Haggis’s film conveys also what is assumed but concealed in programmes such as Highway Watch: that this watchful look plays out as profiling, the usually racialized stereotyping of “Arab”, “Muslim”, “immigrant” and so on. Life in cities that “use freeways and wide boulevards to divide people by race and class”, says Haggis (2006) in an interview, is a life “encased behind glass: in our home, our cars, at work”. In our quest for safety and security, argues Haggis, “we no longer truly feel the touch of strangers as we brush past them on the street”. Following a car accident in the opening scene of the film, a character muses on how the sense of touch is lost, so much missed that “we crash into one another just so we can feel something” (Haggis and Moresco, 2005). Crash reminds us that the fearful and vigilant watching for the “Other” through the windshield is one specific and particular representation of visuality that asks that we look but do not touch. Indeed, so dependent is it on the drawing of lines between Self and Other, homeland and strangeland, safe and unsafe, ordinary and suspicious, that it must feign the prohibition of touch if it is to sustain the distancing that allows the dividing practices to continue. Haggis’s film succeeds in capturing the ambivalence of contemporary watchful politics, conveying both the bleak and violent realities of racist suspicions and the sense that even this looking is ambivalent, it also craves the touch of others. In this article I engage with a form of visual culture that is, W. J. T. Mitchell (2002, p. 170) reminds us, “not limited to the study of images and media”, but extends also “to everyday practices of seeing and showing”. In the spirit of this openness to multiple manifestations of the domain of the visual and visual practices, I will explore two questions. The first asks how a particular mode of vigilant or watchful visuality has come to be mobilized in the “home-front” of the so-called war on terror. Has vision come to be represented not only as the primary sense (more “reliable” and verifiable than touch, taste, smell, etc.), but also as, precisely, the sovereign sense – the sense that secures the state’s claim to sovereignty and legitimates violence on its behalf? Second, taking our illustrative cue from Crash, and from a body of work that conceives of touch as “integral to” and “active in” seeing (Cooley, 2004, p. 133), how might we subvert watchful politics by seeing seeing differently?

Because we see: sight and sovereignty

In his discussion of the myth that is a “purely visual” media, W. J. T. Mitchell asks how the visual “became so potent as a reified concept” (Mitchell, 2005, p. 260). Given what he describes as the “intricate braiding and nesting of the visual with the other senses”, how, he asks, did the visual “acquire its status as the sovereign sense?” (Mitchell, 2005, p. 265). Such questions have become the principal concerns of critical writing on visual culture that seeks to expose the assumptions about visuality that render the visual sovereign. How, critical writers on visual culture ask, has visuality come to be represented as “the superior, most reliable” of the senses? (Bal, 2003, p.13). And, why is it that images are afforded status as “natural and visible signs” with an “inherent credibility” that cannot be found in the tactile or aural senses? (Jay, 2002, p. 269; see also Crary, 1990). It is not that the visual is the sovereign sensory domain, then, or indeed that it can be meaningfully abstracted from other sensory perception, but rather that its representation as sovereign has important and material effects. The representation of a sovereign visuality has played an important role, for example, in the making of rational and responsible subjects. If sight is understood to be the sense most nearly and dispassionately under our control, or, as Heidi Rae Cooley (2004, p. 135) puts it, “if people become distanced observers, whose vision is objectified, quantifiable, manageable”, then the act of seeing is central to the defining of a coherent, autonomous subject. The image itself becomes an artefact that is, of all media, least vulnerable to “subjective intrusion or ambiguity” (Daston and Galison, 1992, p. 82).
Mitchell’s claims about the sovereignty of the visual, though, can be usefully extended to thinking about contemporary articulations of state sovereignty. Not only is the visual reified as sovereign among the senses, but it is also uniquely implicated in the representational practices that make state sovereignty possible. As David Campbell (1998) has argued compellingly, a state must “write” its security; it is dependent on representational practices that identify “us” and delineate “us” from the untrustworthy aliens. Understood in these terms, we cannot see ourselves clearly or know who “we” are until we have recognizable representations of the Other. For Michael Shapiro (1997, p. xi) it is “architectures of enmity”, projected via images and stereotypes, that make threat and antagonism possible. “People go to war”, writes James Der Derian (2002, p. 5), “because of how they see, perceive, picture, imagine and speak of others”.
Let me be clear here that I am not arguing that the visual realm is the realm through which the state secures its sovereignty via representations of the Other. Rather, a particular mode of visuality is called up and into being in contemporary articulations of sovereignty: a visuality that categorizes and classifies people into images and imaginaries of many kinds. Sovereignty is performed, as Edkins and Pin-Fat (2004) suggest, through the drawing of lines, and the drawing of those lines is made credible, at least in part, by the reification of the visual:

Where images become the filters through which we recognize/misrecognize ourselves and the Other, Mitchell’s insights are critical for thinking about what I will call here an emergent watchful politics. The watchful politics of the war on terror has come to use the image – broadly defined as a picture of a person, a visuality of a scene, or the pattern of data on a screen – precisely as the primal scene in which it recognizes and misrecognizes “Other” people.
Recognition has become pivotal to the watchful technologies of sovereignty deployed in this war. Facial- and gait-recognition surveillance, biometric identity cards, and expedited airport-security clearance programmes, for example, are at the forefront of the drive to secure the state from the threat of the Other, and to do so via identity (see Sparke, 2005; Amoore, 2006). Predicated on the assumption that the image can fix, verify and authenticate identity, and in spite of evidence of clusters of false hits among black and Asian faces, the visual is used here as the basis for belonging and expelling. Similarly, the Highway Watch programme’s appeal to “report suspicious activity” is only intelligible if there is some basis for recognizing who or what is alien, who or what is unusual, who or what is outside. In this sense, the identity of a “we” – whether patriotic truck drivers, or “we” the nation, or “we the people” – is realized via the claim to be able to see and to recognize a threat.
Of course, any claim to recognize is always also a misrecognition. Since identity is never fully constituted, can never be fixed, is always in process of becoming identifiable, the visual filter is bound to misidentify, to misrecognize. In Haggis’s Crash, a Persian shopkeeper, confronted with vandalism and racist graffiti in his shop, asks his daughter, “Why do they think I am Arab? When did Persian become Arab?” (Haggis and Moresco, 2005). He feels himself to be misrecognized, falsely identified as Arab. Yet, in the terms of a sovereignty that has designated him as Other he is, nonetheless, rendered recognizable. In the context of our “anxiety, rage, and radical desire for security”, writes Judith Butler (2004, p. 39), we shore up the borders “against what is perceived as alien”, engaging in “a heightened surveillance of anyone who looks vaguely Arab in the dominant racial imaginary”. As Butler suggests, misrecognition (of Sikhs, Hindus, Israelis, Arab-Americans, Persians and so on) scarcely seems to matter, as long as “everyone is free to imagine and identify the source of terror”.
In the vigilant mode of visuality, though, it is not simply the recognition of an alien outside that is at stake, but also the demand that the Other inside be recognizable as Other in order to remain inside. In a further scene from Haggis’s Crash, a white television executive asks the black director to retake a scene:


The white executive fears the misrecognition of the black character. In order for the audience to “see” a black character, he must “sound” recognizable. Jamal must, in order to be recognized by the audience, stand clearly behind a line that designates him as Other.
How might we make sense of the implications of these lines, and of the role of the visual in their mapping and delineation? Didier Bigo (2001, p. 115), in his trope for contemporary security, suggests the Móbius strip – the twisted, conjoined figure in which inside and outside are interpenetrated, where one does not know “on which face of the strip one is located”. Rather as in Giorgio Agamben’s (2005, p. 39) sense of sovereignty that is enabled by exception, the Móbius-style entanglement of interior/exterior allows for the inclusion of the Other by means of its very exclusion. As in the Crash scene, the black actor’s character can be incorporated within what can be safely seen so long as he remains recognizably Other. Similarly, the authority of the black director can be tolerated if he does not breach what is recognizably black. What is at stake, then, at least as Bigo sees it, is not strictly the sovereignty performed via the physical borders of the state, but instead that performed via the lines of “a new security device: the monitoring of minorities and diasporas” (Bigo, 2001, p. 115). The watchful politics I analyse here is absolutely at the heart of this monitoring – a constantly vigilant mode of looking that produces what I have elsewhere described as a “ubiquitous border” (Amoore, 2006, p. 348). If, as Bigo (2001, p.112) asserts, the policing of the lines between Self and Other is to become a matter of “everyday securitization” from “the enemy within”, then the visuality of the vigilant onlooker is the very means of the ubiquitous border. The visual becomes the sovereign sense that participates in the emergence of new forms of state sovereignty. It, quite literally, projects the lines through images and stereotypes, and it projects them forward in time. It is to this projection into an uncertain future that I now turn.

Because we see, we can decide: sight and foresight

One week after the London bombings of July 2005, novelist Iain Sinclair (2005) observed the ghostly images and cinematic repertoires of the mobile phones used on the day of the bombings, and the desire of the authorities to somehow apprehend the images:


Sinclair observes here a significant and politically troubling move in the watchful politics of the war on terror: the suturing together of people’s fears, suspicions and everyday visuality with “mobile screenic devices” (Cooley, 2004) – mobile phones, palm computing devices, handheld electronic organizers – that can translate prosaic inexpert visualities into “expert” data. One way to think about this move is to reflect on the role of “eyewitness” accounts in practices of justice and procedures of law. As Martin Jay (2002, p. 269) writes, “the legal distinction between eyewitness evidence and hearsay, or between a photograph of a crime and a verbal account of a crime” performs the very credibility of the visual. Understood this way, the call for people to submit their images and video clips – repeated on posters in the months following the bombings – is primarily about sight, about what has been seen but perhaps missed, about what can be shown from what is seen. Yet, the watchful politics at work here is not simply about sight and the gathering of evidence after the event. Instead, it is primarily about foresight and the anticipation of the event. Commuters were asked to send all images, even those depicting banal and ordinary “normal” activities and events. In the careless or playful pictures taken on the daily commute are the images of the “norm” against which, it is assumed, transgression of the norm – the “out of the ordinary” – can be identified. Contemporary modes of watchful visuality are geared to the anticipation of events, deploying what FranQois Ewald (1991; 2002) considers to be a “precautionary” approach to risk.
Ewald takes his cue from Michel Foucault’s (1975/2003, pp. 19–20) work on techniques of normalization and their incorporation into expert knowledge and judicial power that reveals “how the individual already resembles his crime before he has committed it”. Here, calculative practices are deployed, in effect, before a crime takes place, in order to see or to envisage the individual as criminal. The act of seeing thus becomes an act of foreseeing, preempting or anticipating. Ewald (2002, p. 288) depicts the contemporary manifestation of such preemptive practices in the “precautionary principle” that “invites one to anticipate what one does not yet know, to take into account doubtful hypotheses and simple suspicions”. This does not mean that suspicions or prejudices or stereotypes are deployed randomly to anticipate an uncertain or unknown future. Rather, the profiling of a “norm” of behaviour – whether via mobile-phone images or by CCTV footage or by passenger manifests on transatlantic flights – becomes algorithmic, it becomes encoded so that deviations can be identified and decisions can be taken.
In a war on terror that actually redeploys “dataveillance” systems in operation long before 9/11, it is this algorithmic logic that is distinctly novel. When suspicions and fears are called in to the Highway Watch or Life Savers hotline, the call centre uses an algorithm on a screen to turn simple suspicions into “actionable intelligence” with the gloss of technoscience. As Ewald (2002, p. 294) notes, the precautionary logic implies that action is taken precisely on the basis of uncertainty: “decisions are therefore made not in a context of certainty, nor even of available knowledge, but of doubt, premonition, foreboding, challenge, mistrust, fear, and anxiety.” In the logic of the precautionary principle, we might say decision is taken on the basis of a particular mode of visuality, one that is captured by the “screening” of the entire population to establish algorithmic norms and anomalies. From financial transactions to air-passenger data, “smart” electronic profiles of daily life are deployed as though they could anticipate an uncertain future (Amoore and De Goede, 2005). In these systems, the assumption is that it is possible to “build a complete picture of a person”, to quite literally see who they are before they board a plane or transfer money, by relating them to the norms of a wider population and identifying their degree of deviance (De Goede, 2006). Indeed, for those who are able to exempt themselves from screening practices, there is now a booming market in prescreening via programmes such as Privium Plus or PEGASE. In these instances, the electronic visualization of a person becomes a specific kind of foresight, mediated via the screen as a “touch screen”, a biometric scanner or the computer-based results of data mining.
Is it the case, then, that the screens of post-9/11 dataveillance techniques extend the vigilant visuality across the more prosaic screens of the windshield and the mobile phone? Sinclair’s observation that “we have a new cinema” in the flickering images of the mobile phone echoes Anne Friedberg’s (2002, p. 184) sense that “the automobile is a viewing machine . . . the visuality of driving is the visuality of the windshield, operating as a framing device”. Tracing the conjoined Californian histories of widening cinema screens and the wraparound windshield, Friedberg’s (2004, p. 190) work considers the screen as an interface between observer and observed and a site of projection. Citing Paul Virilio’s claim that “what goes on in the windshield is cinema”, Friedberg is concerned with the screen as a boundary in and of itself. “Driving”, she writes, “transforms the mobilized pedestrian gaze with new kinetics of motored speed and with the privatization of the automobile capsule sealed off from the public and the street” (Friedberg, 2002, p. 184). The screen as boundary produces a capsularized experience of being able to see, watch and look out, but not to be seen, watched or looked at.
In one sense, the screen here precisely enables the distancing of Self from Other, provides a domain in which the precautionary principle can flourish and suspicions can become the basis of decisions. Certainly, the US homeland security project is coupling the windows of neighbourhood watch to the screens of automobiles and mobile phones in new and troubling ways. Sponsored by Microsoft, Hewlett Packard and Nextel, for example, the Department of Homeland Security has toured US town halls calling for citizens to “report questionable incidents and circumstances” via the communications companies and networks who “have become trusted partners in a global neighborhood watch” (DHS, 2004b, pp. 12–14). The personal computer or mobile-phone screen thus becomes a means of shoring up the visual as the sovereign sense, giving the appearance of a detached, smart and data-driven basis for decision. Yet, as in Sinclair’s observation of the careless and magical quality of the underground images, there is a kind of cinematic conceit associated with the screen – a theatrical suspension of disbelief. “For the film spectator”, writes Anne Friedberg (2002, p. 188), “the darkness that surrounds the frame calls us to play upon its boundaries”, invites us to immerse ourselves in an “imaginary, endless space”. The screen is ambivalent: it performs borders and boundaries, but it also invites us to play upon them.

Other ways of seeing

The vigilant visualities dominating the war on terror rely upon particular representations of sight and seeing. The visual is sustained as the “sovereign sense” through which the Self is secured and the Other identified. In order to sustain the visual in this way, however, the vigilant and watchful mode must occlude the possibility of seeing differently. Specifically, it must say “look, but don’t touch”. To clarify my point here, it is not that vigilant visualities successfully decouple one sense (sight) from one other (touch). Instead, they feign the detachment of sight from all other senses that allow us to be “touched” or moved. Not only is tactility inherent to visuality, but touch itself is to allow the taste, the sound, the scent of an experience, literally to “move” or “displace” oneself (BAL, 1997, p. 61).
To explore the tactility of visuality is to begin to locate some of the vulnerabilities of the vigilant mode of looking. “On closer inspection”, writes W. J. T. Mitchell (2005, p. 257), “all the so-called visual media turn out to involve the other senses, especially touch and hearing”. Sustaining the impression of a purely visual experience, then, involves the prohibition of touch. That is why, suggests Mitchell (2005, p. 260), in the space of the gallery “we are so rigorously prohibited from actually touching the canvas ourselves”. Touching the image would in some way expose the elevated fiction of a purely visual experience. Like the representational practices necessary to elevate the myth of the state as untouchably sovereign, the visual is dependent for its sovereignty on representations that exclude the tactility of vision.
Understood in these terms, the images called for and collected by the authorities after the London bombings, or those embodied in the Highway Watch programme, though they demand a distancing of observer from the observed, are actually both optical and tactile; they are as much about multisensory feeling as they are about seeing. Heidi Rae Cooley (2004, p. 143) analyses what she calls “mobile screenic devices” in terms of the “fit” between visuality and the involuntary tactility of the screen in the palm:


Cooley introduces ambivalence into our sense of the screen as the means of distancing the viewer from what is viewed or creating a privatized, capsularized space of invulnerability. For Cooley, “screenic seeing” should be understood as co-present within “windowed seeing”, precisely because the screen offers an interface, a possible means of breaching its own boundary. The screen, in Cooley’s (2004, p. 145) terms, is a liminal space that is “permeable and without definitive borders”. Experiencing this liminality, she argues, involves a temporality “without linear historical time”, an instinctive experience “of flowing and spreading, not one of focused concentration” (Cooley, 2004, p. 149). The mobile screenic device that has become so central to the watchful politics of the war on terror, then, potentially involves an ecstatic experience of being “transported beyond oneself”, or “implicated in lives that are not our own” (Butler, 2004, pp. 24–25).
Read alongside Judith Butler, Cooley’s sense of the “absent minded” and “distracted” screenic way of seeing evokes the experience of a decentred self. This has significant implications for the vigilant visuality that would seek to appropriate what is captured via the screens of consumer technologies and automobiles. The assumption that the screen allows for rational, distanced and detailed seeing is seriously challenged by the tactility of visuality that is suggested by Cooley and others. Tactile vision such as that involved in screenic seeing disrupts the precautionary logic of vigilance and subverts the sense of sight that is necessary to foresight. By contrast with a mode of visuality that claims to be able to capture and profile everyday images in order to anticipate the future, screenic seeing reveals precisely the indeterminacy of the future. As Cooley (2004, p. 151) understands the “distracted” viewer’s capacity for surprise when “the hand’s engagement allows for a resonance in seeing”:


In an exposure of the ambiguity of the vigilant visuality’s deployment of everyday seeing to prevent shock, accident, violence and surprise, here the tactility of vision precisely allows one to see the ordinary anew, to see the banal or trivial as extraordinary, to see our relationship with the Other differently.
In a scene from Haggis’s Crash, there is something of a momentary tear in the distantiated windowed seeing of stereotypes and profiles, revealing the screenic seeing of how things might be otherwise. A police officer arrives at the scene of a road traffic accident where a woman is trapped inside a car that has flipped onto its roof. Petrol is leaking from the vehicle and firefighters are tackling blazing fires nearby. The officer crawls through the broken window of the car to try to rescue the woman. In the moment that they see one another they appear to know exactly who they see. In a previous scene, after stopping and searching a car driven by the woman’s African-American husband, the officer has sexually assaulted the woman. In linear time she sees the villain, the perpetrator and her abuser, a white racist cop and a threat to be feared. He sees his victim, the light-skinned wife of a black, wealthy, SUV-driving film director:

The shock and trauma of the accident produces a different kind of looking in Haggis’s scene. Inside the burning car they momentarily see one another differently. He uses “look at me” to reassure and to breach the lines that were drawn between perpetrator and victim. Officer Ryan’s colleagues pull him back through the window as the flames reach the car, but he goes back and pulls the woman from the car. Haggis reveals here some of the ambivalence of visuality that Cooley suggests in her study, as well as the potential for trauma to “tear us from ourselves” that Butler identifies. The windowed seeing that dominates much of the film – distantiated, privatized, suspicious – is broken in the scene’s shattered windshield, revealing the copresence of something more akin to Cooley’s screenic seeing – tactile, surprising, immediate and affective.
In Jenny Edkins’s work on the memorialization of trauma in political community, she depicts a similar momentary rupture and emergence of the real in traumatic events. “In the rational west”, writes Edkins (2003, p. 13), “we tend to seek certainty and security above all. We don’t like not knowing. So we pretend that we do”. For Edkins (2003, p. 12), this feigning of control serves to conceal or forget the traumatic real so as to protect “the imaginary completeness of the subject”. It is precisely this feigning of control that is attempted by the windowed seeing of the watchful politics of the war on terror. Indeed, the call for people’s prosaic images of the trauma of the London bombings seeks exactly to incorporate what cannot be readily spoken or expressed into a narrative of bearing witness. Rather as Cooley’s windowed seeing captures the illusion of the complete watchful subject, Edkins shows how the state moves to quickly incorporate trauma or violence into a linear narrative that designates heroes, victims and perpetrators.
Similarly, Edkins’s use of trauma time produces the sense of distraction and shock that Cooley suggests with screenic seeing. “Trauma”, writes Edkins (2003, p. 59), “demands an acknowledgement of a different temporality, where the past is produced by – or even takes place in the present. We need to find a way of remaining faithful to its different temporality, a way to mark or encircle the trauma”. Edkins finds, in the architecture of some memorials to war, genocide or famine, exactly this capacity to encircle the trauma, to bear witness in a way that remains open and resists incorporation. Of course, trauma is not necessarily or inherently a condition for what Kaja Silverman (1996, p. 2) calls “excorporative” processes of identification with “bodies we would otherwise repudiate”. For Silverman, as for many critical analysts of visual culture, “visual violence” can be undone in the architectures of the cinema screen and the windshield that “encircle the audience”, inviting an escape from “the binds of time” (Silverman, 1996, p. 11; Friedberg, 2002, p. 188). The flashbacks that Haggis’s characters experience inside the burning car, flashbacks that Edkins (2003, p. 39) describes as images “we do not have access to other than as images”, find their expression also in the “flashback fluid temporalities of cinema” (Friedberg, 2002, p. 189). Far from producing a visuality that is distanced and dispassionate, or images that can be readily appropriated and interpreted, then, the screen is potentially also a means of flashback visuality, in which the image can be accessed only as image.
To put the argument simply, watchful modes of visuality, in their many guises, assure us that an image secures the presence of a rational observer. What I want to suggest here is that there are ways of seeing that absolutely invert this: seeing can be about the absence of an observer in the sense that the “usual” everyday order is disrupted. It is not the case that screenic ways of seeing offer political alternatives to vigilant visualities, but that they are always already present. Vigilant visualities of the war on terror rely upon the rational screening of normality in order to identify transgression. If the screen is positioned as the impenetrable boundary, as it is in the dataveillant and vigilant seeing of the war on terror, then we “foreclose vulnerability, banish it” (Butler, 2004, p. 30) in such a way that we crave the touch of passers-by. In the tactility of visuality, though, the viewer is transported out of herself (whether by trauma, by the tactile fit of hand to technology, by cinematic experience . . .) and the anchor in a complete subject is threatened. The prohibition of touch in watchful and vigilant modes is revealed to be an impossibility.

The unseen, the unforeseen and the undecidable

As I have argued here, central to the vigilant mode of visuality is an anticipatory logic. Such a logic implies that the unforeseen can be made foreseeable, can be somehow folded into present decision. As for FranQois Ewald (2002, p. 289), “one must take all hypotheses into account, even the most dubious; one must be wide open to speculation, to the craziest imagined views”. The gap between knowing and deciding, considered by Ewald to be central to precautionary logics, in the visual domain appears as a gap between seeing (so that one can know) and deciding (on the basis of what is foreseen). Anticipatory modes of looking seek to act in advance of an event so that future events might be preempted.
One possible route into questioning the anticipatory logic of watchful visualities, as I have discussed, is to think about seeing differently. Another is to question the capacity to decide on the basis of what is seen or foreseen. How do images of commuters on a subway platform have to be seen, ordered and categorized in order for a decision to be taken? If suspicion is central to the vigilant visuality, who is to decide and how, on the basis of whose suspicions? In the case of Highway Watch, the suspicions, hypotheses and “crazy imagined views” of the truck drivers appear on the screens of the “highly trained analysts at the Highway Watch call centre” (Highway Watch, 2005). Here, the gap between what is seen and what is later decided is filled with apparent “expertise” and designated as a data-driven process. Indeed, even where the war on terror’s dataveillance deploys complex algorithms to encode patterns of behaviour and identify suspicious activity, this only conceals the raw stereotyping of what can be seen. The technologies deployed to translate the unforeseen into the foreseeable, of course, fail everywhere and for much of the time. When they do, we see a glimpse of what they otherwise conceal: the absolute undecidability of all decision and the unforeseeability of most of what we think is seen.
Haggis’s Crash beautifully exposes the twists and turns that take place in the attempts to manage uncertainty by identifying stereotypes and securing against them. The Persian shopkeeper, who is suffering racial violence, buys a gun to secure himself and his property. In the firearms store he faces further abuse by the store owner, who says: “Hey, Osama, we’re not planning the Jihad here” (Haggis and Moresco, 2005). An argument ensues and the man is ejected from the store, leaving his daughter to buy the ammunition. Later in the film, the shopkeeper calls in a locksmith to repair a troublesome lock in his shop. When the Latino locksmith tries to tell him that it is the door that needs replacing, the shopkeeper is angry: “I suppose you got a friend who can sell me a new door, you cheater”. Of course, the weak door is smashed again and, this time, because the lock company has reported that the advice was ignored, the insurance company considers the shopkeeper negligent and refuses to pay out for the vandalism. The shopkeeper sets out with his loaded gun and waits outside the locksmith’s house.
The locksmith, meanwhile, has assured his five-year-old daughter that there will be no more bullets through her bedroom window – he comforts her with the gift of an invisible and impenetrable fairy cloak. When the child sees the man pointing his gun at her father, she moves from the bedroom window to stand behind the porch screen: “It’s OK, daddy. I’ll protect you. . . . It’s a really good cloak”. She runs into her father’s arms as the shopkeeper fires his gun. As the Latino locksmith checks his daughter’s body for what he thinks must be a fatal wound, he finds that she is unharmed. The gun that the shopkeeper bought to secure himself was loaded with the blanks that his daughter bought as a precautionary measure. In this scene, Haggis suggests to his audience the very impossibility of foreseeing the effects of actions taken to protect and to secure. We cannot be sure, he seems to say, which acts will protect or secure and which will further endanger. Neither can we be sure, throughout the film, of the identities of those whom we think threaten us – we misidentify at every turn. In a final scene, the Persian shopkeeper says of the young daughter of his apparent enemy, “She is my angel. She came to protect me, to protect us” (Haggis and Moresco, 2005). Each anticipatory move in the film, made on the back of suspicion and racial prejudice, fails somehow along the way: the SUVs crash or are carjacked; the gun fires blanks; the carefully manicured and gated home becomes an isolated place where accidents happen. Meanwhile, the apparently fictitious talismans – the girl’s invisible cloak, a Saint Christopher on the dashboard, a lucky charm on a key ring – produce a kind of fragile justice of their own.
The moments of decision in Crash serve to disrupt the precautionary logic. To act on the basis of what one thinks one sees in order to anticipate future threat is revealed to be intensely problematic. As Jacques Derrida reminds us, “the decision, if there is to be one, must advance towards a future which is not known, which cannot be anticipated”. In these terms, a decision that is anticipatory is not a decision at all, but merely “the application of a body of knowledge” (Derrida, 1994, p. 37; see also Derrida, 1992). To be a decision, then, an act must invoke that which is somehow outside of the subject’s control, it must be made in the face of a future that cannot be known. In the context of sight deployed as foresight, the apparently informed decisions taken on the basis of norms are not actually decisions at all, but simply rearticulated calculations. What is seen on the highway, in the subway, or on the screens of tech-enabled border guards, can never be the basis for decision, much less responsible or just decision.
Visual culture offers useful ways of thinking about how visuality might remain open to an unknown future and “advance where it cannot see” (Derrida, 1994, p. 39). By contrast with modes of visuality that use what is seen to anticipate the unforeseen, critical visual culture does not tightly tie knowledge to decision (we have the visual data, so we can know, so we can decide). Instead, it opens up to allow in the aporia, the undecidable. In photography, for example, the opening and closure of the shutter is simply one moment of decision, but the image that is produced is open to other future possible decisions. A photograph is a means of “translating the unseen or unseeable into something that looks like a picture of something we could never see” (Mitchell, 2005, p. 260). The intelligibility of the use of the visual in watchful politics is disrupted by this sense of the photograph. The visual in the war on terror is offered as a means of translating the unseen or unseeable into an actual picture of something, or someone, we think we see all of the time – an Arab, an immigrant, and so on. Mitchell’s picture is of “something we could never see”, someone always beyond our reach. A critical ethics of vision, as Mieke Bal (2005, p. 160) has articulated it, would engage a “commitment to look”, an act of “seeing through”, not in the sense of deciphering some deeper truth behind the surface of an image but in “halting to see the overlooked”.
Installation artworks have become particularly significant in this pausing to see what is overlooked. Perhaps because they can be physically positioned to make us stop, walk around them, or see them in our peripheral vision as we pass by, installations seem to demand a commitment to look through and to see what is overlooked. What is interesting to me about the potential for installation artworks to open up the unseeable/undecidable is the way that they intervene in our mundane routines and trivial everyday journeys. Where the watchful politics of Highway Watch, for example, use the visuality of daily routines in order to settle out the norm and identify suspicious deviation, artists are using the everyday visuality of the highway precisely to unsettle what comes to be normalized.
Tijuana artist Marcos Ramirez sought to make his 2003 installation “Road to Perdition” a daily intervention on the highways of Reading, Pennsylvania. Commissioned in the “Mexico Illuminated” series, Ramirez’s work was to be mounted on a highway sign on the Bingaman Street bridge. “Road to Perdition” lists eight cities bombed by the United States, alongside their distances from Reading and the dates of the bombings. The sign was wholly ambiguous: it looked like something we see every day, and yet it also looked like something we could never see. The raw data of the distances to sites of violence is a calculation that is always already assumed but rarely seen. In the event, Ramirez was refused permission to site his work on the highway, and exhibited a projected image of the Bingaman billboard at Albright College. Ramirez’s work, rather like Haggis’s Crash, represents a refusal to stand behind the lines that segregate the glass-walled privatized visualities of the highway from the violent actions that make them possible.

Conclusions: visuality before justice

I do not wish to overemphasize the novelty of the watchful politics in the war on terror. Of course, in many ways, the categories and judgements of the vigilant visuality only rearticulate prejudicial classifications long in operation. It does seem to me, though, that vigilant visualities are extending, via algorithmic screening, to incorporate ever more prosaic and playful uses of the screen in our daily lives – and this is deserving of urgent critical attention.
At the time of writing, the DHS has announced the primary bidders for the Secure Border Initiative (SBI) – a “virtual fence” connecting the USA’s land border guards to technological infrastructures – including “smart weapons” manufacturers Raytheon and Lockheed Martin, alongside Swedish telecommunications company Ericsson. The SBI illustrates precisely what I have been concerned with here: it conceals the raw stereotyping of what is assumed can be seen in a cloak of high-tech, data-driven precaution. A close reading of the DHS’s (2006) outline for contractors reveals the vision of a system that can “classify the level of threat prior to the point of interdiction/ encounter by law enforcement personnel”. At the point that the border guard appears to decide, the calculation is already made within the algorithm. The data-led watchful politics of the SBI, then, seek the visualization of a threat in advance of law, prior to justice in the strict sense.
It is this visualization before justice that is so integral to the watchful politics I have discussed here. The etymological sense of prejudice – praejudicare, or “to judge before” – embodies the anticipatory logic of judgement before justice, or prior judgment. The screening of Highway Watch and the mobile-phone images from the London bombings, then, are given the appearance of distantiated, data-led objective knowledges, a basis for anticipatory governance and prejudgement. Yet, the suspicions and stereotypes essential to vigilant visualities are, strictly, pre-judicial, pre-justice. The call for images to support justice or to provide evidence for decisions is, in fact, a call for pre-justice, even prejudice.
If Ericsson’s mobile screens become the visuality of the SBI, will they institute only the windowed and distantiated seeing of watchful politics? “Sight, no matter how disincarnated it may appear”, writes Martin Jay (2002, p. 276), “never loses its links with the flesh in which it is embedded”. Windowed seeing is the sovereign fiction of securability; it is illusory. Far from two different incarnations of the screen, the vigilant visuality and the screenic touch are always co-present. Where the technologies of vigilant visuality apparently fail, we see momentarily the concealed logic of pre-justice and feel momentarily the touch of the Other. As Judith Butler (2004, p. 24) writes:


The vigilant mode of visuality demands that we somehow remain intact, that we can be are not ever only “shields” or windshields, but always also screens or windscreens, in which we see multiple reflections and shifting views, through which we miss the touch of others. To keep open this alternative visuality, even and especially in the face of the drive to recognize, identify and secure, is to expose the incompleteness and contingency of the vigilant homeland security programmes.


Agamben, Giorgio, State of Exception, Chicago (IL): University of Chicago Press, 2005.

Amoore, Louise, «Biometric Borders: Governing Mobilities in the War on Terror», In: Political Geography, vol. 25, no. 3, 2006, p. 336-351.

Amoore, Louise and De Goede, Marieke, «Governance, Risk and Dataveillance in the War on Terror», In: Crime, Law and Social Change, vol. 43, no. 2, 2005, p.149-173.

Bal, Mieke, «Looking at Love», In: Diacritics, vol. 27, no. 1, 1997, pp. 59-72.

Bal, Mieke, «Visual Essentialism and the Object of Visual Culture», In: Journal of Visual Culture, vol. 2, no. 1, 2003, p. 5-32.

Bal, Mieke, «The Commitment to Look», In: Journal of Visual Culture, vol. 4, no. 2, 2005, p. 145-162.

Bigo, Didier, «The Móbius Ribbon of Internal and External Security(ies)», In:
Albert, Mathias, Jacobson, David and Lapid, Yosef (ed.), Identities, Borders, Orders, Minneapolis (MN): University of Minnesota Press, 2001, p. 91-116.

Boston Herald, «Trucker’s Eyes Are on the Road – and You!», In: Boston Herald, 9 July 2004.

Butler, Judith, Precarious Life, London: Verso, 2004.

Campbell, David, Writing Security: United States’ Foreign Policy and the Politics of Identity, Minneapolis (MN): University of Minnesota Press, 1998.

Cooly, Heidi Rae, «It’s All About the Fit: The Hand, the Mobile Screenic Device and Tactile Vision», In: Journal of Visual Culture, vol. 3, no. 2, 2004, p.133-155.

Crary, Jonathan, Techniques of the Observer: On Vision and Modernity in the 19th Century, Cambridge (MA): MIT Press, 1990.

Daston, Lorraine and Galison, Peter, «The Image of Objectivity», In: Representations, vol. 40, 1992, p. 81-128.

De Goede, Marieke, «The War on Terrorist Finance as Biopolitical Practice», paper presented at workshop on «Governing by Risk in the War on Terror», San Diego (CA): International Studies Association, 21 March 2006.

Department of Homeland Security (DHS), «TSA Teams up with American Trucking Associations», press release, Washington (DC), 23 March 2004a.

Department of Homeland Security (DHS), «Homeland Security from the Citizens’ Perspective», Council for Excellence in Government, Washington (DC), 2004b.

Department of Homeland Security (DHS), «Solicitation, Offer and Award», HSBP1006R0463, Washington (DC), 2006.

Der Derian, James, «The War of Networks», In: Theory & Event, vol. 5, no. 4, 2002, available at: <http://muse.jhu.edu/journals/theory_and_event/v005/5.4derderian.html>.

Derrida, Jacques, «Force of Law: The Mystical Foundation of Authority», In: Cornell, Drucilla, Rosenfeld, Michel and Carlson, David Gray (ed.), Deconstruction and the Possibility of Justice, New York: Routledge, 1992.

Derrida, Jacques (in conversation with Richard Beardsworth), «Nietzsche and the Machine», In: Journal of Nietzsche Studies, no. 7, 1994, p. 7-65.

Edkins, Jenny, Trauma and the Memory of Politics, Cambridge: Cambridge University Press, 2003.

Edkins, Jenny and Pinfat, Véronique, «Life, Power, Resistance», In:  Edkins, Jenny, Pinfat, Véronique and Shapiro, Michael (ed.), Sovereign Lives: Power in Global Politics, London: Routledge, 2004, p. 1-22.

Ewald, FranQois, «Insurance and Risk», In: Burchell, Graham, Gordon, Colin and Miller, Peter (ed.), The Foucault Effect: Studies in Governmentality, London: Harvester Wheatsheaf, 1991, p. 197-210.

Ewald, FranQois, «The Return of Descartes’s Malicious Demon: An Outline of a Philosophy of Precaution», In Baker, Tom and Simon, Jonathan (ed.), Embracing Risk, Chicago (IL): Chicago University Press, 2002, p. 273-302.

Foucault, Michel, Abnormal: Lectures at the Colege de France 1974–1975, New York: Picador, 2003 (1975).

Friedberg, Anne, «Urban Mobility and Cinematic Visuality», In: Journal of Visual Culture, vol. 1, no. 2, 2002, p. 183-204.

Friedberg, Anne, «Virilio’s Screen», In: Journal of Visual Culture, vol. 3, no. 2, 2004, p. 183-193.

Haggis, Paul, «On the Origins of Crash», 2006, available at: <http://www.movienet.com /crash.html> [accessed March 2006].

Haggis, Paul and Moresco, Bobby, Crash, Santa Monica (CA)/New York/Vancouver: Lions Gate Productions, 2005.

Highway Watch, «FAQ», 2005, available at: <http://www.highwaywatch.com/eweb/DynamicPage.aspx?site=HWWHome&webcode=FAQ> [accessed January 2008].

Jay, Martin, «Cultural Relativism and the Visual Turn», In: Journal of Visual Culture, vol. 1, no. 3, 2002, p. 267-278.

Mitchell, W. J. T., «Showing Seeing: A Critique of Visual Culture», In: Journal of Visual Culture, vol. 1, no. 2, 2002, p. 165-181.

Mitchell, W. J. T., «There Are No Visual Media», In: Journal of Visual Culture, vol. 4, no. 2, 2005, p. 257-266.

Shapiro, Michael, Violent Cartographies: Mapping Cultures of War, Minneapolis (MN): University of Minnesota Press, 1997.

Silverman, Kaja, The Threshold of the Visible World, Routledge, New York 1996.

Sinclair, Iain, «The Theatre of the City», In: The Guardian, 14 July 2005.

Sparke, Matthew, In the Space of Theory: Postfoundational Geographies of the Nation-State, Minneapolis (MN): University of Minnesota Press, 2005.


1. This text was previously published in Security Dialogue, vol. 38, no. 2, June 2007.