Security technology,urban prototyping,and the politics of failure

Introduction

In cities across the world, we see an enthusiasm among municipal governments, police forces, and citizens for introducing new technologies in the fight against crime or the war on terror: from digital sensors and algorithms to the ubiquitous WhatsApp neighborhood groups. It is important to examine this ‘technological turn’ in urban security and policing critically. While urban security issues tend to be complex social problems, such a turn to technologies offers the illusion of a simple fix that is much easier to implement than long-term social programs. It is always more difficult to reform a police force or to sustain a neighborhood watch than it is to install CCTV cameras or biometric access control. The proponents of such security technologies often present them as neutral, impartial, and cost-efficient, suggesting that they are less prone to error or corruption than security and policing strategies in which human agents play a more visible role. Yet they evidently have their own politics: technology is both produced in specific political contexts and productive of other politics (Winner, 1980).

Recent research in critical security studies has emphasized the role of security technologies in advancing a range of – frequently illiberal – political projects. Digital technologies have come under particular scrutiny. Where governments and corporations promise that algorithms, smart cameras, and other hi-tech innovations are the most effective and efficient way to make nations and cities safer, critical researchers and activists warn that these same technologies enable unprecedented forms of surveillance and control, while their proprietary nature inhibits democratic oversight (see, for example, Ferguson, 2016). We largely concur with such critiques and hold that close academic and activist scrutiny of the technopolitics of security is essential. In this article, however, we propose a shift in analytical focus, towards the many ways in which security technologies – especially when they are actually implemented – fail to live up to both their promise and their threat.

In this article, then, we focus on the politics of technological failure. Drawing on approaches from science and technology studies, we explore these politics through a process-oriented approach that attends to the everyday interactions between technologies and their social and built environments. Such an engagement with the social life of technology over a longer period of time can highlight how and why technologies fail to live up to their alleged potential, to achieve their stated goals. It also pushes us to reflect on what political outcomes sociotechnical ‘failure’ enables. Our analysis suggests that these outcomes may involve the blurring of concepts of failure and success, as ‘prototyping’ and experimentation – rather than, for instance, accountability – become an increasingly powerful logic of urban governance. Drawing on a combination of media analysis and ethnographic research in urban Jamaica, we analyze the promotion and/or introduction of three distinct technological devices: ¹ a communication channel connecting police and private security, a public–private CCTV network, and a smart electricity grid. While all these cases represent technology-focused security interventions, they reflect a diverse range of technologies, governance actors, and publics. In all three instances, technology is promoted as enabling an efficient response to the multiple challenges that urban governance actors face in addressing security issues, from public and private actors for whom security is their core business, to a private electricity provider facing the combined challenges of urban violence and financial losses. This inclusion of governance actors beyond the state – reflecting ongoing processes of security privatization and pluralization ² – enables an analysis of the various interests that technological interventions may serve within the broader urban security landscape.

Our analysis emphasizes the politics of the narratives that emerged around the introduction of each device, highlighting the performances and contestations of sovereignty at stake in such narratives. Beyond a discursive analysis, we seek to complicate such narratives by tracing the fate of these technologies over time and in practice. For each device, drawing on interviews, observations, and media reports, ³ we explore how and why it failed to deliver on its initial promise, and seek to unpack the politics of these failures by asking which new political or economic pathways – or successes – are enabled by apparent failures.

We start the article with a reflection on the tendency towards alarmist narratives within critical studies of security technology and outline a sociotechnical approach that focuses on technological failures and their politics. Next, a background section presents a brief outline of security governance challenges and the turn to technology in urban Jamaica. This is followed by an analysis of the three security-related technologies. We conclude by connecting this analysis to a discussion of the ways in which failure has been reconceptualized in urban governance.

A sociotechnical approach to urban security politics

Critical studies of security technology have tended to be dominated by political science and international relations research, focusing on national security, border regimes, and anti-terrorist measures. Within this field of study, much attention has gone out to hi-tech digital surveillance, a multi-sensory panopticon that combines military and corporate logics to track our every move while reinforcing discriminatory logics (Bigo, 2006; Muller, 2010; Shaw, 2016). As this security technology also creeps into urban policing and smart city strategies, we see critical urban studies expressing a related concern, focused more specifically on the militarization of cities and discriminatory urban policing (see, for example, Graham, 2011; see also Ferguson, 2019).

Academic and activist engagements with security technology have presented strong critiques of what Evgeny Morozov terms ‘techno-solutionism’. Discussing the recent embrace of ‘big data’ and the digital quantification of behavior as the solution to a broad range of social problems, Morozov (2013: 5) describes how ‘complex social situations [are recast] either as neatly defined problems with definite, computable solutions or as transparent and self-evident processes that can be easily optimized – if only the right algorithms are in place!’ He calls this optimism ‘technological solutionism’, arguing that digital technology presents highly efficient but dehumanizing solutions for phenomena that are either not a problem at all or much too complex for an easy fix. He also points to the shift in power that results when digital technology is pitched as the solution to everything: technology companies, rather than elected governments, will shape the future of our cities and societies.

While these critiques are urgent, they can sometimes slide into alarmist narratives, which have multiple limitations. First, an overly dystopian view of security technology can be politically problematic: By suggesting that security technologies live up to their claims of efficiency and effectiveness, such narratives risk empowering the corporations and politicians that promote them. When alarmism slides into fatalism, it also risks disempowering the citizens who interact with these technologies, paralyzing us rather than spurring us to action. Representing digital technologies as all-encompassing or terrifyingly successful in their efficacy, then, runs the risk of amplifying the techno-solutionism of corporations and governments. Second, totalizing narratives of technological doom – whether attributed to the surveillance state or robot overlords – are frequently too universalist and imprecise to help us identify effective action. Such narratives are not always based on in-depth empirical analyses of technologies following their actual implementation and everyday use. In fact, many studies of security technologies that are actually in place in the real world find that the technological devices involved do not live up to their promise at all (e.g. Adelman, 2018; Andersson, 2016; Magnet, 2011). It is tempting to conclude that many technologies of urban policing and security are complete failures.

Conceptually speaking, both technological solutionism and its critical counterparts risk invoking a technological determinism that overemphasizes the role of security technology in shaping society, without necessarily developing detailed empirical accounts of what technologies actually end up doing (or not doing) in practice. ⁴ We suggest that making academic knowledge of technology actionable requires more precise accounts of how technology is utilized, appropriated, and adapted in specific contexts. Such accounts are well served by a sociotechnical approach: drawing on insights from science and technology studies scholars such as Wiebe Bijker, Stephen Woolgar, and Bruno Latour, such an approach understands the political effects of technologies as emerging in non-linear ways from dynamic and contingent relations between humans and non-human entities. While such a sociotechnical approach is increasingly applied to the development and implementation of security technologies, not least in this journal (see, for example, Bellanova et al., 2020; Bourne et al., 2015; Jeandesboz, 2016), overall, authors working in this tradition have spent less time unpacking the political and economic logics of failure and success as they emerge in practice.

Our interest here is not so much to distinguish between ‘good’/successful and ‘bad’/failed technology, but to understand what strategic work failure does, for whom, and how. Rather than understanding failure as a condition that can be identified objectively, we approach it as a contingent outcome of ongoing, contested processes of valuation. Technologies possess ‘interpretative flexibility’ (Pinch and Bijker, 1984: 409): the meaning of an artefact is unstable and varies across its relations with different social actors, extending far beyond the intentions of its designers. ⁵ For example, a security technology that its engineers consider a success in terms of the learning it enables may well end up being viewed as a total failure by the security officers who use it in practice (Lisle, 2018: 893). Accordingly, we understand technological failure and success not so much as objectively identifiable, polarized states, but as sociotechnical constructs that not only derive from the position of evaluators but also often serve to legitimize their situated interests. The labels of ‘failure’ and ‘success’ can function as core elements within larger political narratives and performances (see Fincham, 2002: 7). ⁶ Indeed, in line with the focus of this special issue on technopolitics – that is, in scrutinizing ‘the ability of competing [security] actors to envision and enact political goals through the support of technical artefacts’ (Gagliardone, 2014: 3) – we seek to interrogate the political work that the sociotechnical constructs of failure and success do in the world.

The technopolitics of failure and the logic of urban prototyping

In our analysis of the technopolitics of urban security and policing, then, we propose avoiding technological determinism by dwelling a little longer on the various failures of urban security technology. We are interested in discussing, first, what is understood as failing and why it is seen as such. In so doing, we also take note of Peter Adey and Ben Anderson’s (2012: 113) call to pay more attention to how ‘apparatuses of security fall apart, fail, are disrupted, or are held together’, while placing our emphasis on the role of specific technological devices within such wider apparatuses. There are various ways in which security technologies can fail, which emerge from the inevitable interaction of those technologies with their social and material surroundings: there are failures related to subversion, hacking or dual use; those caused by people’s capacity to bypass or circumvent the technology; but also failures related to less intentional, more mundane types of social or material breakdown connected to off-script uses and disrepair. Getting security technologies to ‘work’ requires various type of labor (Vukov and Sheller, 2013; see also Graham and Thrift, 2007): the labor of various security and technology professionals, from software designers to police officers and security guards and their supervisors. It often also involves the labor of those being policed or ‘secured’; if they actively circumvent or subvert the technology, it is also less likely to work.

Specifically, though, in thinking about how and why different types of failure emerge, we concentrate on the politics of technological failure. Just noting that a security technology has failed in practice is evidently not enough – we have to ask for whom it failed, for whom it might still have been a success, and what new political or economic possibilities it may have established. We are interested in scrutinizing the political dimension of what Debbie Lisle (2018: 891) calls the agency of failure, tracing ‘the work that [failure] does in the world to guide practice, change behaviour, recalibrate relations, alter materialities and mobilize futures’. As our analysis of Jamaica’s urban security landscape shows, some of the futures made newly possible through technological failure involve easily recognizable political and economic benefits to individuals or organizations. In addition, however, in the process from idea to design to (non-)implementation to reconceptualization, the various technologies we discuss in this article also work as de facto (rather than formal) ‘prototypes’ for how to govern urban spaces and populations.

The logic of prototyping, which has become increasingly popular in urban governance, blurs longstanding understandings of failure and success. As Alberto Corsín Jiménez (2014: 381) notes, ‘An important feature of prototyping . . . is the incorporation of failure as a legitimate and very often empirical realisation.’ Within urban governance, failure is increasingly anticipated as a natural outcome of experimentation, while ‘design thinking’ replaces older logics of planning. New technologies, we suggest, may act less as ‘solutions’ or ‘fixes’, and more as experimental forms that facilitate learning by doing, a process in which trial-and-error is taken for granted rather than necessarily viewed as a waste of money or cause for political scandal. ⁷ However, this sociotechnical process should be scrutinized carefully in terms of the political relations it establishes or consolidates. As Martin Tironi (2019: 504) suggests in his analysis of urban prototyping as a political device, ‘One distinctive element of the experiments undertaken by civic authorities and urban agencies is that the design, dramaturgy and mise en scène of these interventions are just as important as the results of the experience or even more so.’ Analyzing these interventions as performances involves taking the materiality of the security technologies seriously, while also attending to the political narratives – of problems and solutions, authority and responsibility, success and failure – that surround them as they enter and exit the stage of urban security governance.

Below, drawing on our research in urban Jamaica, we discuss a number of technological failures in urban security and policing interventions. To provide the context needed to understand the political work that various ‘failed’ security technologies do, we first provide a brief background to urban security governance and the technological turn in Jamaica.

Security governance and technology in urban Jamaica

Jamaica suffers high levels of crime and violence, connected to a history of political turmoil, high levels of inequality, and other social factors. Homicide rates increased sharply in the second half of the 20th century, climbing from 17.6 homicides per 100,000 population in 1976 to 43 in 2001 (Harriott, 2003: 7). Over the past two decades, this rate appears to have stabilized somewhat, hovering around 47 per 100,000 in 2019, still three times higher than the average for Latin America and the Caribbean. ⁸ Jamaica’s public security structure faces multiple challenges, with many citizens viewing formal institutions, and especially the police force, as ineffective, abusive, and corrupt. This has contributed to a crisis of public safety, manifest not only in the high levels of violence and the ineffective response of public institutions, but also in the rise of multiple security providers. In addition to the state’s deployment of both the Jamaica Constabulary Force (JCF) and the Jamaica Defence Force (JDF) in urban policing, residents have turned to a growing number of formal and informal non-state security actors.

The capital of Kingston, where much of the violent crime is concentrated, with a homicide rate of 169 per 100,000 population in 2018, ⁹ represents an especially heterogeneous landscape of security governance. Corporate actors and residents of wealthier neighborhoods in uptown Kingston tend to turn to private security companies for protection, relying on high walls, electronic alarm systems, and armed-response guards for security. In contrast, in low-income neighborhoods in downtown Kingston, so-called dons – community leaders often involved in criminal activities – are often central to more informal systems of non-state security provision (Campbell, 2020). Over several decades, these dons have increasingly assumed an informal governance role in what are known as garrison communities, ‘geographically discrete, fortressed urban areas marked by poverty, gang violence, political manipulation and confrontational relationships with law enforcement institutions’ (Mullings, 2019: 141). In these areas, residents often have no choice but to turn to criminal organizations for security.

Within this fragmented security governance landscape, the most emblematic and mediatized attempt of the state to (re-)establish a primary role in security governance was the 2010 ‘Tivoli Incursion’. This security operation, during which the Jamaican military and police killed 69 citizens, concentrated on the garrison community of Tivoli Gardens and was aimed at capturing and extraditing Christopher ‘Dudus’ Coke, the country’s most powerful don. The 2010 security operation was accompanied by a state of emergency in sections of the country that lasted more than a year, during which curfews and other ‘anti-gang’ measures were implemented in low-income areas. These forms of exceptional policing – resembling urban ‘pacification’ efforts in Rio de Janeiro and other cities – were expanded in subsequent years. In 2018, the Jamaican government established new states of emergency in the most urbanized parts of the island and introduced Zones of Special Operations (ZOSOs), special security zones where curfews enforced by joint military–police operations were to be combined with community development efforts.

In combatting crime, state and commercial security actors have frequently highlighted the potential of incorporating new technologies (including new forms of digital surveillance) in enhancing their capacity to detect, deter, apprehend, or prosecute criminals and to work together effectively across public–private divides. This is the case both in low-income neighborhoods and in wealthier urban areas. The Ministry of National Security’s recent Five-Pillar Strategy for Crime Prevention and Citizen Security places particular importance on the acquisition and use of technology, asserting that ‘in engineering a seamless, effective, all-of-Government anti-crime machine, few factors are as important as the procurement, and full deployment, of appropriate technologies’ (Ministry of National Security, 2017: 106). More recently, the commissioner of police, Major General Antony Anderson, proclaimed that the JCF was ‘in the midst of an aggressive technology-based drive to create a modern, connected and highly efficient police force’ (Smith, 2020). In a parallel move, private security companies have also been expanding the range of technologies they offer to clients, from CCTV surveillance and smart home systems to biometric access control and GPS-based vehicle trackers (see, for example, McKinson 2017).

What technopolitics can we discern in the narratives surrounding specific technological devices? To what extent do these various technological devices achieve their stated goals, and what politics are at work when these security interventions end in sociotechnical failure? Below, we address these questions in an analysis of three distinct devices. First, we discuss attempts to introduce a radio communication channel connecting Jamaica’s private security guards to the police. ¹⁰ This analogue communication technology was supposed to act as what is sometimes called a ‘force multiplier’, extending the reach of the JCF’s eyes and ears to the many sites surveilled by private security guards. In addition, the communication channel was supposed to improve private security access to the JCF, enhancing guards’ ability to respond to suspicious activities. However, a pervasive social taboo on ‘informing’ – sharing information on criminal activities with the police – meant that this technological alignment of guards with the police presented a serious threat to the lives of the guards.

The second case, the JamaicaEye public–private CCTV network, was a follow-up to these attempts. Again, a technical fix was proposed to enhance ‘interoperability’ between public and private security systems, expanding surveillance while also supporting criminal prosecutions by providing a visual form of legal evidence in the absence of witness testimonies. This time, however, the technology sought to connect security cameras operated by the police and military to those owned by private citizens. While the system is still being expanded, early reports suggest that technical breakdowns and a lack of police follow-up are severely hampering JamaicaEye’s impact.

The third device we analyze is that of the smart grid developed by Jamaica’s electricity utility, the Jamaica Public Service Company Limited (JPS). ¹¹ Following the 2010 Tivoli Incursion, the JPS intensified its efforts to ‘regularize’ electricity consumption in low-income, high-crime neighborhoods (so-called ‘red zones’), ¹² where electricity theft is high and JPS staff face the risk of aggression when checking meters or disconnecting illegal ‘throw-ups’. In this context, smart metering – by enabling the JPS to monitor and control its infrastructure at a distance – functions as a security technology: it is seen as capable of protecting company revenue, infrastructure, and staff. However, communication signal failures and continued on-site tampering mean that the smart grid has had little long-lasting impact on electricity theft.

From technological fix to failure

We analyze the social life of these three technologies over time by tracing three phases. First, we discuss the narratives that presented each technology as a fix to a specific security challenge: we ask how security-related problems were framed to match a technological fix, while exploring which political claims and economic interests this techno-solutionism served. Next, we unpack the social and technical aspects underlying the failure of each technological fix in practice. Finally, we explore the politics of these failures, suggesting the ways in which each failure might still signal a form of success, albeit not in terms of the originally stated objectives, and how this connects to an emergent logic of prototyping in urban security governance.

Promoting a technological fix: Framing problems and solutions

In each of our three cases, governance actors constructed a problem narrative that proposed a specific technological solution to security problems, with both problem and solution connected to political claims. In the first case, this narrative put forward a shared communication channel as the technological fix for the problem of uncoordinated public and private security efforts. In Kingston, the number of private security guards is about twice that of JCF officers. Although formally these guards are responsible only for the safety of a specific property or the people within it, they frequently witness crimes or security threats in public space. At multiple points in time, Jamaica’s Ministry of National Security sought to tap into these eyes and ears on the street. One important attempt involved the use of a dedicated radio communications channel, repurposing an existing type of technology as an anti-crime tool by connecting security guards to the JCF. This channel would allow guards who witnessed anything while on duty to access the police directly through their own radios, without going through the notoriously dysfunctional emergency telephone number 119. The idea was that the radios would seamlessly connect the thousands of security guards dispersed throughout Kingston to the JCF, translating their observations into actionable police information.

The problem, then, was constructed as a lack of communication and coordination between public and private security actors. In this framing, the pluralization of policing was not indicative of institutional failure. Rather, Kingston’s fragmented landscape of security provision was presented as an opportunity: it could be transformed – through enhanced communication technology – into a collaborative effort to tackle crime and violence. This vision has been central to successive security policies, which have underlined the need for public–private partnerships, connecting state security actors to private security companies but also heavily stressing citizen participation in policing. Such a narrative centered on the need for partnerships allows the police and military to perform claims to state sovereignty, coordinating rather than monopolizing security provision while shifting responsibility across a broader network of actors.

The second case, which can be understood as something of a follow-up to the first, involves a comparable promotion of technological integration of state and non-state security efforts. Here, the state security forces proposed the integration of private CCTV feeds – from cameras facing public spaces – into a networked system called JamaicaEye, with the police monitoring the video feeds but the military overseeing the central network. Again, this narrative promoted the idea that technologically mediated information-sharing would herald a major improvement in crime-fighting. In JamaicaEye, however, the security-related problem was framed less as a disconnect between commercial and state security providers and more as the need for citizens to be active participants in crime prevention programs and security provision. As the project’s website proclaimed, ‘Jamaica, YOU can help. Connect your camera system to the JamaicaEye national CCTV network and help to make Jamaica safer.’ ¹³

Successive national security policies and public campaigns have placed a major emphasis on citizen participation. An early iteration is found in the 2007 National Security Policy, which states that ‘there needs to be a radical change in the way in which everyone views his or her responsibilities for national and community safety and security. There needs to be the recognition that “security is everybody’s business”, and not the sole responsibility of the police and other law enforcement agencies’ (Ministry of National Security, 2007: 38). A follow-up policy similarly stressed this responsibility: ‘In keeping with the initiative to develop a stronger partnership between citizens, civil society and all Government organizations involved in delivering security services, it is important for all members of the public to understand the critical role that they are required to play in helping to make Jamaica a more safe and secure place to live and visit’ (Ministry of National Security, 2013: 93). This narrative reflects a broader global tendency towards neoliberal responsibilization in security governance. Where this responsibilization has often focused on the importance of citizens sharing information with the police and testifying against criminals in court, JamaicaEye provides a technological solution to a widespread unwillingness to do so by removing humans from the equation. Calling the cameras a ‘force multiplier’, Minister of National Security Horace Chang announced: ‘It will have the requisite monitoring systems to collect and store high-quality footage that can be used as evidence in our courts. . . . This investment in JamaicaEye will give the police the appropriate counter-strategy to intercept and curtail the movements of criminal gang members, the dons and their facilitators.’ ¹⁴ The CCTV network draws on citizens’ resources to provide the police and courts with evidence of crimes, but shields private citizens from the risks involved in this process of informing and testifying.

In the third case, the problem–solution narrative was framed by the JPS, a public–private utility company rather than a government agency. Here, the problem was commercial and security risks, the two seen as intertwined in low-income, high-crime ‘red zones’ (Interview 1). The commercial risk involved high rates of ‘non-technical losses’, that is, electricity theft: some 18% of the net generation in 2018, but over 70% in most of these zones (JPS, 2019). A JPS official stated that when its agents attempt to cut off illegal connections, they face significant aggression and are often unable to visit these areas without police accompaniment (Interview 2). In addition to residents protesting disconnection, dons and other criminal actors play an important role in deterring JPS from its attempts to reduce theft. The technological fix for this commercial-security challenge was the smart grid, in the form of the ‘Residential Automated Metering Infrastructure’ (RAMI). The security governance aspect of electricity management was exemplified by the first installation of the smart metering in Tivoli Gardens, after the 2010 security operation (Radio Jamaica News, 2011). The reassertion of state sovereignty through this operation and the intensified policing strategies that followed offered the company an opportunity to regularize electricity access, dismantling illegal connections and registering customers. ¹⁵ Since 2010, JPS has used the RAMI system in ‘red zones’ as a technical instrument to regularize illegal connections. Not only did the presence of the state security forces and the initial destabilization of the dons’ governance following Dudus’s extradition reduce the security risks of JPS’s attempts to recover commercial losses, but the installation of the RAMI system was also seen as a preventive technology to manage current and future security and commercial risks. ¹⁶

The RAMI system – which involves integrated digital meters, data management systems, and two-way communication networks – enables important changes in the management of JPS’s relations with its customers. Its features allow JPS to remotely execute a number of important operations that previously required a physical presence, such as measuring electricity consumption levels, connecting and disconnecting customers, detecting tampering, and more accurately identifying variations in individual consumption that indicate electricity fraud. This automation of the different operations minimizes the human interaction between customers and (often subcontracted) technicians and electricians that JPS associates with both corruption ¹⁷ and potentially violent conflicts surrounding disconnections. The RAMI system also features an anti-theft design, with meters installed in locked cabinets located on electric poles rather than in customers’ homes, and medium-voltage distribution lines replacing low-voltage (secondary) lines to discourage tampering by increasing the risk of electrocution (see Figure 1). Alarm systems that detect cabinet tampering shut down all meters in one cabinet, a punitive measure aimed at encouraging horizontal control between customers. These combined features were all aimed at replacing or mediating human interventions and contact with automated technology.

OPEN IN VIEWER

Figure 1.

The RAMI system (photograph by Francesca Pilo’).

Conclusion

Many governments, companies, and citizens are disappointed by the ability of public and private security forces to act effectively. Often, human fallibility – inattention, lack of training, corruption – is seen as the cause for this lack of effectiveness. The assumption that technology eliminates this human element underlies the optimistic turn to quick-fix solutions, with hopes and dreams of safer cities projected on a range of hi-tech and low-tech devices. This optimistic ‘techno-solutionism’ relies on an understanding of technology as neutral, a perspective that is increasingly debunked by more critical accounts, which signal the various threats these technologies entail to social and political rights. In this article, we have proposed a slightly different approach to the politics of technology, by focusing on technological failure and adopting a sociotechnical approach that follows a technology from design to roll-out and (non-)implementation, including its everyday interactions with the urban environments, in order to understand its on-the-ground political effects.

In pursuing such an approach, we take seriously Peter Adey and Ben Anderson’s (2012: 100) call to understand ‘the workings of security apparatuses beyond an exclusive concern with the logics that animate [security] and their apparent success’. Here, we have sought to underline the political work of specific technical elements within such apparatuses, considering their failure in relation to their original animating logics, but still seeking to identify a broader rationality of governance at work – that of prototyping. In a world of experimentation, technological failure has become a mundane event rather than a political problem, much less a spectacle. Failure has come to be seen as productive, and boundaries between failure and success have become blurred within political discourse. Public policymaking in cities increasingly relies on prototyping – a logic previously associated with design – in developing a model of urban governance characterized by flexibility, provisionality, and anticipation. Urban governance, including security governance, involves prefigurative experiments focused on technological innovations and adaptations, which institutional actors – whether the police or a commercial organization – hope will generate a new and improved city of the future, even as they anticipate their failure. Rather than being conceived in terms of delivering measurable outcomes or meeting preset targets, the ‘success’ of technologies or policies may be reframed in terms of increased learning opportunities or the habituation and support of various constituencies. In short, to evaluate the success of security technologies in terms of the results proponents originally claimed they would deliver may be to miss one of the main points of prototyping. Concrete, quantifiable results are less important than attempts to inscribe specific ways of administering urban life. Analyzing ‘failing’ technologies through the lens of prototyping enables an understanding of technopolitics that goes beyond current approaches, while forming a useful counterbalance to studies that highlight other types of technologically assisted anticipatory politics.