"First Monday is one of the first openly accessible, peer–reviewed journals solely devoted to research about the Internet. First Monday has published 1,747 papers in 263 issues, written by 2,434 different authors, over the past 21 years. No subscription fees, no submission fees, no advertisements, no fundraisers, no walls."From First Monday:
Volume 20, Number 7 - 6 July 2015
By Jathan Sadowski and Frank Pasquale
About the authors
Jathan Sadowski is a Ph.D. candidate in the “Human and Social Dimensions of Science and Technology,” in the Consortium for Science, Policy & Outcomes at Arizona State University. His research mostly focuses on social theory/justice and political economy of information-communication technology. He is currently writing a dissertation on the socio-politics of “smart cities”.
E-mail: Jathan [dot] Sadowski [at] asu [dot] edu
Frank Pasquale is a Professor of Law at University of Maryland’s Francis King Carey School of Law. His research addresses the challenges posed to information law by rapidly changing technology, particularly in the health care, Internet, and finance industries. He recently published The black box society: The secret algorithms that control money and information (Harvard University Press, 2015), which develops a social theory of reputation, search, and finance.
E-mail: fpasquale [at] law [dot] umaryland [dot] edu
There is a certain allure to the idea that cities allow a person to both feel at home and like a stranger in the same place. That one can know the streets and shops, avenues and alleys, while also going days without being recognized. But as elites fill cities with “smart” technologies — turning them into platforms for the “Internet of Things” (IoT): sensors and computation embedded within physical objects that then connect, communicate, and/or transmit information with or between each other through the Internet — there is little escape from a seamless web of surveillance and power. This paper will outline a social theory of the “smart city” by developing our Deleuzian concept of the “spectrum of control.” We present two illustrative examples: biometric surveillance as a form of monitoring, and automated policing as a particularly brutal and exacting form of manipulation. We conclude by offering normative guidelines for governance of the pervasive surveillance and control mechanisms that constitute an emerging critical infrastructure of the “smart city.”
Contents
I. Introduction
II. What is a smart city?
III. The ideology of the smart city
IV. Smart cities in societies of control
V. The soft power of biometric surveillance
VI. The hard power of policing technologies
VII. Cyborg urbanization, blurred boundaries
VIII. Taking back control
I. Introduction
There is a certain allure to the idea that cities allow a person to both feel at home and like a stranger in the same place. That one can know the streets and shops, avenues and alleys, while also going days without being recognized. But as government and corporate actors, often in close partnership with each other, fill cities with “smart” [1] technologies — turning them into platforms for the “Internet of Things” (IoT): sensors and computation embedded within physical objects that then connect, communicate, and/or transmit information with or between each other through the Internet — there is little escape from a seamless web of surveillance (cf., Hollands, 2008; Townsend, 2014; Neirotti, et al., 2014). Soon, for example, shoppers and viewers will be as “known” by a store or gallery as they are able to know it (Arnsdorf, 2010). Facial recognition software, or smartphone emanations, can project your identity, likely spending habits, and reputation: shoplifter or big spender, “Mortgage Woes” or “Boomer Barons” (to use actual categories from marketers) (Castle Press, 2010).
“Big data” is the new currency of commerce, but like money, some have far better terms of access to it than others. In finance, the average borrower must turn over detailed, personal records to receive a loan; the bank is under no parallel obligation, though, to explain its own internal decision-making in nearly as much detail (Pasquale, 2015). The same dynamics are emerging in the IoT: powerful entities centripetally attracting more data from their users, but denying access to users and regulators, even when very troubling data uses and breaches occur. It no longer makes sense to think of “the Internet” as a thing that one accesses via a computer. Not when the city itself is reimagined and reconstructed as a platform for and node within networked information-communication technologies (ICT).
Wired’s flagship article on the IoT asks, “Have you ever lost an object in your house and dreamed that you could just type a search for it, as you would for a wayward document on your hard drive?” (Wasik, 2013). Well you can now, we are assured, thanks to a startup called StickNFind Technologies that sells cheap, small, “sticker” sensors. Lose a child at the mall? “Smart fashion” RFID tags will keep him or her plugged into the network and tracked at all times. And why stop with kids when making sensor-laden sartorial choices? Before long your car, house, appliances, and every other part of your environment will be engaging in a constant stream of networked communication with each other. Taken at the urban scale, the city becomes a cocoon of connectivity that engulfs us — or, alternatively, it becomes a web that ensnares us — as smart technologies are integrated into our everyday lives. These technologies are billed as modes of finding, of wayfaring. They are technologies of search (when we apply them) and technologies of reputation (when used to evaluate us) (Pasquale, 2015). They map, categorize, and classify — and what could be more innocuous than mere information?
Calculating the costs and benefits of the innovation is a Sisyphean, and deeply ideological, task. Who knows what sinister or spectacular applications may emerge? Scenario analysis and planning could be a valuable alternative to cost-benefit studies (Verchick, 2010): these methods acknowledge the incommensurability of the gains in convenience, and losses of privacy, portended by the IoT. But corporate and government discourse on IoT has tended to marginalize the most important negative scenario analyses, downplaying them as paranoid projections. Technocrats distort policy evaluations of pervasive surveillance and control in urban environments. Moreover, their normative tools of evaluation, focusing on consumer and citizen “consent” to surveillance, are manipulable enough to embrace even the most disturbing technologies of control — such as drone-driven crowd control directed at protesters, or automobile loan technology that disables cars mere minutes after a payment is late — as expressions of democratic will and market rationality.
Technocrats’ convenient blindness to the most worrisome aspects of the “smart city” invites a more balanced theoretical response. We propose one such response that lays out the characteristics and consequences of a dominant socio-political logic that courses throughout and ties together many of the various practices and ideologies related to “smart cities.” We begin by providing a contextual overview of the “smart city,” building from the burgeoning analytical work on the topic. This leads into a critical introduction to the ideology of the “smart city,” focusing on the stated aspirations of some of its most notable corporate, governmental, and academic exponents. We then offer a Deleuzian alternative, outlining a social theory of the “smart city” in service to capital as a form of control (rather than emancipation) of its subject-citizens. Next, we present two illustrative examples along the resulting spectrum of control: biometric surveillance as a form of monitoring, and automated policing as a particularly brutal and exacting form of manipulation. Our penultimate section makes explicit the stakes of the deep integration of person–machine — city in our “post-digital-dualist era” (Jurgenson, 2012). And we end by offering some normative guidelines for governance of the pervasive surveillance and control mechanisms that constitute the emerging critical infrastructure of the “smart city.”
II. What is a smart city?
Globally — in terms of market valuation, expendable capital, technological development, and transformative influence — the smart city movement has been growing at a rapid pace. A 2013 report, released by the United Kingdom’s Department for Business, Innovation and Skills, estimated that “the global market for smart city solutions and the additional services required to deploy them [will] be $408 billion by 2020.” Linked to this growth is the exponential expansion of the IoT. According to commonly cited numbers from telecommunications giant Cisco, one of the major industries involved in the IoT and smart cities, billions of things are already connected — “over 12.5 billion devices in 2010 alone.” And they predict, “Some 25 billion devices will be connected by 2015, and 50 billion by 2020.” [2] Less conservative estimates place the smart city market into the trillion(s) of dollars over the next five to ten years, with the IoT market being worth even more. Case in point, IBM recently announced it would be investing US$3 billion over the next four years in creating a new IoT unit (Reuters, 2015) — an investment that will surely boost IBM’s already lucrative, multi-billion dollar “Smarter Planet” initiative. As an urban planning and governance movement, a lot of effort is expended on pushing and pulling “smartness” — the major corporate players work hard to push smartness as an ideal and to pull city leaders and investors into the smartness orbit. These corporations did not just stumble upon an existing market for which they could fill the needs. They, rather, have worked hard to create this market and to shape it in certain ways.
Yet, with this massive growth and capital investment, the label “smart city” is nebulous. There’s not a single definition that can be called up and applied anytime the label is invoked (Hollands, 2008). This ambiguity does a lot of work for smart city proponents and purveyors. The label is treated like a floating signifier that can change referents whenever needed. Allowing for a flexible, dynamic space in which to plug a variety of products, practices, and policies. Giving them discursive cover in case they need to distance themselves if something goes wrong or doesn’t deliver on a promise.
One important and constant characteristic of these different visions, however, is that they aim to evoke positive change and innovation — at least as the proponents see it — via digital ICT; essentially, building an IoT at the city-scale by installing networked objects throughout the urban environment (and even human bodies) for a wide range of different purposes. The typical examples used to illustrate an IoT-filled world are consumer products — like the ever-present smart fridge that tells the store when you need milk. But, Bruce Sterling argues, this is a “fairy tale,” instead “the genuine Internet of Things wants to invade that refrigerator, measure it, instrument it, monitor any interactions with it; it would cheerfully give away a fridge at cost” [3]. Restricting our focus to the consumer devices poses a red herring that keeps our attention at the surface level, halting analyses that should go beyond the alternating currents of absurd farce and gee whiz excitement. “These grand, world-scale [corporate] alliances did not form in order to sell the reader a smart refrigerator. Most of them would really like the reader to dwell in a ‘Smart City’ where they supply the ‘smartness’ on their own terms — and they’re not much concerned about the reader’s consent as a citizen” [4]. The smart city is not just a linearly scaled version of the smart home where all of our personal devices and domestic appliance are networked, automated, and good communicators. It is fundamentally about infrastructural and civic applications — the kind of things that constitute the techno-political ordering of society — and it is about the data and control those applications generate. To be sure, not all “smart cities” are implemented in the same way; we see three main types.
First, by far the most common ‘actually existing’ smart cities are those that are retrofitted and renovated with upgrades that transition current cities from dumb to smart. Many estimates place the number of cities and towns with smart initiatives into the tens or hundreds of thousands around the world. In these cases, “the smart city is assembled piecemeal, integrated awkwardly into existing configurations of urban governance and the built environment” [5]. Typically the underlying motivations are political economic, the result of an increasingly entrepreneurial form of urban governance that seeks to make the city into a center of (regionally or globally) competitive economic growth and activity (Harvey, 1989). Getting smart is the handy panacea for overcoming austerity, managing the urban system, and becoming an attractive place for capital to flow into — all by using “networked infrastructures to improve economic and political efficiency and enable social, cultural and urban development” [6]. Hence, smart initiatives promise to provide city leaders with the means necessary for achieving their entrepreneurial ends.
Second, there is the ‘shock therapy’ method — or, what we might call smart shock — wherein a city undergoes a quick, large-scale integration of ‘smart’ ideals, technologies, and policies into an existing landscape. There are not as yet any cities that have experienced a full shock, but rather there are examples where the smart city transition has occurred to a greater degree and at a more rapid pace than the typical retrofits. Perhaps the best example is the Intelligent Operations Center built in 2010 by IBM for the city of Rio de Janeiro, which “draws together data streams from thirty agencies, including traffic and public transport, municipal and utility services, emergency services, weather feeds, and information sent in by employees and the public via phone, Internet and radio, into a single data analytics centre” [7]. With this NASA-esque control room, the city of Rio is turned into a system for optimization and securitization. Different parts of city life can be scrutinized and managed at a more exacting level, thus amplifying the already existing practices of militaristic urban control (Wacquant, 2008). IBM and other technology corporations have created similar data centers elsewhere for single agencies like police departments, but none have yet reached the magnitude of Rio’s Intelligent Operations Center. Though, there is plenty of indication that Rio foreshadows the type of systems we can expect to see being rapidly built and deployed in other cities.
Third, the idealistic models for the smart city are the built from scratch projects that are being constructed where nothing existed before. A canonical case is New Songdo in South Korea, which serves as a global test-bed (Halpern, et al., 2013) and urban laboratory (Gieryn, 2006) for implementing large-scale smart systems in the wild. At a cost of approximately US$40 billion, Songdo’s corporate and government backers hope to make it the world’s first fully smart city. As Christine Rosen (2012) remarks, “Songdo claims intelligence not from its inhabitants, but from the millions of wireless sensors and microcomputers embedded in surfaces and objects throughout the metropolis.” This type of implementation represents a zone of futurity. That is, a window into a grand, but plausibly potential, urban future. Furthermore, this type also reveals striking historical similarities that exists between the smart city ideology and the ideology of twentieth century high-modernist architecture. Consider that Brazil’s federal capital of Brasília — a monument to the high-modernism ideals of technocratic administrative ordering — was built, in only 41 months (1956–1960), by clearing out a plot of land in the Amazon rainforest (Scott, 1998). “Point by point,” writes Adam Greenfield [8], “whether they do so out of ignorance, ahistoricity, heedlessness or hubris, the designers of Songdo and Masdar and PlanIT Valley [other canonical smart cities] recapitulate the overspecification, overweening scientism and ponderous authoritarian pomposity of Chandigarh and Brasília, right down to the grand ceremonial axes.”
Even with this plurality of methods and motivations, we believe it is possible and necessary to begin parsing out the underlying socio-political logics that these smart city initiatives hold in common. As we have shown, there’s no sign that the smart city is slowing down. The ideals and practices of the movement — in the various styles they are implemented — continue to colonize the urban landscape and political imaginations of city leaders. Given the constraints of this paper, our overview is only meant to set the stage for the critical social theory at the heart of this paper. For a more exhaustive genealogical analysis of the dominant discourses and ideologies that are driving these sociotechnical systems and policies — specifically those emanating from the major corporate actors of IBM, Cisco, and Siemens — we point the reader to Adam Greenfield’s thorough pamphlet, Against the smart city (2013). What’s more, we should be clear that our generalized use of “smart city” in the rest of the article is meant to be a shorthand for technologies and techniques that align with both the practices and ideologies of the “smart city” label — no matter what their scale or style of implementation. We don’t intend to homogenize or flatten out the differences in what the “smart city” means for different cities, policy-makers, and corporations. Rather, our hope is to draw attention to the ways in which seemingly disparate technologies and techniques have origins in and reproduce common socio-political logics — and we will do this by discussing specific initiatives. But first, the next section introduces the ideologies — updating and adding depth to Greenfield’s own study — that are embedded within and enacted by smart city initiatives.
III. The ideology of the smart city
In more formal spaces of policy advocacy, a stark meliorism informs a Whiggish imaginary of technological progress via the IoT. In a widely cited article for Foreign Affairs, two chief executives for Cisco trumpeted the benefits of applying the “Internet of Everything” to nearly all aspects of city infrastructure and governance (Chambers and Elfrink, 2014). They promised “intelligent and efficient stewardship of growing cities” to reduce “traffic, parking congestion, pollution, energy consumption, and crime.” Who could be against such a program? The only cost, the executives assure readers, would be a slight reorientation in governance and IT procurement strategies. First, “the world must rethink IT investments” by “moving away from purchasing isolated services and instead focusing on end-to-end solutions that are integrated across disparate or siloed systems.” Second, “hyper collaborative partnerships between the public and private sectors” with strict “adherence to deadlines” is essential. As one of their principles for making smart cities the global “norm” proclaims, “the world can’t be afraid of embracing technology in new ways. This means rethinking the contract with citizens and the services IT firms and governments provide them” (Chamber and Elfrink, 2014).
The shift in political language — wherein the social contract is replaced by the corporate contract — is subtle, but critical for understanding the politics smuggled into the technocratic agenda of smart cities (cf., Sadowski and Selinger, 2014). This explains why the six principles they propose are all based on admonishing “city leaders” for not valorizing (enough) the products and services offered by the ICT sector. Like savvy businessmen, the authors recognize the asymmetry of public-private partnerships in an era of neoliberalism. When top managers at firms earn many multiples of top civil servants, the latter readily allow the private sphere to reshape the public sphere in its own image. Corporations can afford a phalanx of economists, designers, attorneys, and public relations specialists, all skilled in presenting one possible future for the city as a technocratic pensée unique. Indeed, other than the corporate model, “there exist no large-scale alternative smart city models, partly because most cities have generally embraced a pro-business and entrepreneurial governance model of urban development” [9].
Of course, Cisco has a commercial interest here: designing, manufacturing, and installing the hardware for these networks is Cisco’s lifeblood, and future profit margins may depend on the firm’s ability to craft seductive narratives of ‘smartness.’ But numerous municipal leaders and non-profit foundations have jumped on the bandwagon, as well. There are material motivations here, too, as politico-economic analyses of revolving door employment patterns between private, public, and “third” sector concerns reveal. When civil servants can easily multiply their pay by moving from government to corporate offices, as long as they are pliable and cooperative, few have an incentive to ask hard questions (Carpenter and Moss, 2013). The boundaries between public office and private consulting are porous.
Just as important as material motives and career ambition, the narrative of the smart city, as an interpretation of technological systems, rationalizes these urban transformations (Söderström, et al., 2014). In a commentary on smart cities research, geographer Rob Kitchin argues that it is problematic the way in which “much of the writing and rhetoric about smart cities” — whether stemming from business, academia, or government — “seeks to appear non-ideological, commonsensical and pragmatic” [10]. This is an outgrowth of a technocratic neoliberal ideology, and a broader political economic imaginary of stable extraction of profits and taxes. Advocates of the smart city style themselves as hard-headed problem solvers who transcend the zero-sum politics that cause other to become embroiled in gridlocked conflict. Yet, they all too often slip into the attitude memorably parodied by Clifford Geertz as “I have a social philosophy; you have political opinions; he has an ideology” [11]. Here the “I” might be smart city contractors; the “you,” city leaders; and “he” the various interest groups raising deeper concerns about the implementation of mass surveillance, data processing, and control. Take, for instance, a speech by Samuel Palmisano (2010), then the Chairman, President, and CEO of IBM, in which he asserted, “Building a smarter planet is realistic precisely because it is so refreshingly non-ideological.” However, as Geertz advises, the deployment of the term ideology is one of the most ideologized practices of modern rhetoric, a way of concealing the more contestable values and assumptions driving those dismissing their opponents as ideological. In this paper, we do not use the term ideology as an a priori accusation, but rather in its descriptive capacity — and somewhat ironically since many technologists and neoliberals alike expend so much energy claiming that their practices are the results of a value-free, Progress-driven, extra-human force (e.g., technology and markets).
To better understand the invariably political character of the smart city, consider a logical extension of some current smart city thinking, proposed as a thought experiment by philosopher and legal theorist Lawrence Solum. Singapore has “smart intersections that var[y] their red/green cycles according to traffic” (Baum, 2001), and one can imagine far more elaborate methods of controlling the flow of automobiles. Solum posits the development of an “Artificially Intelligent Traffic Authority (AITA),” which could “adapt itself to changes in driver behavior and traffic flow” [12]. The system would be designed to “introduce random variations and run controlled experiments to evaluate the effects of various combinations on traffic pattern” [13], recalling Jim Manzi’s (2012) recommendations for far more experimentation in public policy. But the system would not be very forgiving of individual experimentation with, say, violating its rules. Rather, as imagined by Solum, “[v]iolations would be detected by an elaborate system of electronic surveillance” and offenders would be “identified and immediately would be removed from traffic by a system of cranes located at key intersections” [14].
Solum uses this example to break down the usual distinctions between human and artificial meaning in law, rather than as a policy proposal for the future of traffic. The scenario is just as useful to flag the inevitably legal and political aspects of automated law enforcement, even in an area as seemingly technical as traffic. Would the cranes posited in Solum’s hypothetical surgically remove protesters, like the Ferguson marchers, who blocked highways (Harcourt, 2012)? Would anyone with an expired license or tags be plucked away as well — in a vision already half-realized by subprime lenders who stop cars remotely as soon as a payment is late (Sadowski and Pasquale, 2014)?
The problem for smart city advocates is one of overcoming several tensions, if not outright contradictions, in their ideal-type of corporatized governance. Who is ultimately in charge of “hyper collaborative partnerships between the public and private sectors?” What are the penalties when, say, deadlines are not met? Who imposes them? What are the problems that the smart city will use “end-to-end solutions” to solve? How will the imposition of such “solutions” be sequenced?
To take some obvious examples: should new forms of surveillance focus first on drug busts, or evidence of white-collar crime, or unfair labor practices by employers? Wage theft is a massive problem, but rarely taken seriously by authorities (Bobo, 2011). Do the cameras and sensors in restaurants focus on preventing employee theft of food, stopping food poisoning, and/or catching safety violations? Does “traffic control” include efforts to stop honking of horns and loud motorcycles late at night in urban neighborhoods, or is that health-damaging noise deemed just as unworthy of computational scrutiny as it is casually excused by millions of small acts of policing discretion each year — as opposed to the charge of “blocking pedestrian traffic” that is commonly used by police as an excuse to harass African-Americans standing on empty sidewalks (Taibbi, 2014)? Would autonomous car control systems prioritize preventing pedestrian deaths, or merely aspire to smooth flows of cars into and out of the city?
The ideology of neoliberalism all too often provides rapid, “obvious,” and unchallenged answers, based on dubious cost-benefit analyses. Its summum bonum is to improve the business environment and spread market logics to all dimensions of human life. Yet problems multiply even within the neoliberal framework, particularly as it expects state actors to realize business goals (and vice versa). The state itself must capitulate to (and coordinate) its subjects’ purported emancipation from it. So, as Philip Mirowski argues, there is a neoliberal pattern of “hav[ing] it both ways: to stridently warn of the perils of expanding purview of state activity while simultaneously imagining the strong state of their liking rendered harmless” [15]. These tensions are a formal feature of ideological thought: it is a way of containing and coordinating commitments that are contradictory either in theory or practice (Geertz, 1973).
Although these ideological beliefs are most often pegged to Wall Street and Silicon Valley, they can be found, without much difficulty, in even our highest legislative bodies. In February 2015, the United States Senate held a hearing called “The Connected World: Examining the Internet of Things” [16]. The hearing featured statements from senators and testimony from a panel of five witnesses. The attitudes throughout were overwhelmingly excited for the smarter lives we will all be leading thanks to the IoT. While there were occasional mentions of basic issues related to security and privacy, most of the concern stemmed from worries about “over regulation,” which meant anything more than a “light touch” approach. In his statement, U.S. Senator Cory Booker (D–NJ) neatly encapsulated the political economic ideology on display in the hearing — and while he was more enthusiastic and explicit in tone than others, his remarks are representative and worth quoting at length:
“This is a phenomenal opportunity for a bipartisan, profoundly patriotic approach to an issue that can explode our economy. I think that there are trillions of dollars, creating countless jobs, improving quality of life, [and] democratizing our society in ways that gives advantages to people who are being marginalized on the edges, breaking down barriers of race and class. We can’t even imagine the future that this portends of, and we should be embracing that ... And so a lot of my concerns are really what my Republican colleagues also echoed — which is, we should be doing everything possible to encourage this, and nothing to restrict it ... But for us to do anything to inhibit that leap in humanity to me seems unfortunate ... And I also believe that this should be a public-private partnership. We all have a role.”
Booker’s statements are not radical. He is in fact channeling the mainstream views about innovation in society. The least we can do is get out of the way. At best, our duty is to provide all the legal, material, and ideological support we can for innovations — and their innovators — like the IoT. Anybody who wishes to ask critical questions about the future, let alone actually constrain and slow down technological development, is de facto extinguishing an exploding economy and standing in the way of a democratizing force for justice.
Booker’s language recalls the puffery of finance capital — the same group he vigorously defended in 2012 after the leader of his political party (Barack Obama) gently suggested the possibility of ending private equity tax loopholes. Overclaiming the value of the smart city is vital to contemporary capital markets, since extreme inequality in wealth allows rentiers to live well even on the very low interest rates offered by nearly risk-free sovereign debt. The “smart money” probably will understand the “smart city” as an even more speculative bet if it peruses security experts’ warnings about the security problems now endemic in the Internet of Things. (As Bruce Schneier (2014) has observed, when computing is embedded into hardware (as is the case in most of the IoT), sensors and routers are “riddled with vulnerabilities, and there’s no good way to patch them”). The riskier the investment, the more spectacular the potential gains must be: thus the proliferation of characterizations of smart city technology as epochal, groundbreaking, world-making.
Of course, the rhetoric is not always so grandiose — there are cross-cutting, technocratic pressures to sound cool, analytical, and mechanically objective when describing new technology. Bland bipartisanship is also a favored rhetorical mood. Boosters lard manifestos, manuals, and exhortatory books with simple, straightforward examples of problems all can agree need fixing (Newsom, 2013; Townsend, 2014), in order to obscure the stakes of automated surveillance and regimentation of every moment and place. A pothole-spotting app, for instance, is a step toward at least informing (if not guaranteeing the filling of) an unmitigated, car-harming bad. But not everyone agrees with, say, Goldsmith and Crawford when they argue for “postprogressive” city management that focuses on “results not compliance” [17], once the “results” desired move far beyond fast trash pickup or smooth roads. Indeed, the very choice to deploy resources for road smoothing (rather than, say, train or bus air conditioning, or green spaces) is an inherently political one. Goldsmith and Crawford celebrate a new, “smart” fingerprinting initiative aimed at criminals [18], with nary a reflection on the ways in which these records databases create underclasses of effectively unemployable individuals.
Smart city advocates may counter that such conflicts over resource allocation are inevitable in any political order, and stress that their own deployment of sensors, apps, open data, and progress reporting cannot be expected to unravel them. But realities of scarcity apply to political attention, problematization, and action as well. Time spent organizing to deploy a “platform for citizens to engage city hall, and each other, through text, voice, social media, and other apps” [19], is time not spent on highlighting the role of tax resistance by the wealthy in creating the very shortage of personnel that smart cities are supposed to help cure by “force multiplication” of the cities’ remaining workers (Winters, 2011; Bady, 2013). Would Newark, New Jersey, need Mark Zuckerberg’s donation of US$100 million to its school system, if so many others in the billionaire class had not fought so hard to reduce their own (and corporate) taxes, shelter wealth abroad, and defang regulation? Each time a “quantrepreneur” proposes ingenious new ways of measuring and maximizing the “output” of government workers, a critical citizenry should ask: how did we come to this pass? Where has the constant pressure to “do more with less” come from? Focusing on the tech of “doing more” displaces critical debate on the why of “less” governmental resources and employees.
The corporate and governmental actors behind the smart city ideal have distorted debate in two ways. First, focusing on the narrow goals of promoting transparency and efficiency, they have obscured the revolutionary changes in law enforcement’s intensity, scope, and punitive impact portended by pervasive surveillance systems that are easily embedded into a regime of ambient law. Second, they offer a doubly crabbed view of the politics and ethics of digitizing space via the IoT: as a post hoc constraint imposed on technical systems, primarily to encourage “privacy,” in the individualistic sense of the right to control the collection of information about oneself.
By applying a hermeneutics of suspicion, a more complete — and troubling — social theory of the smart city emerges. Even at the least intrusive end of the spectrum of control enabled by the IoT, there is far more at stake than the nebulous set of concerns about perception and reputation traditionally encapsulated in the umbrella term “privacy.” And at the far end of control, the stakes are very high. The IoT is not simply a chance to watch people, but to produce and reproduce certain patterns of interaction (Bogard, 1996), and to replace people with robotic agents once data about them has been so pervasively recorded that it can be downloaded into an automaton to simulate their actions.
IV. Smart cities in societies of control
What will a social theory of the smart city demand? As opposed to the ideology of advocates, social theory is a “systematic, historically informed and empirically oriented theory seeking to explain the nature of ‘the social,’” where the social “can be taken to mean the general range of recurring forms, or patterned features, of interactions and relationships between people” [20]. To take on ideal-types of interactions in urban environments, critical patterns include relationships of allocation/extraction, oppression/emancipation, and recognition/misrecognition (Fraser, 1995). Close examination of the phenomenology of being a surveilled subject, a data subject, reveals the vulnerability of each resident of the smart city to extraction, oppression, and misrecognition.
In many ways, Foucault’s concept of biopower has explanatory fit. One form of biopower is, he writes, “centered on the body as a machine: its disciplining, the optimization of its capabilities, the extortion of its forces, the parallel increase of its usefulness and its docility, its integration into systems of efficient and economic controls, all this was ensured by the procedures of power that characterized the disciplines: an anatomo-politics of the human body” [21]. In contrast to the modes of sovereign power that exercised the right “to take life or let live” [22], the modes of disciplinary biopower exercise the ability to administer and manage bodies and populations. The smart city not only operates on people in this way — for instance, viewing citizens as analog-cum-digital information nodes, or “citizen sensors” [23] — but it also reimagines and reconstructs the city, in itself, as a machine, which can and must be administered and managed. One theorist, inspired by Foucault’s concept of “governmentality,” has deemed this type of disciplining “smartmentality” (Vanolo, 2014).
While the concept of biopower is certainly illuminating, it doesn’t give us the full picture. We can reveal more about the smart city by applying a different social theory — one that explicitly sought to succeed Foucault’s disciplinary societies, just as Foucault’s model succeeded the “societies of sovereignty” — namely, Gilles Deleuze’s (1995) notion of “societies of control.” If the sovereign power was, as Foucault points out, symbolized by the sword, and disciplinary biopower was represented by industrial machines, then control corresponds to computer networks (Deleuze, 1995). Now, of course, the existence of one mode of power does not abolish the others. Rather, it is a question of which one is the dominant operational logic. And, when applied to ICT, especially the networked technologies of smart cities, Deleuze’s framework makes clear the common logics underlying these practices and ideologies. We will provide a preliminary application of this framework to demonstrate its merit as a social theory of the smart city.
A Deleuzian “society of control” has at least three crucial components — dividuals, rhizomes, and passwords — which come together to form a continuously acting logic.
When one person observes another, a basic perceptual apparatus of sight and vision demands at least some minimally holistic assessment. It is hard to register what a walking person is wearing, for example, without also noticing gender, if the person limps or strides, is tall or short, among the hundreds of other bits of tacit knowledge that may be conveyed by an appearance. Monitored by sensors, by contrast, city dwellers are becoming less individuals than “dividuals”: entities ready to be divided into any number of pieces, with specific factors separated, scrutinized, and surveilled. What the person does becomes less important than the consequences calculated in response to emanated data streams. For example: the metadata from a phone call may be far more fateful than the talking which we usually take to be its purpose.
With digital technologies, the individual is atomized, blown apart into streams of data fed into processors. And as these sensors gain immediate influence over physical objects like doors, fences, and automobiles, there is little to no chance of the communicative dialogue that is a hallmark of human interaction. Instead, these relations are at their core strategic, in the Habermasian sense, rather than communicative (Habermas, 1984). Consequences will result not from the “unforced force of the better argument,” or even coaxing and cajoling, but rather, by force alone, as programmed by a set of managers and software developers far removed in time and space from particular implementations of programmed rules [24].
For example, facial recognition software enrolls a person’s face, and by extension the person it is associated with, into a network, whether the person wants to be enrolled or not. Hackers now claim they can even use photographs to identify fingerprints as well (Santus, 2014), a potentially massive boon for law enforcement. The health wristband paints a picture of a self by collecting and analyzing somatic data. The location-tracking sensor registers geospatial coordinates. The Department of Homeland Security’s Cell-All initiative senses “deadly” chemicals. The RFID reader only cares about the chip in your wallet. The biometric lock is only concerned with your fingerprint or irises. The list of ways that people are dividualized goes on. It is identity via synecdoche, where a factor — which factor depends on the system — becomes representative of the whole and becomes all that matters.
The array of underlying technical systems, which are often hidden from sight and mind, can be conceptualized as what Deleuze calls a “rhizome” — like the roots and shoots of a persistent, massive set of plants, it seems to pop up everywhere. Rhizomes are assemblages of concepts, relationships, materials, and actions. They have no distinct boundaries; rather, they are fluid fields, always acting, pulsating power, emanating from multiple directions with varying intensities. The city’s networked, ‘smart’ technological apparatus can simultaneously be: sensing chemicals in the atmosphere; tracking bodies as they move through space; surveilling the types of faces on the street; sending police to remove unwanted people; moving traffic along the roads; and more.
Even as a swarm of disconnected, “dumb” machines, this emerging rhizomatic apparatus of monitoring and control can be intimidating. No one wants to be on the wrong side of its algorithms. As urban technological networks grow vaster and more interconnected, secondary uses of data barely imaginable at the time “users” begin participating in the IoT may well become commonplace (Hoofnagle, 2003). Data gathers and brokers — from corporations to governments — will find a plethora of uses for the information. Consider the biometric lock: Surely the times, places, and identities of who is granted access will be categorized and logged, but what might be even more interesting to authorities is the data for who is denied access.
And people-qua-dividuals have freedom only insofar as all their “passwords” — the products of dividualization that mark access or restriction, allowing one to move freely through or be stymied by the rhizomatic system — are in working order. (Do you wish to enter through a keypad lock? Your PIN is the password. Do you wish to purchase something? Your credit card is the password.) Life is filled with these passwords. Yet, at any moment a password could be rejected — rightly or wrongly, with or without your knowledge — and the amount of control the array of underlying mechanisms have over you become bluntly apparent. Deleuze asked us to imagine “a city where one would be able to leave one’s apartment, one’s street, one’s neighborhood, thanks to one’s (dividual) electronic card that raises a given barrier; but the card could just as easily be rejected on a given day or between certain hours” [25]. As infrastructure decays and the rhizomatic tendrils extend further, city dwellers increasingly feel the Kafkaesque frustration such a scenario entails.
Technology critics often portray these unexpected developments in technological control as a kind of Frankenstein’s monster or sorcerer’s apprentice, one that “we” have unleashed via thoughtless adoption of technology [26]. Social theorists must push the question of causation and agency further, identifying the powerful actors who remain above the fray of dividualization, weaving a web of forces that increasingly constrain the time and space of city dwellers (Krieger, 1994). Masses may be consenting to be dividualized, but only a few wrote those terms and enforce them (Rothkopf, 2009).
Through Mirowski’s detailed analysis of the cause and context of the financial crisis, we can see how these ‘smart’ initiatives plug into the ideologies and tactics of neoliberal political economy: “Technocratic elites could intently maintain the fiction that ‘the people’ had their say, while reconfiguring government functions in a neoliberal direction. These elite saboteurs would bring about the neoliberal market society far more completely and efficaciously than waiting for the fickle public to come around to their beliefs” [27]. The distinction between control and consent is important to several recent initiatives toward the creation of smart cities. Pervasive interlinking of surveillance/sensor arrays, computational processing, and virtual databases into the physical structure of cities is only legitimate if citizens can, both politically and in individual encounters, can be said to have “consented” to it. But when that consent is remote or indirect, its force, validity, and scope should be vitiated. Internet “terms of service” are the ideal-type of desiccated, hollow, pro forma “consent” that is better termed obeisance, acquiescence, or learned helplessness. Thus the overall pattern of relationships in the smart city results in a seamless “spectrum of control,” with meritorious or merely creepy technologies directly imbricated with deeply disturbing ones.
The idea of a “spectrum of control” is more than a turn of phrase [28]. It serves as a symbolic visualization of an interpretation of a text — here, the text is the city, considered simultaneously as a kind of aesthetic object and software program. As Charles Taylor has stated, in canonical work on interpretive social science, interpretation “is an attempt to make clear, to make sense of an object of study” that is in “some way confused, incomplete, cloudy, seeming contradictory — in one way or another, unclear” [29]. Just as skilled commentators can interpret literature, judicial opinions, and works of art, we should take the city’s increasingly technologized systems of governance as expressive texts in need of interpretation.
Both the rhetoric of the “smart city,” and the actual city itself, are texts and text-analogues. They are cloudy and confused now because so much theoretical effort has gone into separating out wheat and chaff, to minimize “coercion” and maximize opportunities for “consent.” This well-meaning, but ultimately futile, normative agenda contributes to confusion and contradiction because the IoT and all-pervasive surveillance are building less a smart city than a cyborg city (Gandy, 2005) — urban places where the stakes of access to certain prosthetic extensions of the self are ever rising. In such cyborg cities concepts like consent vs. coercion, control vs. autonomy do not exist as binaries — but rather they exist on a continuum. Shoehorning the daily experience of the smart city dweller into such binary choices will only further falsify the lived experience of urbanites. Economic pressure toward a “full disclosure future” [30] makes opting out a luxury good (Angwin, 2014).
By theorizing in terms of a spectrum of control we can draw connections between technologies that were before thought of as discrete and independent. The innocuous is enfolded with the menacing. Any significant technology of the smart city becomes a tool to be repurposed for later, often-unforeseeable goals. Claude Lévi-Strauss has compared human thought processes to the work of the handyman, or bricoleur, who fixes problems as best he can with whatever tools or materials are lying at hand (Lévi-Strauss, 1966). A similar process of bricolage will embed technologies of the smart city into solutions proposed for problems large and small — and will, in turn, help define what is viewed as a problem properly solved by the polity....MUCH MORE
