Computerizing carceral space: Coded geographies of criminalization and capture in New York City

Abstract

This article examines how computer software mediates the way cities conceptualize relations between corrections, law enforcement, and urban space. Focusing on New York City, it builds on concepts from carceral geography and geographic studies of software to decipher the “coded spaces” of corrections and law enforcement, or spaces whose existence and morphology are partially determined by software. It does so through discourse analysis of technical documents from New York City criminal justice agencies and specialized publications on applications of software in corrections and law enforcement. In analyzing these materials, the article demonstrates how the New York City Department of Corrections and Community Supervision envisions software as a medium for embedding coded spaces of continuous surveillance, discipline, and capture across the city. It also chronicles how the New York City Police Department articulates software as a means of establishing functionally similar spaces of its own, albeit on larger scales. In each instant, authorities perceive software as a means of establishing intracity coded spaces in which correctional supervision looks more like policing, and policing more like correctional supervision. The article suggests that this coded dimension of surveillance, discipline, and human capture indicate new horizons for the geography of the carceral state, and merit further empirical investigation to understand their technicity.

Keywords

Carceral geography digital geography software studies coded space policing

The phenomenon of capture is deeply ingrained in the practice of computer system design through a metaphor of human activity as a kind of language. Within this practice, a computer system is made to capture an ongoing activity through the imposition of a grammar of action. (Agre, 1994: 113)

Cities, computers, carceralization

Computer code, understood as “instructions and rules that, when combined, produce programs capable of complex digital functions that operate on computer hardware” (Dodge and Kitchin, 2005: 162), is playing an increasingly prominent part in the everyday management of urban space. Code helps regulate flows of electricity, traffic, waste, water, and an array of other vital urban processes (Graham and Marvin, 2001; Kitchin and Dodge, 2011). Criminal justice agencies have also embraced it to manage the monitoring (Kilgore, 2013), biometric profiling (Kaufman, 2016), geographic profiling (Vigneswaran, 2014), real-time analysis (Joh, 2014), and satellite imaging of urban people and places (Graham, 2011). But what does the rapid proliferation of these coded processes tell us about how city authorities conceptualize relations between criminal justice and urban space? What implications do they have for geographies of the carceral state?

To answer these questions, the article builds on theories from carceral geography and software studies. Carceral geography offers tools to excavate the criminal justice apparatus’s obscure spatial formations dispersed throughout the social field (Allspach, 2010; Gilmore, 2007; Loyd et al., 2012; Moran, 2013; Moran et al., 2017; Shabazz, 2009; Turner, 2016). In a similar vein, geographic woks on software help identify hidden “coded spaces” (Kitchin and Dodge, 2011), or spaces partially determined by computer software (and hardware) that permeate the social field as well (e.g. Graham, 2005; Thrift and French, 2002). The article assembles an analytical framework from these literatures to study the spaces that emerge at the intersection of criminal justice and computer software. It finds that corrections and law enforcement software is being designed in aims of establishing intracity spaces to continuously surveil, constantly discipline, and swiftly capture criminalized individuals identified through software. The article demonstrates that in instances where corrections and law enforcement envision computer software as a means to “hold human beings without consent” (Martin and Mitchelson, 2008), it must be theorized as a conduit of an emergent form of carceral space.

In constructing this argument, the article directs attention to New York City, which provides fertile terrain to explore applications of computer software to corrections and law enforcement. Since the early 1990s, the city has assembled a vast network of audio detectors, closed circuit cameras, datacenters, environmental sensors, and ticketing terminals, each partially operated by software. Moreover the city is widely regarded as a global leader in the application of computing software to crime control (Henry, 2003; Silverman, 1999). The article therefore conducts discourse analysis on technical documents from New York City’s Administration for Children’s Services, Corrections and Community Supervision Department, Criminal Justice Agency, Mayor’s Office, Data Analytics Office, Police Department, and Public Safety Council.¹ It also analyzes specialized criminal justice publications including Corrections Compendium, Corrections Today, and Law Enforcement Technology to reveal the perceived advantages of software from the perspectives of corrections, law enforcement, and programmers. While discourse analysis cannot provide grounded accounts of the implementation of software like ethnographies or interviews, it can reveal the purpose for which the software was developed in the first place.² As such these obscure materials cast light on the perceived benefits of computer software in the eyes of the criminal justice system's architects.

The article shows that New York City authorities developed software to a great extent as a means of creating coded spaces across the city in which correctional supervision looks more like policing, and policing looks more like correctional supervision. It suggests these coded spaces are emergent, digitized forms of carceral space which are quietly metastasizing throughout urban environments. Understanding the spaces that emerge from the meshing of criminal justice and computer software is particularly important in the context of US cities, whose number of parolees and probationers have exploded over the past three decades, which has in turn detonated a multiplicity of mechanisms of criminal justice surveillance and discipline (Peck and Theodore, 2008). The paper contributes through its chronicling of the computerized dimension of this eruption, and its wider implications for geographies of carceral governance.

Distributional geographies of carceral governance

While the software-mediated geographies of corrections and law enforcement are unexplored, carceral geographers provide a conceptual repertoire for such an exploration. Carceral geography has emerged largely in response to the meteoric rise of mass incarceration and the subsequent emergence of carceral cities (Peck and Theodore, 2008), or cities with masses of formerly incarcerated people and people under correctional supervision. These developments have demanded new ways of conceptualizing the boundaries of “carceral governance,” or governance in which the confinement of large segments of the population figures prominently (Gilmore, 2007; Wacquant, 2002).

On one hand, carceral geographers shed light on the unique spatialities and temporalities created inside detention centers, jails, and prisons (Jared and Lee, 2006; Martin and Mitchelson, 2008; Milhaud and Moran, 2013; Sibley and van Hoven, 2009). Prison architecture, sensory deprivation, surveillance, and the regulated dispensation of food, hygienic provisions, and medical attention are all shown to shape configurations of space and time inside custodial institutions (Milhaud and Moran, 2013; Moran, 2012). Geographers thus stress how carceral spacetime determines aspects of an inmate’s daily routines, subjectivity, and fundamental relation to space and time, even though its effects are never totalizing (Dirsuweit, 1999; van Hoven and Sibley, 2008).

On the other hand, geographers are paying increasing attention to “distributional” carceral geographies (Moran, 2012), or the geographies of detention and incarceration that overflow institutional settings (Allspach, 2010; Loyd et al., 2012; Moran, 2013, 2014b; Nevins, 2012; Shabazz, 2009; Turner, 2016). Allspach (2010) probes the “widened network of social control” of criminalized women comprised of halfway houses, parole offices, and rehabilitation centers that function as transcarceral spaces; Turner (2016) develops a theory of carceral patchworks to elucidate the mesh of activities, individuals, movements, organizations, and policies that form the boundaries of imprisonment; and Moran (2014a) and Gill et al. (2018) draw on Marx and advance the notion of carceral circuitry to capture the networked flows of discourses, people, material objects, and practices that make up the custodial apparatus. Together these works challenge taken for granted boundaries of detention and imprisonment by highlighting the vast network of spaces and flows that presuppose them.

One key concern for carceral geographers pertains to how convicts and detainees are transported across space (Gill, 2009, 2013; Moran et al., 2012; Mountz et al., 2012). Several studies consider the city-rural-city-rural³ feedback loop of recidivists traveling between urban communities and rural prison facilities as a central component of mass imprisonment (Bonds, 2009; Gilmore, 2007). Kurgan (2013) describes prisons as “urban exostructures” since cities feed these mostly rurally located facilities continuous streams of convicted people. This dynamism is captured in Follis’s (2015) analysis of prison transformation systems, which are characterized by intractable tensions between prisoner entry, queuing, overcrowding, and relocation. On a more granular scale, Moran (2014b) traces how prison stigma follows formerly incarcerated persons throughout the domestic sphere, health clinics, housing market, and labor market. From this view, prisons are seen as mere relay points within extensive networks that ripple across space and time.

Geographers also chronicle how prison construction alters the physical environments that surround correctional facilities (Bonds, 2013; Che, 2005; Gilmore, 2007). Prison development modifies nonhuman nature, moves groups of humans around, and requires the constant transport of resources to and from custodial facilities. Oftentimes prison development is accompanied by the construction of new businesses and residences, which all require infrastructure (Norton, 2015). But while some geographers have explored the rising role of technology in the carceral apparatus (Gill et al., 2014; Moran and Etchegoyen, 2017), what remains unclear is the significance that “coded infrastructures,” or networks of software-operated devices have for the distributional dimension of the carceral state. Moreover, the sociospatial function that these infrastructures perform also remains dimly understood.

Coded spaces and infrastructure in cities

Geographers who draw on software studies offer conceptual resources to probe how computing software is increasingly embraced by urban administrations to “measure, monitor, manage, and control populations and the space-time of cities” (Coletta and Kitchin, 2016). For nearly two decades, geographers have explored what Kitchin and Dodge (2005) term coded space, or “spaces where software makes a difference to the transduction of spatiality” (p.18). These largely invisible spaces are becoming more prevalent in cities whose energy, security, traffic, and water/waste systems are operated in part by computers (Amin and Thrift, 2002; Graham and Marvin, 2001; Kitchin and Dodge, 2011; Thrift and French, 2002). Though coded spaces are ontogenetic, or continuously reproduced within embodied, contingent, and uncertain conditions, they are nonetheless covering progressively large sections of urban environments. But while these spaces are becoming more prevalent in the realm of criminal justice, their essential properties remain for the most part unexamined.

As with carceral spaces, understanding how city authorities conceptualize software-mediated spaces demands attention to infrastructure. Interlinked and automated traffic controls, logistics systems, surveillance, and sanitation each make up coded infrastructures, or “networks that link coded objects together and infrastructures that are monitored and regulated, fully or in part, by software” (Kitchin and Dodge, 2005: 6; Thrift and French, 2002). Such infrastructures have grown so vast that they have provoked new ways of conceptualizing entire cities: computable cities (Batty, 1997), cybercities (Boyer, 1996), informational cities (Castells, 1992), networked urban infrastructures (Kitchin and Dodge, 2011), programmable cities (Kitchin, 2011), splintering urbanism (Graham and Marvin, 2001), and quantified self-city nations (Wilson, 2015) are a few examples. Kitchin (2011) represents interactions between software and urban processes in these cities thus:

Figure 1 provokes several questions that relate to the computerization of criminal justice: what types of knowledge and models are created by translating criminal justice into computer code (as represented at the top of Figure 1)? What type of software is used in translating these knowledges and models into computerized form?

Figure 1.

Programmable urbanism model. Adapted from Kitchin (2011).

Several works in software studies show how the state uses video analytic and geospatial software to quantify civilian mobility and turn human movements into machine-readable identification codes (Cresswell, 2010; de Souza e Silva, 2013). Dodge et al. (2004) examine how this newfound ability to monitor urban environments in real-time is altering the way people navigate the city (see also de Souza e Silva and Frith, 2010). In point of fact some geographers show how digitized forms of state surveillance can discourage people from moving about parts of the city (de Souza e Silva, 2004; Graham, 1998b).

Similar to carceral geography’s studies of carceral space, these studies of coded space draw attention to how obscured configurations of space serve as mediums of circumscribing mobility, intensifying surveillance, modifying behavior, and enhancing the capturability of targeted subjects. Moreover, like carceral geography, software studies examine how spaces materialize in the form of infrastructure and quietly permeate urban landscapes. But little is known about the extent to which these two permutations of space intersect and the spacetime properties that emerge from such interactions.

Software-mediated spaces of corrections

Since the late-1999s, computer programmers and city officials have worked to develop software programs to augment the presence of correctional departments beyond the boundaries of jails and prisons. Database management software, geographic information software, statistical analysis software, and video analytic software have been cross-fertilized with the intent to amplify correctional agencies’ ability to manage inmates, parolees, and probationers no matter their location. This section traces how parole officers, prison managers, probationers, and programmers conceptualize the virtues of such software.

Toward the end of 1990s, the American Probation and Parole Association convened in New York to explore how geographic information software could be used in corrections (Harries, 1999). Inside correctional facilities, one of the products of the summit arrived in the form of GIS-based Correctional Mapping (CORMAP), which was developed to monitor and regulate the movement of inmates (Miller, 2004). When linked into prisoner databases, CORMAP was designed to classify inmates by risk factor to aid authorities in determining cell and housing arrangements. This was also designed to help prison authorities track the age, ethnoracial category, religion, and gang affiliation composition of each cell and housing unit. Some versions of CORMAP kept logs of every person inmates have been in contact with during incarceration to help preempt conspiracies (TechBeat, 2002).

CORMAP was also made to track the location of gang members and sites of repeated disciplinary incidents across all correctional facilities. When linked to surveillance equipment (e.g. ankle bracelets, closed circuit cameras), it could receive and map the locational data of an inmate’s every movement. Other versions predicted potential escape routes based on the architectural layout of the facility, group affiliation data, inmate locational data, and inmate interaction records (Hart, 2003). In short, the development of CORMAP reflected efforts to overlay prisons with coded spaces of social differentiation, continuous geosurveillance (Kitchin, 2015), and spatial containment through ensembles of database management, geographic, and geospatial software.

GIS-based corrections software was also devised to augment the role of correctional departments outside of jails and prisons. The state of California was the first to use GIS database software to provide data about correctional facilities to local planning departments (Karuppannan, 2005). The database was constructed to apprise local planners of the department’s land holdings including agricultural pastures, buffer zones, prisons (current and future), and wastewater spraying fields (Jouganatos and Goodfellow, 2001). This was part of a bigger GIS-driven project to help planners optimize resource allocations across state bureaucracies and dictate the amount of resources that flowed into correctional land holdings. Locational modeling software was also devised by local governments to identify promising landscapes for future prison and halfway house development (Lawrence and Travis, 2004). Here, code was written to help carceral architects determine the coordinates at which the prison-industrial-complex’s future facilities will materialize.

To coordinate the geographies of correctional supervision, geographic software has been developed to optimize the distribution of parole and probation officers throughout the public realm. Near the turn of th century, a specialized version of MapInfo Redistricter software was designed to generate districts based on caseload distributions of parolees and probationers (Harries, 1999). The computer-generated districts were intended to help probation offices select caseloads for individual officers and home visitation assignments. This form of districting mirrored the way that law enforcement used mapping software to generate “beat structures,” or patrol assignments based on geographic distributions of caseloads (Mitchell, 1972). Furthermore, an ArcView Network Analysis package was crafted to determine the routes and sequences with which parole and probation officers conducted home visits. Both applications were built to position correctional supervision officers throughout the public sphere in ways that minimized the amount of time it took for officers to establish contact with parolees and probationers and thereby close the gap between home residences and the carceral personnel.

In a similar vein, specialized geographic software has been created to help correctional agencies in Denver, Hartford, Milwaukee, Oakland, Seattle, Washington DC, and other cities to determine where parolees and probationers may live and travel (La Vigne, 2004). Corrections mapping software has been devised to aid decision makers in determining desirable re-entry locations for parolees, prison releasees, and probationers. Many of these performed spatial analysis to chart the proximity of potential re-entry sites to labor markets, social service officers (rehabilitation centers), and schools (Rose, 1998). In the case of spatial analysis, software was crafted to assign inmates to houses distant from their communities of origin lest they (re)connect with criminal elements (Russo, 2001). These analyses were also designed with intentions to aid courts and parole boards when considering prisoner releases. During early phases of development, correctional databases were linked to police databases for the purposes of aiding law enforcement in identifying previously convicted persons when crimes occur near their residence. A common characteristic among these softwares is that they were all designed to bolster the capacity of corrections offices to determine and intervene in the living spaces of convicted subjects (current and former), and disperse the presence of these offices through heavily invisibilized coded geographies.

Location-aware software has also been cast by city officials in Chicago, Los Angeles, New York City, as fundamental to extending correctional offices’ ability to monitor parolees and probationers throughout the public sphere. While public housing units have been subject to closed circuit surveillance, hallway patrols, and ID checkpoints for decades (Shabazz, 2009), GPS navigation software has been designed to prisonize the apartment unit itself. To be sure, corrections departments have used radio frequency software to track parolees for decades. It was not until the late-1980s that such forms of monitoring were put to widespread use on pretrial defendants, parolees, probationers, and work-releases (Kilgore, 2014; Pew, 2016). House arrest is the default position of these forms of “electronic monitoring” (Kilgore, 2013), which use mobile GPS⁴ devices attached to ankle bracelets worn by parolees and probationers. Approximately every 10 seconds, bracelets transmit coded signals through radios, satellites, or Wi-Fi to a receiver box, which in turn relays the data to parole officers and police patrols. Here, the GPS software’s value is perceived to lie is in its ability to extend the gaze and reach of correctional supervisors and further discourage the mobility of parolees and probationers where it is disallowed.

Like other forms of correctional surveillance explored above, electronic monitoring is presupposed by a software-operated infrastructure of fiber optic cables, location-aware mobile devices, and telephony. This infrastructure was designed to allow parole and probation officers to monitor offenders as they move across rehabilitation centers, residences, schools, streets, and workplaces in real-time. The ensemble of GPS, machine-to-machine (M2M), and satellite receiver software used in this type of monitoring reflects concerted efforts of computer programmers, parole boards, and probationers to overcome the obstacles that public space poses to corrections’ ability to monitor and manage convicted individuals. Through the medium of coded space, the boundaries of continuous surveillance, evaluation, and capture of criminalized subjects appear wide open. While the panopticon originated in the penitentiary (Foucault, 1977), correctional software is designed so that it can materialize anywhere in the city.

Nonsecure detention and location-aware supervision

In New York City, the Criminal Justice Authority (CJA) embraced variegated packages of software as means of coordinating the citywide disprsion of correctional supervision. These packages made up the coded substrate of electronic monitoring programs, which authorities describe as future horizons of incarceration (Bagaric et al., 2018). The state of New York first introduced electronic monitoring in 1999, when it unveiled the Juvenile Electronic Monitoring Project for 115 adjudicated juvenile delinquents (Harig, 2002). The Project found that juvenile offenders classified by schools as emotionally disturbed or in need of special education fared best under surveillance programs as opposed to incarceration. Evaluators of the Project also praised it for saving the state over 100,000 dollars per juvenile (ibid). As such, the Department of Corrections and Community Supervision recommended expanding it by increasing the number of categories of high-risk people who were eligible for electronic monitoring. It was also designed with intentions of equipping probation officers with more tools to control high-risk persons in public settings including the authority to legally conduct arrests, warrants, and weapons searches, thus resembling law enforcers to a greater degree. The results were deemed so successful that the Project was brought to the city 2 years after launching.

Details about the New York City’s electronic monitoring initiative emerge from its Positive Alternative Towards Home (PATH) program (NYCACS, 2012). Touted as the first electronic monitoring program in a major US city, PATH was a component of the Juris Monitor initiative, which was intended to facilitate criminal justice institutions with “electronic teeth” and forging “electronic barriers” throughout the city (Mayor’s Office, 1999). Its main objective was to divert approximately 350 court-involved juveniles (aged 13–15) from secure detention in correctional facilities to nonsecure detention in the city. Nonsecure sites included halfway homes, public housing, private residences, and rehabilitation centers equipped with “robust and reliable monitoring mechanisms” (Busching, 2012). The initiative sought to achieve the vision of nonsecure detention through a coded infrastructure of automated alert systems, CAD, location-aware devices, statistical software, and surveillance cameras. This infrastructure, authorities postulated, would decentralize the department of corrections, and engender a more versatile form of correctional supervision capable of traversing the city.

Juris Monitoring cases were first conceived in statistical software. Since 1997, the city’s Criminal Justice Agency (CJA) has determined eligibility for Alternative to Incarceration (ATI) monitoring programs using software that classifies different types of felons and rates their risk of recidivism (Porter et al., 2002). The procedure, dubbed “post-arraignment computer targeting,” generates daily target lists of ATI-eligible subjects across the city, which appear on court-date lists for judicial consideration. The software runs multivariate statistical analysis that analyzes data concerning a convict’s education level, English proficiency, employment history, charge type, drug test results, gender, medical history, prior record, psychological evaluation, and residential status among other factors (Revere and Curbelo, 2001). Qualified subjects are usually those who have pending cases, are presently in detention facilities, are determined by computer targeting to be low risk, and demonstrate the potential to conform to certain quantifiable behaviors (Siddiqi, 2001).

Once admitted into an electronic monitoring program, subjects are given a radio frequency or GPS device attached to an ankle bracelet. Irrespective of the device, the bracelets are designed to feed continuous streams of locational data to the CJA’s Motion Control and Communications Unit. The Unit is tasked with scanning for movement violations and equipment tampering, and notifying police patrols near the subject’s last known location in instances of parole violation. According to program guidelines, radio frequency devices were to be used in cases where curfews were imposed to confirm whether the subject is at home. GPS-based programs were designed to incorporate combinations of GPS, M2M, and location-based service software to delineate exclusion zones, or parts of the city that monitored individuals are barred from entering (PJI, 2012). GPS monitoring was thus conceived as a way of multiplying the spaces in which corrections agencies can “physically segregate, contain, and exclude” (Martin and Mitchelson, 2008). This not only reflects the extent to which mobility is a “constant practical concern in the management of penal systems” (Moran et al., 2012), but also how software was conceptualized as a medium through which this practical concern could be addressed anywhere in the city.

Moreover, electronic monitoring programs in New York City are not limited to house arrest. They can also include compulsory community service, labor market participation, rehabilitation, and/or school attendance. This compulsory dimension is designed to project the “embodied temporality and spatiality of [the] carceral” (Moran, 2012) through a specific form of “disciplined mobility” (Moran et al., 2012). That is, subjects who are enrolled in such programs are stripped of their agency in timing, nature, routing, and physical circumstances of travel, which are all acclimatized to the penal apparatus. Once a person is determined eligible for electronic monitoring, s/he is required to sign a behavioral contract that outlines a daily regimen. Under the program, probationers are required to remain at home unless they obtain permission to travel to the doctor, school, or work. In assault and rape cases, subjects are forbidden from coming within 500 feet of a victim’s residence. Should the individual breach the exclusion zone, an automatic alert is triggered, which notifies 911 operators and activates a recording device in the victim’s home. Once an alert is activated, computer-aided dispatch (CAD) software automatically classifies and records the violation, which precipitates arrest, prosecution, and incarceration. Ensembles of CAD, GPS, and satellite receiver software are not only considered means of shrinking geographical distance between the monitored subject and the state (see de Souza e Silva, 2013), but also as means of collapsing distance altogether.

The meshwork of software was not only designed to help regulate monitored subjects, but also to modulate their lifestyles (Young et al., 1999). Juveniles in the program were to be subjected to constant computer-aided evaluation, the results of which were to determine their permissible activity space. Data about a juvenile’s “status of life” as measured by drug tests, living situation, time spent in job training, school, or work, and psychiatric diagnoses were to be continuously collected, stored, and organized into digital profiles. The profiles were produced for purposes of analyzing them on 30-, 60-, and 180-day intervals using software that runs logic regression analysis and propensity score matching to determine risk factors (Porter et al., 2002). The results were intended to be used by CJA personnel to evaluate subjects currently under supervision and decide future cases (Solomon and Ferri, 2016). This ceaseless evaluation was devised to ensure that monitored subjects conformed to state-proscribed behaviors, no matter their location. Electronic monitoring was effectively designed to pave the way for new modes of correctional governmentalization that travels through urban networks of alarm, geospatial, and statistical software and hardware.

Due to Juris Monitor’s perceived success, CJA sentencing experts have led efforts to extend the program to noncompliance cases involving high risk individuals (Howell et al., 2008). Electronic monitoring was also expanded in the aftermath of a 2015 federal report exposing the excessive force, rape, and solitary confinement in Riker’s Island prison, which prompted the relocation of young inmates from prisons to nonsecure detention sites using location-aware ankle bracelets (McKinley, 2015). Furthermore, electronic monitoring has been adopted by New York City’s Immigration and Customs Enforcement (ICE) branch, to act as a central component of its Alternatives to Detention (ATD) initiative. Used primarily on undocumented immigrants from Central America (Stuart, 2014), the initiative is part of a nationwide supervision appearance program used on over 40,000 illegalized immigrants (ACLU, 2014). In addition to ankle bracelets, the program also uses voice recognition software so that employers may verify a monitored immigrant’s presence at workplaces (ibid). People in this program are also required to make weekly visits to ICE offices, and are subjected to unannounced home visits. This suggests geospatial software is also perceived as a conduit of border relations, if not a conduit of something resembling prison labor.

Inasmuch as corrections personnel and programmers collaborate to design software to determine where parolees, probationers, and in some instances prison releases may travel, the geographic coordinates of correctional populations are poised to be partially determined by computer code. Moreover, the forms of surveillance envisioned by the architects of electronic monitoring are only possible with the aid of software. In Juris Monitor, the CJA has conceptualized a form of correctional control that permeates the urban landscape. This vision of coded correctional surveillance, I show below, shares striking similarities with the coded spaces envisaged by law enforcement, as it reveals blueprints for transducing urban space through computer-operated spaces of surveillance, discipline, and human capture.

The NYPD’s coded infrastructures

Assemblages of database, statistical, spatial, and video analytic software are not only viewed by corrections and immigration departments as means of monitoring and managing human movement. Police also regard these assemblages in the same light. This section explores the properties of the NYPD’s coded spaces, drawing special attention to their isomorphic relations with coded correctional space.

The NYPD’s coded infrastructure crystallizes in its real-time crime center (RTCC), a datacenter that allows computer scientists to focus on the “information processing and decision-making aspects of policing” (Murray and Joyce, 2017: 12). Founded in New York City, RTTCs are centralized hubs in which software-operated alert programs, closed circuit cameras, dispatch systems, data analytics, geographic information systems, and other systems converged. They were designed to help commanders determine geographic distributions of patrols units and organize how patrols navigate the city (NYPD, 2010a). They were also devised to upgrade police surveillance to circumscribe the mobility, monitor the movement, and capture targeted subjects identified through geographic, statistical, and video analysis software. These datacenters, I show, were envisioned as ways of establishing spatial relations similar to those of correctional mapping and correctional monitoring, albeit on much larger scales.

In retracing the origins of the NYPD’s datacenter, it is important to note that crime was not a determinate factor of its inception. At its launch, the city was trending toward the lowest rates of serious crime in half a century (NYPD, 2016). The RTCC was instead catalyzed by 9/11. It was originally built to make the city more surveillable for the Department of Homeland Security. It was introduced in 2005, when the NYPD’s Counterterrorism Bureau approached Microsoft to assist to develop software to help securitize the Financial District and other civic spaces (NYCLU, 2006: fn. 36). The RTCC, along with the Lower Manhattan Security Coordination Center, served as a basis for the 2005 Lower Manhattan Security Initiative (LMSI). The LMSI was originally made up of 3,000 closed circuit cameras designed to scan automobiles and passersby across the city. It also included 1.7 miles of biological, chemical, and nuclear detectors across the southern part of Manhattan (NYPD, n.d.). The cameras and sensors were equipped with recognition software designed to detect furtive movements, illegal activities, or wanted persons and alert the closest patrol unit. While the effectiveness of this software was far from perfect (Meyer, 2015), it represented a vision of ceaselessly monitoring and assessing the city’s population and maximizing the visibility and capturability of targeted subjects.

The LMSI was possible due to the city’s RTCC, a datacenter that ran primarily on data warehousing, data analysis, and video analysis software. The RTCC took data from the NYPD’s sprawling network of cameras, detectors, and databases from multiple bureaucracies and produced targets for police on computer-generated maps (Lueck, 2007). It was able to manage huge quantities of data due to advancements in data management systems. Before the RTCC, information about arrests, crime suspects, ex-convicts, precinct crime rates, and warrants were siloed on disparate city, state, and federal databases. But with the adoption of IBM’s relational database, the Crime Information Warehouse, the RTCC was mining data from over 5 million New York State criminal records, parole, and probation files; 20 million criminal complaints, emergency calls, and summonses; 31 million national crime records; and an astounding 33 billion public records just 5 years after launch (NYPD, 2017). The NYPD has had to employ analytical software that continuously mines these databases, and ceaselessly evaluates streams of criminal history data similar to the way parole offices evaluate streams locational data. Indeed different scholars have suggested such colossal fusions of datasets blur distinctions between state bureaucracies (Graham, 1998a; Kitchin, 2015). This has given the NYPD access to unprecedented quantities of data, increasing its ability to surveil private lives (see Joh, 2014). Thus, privacy is warped by the LMSI’s coded space, which is a central characteristic of carceral space (Milhaud and Moran, 2013).

Like the Juris Monitor Initiative, the NYPD’s coded surveillance apparatus slowly expanded across the city. In 2010, the NYPD extended the LMSI to midtown Manhattan through a network of 500 closed circuit cameras in Grand Central Station, Penn Station, and Times Square (NYPD, 2015). This Midtown Manhattan Security Initiative (MMSI) was built with the intentions of establishing an automated surveillance network to preside over centers of commerce, finance, government, tourism, and transportation (NYPD, 2010b). By the time, the MMSI was complete the total number of cameras linked into the RTCC had risen 160 percent (ibid), thickening its infrastructure of software-operated data generation, surveillance, and evaluation.

The NYPD’s surveillance infrastructure now traverses public housing buildings, roof-tops, skyscrapers, squad cars, and traffic signs. In its infancy, it was made up of 3,000 cameras (Verton, 2012). The number tripled within 5 years. In fact the surveillance apparatus grew so vast a former commissioner remarked that if the NYPD is “looking for a person in a red jacket, we can call up all the red jackets filmed in the last 30 days. We’re beginning to use software that can identify suspicious objects or behaviors” (Raymond Kelly quoted in Signore, 2010). The rapid expansion of the NYPD’s software-operated surveillance system resembles the rapid expansion of electronic monitoring, and both demonstrate a vision in which the criminal justice apparatus permeates the material environment to track the movement, assess the behavior, and capture individuals who were identified, to some degree, by software programs as having criminal inclinations.

Meshing law enforcement and correctional supervision

New York City’s efforts to replicate its coded spaces of surveillance, analysis, and capture in public space are increasing via the police department. Toward the end of 2012, Mayor Bloomberg (2002–2013) and Police Commissioner Raymond Kelly (1992–1994, 2002–2013) introduced an upgraded RTCC, the Domain Awareness Center (DAS) (Mayor’s Office, 2012a). DAS emerged in response to the deluge of data generated from the original RTCC. It was intended to establish universal technical standards across the NYPD software-operated platforms, which came to include environmental sensors, mobile and stationary cameras, and license-plate readers connected to DAS’s centralized datacenter (NYPD, 2017). It did so by standardizing data classification, data retention, and coding procedures, and integrating documentation procedures, hardware, security protocols across all NYPD platforms (NYPD, 2015). The mayor explained:

state-of-the-art technology that builds on years of NYPD success in ensuring the security of some of New York’s highest-profile areas: The Financial District and Midtown Manhattan. This new system … enables officers to instantly assemble, analyze, and respond to information streaming in from a wide range of sophisticated sources, including closed-circuit camera networks and mobile license plate readers and radiation detectors, as well as law enforcement databases. (Mayor’s Office, 2012b)

With DAS, the Financial District and segments of Midtown Manhattan were modulated to augment NYPD surveillance, evaluation, and in some instances intervention. It was made up of computr-operated networks of audio detectors, closed circuit television cameras, license plate readers, and radiation sensors. The network was devised to generate continuous “surveillant-simulations” (Graham, 1998b) of the cityscape, meant to hasten the police’s ability to descend upon and detain people flagged by detectors, cameras, and sensors (Figure 2). Taken together this network is the result of efforts to shrink the city for patrol units and thereby enhance their ability to swarm and capture computer-identified targets (Levine et al., 2017). By compressing urban spacetime for patrol units, DAS is designed to create coded spaces that have the same basic properties of electronic monitoring (i.e., continuous surveillance, analysis, and capture). Although the patrol destinations and routes of DAS differ from those of electronic monitoring initiatives, both share much of the same infrastructural grid, which is becoming increasingly central to the topography of surveillance, analysis, and human capture established across branches of the criminal justice apparatus.

Figure 2.

DAS’s automated alert system. Adapted from Levine et al. (2017).

DAS’s coded geography was also meant to encompass the traffic system and establish a panoptic overview of all of the city’s incoming and outgoing traffic. Its network of license plate readers is comprised of 250 mobile units mounted on patrol cars and stationary units that preside over “every lane of traffic on every bridge and tunnel” leading in and out of Manhattan (NYPD, 2015). The stationary cameras transmit geocoded images of each vehicle that passes through the island of Manhattan. Both mobile and stationary cameras are made to feed visual data into predictive algorithms that search for abnormal patterns. The algorithms are designed to probe for irregular spatiotemporal patterns, which are determined by the frequency with which a given license plate is captured by the cameras. They also probe for routing irregularities, which are recognized when a given plate passes through multiple cameras at rates exceeding a predetermined threshold (Levine et al., 2017).

Much like the NYPD’s other forms of software-operated surveillance, traffic surveillance was made to bring people who have not even made contact with law enforcement into its digital corpus. By 2013, the department’s database housed 16 million data points on license plates (Goldstein, 2013). Just 2 years later, the number of data points increased to two billion (Levine et al., 2017). These datasets, which can be mined at any point or time for illegality, add a new temporal dimension of identification and criminalization into the equation that is not present in electronic monitoring. But the fundamental logic that undergirds traffic surveillance deeply resembles that of electronic monitoring in that mobility is the object of surveillance, analysis, and regulation.

Similar to the districting software of parole and probation, DAS was also created to optimize patrol responses to emergency calls, cutting down the time it takes to execute arrests, fines, summons, and warrants (Mayor’s Office, 2012b). Upon receiving an emergency call, CAD software presents officers with the emergency caller’s phone number, transcriptions of the call, and police records related to the area, caller, and potential suspects. The software obeys a logic-based rule set in which, for instance, calls to originating near high-risk facilities (e.g., financial buildings) trigger priority alarms. Though high-value targets have long been subjected to differential surveillance, DAS introduces a form of unceasing monitoring by integrating the NYPD’s coded infrastructure deeper into the material environment.

The NYPD’s crime mapping and analysis system, CompStat 2.0, provides a clear view into the way urban space is comprehended in these software-mediated of law enforcement; that is, as a field of law enforcement anticipation, analysis, detection, and capture. CompStat 2.0 is the product of interlinked database management and geographic information software, which generate charts, graphs, and maps of NYPD data. It is designed with algorithms to mine these data for recurrent spatiotemporal patterns and distribute patrol units accordingly (NYPD, 2013). This form of algorithmic law enforcement is not only designed to detect abnormal people, but also abnormal places (Henry, 2003). In CompStat 2.0, the city is rendered as a coded field of NYPD evaluation and intervention. Such modeling of the urban form relies on proximity analysis, which designates neighborhoods, precincts, and microspaces as categorically criminogenic and therefore subject to enhanced police monitoring, assessment, and encroachment. CompStat 2.0 bears resemblance to CORMAP in that its core objective is to measure, analyze, and differentiate for sake of neutralizing anomalous behaviors logged in databases. But whereas CORMAP is applied to prison facilities, CompStat is applied to the entire cityscape.

What is more, CompStat 2.0 mustn’t be reduced to geographic software. It required the construction of a new propriety fiber-optic network composed of hundreds of miles of lines (NYPD, 2015). It was connected to every NYPD facility in the city and increased precinct bandwidth by a factor of 100. The NYPD also linked the network of over one thousand closed circuit feeds in two hundred public housing buildings (NYCHA, 2016), in addition to an unspecified number of Mobile Utility Surveillance Towers and 25-foot-tall SkyWatch towers. A key objective of Compstat 2.0 is to guide where police are placed, where civilians should avoid, and how communities with high crime indexes are portrayed in public discourse (Frost, 2017). CompStat therefore evinces the symbolic and material dimensions of the NYPD’s coded space, which were established to monitor, predict, and intervene in the behavior of persons and places as if they were enrolled in a form of electronic monitoring.

The objectives expressed during the development of CompStat 2.0 and DAS also signify how software can morph conceptions of rank-and-file law enforcement, which now includes images of patrol units constantly analyzing data just like the Juris Monitor’s Motion Control and Communications Unit constantly analyzed locational data (Eterno and Silverman, 2010; Mayor’s Office, 2012b). In point of fact the construction of DAS took place alongside a 140 million dollar mobility initiative to connect patrol units to DAS through mobile devices. The NYPD delivered DAS-linked smartphones and tablets to over 41,000 personnel to how police officers perform their duties. Some scholars have suggested that the increasing prominence of database software in law enforcement has added a burden of “producing numbers” on each patrol officer, which has in turn cultivated a quota-driven culture of hyperaggressive policing (Eterno and Silverman, 2010; Ferguson, 2017; Sciarabba, 2009).⁵

Mobile data platforms are the “single largest driver of information technology growth in the [Police] Department” (NYPD, 2015), and have, from an administrative standpoint, induced police executives and software developers to reimagine law enforcement in ways that parallel how software has induced corrections executives to reimagine parole and probation. In the code-mediated vision of law enforcement, patrol units are expected to check mobile applications constantly for software-generated alerts that were designed to dictate how patrols navigate the urban terrain (Evans, 2012; IBM, 2005). Whether or not a causal relation exists between generated alerts and the distributional geographies of patrol units, the NYPD intends to further expand the data-generation tasks of officers by equipping them with mobile fingerprint scanning devices for in-field identification and warrant checks (NYPD, 2015). This would only further decentralize criminal processing procedures, moving them from precinct houses and courtrooms into patrol units dispersed across the city. In short software is envisioned by city authorities as tools for reconfiguring geographies of criminalization and confinement.

Carceral horizons

The technical materials analyzed in this article evidence the ways in which corrections personnel, police, and programmers envision computer code as medium for diffusing its field of action across urban space. More specifically, corrections and law enforcement software is articulated by public officials, computer programmers, and criminal justice bureaucrats as means of enhancing the ability to geosurveil, discipline, and capture computer-identified persons throughout urban environments. For corrections, software has been embraced to afford parole and probation officers the ability to analyze the behavior, monitor the movement and location, restrict the mobility, and capture the persons of probationers and parolees with greater efficiency. For police, it has been embraced as a medium of monitoring, evaluating, and capturing persons flagged by computer programs in a similar vein. This vision of computerized criminalization has emerged and coalesced for the most part outside the realm of public debate, even though its core objectives include reconfiguring urban geographies to accommodate current mutations in the carceral state.

These developments, interpreted through the lens of coded space, have a great deal of significance for carceral geography and geographic studies of software. Conceptually, computerized carceral space presents a new analytic for geographers. Further exploration of this dimension of carceral space presents the conceptual challenges of synthesizing core theories about digital space from digital geography and media studies among other fields. For instance, while geographers have shed considerable light on the centrality of surveillance in carceralized spaces, the practice of dataveillance, or the continuous monitoring of online activity (Rayley, 2013), must also be explored by geographers, especially given the rise of geotagged social media. Furthermore, a theory of carceral cyberspace is urgently needed given the proliferation of online criminal records and registries. These developments also raise urgent questions concerning the current state of carceral governances for geographers. Is it possible for cities themselves to one day become as “carceral” as prisons, if not more? In addition to public housing, street networks, and tourist attractions, what other places in cities are superimposed with coded spaces of criminal justice? These questions are especially pertinent in the US context, where post-recession budget cuts have triggered a multiplicity of nonsecure detention apparatuses to defray the costs of mass criminalization.

Recognition of the coded spaces of criminal justice also raises a host of questions for geographic studies of software. Drawing on Kitchin and Dodge’s (2011) work, questions concerning how software has “seduced” city authorities emerge. Additional analysis of discourse around criminal justice software in the political sphere is therefore needed. What discursive strategies are used to legitimize the development and implementation of such software to the tune of billions of dollars? Also, inasmuch as coded spaces are ontogenetically produced—and therefore subject to human and mechanical error—empirical analysis of the implementation of correctional and policing coded spaces is also urgently needed. Various questions emerge in such an inquiry: what extent do parole, police, probation officers utilize software applications they way programmers and decision makers want them to? What unintended effects emerge from the implementation of criminal justice software? Answers to these lingering questions would be of interest in critical criminology, political geography, urban politics, and science technology society studies among other fields.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

ACLU (2014) Alternatives to Immigration Detention: Less Costly and More Humane than Federal Lock-up. Washington, DC: American Civil Liberties Union.

Agre

(1994) Surveillance and capture: Two models of privacy. The Information Society 10: 101–127.

Allspach

(2010) Landscapes of (neo-)liberal control: The transcarceral spaces of federally sentenced women in Canada. Gender, Place and Culture: A Journal of Feminist Geography 17(6): 705–723.

Amin

Thrift N (2002) Cities: Reimagining the Urban. London: Polity.

Bagaric

Dan

Gabrielle

(2018) Technological incarceration and the end of the prison crisis. Journal of Criminal Law and Criminology 108: 73.

Batty

(1997) The computable city. International Planning Studies 2(2): 155–173.

Bonds

(2009) Discipline and devolution: Constructions of poverty, race, and criminality in the politics of rural prison development. Antipode 41(3): 416–438.

Bonds

(2013) Economic development and relational racialization: ‘Yes in My Backyard’ prison politics and the racialized reinvention of Madras, Oregon. Annals of the Association of American Geographers 103(6): 1389–1405.

Boyer

(1996) CyberCities. New York: Princeton Architectural Press.

10.

Bratton

(1998) The Turnaround: How America's Top Cop Reversed the Crime Epidemic. New York, NY: Random House.

11.

Bratton

(2015) Broken Windows and Quality-of-life Policing in New York City. Edited by The New York City Police Department. New York, NY: City of New York.

12.

Busching

(2012) Oversight – Examining the Division of Youth and Family Justice's Positive Alternative towards Home Program. New York, NY: Committee on Juvenile Justice.

13.

Castells

(1992) The Informational City: Economic Restructuring and Urban Development. New York, NY: Wiley-Blackwell.

14.

Che

(2005) Constructing a prison in the forest: Conflicts over nature, paradise, and identity. Annals of the Association of American Geographers 95(4): 809–831.

15.

City Council Public Safety Committee (2012) Marijuana Possession Arrest, Illegal Searches, and the Summons Court System. New York, NY: City of New York.

16.

Coletta

Rob

(2016) Algorhythmic Governance: Regulating the ‘Heartbeat’ of a City using the Internet of Things. The Programmable City Working Paper. Maynooth, Ireland: National University of Ireland Maynooth.

17.

Cresswell

(2010) Towards a politics of mobility. Environment and Planning D: Society and Space 28: 17–31.

18.

de Souza e Silva

(2004) Mobile networks and public spaces: Bringing multiuser environments into the physical space. Convergence: The International Journal of Research into New Media Technologies 10(2): 15–27.

19.

de Souza e Silva

(2013) Location-aware mobile technologies: Historical, social and spatial approaches. Mobile Media & Communication 1(1): 116–121.

20.

de Souza e Silva

Frith

(2010) Locational privacy in public spaces: Media discourses on location-aware mobile technologies. Communication, Culture & Critique 3(4): 503–525.

21.

Dirsuweit

(1999) Carceral spaces in South Africa: A case study of institutional power, sexuality and transgression in a women’s prison. Geoforum 30: 71–83.

22.

Dodge

Kitchin

(2005) Code and the transduction of space. Annals of Association of American Geographers 91(5): 162–180.

23.

Dodge

Michael

Rob

(2004) No Longer Lost in the Crowd: Prospects of Continuous Geosurveillance. Philadelphia: Association of American Geographers Conference.

24.

Eterno

Silverman

(2010) The NYPD’s Compstat: Compare statistics or compose statistics? International Journal of Police Science and Management 12(3): 426–449.

25.

Evans

(2012) New York City Police Department and Microsoft Partner to Bring Real-Time Crime Prevention and Counterterrorism Technology Solution to Global Law Enforcement Agencies. Microsoft News, 8 August.

26.

Ferguson

(2017) The Rise of Big Data Policing: Surveillance, Race, and the Future of Law Enforcement. New York, NY: NYU Press.

27.

Follis

(2015) Power in motion: Tracking time, space, and movement in the British Penal Estate. Environment and Planning D: Society and Space 33(5): 945–962.

28.

Foucault

(1977) Discipline and Punish: The Birth of the Prison. New York: Vintage Books.

29.

Frost

(2017) First look at NYC’s new CompStat 2.0: Eight Brooklyn crime trends. Brooklyn Daily Eagle, 30 June.

30.

Gill

(2009) Governmental mobility: The power effects of the movement of detained asylum seekers around Britain’s detention estate. Political Geography 28: 186–196.

31.

Gill

(2013) Mobility versus liberty? The punitive uses of movement within and outside carceral environment. In Moran

Gill

Conlon

(eds) Carceral Spaces: Mobility and Agency in Imprisonment and Migrant Detention. Farnham: Ashgate, pp. 19–36.

32.

Gill

Conlon

Tyler

et al . (2014) The tactics of asylum and irregular migrant support groups: Disrupting bodily, technological, and neoliberal strategies of control. Annals of Association of American Geographers 104(2): 373–381.

33.

Gill

Deirdre

Dominique

et al . (2018) Carceral circuitry: New directions in carceral geography. Progress in Human Geography 42(2): 183–204.

34.

Gilmore

(2007) Golden Gulag: Prisons, Surplus, Crisis, and Oppression in Globalizing California. Berkeley: University of California Press.

35.

Goldstein

(2013) Weekly police briefing offers snapshot of department and its leader. NY Times, 10 February.

36.

Graham

(1998a) The end of geography or the explosion of place? Conceptualizing space, place and information technology. Progress in Human Geography 22(2): 165–185.

37.

Graham

(1998b) Spaces of surveillant-simulation: New technologies, digital representations, and material geographies. Environment and Planning D: Society and Space 16: 483–504.

38.

Graham

(2005) Software-sorted geographies. Progress in Human Geography 29(5): 562–580.

39.

Graham

(2011) Cities under Seige: The New Military Urbanism. London: Verso.

40.

Graham

and Marvin

(2001) Splintering Urbanism: Networked Infrastructures, Technological Mobilities and the Urban Condition. London: Routledge.

41.

Harig

(2002) The Juvenile Electronic Monitoring Project: The Use of Electronic Monitoring Technology on Adjudicated Juvenile Delinquents. Albany, NY: New York State Division of Criminal Justice Services.

42.

Hart

(2003) Making prisons safer through technology. Corrections Today 65(2): 26–26.

43.

Harries, K (1999) Mapping Crime: Principle and Practice. Washington, DC: National Institute of Justice.

44.

Henry

(2003) The Compstat Paradigm: Management Accountability in Policing, Business and the Public Sector. Flushing, NY: Looseleaf Law Publications.

45.

Howell

Trent

Olinda

et al . (2008) Intensive Supervision/Electronic Monitoring/GPS. In: Symposium on alternatives to incarceration, Washington, DC, July 14, pp.89–107.

46.

IBM. (2005) IBM Crime Information Warehouse: Unlocking Data to Fight Crime. Somers, NY: IBM Corporation.

47.

James

(2015) Electronic Monitoring Is Not the Answer: Critical Reflections on a Flawed Alternative. Urbana-Champaign, IL: Urbana-Champaign Independent Media Center.

48.

Joh

(2014) Policing by Numbers: Big Data and the Fourth Amendment. Washington Law Review 89(35): 35–68.

49.

Jouganatos

Goodfellow

RAH

(2001) An Inventory Study of the State of California's Land Holdings. Sacramento, CA: California State University, Sacramento.

50.

Karuppannan J (2005) Mapping and Corrections: Management of Offenders with Geographic Information Systems. Corrections Compendium 30(1): 1–10.

51.

Kaufman

(2016) Policing mobilities through bio-spatial profiling in New York City. Political Geography 55: 72–81.

52.

Keith

(1999) Mapping Crime: Principle and Practice. Washington, DC: National Institute of Justice.

53.

Kilgore

(2013) Progress or more of the same? Electronic monitoring and parole in the age of mass incarceration. Critical Criminology 21: 123–129.

54.

Kilgore

(2014) Repackaging mass incarceration: The rise of carceral humanism and non-alternative alternatives. Counterpunch, 6 June, 6–8.

55.

Kitchin

(2011) The programmable city. Environment and Planning B: Planning and Design 38: 945–951.

56.

Kitchin

(2015) Continuous Geosurveillance in the “Smart City”. DIS Magazine, 15 November.

57.

Kitchin

and Dodge

(2011) Code/Space: Software and Everyday Life. Cambridge, MA: The MIT Press.[WorldCat]

58.

Kurgan

(2013) Close up at a Distance: Mapping, Technology and Politics. New York: Zone Books.

59.

Kurgan

and Cadora

(2006) Million Dollar Blocks: Justice and the City. New York: Spatial Information Design Lab.

60.

La Vigne

(2004) Why map prisoner reentry? Crime Mapping News 6(4): 1–12.

61.

Lawrence

Jeremy

(2004) The New Landscape of Imprisonment: Mapping America’s Prison Expansion. Edited by Justice Policy Center. Washington, DC: Urban Institute.

62.

Levine

Tisch

Tasso

and Joy

(2017) The New York City Police Department’s Domain Awareness System. Interfaces 47(1): 70–84.

63.

Levine

and Siegel

(2012) Summonses Issued by NYPD, 2004–2010. New York, NY: Marijuana Arrest Research Project.

64.

Loyd

Matt

Andrew

(eds) (2012) Beyond Walls and Cages: Prisons, Borders, and Global Crisis. Athens, GA: University of Georgia Press.

65.

Lueck

(2007) From database to crime scene: Network is potent police weapon. New York Times, 7 June.

66.

Martin

Mitchelson

(2008) Geographies of detention and imprisonment: Interrogating spatial practices of confinement, discipline, law and state power. Geography Compass 3(1): 459–477.

67.

Mayor’s Office (1999) Mayor Giuliani Builds on Successful Domestic Violence Campaign with Three Innovative New Programs. Edited by Mayor’s Office. New York, NY: City of New York.

68.

Mayor’s Office (2012a) Mayor Bloomberg Discusses How City’s New, State-of-the-Art Law Enforcement Technology and New York’s Finest Keep New York the Safest Big City in the Nation in Weekly Radio Address. Mayor’s Office. New York, NY: City of New york.

69.

Mayor’s Office (2012b) Mayor Bloomberg, Police Commissioner Kelly and Microsoft Unveil New, State-of-the-Art Law Enforcement Technology that Aggregates and Analyzes Existing Public Safety Data in Real Time to Provide a Comprehensive View of Potential Threats and Criminal Activity. Mayor’s Office. New York, NY: City of New york.

70.

McKinley

Jr (2015) New monitoring program aims to keep youths out of Rikers Island. New York Times, 14 August.

71.

Meyer

(2015) Who owns your face? Atlantic, 2 June.

72.

Milhaud

Moran

(2013) Penal space and privacy in French and Russian prisons. In: Moran

Gill

Conlon

(eds) Carceral Spaces: Mobility and Agency in Imprisonment and Migrant Detention. Farnham: Ashgate, pp.167–182.

73.

Miller

(2004) Using GIS in corrections management. Law Enforcement Technology 31(6): 168, 170–173.

74.

Mitchell

(1972) Optimal selection of police patrol beat. The Journal of Criminal Law, and Criminology and Police Science 63(4): 577–584.

75.

Moran

(2012) “Doing time” in carceral space: Timespace and carceral geography. Geografiska Annaler: Series B, Human Geography 94(4): 305–316.

76.

Moran

(2013) Between outside and inside? Prison visiting rooms as liminal carceral spaces. GeoJournal 78: 339–351.

77.

Moran

(2014a) Carceral Geography: Spaces and Practices of Incarceration. London: Routledge.

78.

Moran

(2014b) Leaving behind the ‘total institution’? Teeth, transcarceral spaces and (re)inscription of the formerly incarcerated body. Gender, Place and Culture: A Journal of Feminist Geography 21(1): 35–51.

79.

Moran

Etchegoyen

(2017) The virtual prison as a digital cultural object: Digital mediation of political opinion in simulation gaming. Environment and Planning A 49(2): 448–466.

80.

Moran

Jennifer

Anna

(2017) Conceptualising the carceral in carceral geography. Progress in Human Geography.

81.

Moran

Piacentini

Pallot

(2012) Disciplined mobility and carceral geography: Prisoner transport in Russia. Transactions of the Institute of British Geographers 37: 446–460.

82.

Mountz

Coddington

Catania

et al . (2012) Conceptualizing detention: Mobility, containment, bordering and exclusion. Progress in Human Geography 37(4): 522–541.

83.

Murray

and Joyce

(2017) The future of law enforcement. KM World, 31 March

84.

Nevins

(2012) Policing mobility: Maintaining global apartheid from South Africa to the United States. In: Loyd

Mitchelson

Burridge

(eds) Beyond Walls and Cages: Prisons, Borders, and Global Crisis. Athens, GA: The University of Georgia Press, pp.19–27.

85.

Norton

(2015) Little Siberia, star of the North: The political economy of prison dreams in the Adirondacks. In Morin

Moran

(eds) Historical Geographies of Prisons: Unlocking the Usable Carceral Past. New York, NY: Routledge, pp.168–184.

86.

NYCACS (2012) New York City Administration for Children’s Services Close to Home: Plan for Non-Secure Placement. Edited by New York City Administration for Children’s Services. New York, NY: New York State Office of Children and Family Services.

87.

NYCHA (2016) New CCTV cameras installed at 31 developments. The NYCHA Journal 46(2): 1–9.

88.

NYCLU (2006) Who's Watching? Video Camera Surveillance in New York City and the Need for Public Oversight. New York, NY: New York Civil Liberties Union.

89.

NYCLU (2016) Stop-and-frisk data. Available at: www.nyclu.org/content/stop-and-frisk-data (accessed 20 November 2017).

90.

NYPD (2010a) Best Practice: Real Time Crime Center: Centralized Crime Data System. Edited by New York City Police Department. New York, NY: City of New York.

91.

NYPD (2010b) Mayor Bloomberg, Police Commisioner Kelly and MTA Chairman Walder Activate Security Cameras inside Times Square, Penn Station and Grand Central Subway Stations as Part of NYPD's Midtown Manhattan Security Initiative. New York, NY: New York Police Department.

92.

NYPD (2013) Policing by Place: A Proposed Multi-Level Analysis of the Effectiveness of Risk Terrain Modeling for Allocating Police Resources. Edited by National Institute of Justice. Washington, DC: National Institute of Justice.

93.

NYPD (2015) Developing the NYPD’s Information Technology. Edited by New York Police Department. New York, NY: City of New York.

94.

NYPD (2016) New York City’s low crime rate, 2015. In: New York Police Department (ed.) NYPD News. New York, NY: City of New York.

95.

NYPD (2017) NYPD technology: Helping the finest keep NYC safe. In: NYPD News. New York, NY: New York Police Department.

96.

NYPD (n.d.) New York City Police Department Releases a Draft of the Public Security Privacy Guidelines for Public Comment. New York, NY: New York City Police Department.

97.

Peck

Theodore

(2008) Carceral Chicago: Making the ex-offender employability crisis. International Journal of Urban and Regional Research 32(2): 251–281.

98.

Pew (2016) Use of Electronic Offender-Tracking Devices Expands Sharply. Washington, DC: Pew Research Center.

99.

PJI (2012) Using Technology to Enhance Pretrial Services: Current Applications and Future Possibilities. Edited by Bureau of Justice Assistance. Washington, DC: US Department of Justice.

100.

Porter

Sophia

Mary

(2002) Balancing Punishment and Treatment: Alternatives to Incarceration in New York City. New York, NY: Vera Institute of Justice.

101.

Rayley

(2013) Dataveillance and countervailance. In:

Gitelman

(ed.) 'Raw Data' is an Oxymoron. Cambridge: MA: MIT Press, pp.121–146.

102.

Revere

and Curbelo

(2001) Alternative-to-Incarceration Information Services End of Year Report: Fiscal Year 2001. New York, NY: New York City Criminal Justice Agency.

103.

Rose

(1998) Mapping technology takes a bite out of crime. Business Geographics, June, 98.

104.

Russo

(2001) Helping corrections inside and outside prison walls. Corrections Today 63(7): 140–144.

105.

Sciarabba

(2009) Community oriented policing and community-based crime reduction programs: An evaluation in New York City. Professional Issues in Criminal Justice 4(2): 27–41.

106.

Shabazz

(2009) So high you can’t get over it, so low you can’t get under it: Carceral spatiality and black masculinities in the United States and South Africa. Souls: A Critical Journal of Black Politics, Culture, and Society 11(3): 276–294.

107.

Sibley

van Hoven

(2009) The contamination of personal space: Boundary construction in a prison environment. Area 41(2): 198–206.

108.

Siddiqi

(2001) Prediction of Pretrial Failure to Appear and Alternative Pretrial Release Risk Classification Schemes in New York City: A Validation Study. Edited by New York City Criminal Justice Agency. New York, NY: City of New York.

109.

Signore

(2010) NYPD tightens surveillance in subway’s “Ring of Steel.” Gothamist, 21 September.

110.

Silverman

(1999) NYPD Battles Crime: Innovative Strategies in Policing. Boston: Northeastern University Press.

111.

Solomon

and Ferri

(2016) Community Supervision as a Money Bail Alternative: The Impact of CJA’s Manhattan Supervised Release Program on Legal Outcomes and Pretrial Misconduct. Edited by Criminal Justice Agency. New York, NY: City of New York.

112.

Stuart

(2014) ICE Is Fitting Immigrant Mothers with Ankle Monitors When They Arrive in New York. The Village Voice, 15 August.

113.

TechBeat (2002) CORMAP It. Edited by National Law Enforcement and Corrections Technology Center. Washington, DC: National Institute of Justice.

114.

Thrift

French

(2002) The automatic production of space. Transactions of the Institute of British Geographers 27: 309–335.

115.

Turner

(2016) The Prison Boundary: Between Society and Carceral Space. London: Palgrave Macmillan.

116.

van Hoven

Sibley

(2008) ‘ Just duck’: The role of vision in the production of prison spaces. Environment and Planning D: Society and Space 26: 1001–1017.

117.

Verton

(2012) Cybersecurity: NYPD, Microsoft Launch Domain Awareness System. Homeland Security Today, 13 August.

118.

Vigneswaran

(2014) The contours of disorder: Crime maps and territorial policing in South Africa. Environment and Planning D: Society and Space 31: 91–107.

119.

Wacquant

(2002) Mass Incarceration: Rethinking the ‘race question’ in the US. New Left Review 13:41–60. “From Slavery to

120.

Wilson

(2015) Flashing lights in the quantified self-city-nation. Regional Studies, Regional Science 2(1): 39–42.

121.

Young

Rachel

Gail

et al . (1999) Alternative to Incarceration Programs for Felony Offenders in New York City. New York, NY: Vera Institute of Justice.