Why policing the risk society became a footnote in American police studies: A missed opportunity to move police theorizing forward

Introduction

As of 2019, Policing the Risk Society (hereafter, PRS) has been cited over 2,000 times according to Google Scholar. PRS is cited, usually, whenever the phrase ‘risk management’ and ‘policing’ appear together, almost as a perfunctory acknowledgment of its place in police scholarship. Despite its pivotal role in the shift in theoretical framework from a post-crime era to a pre-crime one where uncertainty, security, and risk management precede the reactive character of criminological enquiries in general (Zedner, 2007) and police studies in particular, the potential influence of PRS has been absent in American literature on policing (see Bayley and Shearing, 1996). The trend is almost opposite in Canada, Australia, New Zealand, and the United Kingdom where criminological enquires and police studies have moved toward the incorporation of risk, governmentality, and security into their research questions and analyses (e.g. Mazerolle and Ransley, 2005; Shearing, 2016; Young, 2007, 2011). Moreover, the significance and impact of PRS—its potential to inform the theory, function, and role of the police for the future—rarely appear in the mainstream, positivistic accounts of police studies in criminal justice scholarship (O’Malley, 2015). Despite being an integral part of police theorizing, the idea of risk has at best been only a tangential focus of academic research on the police—particularly in the United States (for exceptions, see Brayne, 2017). Written by two scholars who have been trained and/or taught in criminology and criminal justice departments in US institutions of higher learning, we provide a theoretical speculation as to why this omission may have occurred in late 20th century American police scholarship.

With the exception of Ericson and Haggerty’s (1997) book, Policing the Risk Society, which focused on Canadian police, few researchers have attempted to fully incorporate the idea of risk into a more robust understanding and theory of contemporary police work. Ericson and Haggerty’s book should have provided such an impetus for a risk focused research agenda and theoretical framework. However, as O’Malley (2015: 430) argues, it instead ‘became just a convenient anchorage point for mooring many a minor study that found evidence of risk in police work…[it] was a landmark provocation wasted on an uncritical audience’. In this article, we provide a speculation of why PRS may have found such uncritical reception by policing scholars in America.

According to established canons that attempt to account for broad shifts in the theoretical activities of a discipline, paradigm shifts occur due to outliers that cannot be explained by the existing frameworks. The accretion of significant anomalies wends in uncertainty, followed by trenchant debates before a new paradigm emerges (Kuhn, 1996). The last 20 years since the publication of PRS may reflect the period of dueling paradigms in police studies, although some have argued that we are in the midst of another era (Hooper, 2014; Kim and de Guzman, 2012; Oliver, 2006). This account, however, is implausible as community policing has been a political and rhetorical success. That is to say, the ascendancy and perdurability of a theory depends on factors that are extraneous to its internal coherence or external recognition. For example, the predominance of Gottfredson and Hirschi’s (1990) self-control theory throughout the 1990s and 2000s in criminology is attributable to the ease of its testability, the discipline’s preferred methodology (self-report surveys) using readily available adolescent populations, and the theory’s ability to generate disprovable propositions (Cullen, 2011).

Using Kuhn’s model of change, the emergence of community policing as the logical step after the reform/professional era is anticipatable. The professionalization of major police departments ought to have ameliorated the problems of the patronage system and corruption of the preceding era. Yet, the reforms that were introduced to improve police service delivery and police–community relations, largely through technological advances—radios and telephones that facilitated disinterested and impartial contact between the public and the police; police cars that standardized rapid response as a default metric of effective police service; and rigorous testing in their recruitment of personnel—failed to materialize, resulting in urban disorder throughout the 1960s and 1970s. Such failures in police–public relations, especially between minorities and big-city police, would have paved the way for the emergence of a new paradigm in policing, as Kuhn’s model of theoretical change would have predicted.

The implications of PRS should be fully teased out, for the emerging technologies Ericson and Haggerty described in their work has altered policing in exponential ways due to the application of ‘big data’ in police departments as well as other government and social services (Brayne, 2018). PRS should be interpreted, appreciated, and expanded against the past technological infrastructure that is described in their work as well as the emerging predictive analytics that guide police work today (Brayne, 2017; Joh, 2016). In this article, we provide several factors that may have contributed to the omission of PRS as a new paradigm in police studies in the United States. We argue that the glossing over of PRS by contemporary police scholarship in the 1990s can be attributed to following three reasons: (1) incommensurable theoretical frameworks, (2) timing and aim of the book’s publication, and (3) the intrusiveness of deductive surveillance technologies in the policing of identities. The implications of PRS on contemporary policing are discussed.

Incommensurable theoretical frameworks

PRS does not fit into any of the existing theoretical frameworks that account for the origin, function, and evolution of the police. The three eras of policing that have been couched as a theoretical framework to understand the evolution of the police is historically incorrect. The three eras describe the origins, changes, and practices related to American policing, but they do not provide the rationale and cause of those changes. That answer does not exist because the existing frameworks describe the changes in police practices throughout American history, not the theories that explain them.

The gradual evolution of American policing is masked as theories of change. In a simplified version, the police changed due to the excessive graft and corruption that occurred in 19th-century America (political era), the urban unrest in minority communities caused by police brutality (reform era), which ushered in the period of partnerships (community policing era). Although some have begun to argue for the emergence of a fourth era in policing—information era (Hooper, 2014), Homeland Security era (Kim and de Guzman, 2012; Oliver, 2006)—the proliferation of adjectives in policing represents changes in tactics related to policing, not its theories. A more historically cogent account of police change has been proffered in recent works, in a way that takes into consideration the role of the state (O’Connor and Shon, 2019). We argue that PRS fits most cogently into this model of change, as the findings in PRS support the notion of police being governed by the needs of other institutions in ways that have transformed policing.

PRS does not fit into the existing theoretical frameworks because it is already embedded in a theory of macro-level change. Namely, Ericson and Haggerty’s work on policing draws upon the risk society thesis put forward by Ulrich Beck (1992, 1997). In societies marked by risk, traditional social structures guiding people’s lives begin to dissolve and people become increasingly impacted by uncertainties and social changes. That is, people face increasing economic and ontological insecurities (Young, 2007) as the state retreats from the business of governing. For example, gone are the days where employment could be considered stable and lifelong (Beck, 2000). Instead, in risk society, social structures change and work becomes temporary, precarious, and part time (Sassen, 2001). To navigate this uncertainty and social change, individuals and institutions begin to think and organize themselves in terms of risk. PRS argues that changes in policing are brought about by the changes in society, that a risk society brings new contingencies and risks that must be palliated and managed despite the radical changes brought on by technology, immigration, and pluralism. Thus, Ericson and Haggerty’s analysis of police behavior occurs after they have already assumed a theory of social change. In order for the canonic three eras of policing to work, Kelling and Moore (2006) must first clarify this presupposition about social change in America, circa 1840s to 1990s, and then show why the change that affected America and other social institutions also affected the police, in a way that is theoretically parsimonious. That is precisely what PRS does: it draws on the risk society thesis to illustrate the changes that are occurring in police organizations in ways that reflect the needs of a society that is risk-based.

The essence of Ericson and Haggerty’s (1997) argument, as it applies to police, is that in risk society the police become risk communicators and managers. To accomplish this transformation, the police increasingly focus their attention on managing information and knowledge. Demands from outside of policing for risk information position the police as experts in risk knowledge. Thus, these outside institutions—such as financial, education, and health—begin to shape the character and practices of contemporary policing (Buerger and Mazerolle, 1998). Enforcing laws, fighting crime, and maintaining social order become less and less important as entering, collecting, distributing, and managing risk data, primarily for other institutions, ascends as the driving force behind policing.

This substantial change in the focus of the police marks the shift from the police as a self-governed organization to an other-governed organization (O’Connor and Shon, 2019). This other-governed character is more acute in the age of ‘big data’ as police, like other commercial institutions, purchase private data such as foreclosures, social media, and call data from fast-food businesses, just to name a few, which are then integrated into their analytical platforms for surveillance and countersurveillance purposes (Brayne, 2018; Walsh and O’Connor, 2019). In this sense, the police unwittingly give up their autonomy as they become reliant on data brokers and data firms to populate their databases, which are then integrated across platforms and boundaries—without becoming fully cognizant of the consequences (Brayne, 2018). Other social agencies, private companies, and data peddlers are stepping in to reshape the role and function of the police—contemporary policing—as they define the needs of the police rather than the police themselves. The intrusion of others (i.e. software engineers and programmers) into the role and function of the police is one key way that the retreat of the state from its mandates is canalizing inroads into the other governance of police.

The timing and aim of publication

PRS was published in 1997. By then, the key texts on community policing had already saturated the policing literature, particularly in the United States. Therefore, it had the effect of being crowded out amidst other police scholarship. Although one can argue that important works still rise above their competing narratives, that lack of objectivity on the part of the audience palliates the impact of trenchant arguments such as the one made in PRS, it is precisely the exclusion of such minority doctrines that is effectuated by the putative objectivity of dominant discourses. ¹ Thus, a process of theoretical pigeonholing may have occluded the ascendency of PRS as a theoretical framework and its import in an American context (Bottoms, 2000). As part of the dissatisfaction with the professional model of policing, the thrust of discourse in police studies was already headed in a particular direction. Throughout the 1980s and into the 1990s, key US government and municipal-sponsored publications extolled the virtues of community policing as a revolutionary paradigm that was to logically follow the reform era (Barnett and Bowers, 1990; Freeman, 1989a, 1989b; Hartmann, 1988; Horne, 1991; Kelling and Moore, 1986). The published impetus for a new model of policing emerged in 1982 as Wilson and Kelling (1982) argued for the significance of broken windows on civil order, its impact on crime, and the implicit new role for the police (see also Moore and Trojanowicz, 1988b; Wilson and Kelling, 1989). Other key texts provided a blueprint for implementing the proposed ‘revolution’ in policing: specific techniques and case studies of problem-solving in troublesome neighborhoods; how to decentralize hierarchical chains of command; how to change organizational culture; how to revise performance assessments in ways that were consistent with the values of community policing (Brown, 1989, 1991; Goldstein, 1979; Hudgins, 1990; Kelling, 1988; Moore and Trojanowicz, 1988a; Sparrow, 1988; Spelman and Eck, 1987, 1989; Stephens and Watson, 1990). Kelling and Coles’s (1996) monograph on broken windows policing was published in 1996. The preceding key publications set the stage for the emergence of community policing as the ascendant third paradigm in the evolution of the police.

In addition to the timing of the publication of PRS, there are other theoretical reasons why PRS may have escaped the critical discussion of scholars. There is no mistake about the number of subjects Ericson and Haggerty interviewed for their study: 155. Yet, their study diverged in a key way from previous studies. Although PRS interviewed police personnel in various roles, their study was not concerned with the discretionary decision-making of patrol officers emblematic of the ethnographies of the 1960s and 1970s (Banton, 1964; Bittner, 1967a, 1967b; Chevigny, 1969; Muir, 1977; Rubinstein, 1973); the study differed from the systematic observations of individual police behavior (Black, 1980; Mastrofski and Parks, 1990; Reiss, 1970) or management characteristics of police agencies (Wilson, 1968).

PRS situates policing during global changes in patterns of work, family, technology, and society and their impact on police behavior across the social spectrum, not just one targeted population such as police recruits or juveniles (e.g. Black and Reiss, 1970; Niederhoffer, 1967). Although police personnel were interviewed, they were tangential to the functions they performed in police organizations embedded in a society obsessively creeping toward risk management. Therefore, the unappreciated subject in PRS is the policing communicative system in a risk society. It would be more congruent to view the police officers and civilian personnel who worked in police agencies as subjects who happened to be caught up in an ethnography of a police communication system à la Peter Manning. Manning’s work (1982: 124; 1988) conceptualizes the police as processors of signs containing a ‘structural system for encoding incoming data’. His work, like PRS, provides a way in which the police invoke rules and practices through a semiotic filter with which they organize and interpret the vicissitudes of police work. This point is supported from one of the main conclusions that can be drawn from the book: PRS examines the patterns of dispersal of hierarchy in policing a risk society.

Dispersal of authority from the hierarchy of a paramilitary organization; the decentralization of authority from the chain of command to the patrol officer; and the maximization of autonomous and creative problem-solving skills culminating in discretionary decision-making were the purported aims of this emerging third paradigm centered on community policing. This reconceptualization of the police role, it was argued, occurred due to the limitations of a policing system that was crippled by its reactive character rather than one that addressed the underlying causes of the calls (Goldstein, 1979). In the process, various policymakers and academics advocated for the needed changes in police operations, performance assessment, and police culture. In other words, the community policing texts provided a blueprint for police executives on how to bring about organizational change and implement those changes given the obstreperousness of police culture. They tried to get police executives to do something. In a word, the emerging community policing texts were prescriptive in character.

PRS offered no operational directives, best practices, or guidelines for implementing policy. It did not provide a blueprint for how to alter police operations in the context of increased risk, provide new assessment measures for performance, or suggest ways to change police culture in risk society. Rather, PRS described the way police operate in the context of a shift in social and economic relations on a global scale. Although the setting of PRS was in Canada, the system that Ericson and Haggerty described in 1990s is just as applicable to police agencies—as well as fusion centers in the United States—elsewhere in Western nations.

The findings of PRS also militated against one of the major initiatives of community policing: the erosion of discretion by patrol officers. The patrol officers in PRS were imputed to be mere data entry clerks in the statistical life of populations, policing texts and narratives and checking boxes in fixed response categories, responses demanded and created by the risk needs of other institutions. The principles of community policing required the exact opposite of Ericson and Haggerty’s findings. The primary documents of the 1980s and 1990s extolled the merits of increasing the discretion of patrol officers who are assigned to community policing initiatives, going so far as to change the performance assessment structures to align the police organization’s values toward this new mission. Police executives wanted performance assessment to reflect an absence of crime rather than arrests or calls answered. Therefore, while this community policing rhetoric was aspirational, the aim of PRS was to describe the actual—sociological—character of policing.

The appearance of intrusiveness in the policing of identities

The word ‘surveillance’ is polysemous and encumbered in different ways, with sundry histories, origins, and definitions. The most obvious is the dictionary version: close observation of another. Others have referred to this type of surveillance as directed or inductive surveillance, for it focuses on individuals and places under suspicion (Marx, 2002). This type of state-sanctioned surveillance has a deeply regretful past in American history, conflating police work with intelligence work, which resulted in egregious violations of the law. Federal agencies such as the Federal Bureau of Investigation (FBI), and municipal-level law enforcement agencies, such as the Los Angeles Police Department and Chicago Police Department, engaged in direct surveillance, penetration (i.e. insertion of undercover law enforcement officers in target organizations), and sabotage operations against those they perceived and labeled as a threat to the social order throughout the 1960s and 1970s. They did so in ways that exceeded the boundaries of conventional undercover police work (e.g. Marx, 1992) and adumbrated the predictive interdiction activities of intelligence agencies.

The Counter Intelligence Program (‘COINTELPRO’) carried out by the FBI against political figures, political protesters, and purported ethnic identity groups (e.g. Black Panther Party, Ku Klux Klan) is a notorious legacy in the annals of American law enforcement as paragons of police power gone too far (Final Report of The Select Committee to Study Governmental Operations, 1976). Another notable point is that these types of domestic surveillance of citizens occurred without due cause and warrants, thus violating one of the core principles of search and seizure. Words like ‘government’, ‘control’, and ‘surveillance’ to describe the police operations reek of police excesses that might be unpalatable for American sensibilities (Theoharis, 2004). Moreover, the technologies of surveillance described in PRS connoted the failure of the government to check its own excesses and power, thus contravening one of the foundational principles of its own governance as well as fundamental principles of procedural justice (Tyler, 2006). Simply put, the type of surveillance described in PRS could be seen as being problematic, if viewed from American history and sensibilities.

The second usage of surveillance is best described as governmentality. This body of literature draws on the work of Foucault (2003a, 2003b) and explores the theoretical linkages between sovereignty, discipline, and government as populations are governed and managed in subtle ways that reify the hierarchical relations between the people and the state (Leira, 2009). In this framework, the police act as a technology of discipline alongside other technologies, such as schools and prisons (see Wacquant, 2001). The police are tasked with improving the welfare of populations by encouraging populations to self-discipline. Therefore, the police and other governing bodies encourage individuals to self-discipline and populations to self-govern through the threat of constant surveillance (Rose et al., 2006).

The third type of surveillance described in the literature is deductive. It is not directed at any particular person. Rather, it is similar to a dragnet that swallows anything that happens to come that way. Deductive surveillance captures places and contexts rather than specific people (Marx, 2002). This deductive character of surveillance has been enabled by ‘big data’. Data collection that is deductive is involuntary, warrantless, and co-opts quotidian machines during the course of everyday life (e.g. using an ATM or a smartphone) to amass samples in the millions as a way of constructing ‘data doubles’ which then becomes the objects of surveillance across integrated platforms (Haggerty and Ericson, 2000; Linder, 2019). This deductive surveillance in its incipient stages is what PRS described. This type of predictive analysis based on big data constituted the purview of US intelligence agencies (Bamford, 2001), not municipal police.

The communicative system and the analytical platforms used in the 1990s were the prototypes of the digital revolution to emerge, which has been enabled by big data. PRS suggests that police might be targeting specific groups of people through police knowledge systems and techniques of surveillance, without appropriate oversight or warrants. This impression was created in the minds of readers because a police agent would have had to identify, flag, and disseminate the information to another social control agent. The deductive character of surveillance blurs the applicability of fourth amendment protections as no targets are identified (Ferguson, 2017; Joh, 2014). In the age of big data, such suspicion alerts are outputs created once the search parameters and input criteria are fed into the analytical platform, which are then automatically texted to a patrol officer’s smartphone (Brayne, 2017). In the 1990s, an analyst might have picked up a phone and alerted someone about a suspicious pattern or a person. Deductive surveillance assembles digital traces in a way that scoops up more people, even noncriminals, into databases thereby casting a wider net; it penetrates deep and is able to trace a single individual across a broader array of institutional contexts, in ways that reproduces existing social inequalities (Brayne, 2014; Eubanks, 2017; Ferguson, 2017).

Deductive surveillance then captures the intrusiveness of targeted observations and the illustration of governmental power of the Foucauldian ilk through amorphous data collection and panoptic surveillance. This type of description would have been incredulous to practitioners and academics caught up in the philosophy of community policing. It was based on values such as trust, cooperation, and the formation of partnerships between the police and the public. Values that guide deductive surveillance are much less prosocial. The language of big brother surveillance—inductive and deductive—would not have been meaningful or welcomed in the proliferation of discourses related to community policing. This difference may be another reason that readers in the United States failed to appreciate the findings and implications of PRS.

Discussion

In this article, we have attempted to illustrate why PRS was glossed over by police scholars in the United States in exchange for a focus on community policing. A key tension exists in the current paradigms in policing. Community policing requires the exercise of maximum discretion by patrol officers while the emerging patterns in police work indicate that patrol officers exercise less discretion. The data entry functions performed by patrol officers in PRS captured the beginning stages of a policing system that was inching toward incorporation of big data; the language and the analytical capabilities did not exist yet, but the lower level rank and file began the task of dutifully entering data. That is why they were referred to as knowledge workers. We suggest that PRS provides a more cogent theoretical framework for understanding recent developments in contemporary policing such as the incorporation of big data into policing than community policing. Consequently, we advocate the revisiting and reconsideration of key findings from PRS into the theoretical infrastructure of contemporary police scholarship.

Examining some of the current trends in contemporary policing bolsters Ericson and Haggerty’s arguments made in PRS. The police thinking and acting in terms of risk (i.e. becoming risk managers, risk communicators, and knowledge workers) has only become more entrenched in policing since the book’s publication. In particular, there are several trends that lend credence to PRS’s larger arguments. First, big data, algorithms, and predictive analytics are increasingly being utilized to assess risk and dictate police practices and responses (Brayne, 2017; Hutt et al., 2018). For example, hot spots policing uses data to dictate the amount of time officers spend patrolling specific locations found to be high in crime (Ferguson, 2017; Koper et al., 2015; Lum et al., 2017). The police collect and analyze data to predict risk, and officers are asked to patrol the predicted high-risk areas. One could argue that the alert-based systems that automatically notify police where to go have turned police into errand clerks who respond to the outputs generated by algorithms. There are few remnants of community policing and building partnerships with community members in these techniques. Instead, the partnerships that are formed are with technology companies who dictate the form and substance of data inputs and outputs (Aradau and Blanke, 2017; Cohen and Graver, 2017).

Second, the rise of big data policing portended in PRS and the gradual erosion of the philosophy and practices of community policing ought to be of concern given the history of policing in America. Community policing emerged in the light of the public’s dissatisfaction with the ways that the police exercised their authority against the public, especially in minority communities, as they framed themselves as the singular experts in matters of crime control and citizens as mere passive consumers of police services. According to the presuppositions of policing in the reform era, technology—telephones, radios, and cars—was purported to be the solutions to problems in policing. Yet, technology meant to deliver impartial and bias-free policing exacerbated problems between the police and the public while pushing performance assessment in a quantitative direction. One of the acknowledged mistakes of the reform era of policing was that citizens were omitted from the consultation process as police leaders formulated crime reduction strategies. Put another way, the police governed the citizens (e.g. Baldwin, 1990). As implicitly argued in PRS, and as we have explicitly argued here, the police are in the process of becoming other governed. This trend is significant as the police no longer wield epistemological authority to define the boundaries of their work or frame themselves as the experts in matters of crime and risk.

Third, the encroachment of big data in policing has the potential to reconceptualize the dyadic character of the police as the source of ‘othering’ to the ‘other’. In previous literature, the police have been seen as the wielders of power that have interpellated the identities of men of color as undesirable subjects of policing (Althusser, 1971). We have argued that the trajectory of policing is moving toward other governance, with the police being governed by the needs of other social institutions while the citizens are becoming subjected to unfettered searches and surveillance from private and public institutions alike. The police also are not immune from this process, for the very strokes and searches they perform at work, the record of their interactions with the public, as well as their bank records, major retail purchases, and social media usage have the potential to be sources of data for early warning and intervention systems (Recommendations for Reform, 2016). That is to say, the police themselves may become ‘otherized’ as the targets of deductive surveillance in ways that have yet to be addressed by police management, police scholars, and existing collective bargaining agreements. This epistemological shift has been unwittingly authorized by the emergence of technology where computer programmers and software engineers have assumed the position of technocrats who set the parameters of the definition in search queries. In a way, police no longer ‘define the situation’ (Van Maanen, 1978); the data brokers and software programmers do. This movement toward the police as targets of their own surveillance reifies the Foucauldian notion of panoptic discipline connoted in the governmentality literature.

Fourth, the incorporation of big data into policing also has implications for the public’s relationship with the police. As procedural justice researchers have shown, citizens’ willingness to comply with the law is shaped by factors that are not necessarily related to police effectiveness or their adherence to legal requirements (Meares et al., 2016). Numerous researchers have shown that citizens’ perceptions of legitimacy are related to the presence of opportunities to voice their opinions as well as the impartiality of the police in their decision-making (Sunshine and Tyler, 2003a). Trust and legitimacy in police have been defined by the willingness of the public to ascribe benign intention to the motives of police as well as the quality of treatment they receive at the hands of the police (Hamm et al., 2017). It remains unclear how the intrusion of big data into patrol work in a society that is running full-steam toward risk will affect how the public ascribes benevolent intention to the police or improve the quality of treatment that citizens receive.

In reality, the intrusion of big data into police work has the potential to adversely affect police officers and citizens in similar ways. The reliance on big data removes the agency of the police to refine their tradecraft, becoming mere responders to alerts on their smartphones rather than learning the skills of observation-based policing that has been the essence of good police work (Sacks, 1978). Moreover, the police may become cognizant of the larger patterns of data collection and usage, for they themselves become objects of inquiry and surveillance, in ways that the law has yet to elucidate (Joh, 2014). Moreover, the punishment some police officers have faced for inappropriate social media content illustrates the potential pitfalls of surveillance applied inward. The police may have justifiable reasons to be more distrustful of their own profession given their unique position as insiders.

Similarly, there may be a decrease in public trust and support for police as opportunities for citizens’ involvement dissipates. The sense of empowerment that participating in a community meeting brings may be lost as the voice of the public is relinquished to software engineers who construct the algorithms that dictate where the police can go and how long they should stay. Software engineers are guided by the logic of the programming language, not the voices of citizens living in fear. In this way, the participation framework behind procedural justice may affect the legitimacy of the police as an institution for citizens and police officers alike as the notion of trust may be realigned in the age of big data in a society deluged in risk. The four core principles that shape citizens’ perspectives on procedural justice—voice, impartiality, benevolence, and respect—just may influence how the police view themselves, their work, and their office in the age of big data (Murphy and Tyler, 2017).

Fifth, social media is increasingly being used by police to assess and communicate risk to the public. For example, social media is treated by police as a data source and when combined with data analytics, algorithms, and other data sources, it leads to automated investigations that help identify risky individuals, groups, places, and sites for surveillance (Brayne, 2017; Chan and Bennet Moses, 2016; Ferguson, 2017). The goal of this ‘dataveillance’ (van Dijck, 2014) presumably becomes preventing crime before it occurs (Aradau and Blanke, 2017; Dencik et al., 2018). It is the data points and analytical systems that dictate risk and subsequent interventions. This is far removed from community policing’s rhetoric around collaborative police-community problem-solving. On the public communications side of social media, the police have embraced social media as a tool for risk communication. Approximately 96% of police departments regularly utilize social media as part of their policing practice (IACP, 2016). Social media is particularly useful in disseminating risk information in times of crisis (e.g. natural disasters, terrorist attacks, and riots), given the immediacy of the platform and its amplification capabilities of having citizens repost police content. Social media also allows the police to circumvent the media and position themselves as risk experts. However, most studies have found that social media is predominantly used to push (risk) information to the public (Crump, 2011; Lieberman et al., 2013; Procter et al., 2013) with little of the key aspects that make up community policing being present. That is, two-way communication between the public and the police via social media platforms is atypical (Walsh and O’Connor, 2019).

Sixth, the rise of evidence-based policing (EBP) in several countries in the early 21st century provides additional support for the importance of PRS to police theorizing and research. EBP emphasizes the importance of incorporating research evidence into the everyday practices of policing. Sherman (2013: 377) refers to EBP as using ‘targeting, testing, and tracking’ to direct police resources and strategies. Societies of EBP have recently emerged in the United Kingdom, Australia and New Zealand, the United States, and Canada. These societies place emphasis on the police practitioner conducting research and determining ‘what works’ in policing based on methodologically sound research. While this evidence-based research may be conducted in partnership with academics and other researchers, police personnel are encouraged to lead and develop research projects. PRS identified the early stages of this trend when describing police as knowledge workers. Now, we see the beginnings of police personnel moving from data entry to data analysis and evidence testing. Again, this trend in policing has little to do with community policing and is best explained by the findings laid out in PRS. While EBP does not offer a theoretical base from which to understand changes in policing, we argue that PRS does. Although EBP tends to currently operate with an empirical focus, we encourage future researchers to examine the theoretical connections between PRS and EBP.

Conclusion

We have argued in this article that PRS was overlooked by American policing scholars for three principal reasons: (1) the incommensurability of the risk society thesis with the dominant (inaccurate) model of change in policing. (2) The saturation of the literature with the seminal texts in community policing, which effectively crowded out PRS, pigeonholing it on two fronts: theoretical and national. (3) The distastefulness of the appearance of intrusiveness of deductive surveillance technologies practiced in the policing of identities by Canadian police agencies. Given the ugly history of inductive surveillance and its incumbent illegal acts perpetrated by American law enforcement, we argued that American readers would have found such a trend unpalatable. Using these three reasons as the foundation for our argument, we have tried to show that this glossing over of PRS was a missed opportunity to move police theorizing forward in important ways. The evolution of trends such as big data, social media, and EBP highlight the need to reconsider community policing as the guiding paradigm in American policing. Therefore, we encourage other scholars to reexamine the theoretical presuppositions inherent in the community policing paradigm and the possibilities PRS offers. By tying changes in policing to larger societal changes centered on risk, PRS provides police scholars a theoretical framework grounded in descriptive empirical data. In the light of recent developments in policing, it is now time for American policing scholars to give PRS more serious attention.

There are, however, limitations of our work. We have provided a mere speculation as to why PRS never achieved the ascendancy that it deserves given its import, despite the impressive citation index. It may be worthwhile to simply ask the policing scholars themselves why they never incorporated tenets from PRS into their theoretical and empirical research agendas. It is possible that some may have been turned off by the sheer length of the text and not bothered to read it in its entirety, or American scholars may have dismissed the relevance of PRS because the findings were derived in a Canadian city, situated in Canadian police agencies, written by Canadian scholars, thereby prejudicially pigeonholing the applicability of the findings to their American contexts ab initio (Bottoms, 2000). Without additional empirical work, it is impossible to discern the true reasons. We have proffered a speculation as to why American policing scholars may have missed the significance of PRS.