MSc (Res) Thesis
Polycentrism, illegality and the
urban space
The case of community policing in Jamaica
Max Méndez Beck Student ID: 11126833 Msc (Res) International Development Studies 13 of June, 2017 Contact: maxmendezbeck@gmail.com Word count: 31.268 (excludes References, Appendix, and Footnotes)
Graduate School of Social
Sciences
Research Master International Development Studies
Thesis supervision and evaluation Supervisor: Rivke Jaffe Professor of Cities, Politics and Culture Department of Human Geography, Planning and International Development Studies Centre for Urban Studies Amsterdam Institute for Social Science Research University of Amsterdam Nieuwe Achtergracht 166 1018 WV Amsterdam The Netherlands R.K.Jaffe@uva.nl Second reader: Dennis Rodgers Professor of International Development Studies Department of Human Geography, Planning and International Development Studies Centre for Urban Studies Amsterdam Institute for Social Science Research University of Amsterdam Nieuwe Achtergracht 166 1018 WV Amsterdam The Netherlands D.W.Rodgers@uva.nl
Abstract— In recent decades, aid agencies have increasingly pushed for the international
adoption of a new policing strategy known as community policing. The strategy is said to reduce levels of crime by establishing relations of trust between police officers and residents of marginalized urban communities. This paper examines the assumptions that underlie community policing, explores the origins of the strategy, and then connects policymaking narratives to existing theories of urban governance. By identifying different schools of thought in the literature, it constructs three distinct models of urban governance that may apply to community policing and uses those models to explore the introduction of the strategy in two inner-city communities of Kingston, Jamaica. The paper shows that basic US-centric assumptions regarding the relationship between the police, the community, and, crucially, “extra-legal” governance actors do not reflect the experiences of inner-city residents. Furthermore, the Jamaican police’s embrace and reinterpretation of those US-based assumptions may lead to a misguided and ineffective implementation of the strategy. Instead, the paper argues for the use of more complex urban governance models, such as Ostrom’s polycentric systems approach, in designing urban policing strategies around the world.
Keywords— Urban governance; community policing; police reform; polycentric systems; policy
transfer; Jamaica
Table of Contents
List of Figures ... vi Acknowledgements ... vii Chapter 1 Introduction ... 1 Chapter 2 Theorizing Community Policing ... 8 Introduction ... 8 Community policing: its origin and tenets ... 9 Community policing: a basic theory and its assumptions ... 11 Community policing under the competing orders model ... 14 Community policing under the polycentric model ... 18 Conclusion ... 21 Chapter 3 Navigating the Violent Urban Context ... 23 Introduction ... 23 Downtown Kingston as violent fieldwork setting ... 24 Selecting the case studies ... 26 Participant observation ... 29 Semi-structured interviews and sampling ... 31 Positionality ... 34 Conclusion ... 35 Chapter 4 Policing Social “Deviance” ... 37 Introduction ... 37 The History of Community Policing in Jamaica ... 38 Targeting Youth within Community Policing Strategies ... 42 Correcting Social Deviance within Community Policing Strategies ... 44 Challenging Police Assumptions ... 48 Conclusion ... 51 Chapter 5 Modeling Governance in Naseberry and Maypole ... 53 Introduction ... 53 The community of Naseberry ... 54 The Police and the Community ... 56 Gangs and the community ... 59 The Police and the Gangs ... 64 The community of Maypole ... 66 The State and Gangs ... 68 Conclusion ... 69 Chapter 6 Conclusion and Recommendations ... 71 Segregation or co-optation? ... 74 Ideas for the way forward ... 75 References ... 78 Appendix ... 84Appendix A: List of Respondents ... 85
Appendix B: Participant Observations ... 86
Appendix C: Interview Guide ... 87
List of Figures
Figure 1. The state capacity model...13 Figure 2. The competing orders model...16 Figure 3. The polycentric model...20 Figure 4. Map of Uptown and Downtown Kingston...24 Figure 5. Stages and Foreign Donors in Rollout of Community Policing in Jamaica..40Acknowledgements
The following paper is the result of nearly two years of work and would not have been possible without the generous help and support of numerous individuals and institutions. First and foremost, I cannot thank my thesis supervisor, Rivke Jaffe, enough for her guidance during this process. Rivke helped me with this project in innumerable ways, of which I can only highlight a few here. For example, her deep knowledge of Jamaica along with her extensive experience in Downtown Kingston served as both a resource and a needed check for my assertions and speculations. She was incredibly helpful during the writing process, offering suggestions on theories to explore and ways to structure my thoughts. From the jumble of ideas and data, she continuously encouraged me to find and present a clear and coherent narrative. When, at times, I got bogged down in the nitty-gritty of a paragraph or section she was there to calmly remind me to keep in mind the bigger picture. This work is immeasurably better thanks to her reliable and patient support.
Rivke was also instrumental in my having a successful fieldwork experience. By generously offering to arrange a meeting with a friend and acquaintance of hers, I was able to gain access to one of my case study communities in Downtown Kingston. Without this contact, it would have been impossible for me to conduct the in-depth resident interviews that became central to my analysis. This brings me to the residents of the two Kingston communities that form part of my investigation. This work was only possible thanks to their warmth and generosity. I am thankful for all the evenings on the stoop, the open arms and thumb-flicking fist bumps, and the moments of “reasoning.”
My meetings with top-level security officials as well as beat police officers were possible thanks to the assistance of several individuals at USAID COMET-II. Donaree,
Jamaila, Jhana, and Ian bear special mention in this regard, but everyone at the office was incredibly welcoming and did everything they could to assist me with my investigation. I am also grateful to the Institute of Criminal Justice and Security at the Univeristy of West Indies Mona for providing me with a work space, and especially to Julian for helping with the arrangements. Finally, I would like to thank the police officers and crime experts who generously volunteered their time and provided their insights and perspective.
I hope this work is of some help to both the police officers and community residents of Jamaica as they continue to search for ways to live in peace and security.
Chapter 1
Introduction
“Murder Madness”, screamed the front-page of Jamaica’s largest newspaper, the Gleaner, on September 21st, with the subheading reading: “Jamaica averaging 100 homicides monthly” (Barrett 2016). Such headlines are commonplace in a country that persistently lands near the top of worldwide homicide rate rankings; in 2016, for example, it had the sixth highest murder rate in the world with 50.7 nationwide homicides per 100,000 inhabitants (for the sake of comparison, in Canada that rate was 2 murders per 100,000 inhabitants) (Igarapé Institute 2016). The issue of crime and violence has direct implications for the international image and economic prospects of the country and, as such, has become a perennial preoccupation for successive governments, focused as they are on attracting foreign investment and selling the country as a major tourist destination.
Among Jamaican police officials, government representatives, and crime experts, there appears to be little disagreement over who is to blame for the violence: powerful gangs financed through drug trafficking, extortion, and the now popular lottery scam schemes (where calls are made to elderly Americans informing them of having won a fictitious prize and then asking them to pay a fee to obtain their reward). While most experts seem to agree on the problem, the strategies for addressing it are still hotly debated. Prominent voices continue to call for a hardline approach of busting down doors and imposing curfews. One prominent columnist published a much discussed op-ed in January 2017, complaining of the “well-spoken defense attorneys and other human rights fundamentalists [who] clobber [politicians] if they dare to act decisively and tough” and demanding that civil liberties be curtailed “in the interest of all” (Boyne 2017). Such approaches, which
fade in and out of policy discussions every few years, have in the past led to the rampant use of extrajudicial killings on the part of the Jamaica Constabulary Force (JCF), making it one of the most violent police forces in the world (Amnesty International 2016).
In part due to those excesses and to fears that more draconian policies may actually “exacerbate the routine stigmatisation of gang members, thus preventing their reform and … reintegration into society,” the crime prevention policymaking landscape has shifted in recent years towards “softer” more community-driven approaches (Jütersonke et al. 2009; 11). Of these “softer” approaches, the most popular is known as community policing: a strategy that emphasizes the collaborative nature of creating secure communities and stresses the importance of the police building relationships with local residents. Rather than seeing the police as an external expert authority that imposes order by identifying and apprehending criminals, the community policing approach views the officers as a co-producer of safety, who seeks the guidance and assistance of the local civilian population in detecting problems and resolving disputes (Moore 1992).
Nudged by international donors, who have provided financial and technical support, in recent years the JCF has attempted to roll out community policing nationwide and claim it as the Force’s central philosophy (Chambers 2014). Proponents argue that the approach is necessary for fighting crime in Jamaica and point to the lack of general public support for the police, going so far as to claim that the organization faces a “crises of legitimacy” due to a widespread perception of the JCF as a corrupt, trigger-happy, and incompetent institution (Harriott 1997). These advocates argue that residents, particularly in marginalized urban areas, refuse to cooperate and share information with officers due to misgivings about the effectiveness and trustworthiness of the police (Reisig & Lloyd 2009). The widening distance between officers and residents, they suggest, leads residents of marginalized neighborhoods to turn to the gangs for protection and arbitration of justice.
According to this narrative, community policing can change this dynamic by embedding officers within communities and establishing meaningful ties to local residents. Through greater exposure and one-on-one contact with community members, the image of the police as an institution would improve, encouraging residents to share more information with police officers. The ability to tap into the “trust networks” within neighborhoods would allow the police to more easily identify the perpetrators of crimes and recruit witnesses to prosecute them. In other words, it would take away the protective cover offered to the gangs by a compromised community. The only way to wrest control of these areas away from the gangs, community-policing advocates say, is to get the residents on your side.
Implicit in the narrative are certain assumptions regarding the relationship between the primary actors of the modern urban setting, including community residents, state-backed security institutions (such as the police), and non-state “extra-legal” governance authorities (such as gangs). The accuracy of the assumptions directly
affects the ultimate success or failure of the community policing strategy. For example, the government may assume that a lack of community outreach on the part of the police generates a distance between the officers and residents that leads to the “breakdown” in the relationship, when in reality it may be the ability of gangs to produce more efficient public goods (such as security) that reduces their need to call on the police. Assuming the former may lead the government to focus one set of strategies (centered on improving the image of the police) while the latter would require a focus on a different set of strategies (centered on improving the provision of security).
The main question and central topic of this investigation revolves around these assumptions, the urban governance models on which they are based, and the implications those assumptions have on the effectiveness of the community policing strategy. My main research question, therefore, is as follows:
Main Research Question: What theories and assumptions of urban
governance underlie community policing strategies and how do they influence the strategy’s success or failure in reducing criminal violence?
In order to understand the governance models used to design and justify the community policing strategy in Jamaica, it is important to analyze the origins of the strategy. Given that community policing emerges largely within the United States urban context, the first sub-question of my investigation is:
Sub-question 1: What theories and assumptions dominate in US-based
discussions about community policing?
Beyond the original assumptions, the study also explores the manner in which those assumptions evolve as policies move from one context to another. In this case, community policing strategies were exported from the United States, through US-funded organizations, to the JCF. The transfer process may involve modifications (by both international and local policymakers) to US-based theories and assumptions that attempt to take into account distinctive features of the Jamaican context. The second sub-question addresses this topic:
Sub-question 2: To what extent are US-based theories and assumptions
reproduced by development and security policymakers in Jamaica when community policing strategies are exported?
Finally, the urban governance assumptions used by the different policymaking actors to design the community policing strategy, may be contested (or confirmed) by the experiences of the populations they target or serve. In the case of Jamaica, community policing strategies are focused primarily on marginalized communities where trust in the police is thought to be low. Therefore, the third and final sub-question is:
Sub-question 3: How well do the theories and assumptions of development and
security policymakers in Jamaica reflect the experiences of residents in marginalized communities?
I argue that policymakers implicitly apply certain assumptions about the urban landscape when designing and implementing community policing strategies. I then link these assumptions to current academic debates on urban governance, identifying in the literature several schools of thought. By combining this academic literature with the policymakers’ assumptions, I develop three models of urban governance that can be linked to community policing strategies.
The first of these models sees the police as a direct appendage of a unified state that is tasked with enforcing universally accepted laws and generating essential public goods (such as security) for all communities. As inefficiencies in the delivery of these public goods to certain communities emerge, the ties between police and residents erode, thus preventing the co-production of security. I call this the state capacity model.
The second model assumes a similar dynamic as the first—with a unified state, an inefficient police, and a “breakdown” in police-community ties—but it introduces a new non-state actor into the equation: the gang. The gang is seen as a competing actor that enters and fills the gaps left behind by the state-backed security institutions, generating a separate and independent order that erodes the legitimacy and standing of the state system. I call this the competing orders model.
The third model, inspired by Vincent and Elinor Ostrom’s work, modifies the previous competing orders model by arguing that, rather than representing a separate and antagonistic force, gangs form part of a polycentric system of urban governance where they represent localized decision making centers with intractable linkages to other centers within the system. Gangs, the police, residents, other state actors, and other non-state actors interact in both competitive and collaborative ways contingent on the issue, but always within a common underlying structure. That is to say, gangs and state-backed authorities may collude and conspire in systematic ways that undermine notions about their antagonistic relationship.
By applying these models to recent policy debates, I argue that views on community policing in the United States center on the state capacity and competing orders models, which emphasize the need for improvements in the delivery of state- generated public goods while simultaneously viewing any encroachment by “extra-legal” non-state actors as a direct challenge to the authority of the state-backed order. The perspective assumes a preeminent state, a well-intentioned (yet overwhelmed) police force, and a dissatisfied community exploited by non-state criminal actors.
Once exported to the Jamaican context, where the presence and prominence of gangs is significant, I suggest that the competing orders model becomes embedded
into longstanding culturalist understandings of crime and poverty. The Jamaican police link assumptions about the independent order generated by opportunistic gangs to the problem of culturally “deviant” subgroups. This “culturalist” interpretation of the competing orders model leads to a Jamaican community policing strategy that revolves around altering the social norms of the community in order to correct the cultural characteristics that the police force believe breed a separate order. The perspective assumes a preeminent state, a well-intentioned (yet overwhelmed) police force, and a culturally “deviant” community that contributes to and is embodied by the alternate gang-led order.
The experiences of Jamaican residents in marginalized communities expose a different dynamic that challenges the models and assumptions used by both US policymakers and the JCF. By highlighting the intricate ties between the police and the gangs, as well as the role that the gangs have historically played within the legal political structure, they point to a more complicated web of relations. This complex web undermines US policymakers’ and JCF’s neat picture of a battle waged between two separate governing authorities (one legitimate, one illegal) that vie for control of a territory, with the consent (or social norms) of the residents seen as the principal measure of success. Instead of autonomous, self-financed criminal actors that capitalize on the support of local constituencies, gangs in Jamaica form part of a larger national (and transnational) structure tied to larger business and political interests. Therefore, I argue that the resident narrative aligns with the assumptions of the polycentric model and may offer valuable insights into the potential weaknesses of current community policing approaches, not only in Jamaica but also around the world.
The complex, emergent, and opaque nature of this polycentric system implies that political and economic forces beyond the control of local police officers and community residents dictate the role played by gangs. These forces, if not included in the community policing strategy, can significantly undermine any efforts to delegitimize criminal governance agents. In other words, the work of the police to counter the influence of the gangs by getting residents “on their side” assumes that the power of the gangs is based primarily on the loyalty of the residents. However, the gangs may also be empowered by their relationship to the upper echelons of the system: judges, politicians, businessmen, etc. Furthermore, in situations where gangs are firmly ingrained within the formal political and economic system, attempts to dislodge them without permanently severing crucial ties to the multiplicity of decision centers that facilitate employment, public works, and order in marginalized communities can result in a worse local outcomes and greater distrust in state institutions among residents.
In answering these questions the paper contributes to the growing academic literature on modern urban dynamics in marginalized communities by, first, generating three models of urban governance that synthesize three distinct theoretical positions on the presence of and relationship between state-backed security institutions, “extra-legal” non-state actors, and marginalized communities.
Second, it applies those models of urban governance to current policy debates on policing and security. Third, it builds on the policy transfer literature, by describing the way contextually-based assumptions that underlie security policies travel and are reinterpreted in their new settings. Finally, the study adds to the scarce empirical literature on the introduction of community policing strategies in the Global South and argues that its success can be dependent on implementation issues but also, and more importantly, on entrenched local governance structures.
The paper proceeds as follows: I begin, in Chapter 2, by tracing the origins of the community policing strategy to the United States and discussing the logic used by policymakers and academics to justify the strategy. In doing so, I identify the policymaker’s assumptions and link them to current academic debates on urban governance, developing in the process three distinct models of urban governance; each model is based on different sets of assumptions about the relationship between the various actors involved in community policing. Chapter 2, therefore, attempts to answer the first part of the main research question (on the theories and assumptions of urban governance that underlie community policing strategies) as well as the first sub-question (on US-based theories and assumptions).
Chapter 3 explores the implications of my fieldwork experience, in light of the violent urban context in which it took place. In it I outline the various research methods applied during the five months I spent in Kingston, Jamaica, highlighting the ethical and methodological considerations of investigating a highly sensitive research topic in a marginalized and stigmatized area. This leads to a discussion on the limitations and potential biases of the information gathered, and offers the reader context for interpreting the empirical data.
Chapter 4 uses police interviews, government documents, and international reports to analyze the introduction and current status of community policing efforts in Jamaica. By analyzing the narratives offered by various JCF officers and reviewing the aspects of the strategy that received most attention, I identify the primary assumptions that underlie the strategy and link them to one of the urban governance models. The chapter ends by looking at how some of the challenges faced during the implementation process put in doubt the strength of the police assumptions. The chapter answers the second sub-question on the way US-based assumptions regarding community policing change when exported to Jamaica, and begins to address how those assumptions influence the success or failure of the strategy (the second part of the main research question).
In Chapter 5, I compare the police narrative to the perceptions and experiences of residents from two inner-city communities in Kingston, Jamaica. Using resident interviews, participant observation, and academic literature on marginalized communities in Kingston, I pinpoint the basic assumptions of the residents’ narrative of inner-city governance dynamics and determine the model that best matches those assumptions. The chapter answers the third sub-question on the way policymakers’ theories and assumptions reflect the experiences of residents in
marginalized communities. By showing how the conflicting assumptions of police officers and residents can lead to greater levels of insecurity in marginalized communities, the section also attempts to answer the second part of the main research question.
Finally, Chapter 6, offers a conclusion and policy recommendations based on the arguments of the previous chapters.
Chapter 2
Theorizing Community Policing
Introduction
To understand the popularity of the community policing strategy and its rapid spread across the globe, one must first recognize the power of the narrative that it promotes. These narratives, pushed by policymakers to justify the strategy, are implicitly based on assumptions about the types of actors that occupy the urban space, the legitimacy of those actors, and the ties between them. By exploring the US-based contextual origins of community policing, this chapter attempts to identify those assumptions and link them to pre-existing academic literature on modern urban governance theories.
By combining the latest theories on urban governance with the assumptions inherent in policymakers’ community policing narratives, I develop three distinct models that depict the urban conditions under which community policing might operate. The models describe the relationship between different sets of actors commonly associated with the strategy, starting with the most basic state capacity model that includes only the state, the police, and the community; proceeding to the competing orders model that introduces the gang as an additional governance actor; and concluding with the polycentric model that includes a variety of other state and non-state governance actors. As the chapter demonstrates, each model is linked to a different set of assumptions that have implications for the implementation and success of the community policing strategy.
The goal of the chapter is to understand, first, the origins of the implicit assumptions that underlie community policing strategies and, second, the urban governance
theories that are linked to those assumptions. The chapter, therefore, addresses the first part of the main research question, which deals with the general urban governance theories and assumptions that underlie community policing strategies, along with the first sub-question, which deals with US-based community policing assumptions. In answering these questions, I begin to challenge the reigning US-based assumptions associated with community policing and propose new, more complex models associated with Ostrom’s ideas about polycentricity that may better reflect the reality of the modern urban landscape.
The chapter begins by describing the context that led to the emergence of community policing strategies in the United States and outlining the basic tenets and reforms promoted by the strategy. It proceeds to construct three models of community policing based on the existing literature, those models are: the state capacity model, the competing orders model, and the polycentric model. For each model there is a discussion of the assumptions involved and the implications for the community policing strategy. This theoretical framework sets the stage for my analysis of the transfer of the policy (and its assumptions) to the Jamaican context as interpreted by police officers (in Chapter 4) and residents of two inner-city Kingston communities (in Chapter 5).
Community policing: its origin and tenets
The inability of the United States police force to control the spread of violent urban uprisings in African-American areas of major cities during the middle of the twentieth century led to a period of self-reflection within the institution (Cordner 2014; Pelfrey 1998). According to Jack Greene (2000), the professionalization of the force—which standardized practices, created a top-down bureaucratic structure, and molded the police officer into an “outside security expert”—was seen to have weakened police ties to the communities where they operated. Of course, this “isolation” of the police officer, particularly in marginalized (mostly black) neighborhoods, can also be understood as the result of the police’s role in executing the government’s racist and exclusionary government policies. Either way, the recalcitrance and distrust of large population centers (including an increasingly mobilized anti-war left) was seen as evidence of a “legitimacy crisis” that required a basic rethinking of what it meant to police marginalized communities (Greene 2000).
These circumstances lay the foundations for the introduction of a new policing strategy known as “community policing.” Community policing is a style of policing that includes three main components: (1) consulting with members of a community to “define, prioritize, and address crime problems”; (2) flattening the traditional police hierarchy and delegating decision making to “the frontline officers who directly engage the community”; and (3) applying a problem-oriented approach to crime which prioritizes the identification of the larger underlying patterns associated with individual criminal incidents (Gill et al. 2016: 401). In general terms,
community policing has been described as a preventative and proactive strategy that empowers and encourages local officers to collaborate with community residents in order to identify and resolve the primary causes of the most pressing community issues.
The strategy, as outlined, represents a multipronged, multi-scalar approach, entailing fundamental changes in the philosophy, organization, and every-day practices of the police. First, at the most expansive level—the philosophical realm— the strategy represents a move towards conceptualizing the police officers as subservient to as opposed to separate from the citizenry they patrol (Herbert 2006). The citizen, in addition to the bureaucratic state, acts as a power check and oversight authority to the police’s performance. Second, at the institutional level, community policing proscribes reforms to the organizational structure and decision-making dynamic of the force, which transfers greater freedoms and responsibilities to the local on-the-ground police officers. As a result, the strategy tends to be viewed as a more bottom-up, rather than top-down, approach to policing (Boettke et al. 2016). Finally, at the operational level, community policing emphasizes tactics that facilitate exchange and generate trust between the officer and the community resident. Typical recommendations at the operational level include performing on-foot rather than vehicular patrols, attending neighborhood functions, sponsoring local events, and creating a more “open door” atmosphere inside the police station (Gill et al. 2016).
Given its multidimensional nature and the numerous levels and components involved, it is not surprising to find that the implementation of the strategy has varied and evolved over time and space (Greene 2000; Goldstein 2003). Gill et al.’s (2014: 403) systematic review of the academic literature on community policing notes that “a diverse range of strategies have been employed under the auspices of [community policing] over the past few decades,” with some authors worrying “that the resulting heterogeneity renders the … approach ‘vague and difficult to execute.’” In spite of the approach’s holistic aims, few departments in the United States implement the three previously mentioned components, and the strategy is “often introduced at the unit level, as a set of tactics employed by individual police officers, in specific beat areas, or by specialized teams” as opposed to forming part of a larger organizational realignment (Gill et al. 2014: 403). In certain cases, the language of community policing has been applied without any corresponding structural or operational reforms taking place at all, leading some scholars to question whether the strategy is sometimes used as a mere public relations tool (Ross 1995; Greene & Mastrofski 1988).
The fact that the approach is vague and can be implemented in various ways has two important implications for my research. First, it creates more room for the community policing strategy to be modified and reinterpreted during the policy transfer process. And second, the aspects of community policing that each department (or police force) emphasizes offers evidence of the department-level (or country-level) assumptions about governance dynamics that underlie the policy. In
Jamaica, for example, the emphasis on youth programs and the disregard for the “police decentralization” component, I believe, is indicative of the dominance of “culturalist” assumptions within the JCF. (I explore the prioritization of particular aspects of community policing and its relation to urban governance assumptions in Chapter 4.)
An additional challenge identified by Gill et al.’s (2014) systemic review of community policing efforts in the United States that merits a brief discussion has to do with the issue of defining the term “community.” Gill et al. (2014: 422) state that one of the problems with community policing is that “a clear description of the ‘community’ that was targeted by the intervention was absent from the discussion in most of the studies we reviewed.” The authors go on to emphasize that without a good understanding of the boundaries and dynamics of a “community,” police officers may be susceptible to small groups of vocal residents who may not be representative of the larger target group.
I, therefore, take this opportunity to clarify the use of the term “community” in this paper. The term “community” is used to denote not only a group of individuals that share a common locality but a socio-political unit that is recognized by multiple actors, including outside state authorities as well as the members of the unit itself. This unit is understood to share certain economic ties as well as “history, knowledge, beliefs, morals and customs” (Kepe 1999: 421). As I show in later chapters, this definition applies to certain facets of the two case study localities that form part of my study, but its application could also be easily contested in light of the many divisions and disputes that arise between groups that reside inside those “communities.” While the fragility of the concept1, and the possibility of internal
divisions, is given greater weight in the polycentric model, my analysis still requires “settling” on a specific unit. Given that residents, police officers, and crime experts in Jamaica applied the concept of “community” to the two case study localities, and taking into account the difficulties involved in modeling a greater and greater number of subgroups (the smallest of which would end up being the individual), I ended up “settling” on the term “community” while acknowledging that it does not constitute a single homogenous grouping. Community policing: a basic theory and its assumptions
In the previous section I described the US origins and outlined the principle components of the community policing strategy. In spite of the vagueness of the strategy and the difficulties associated with its implementation, community policing continues to be promoted around the world by law enforcement agencies and international donors (Brodgen & Nijhar 2005). Part of the popularity seems to stem from a general acceptance of the underlying logic of the strategy. The following
1 And this fragility applies not only to the term “community” but also to the other main concepts that form part of the theoretical framework, such as “the state,” “the police,” and “gangs.”
section investigates this logic and presents the first and most basic model of urban governance on which that logic is largely based. One of the principal arguments for the use of community policing strategies involves the idea that the legitimacy of state authorities is key to generating order (Hawdon et al. 2003). Legitimacy, as David Beetham (2001: 107) has noted, is the justification given by those in power for “why their access to, and exercise of, power is rightful, and why those subject to it have a corresponding duty to obey.” According to Beetham’s interpretation of Max Weber, legitimacy is important for power relations as it allows rulers to govern their subjects without need for widespread coercion. Subjects that acknowledge the right of the ruler to govern will feel duty-bound to follow the ruler’s orders. This willful cooperation on the part of the ruled is especially important, Beetham argues, in large, complex societies where the costs of monitoring and physically enforcing rules on all subjects may exceed the ruler’s capacity.
Following this line of reasoning, academics have argued that citizens who accept the legitimate authority of the police not only cooperate more willingly with their directives but also violate fewer laws (Tyler 2006; Tyler & Fagan 2008). Conversely, police work is seen as much more challenging in communities where the legitimacy of the state is called into question. These difficulties arise in large part due to the nature of the public good that the police are said to supply: security. As Vincent and Elinor Ostrom (1977: 20) have noted, certain public goods require levels of “co-production” where “users of services also function as essential co-producers.” In such cases, “collaboration between those who supply a service and those who use a service is essential if [these] public services are to yield the desired results.” In the case of the public good of security, citizens are expected to contribute by, for example, keeping an eye out for violations and passing on information to the police when infractions occur. When police legitimacy is low, residents may refuse to offer their assistance and the police may interpret their silence as a way of intentionally obstructing their efforts. A negative feedback loop is then generated, whereby the inability of the community and the police to co-produce safety leads each side to stigmatize the other, which translates into even less co-production of security. To resolve the vicious cycle of suspicion and recrimination, the community policing strategy was proposed to increase levels of collaboration (and therefore, security co-production) between community members and the police.
Numerous studies have explored the mechanisms through which the police acquire or lose legitimacy in the eyes of the public. In general, these studies have focused primarily on factors such as procedural justice (the manner in which citizens are treated by the police and the justice system during decision-making processes), distributive justice (whether citizens believe laws are enforced similarly between different groups), legality (whether citizens believe the laws themselves are legitimate), and police performance (Mazerolle et al. 2013). Community policing is seen as influencing some of these factors, such as procedural justice and police
performance. For example, Hawdon (2008: 4) describes the mechanism that links community policing to legitimacy in the following way: “community-policing tactics increase resident perceptions of procedural justice … [which] enhances perceptions of police legitimacy … [and] residents who perceive the police as being legitimate are more likely to cooperate with the police and comply with the law.”
The model presented is simple and involves, at maximum, three actors: the state, the police, and the community. The state is assumed to dictate a legitimate and universally accepted social order (codified through laws) while the police are tasked with enforcing that order. The state-imposed social order is disrupted due to a breakdown in the relationship between the police and the community, which is assumed to be a product of a poorly performing and socially distant police force. This breakdown leads to a crisis of legitimacy that then spreads to the state, as the police are viewed as a simple and direct extension of the state. Figure 1 provides an illustration of this model:
Figure 1. The state capacity model
As Figure 1 indicates, the theory posits that the police—rather than being separate from—is encompassed by the state; that is to say, the police are believed to act as a faithful conduit of the state, implementing its directives and providing a direct connection between the state and the community. The dotted line connecting the police and the community symbolizes the tenuous relationship between the poor performing police officers and the residents of the community, who question the legitimacy of the police. Community policing, by restoring the legitimacy of the police as an institution, is intended to shore up this relationship (turning the dotted line into an uninterrupted straight line) and, by default, reestablish the community’s faith in the state. Additionally, in this simplistic model, the state and the community are each assumed to be unitary agents that act with one voice—the police, meanwhile, is assumed to simply channel the wishes voiced by the state. Community Police State
Much of the US-centered policymaking rhetoric on the implementation of community policing can be viewed through this model (Skogan & Hartnett 1997). This rhetoric tends to focus on the idea that the relationship between the community and the police has been disrupted by a failure on the part of the police to properly carry out the functions of the state. It tends to assume that all parties accept the single, unitary, consensual state order, but that distance and/or poor performance (perhaps coupled with intentional deviant acts of discrimination that fall outside the established rules) produces resident distrust in the institution. By improving the delivering of public goods, community policing can facilitate the co-production of safety and the community’s adherence to the rules.
This first model also serves as the basic framework that will be elaborated and expanded on by the two subsequent governance models, as they introduce new actors and relax certain assumptions. For example, the breakdown in the relationship between the police and the community could lead to a scenario where a legitimacy vacuum is created, which is then exploited by a non-state violence specialist. This violence specialist could begin to provide similar services as the police and, as a result, could further undermine the standing and legitimacy of the state. This forms the basis of our next model, the competing orders model, which introduces a new player: gangs.
Community policing under the competing orders model
Gangs and other similar urban non-state violence specialists represent a global phenomenon and, while they do not represent the main driving force for the introduction of community policing in the United States, they have begun to figure more and more in literature connected to the strategy (Decker 2003). Furthermore, early experiments with the strategy took place in marginalized areas with a high gang presence (such as Chicago) and, therefore, the literature could be said to implicitly address the issue (Skogan & Hartnett 1997). This section explores one particular model for community policing strategies in urban contexts that explicitly includes gangs as prominent governance actors. Related to the idea that the police have lost the trust of the citizenry is the notion that non-state actors may sweep in and begin to compete for that trust. The model presented in the previous section ignored this dynamic and restricted its focus to a fraying bond between the police and the community, without questioning the availability of alternatives to that relationship. The competing orders model, however, looks at situations where the breakdown in the police-community relationship leads non-state actors to enter the space normally occupied by the police. As the lack of cooperation and co-production between the community and the police leads to greater insecurity, the community may begin to demand protection services from other violence specialists.
In some cases, these violence specialist actors constitute illicit criminal networks that operate in a similar fashion to governments: they arbitrate disputes within their territories; they enforce economic transactions and protect private property; they even distribute pensions to the elderly and employment to the young. Since Charles Tilly’s (1992) seminal work that concluded that state formation is akin to the consolidation of use of coercion by a single gang (the government), a significant amount of academic literature has been devoted to the idea that criminal organizations are not the agents of chaos that the media paints them out to be, but in fact emerge to provide order under conditions where a central authority (the government) is weak (Skaperdas 2001; Rodgers 2006; Stephenson 2011).
Akerlof and Yellen’s (1994) paper on police-community relations in the urban “ghettos” of the United States offers a simple economic model of how the gang fits into the general urban governance landscape. In the model, the authors specify three distinct players: the community, the gangs, and the police (who function as a stand-in for “the state”). The authors assume a zero-sum logic whereby community residents must “choose” between supporting the anti-gang work of the police and collaborating (explicitly or implicitly) with the gang system. In more formal language, the authors apply a principle-agent model, where community members and gangs cooperate in order to “cheat” or “evade” the principle (the state). Gangs are said to evade the state in order to continue to profit from their illicit enterprises, but the community’s motivations for shielding gang members from police action are less clear. Akerlof and Yellen suggest four potential explanations for understanding the community’s decision to cooperate with the gangs: (1) fear of reprisal from the gangs, (2) uncertainty as to the consequences of weakening the local gang, (3) the perceived fairness or unfairness of penalties imposed by the justice system, and (4) attitudes toward police/social norms. Akerlof and Yellen (1994: 15) conclude that the police’s hardline tactics of “bricks and sticks may be self-defeating,” as they simply increase the perceived unfairness of the justice system and push residents further in the direction of the gang. Instead, the authors suggest adopting “a strategy of community policing with the objective of improving rapport between police and local community members” (Akerlof & Yellen 1994: 19). The picture they paint is stark and clear: a kind of tug of war between criminals on one end, police on the other, and a community that is stuck in the middle and getting pulled closer to one end or the other depending on who can muster up enough power and influence. That is to say, community policing functions as a way for the police to induce the local population to switch its “allegiances” from the criminals to the police officers. The means of switching those “allegiances” and “winning the hearts of minds” of the community are somewhat contested. On the one hand, it could occur mechanically through an improvement in the delivery of state services to the community (a similar mechanism to the one explored in the previous section). The state would have to simply offer up public goods at a more “competitive rate” than that offered by the gang. On the other hand, the issue may go beyond the provision of services or
the “rapport” between the officers and the residents, it may be linked to deeper issues of social and moral order. Jackson and Bradford (2009) have shown that, for the case of the United Kingdom, public confidence in the police may have just as much to do with feelings of security as with local understandings of social order and moral consensus. In other words, what may matter is not the prevalence of crime but definition of “crime.” Different actors may have different understandings of the range of acceptable behavior or of the conditions under which one should be considered a citizen or constituent of a given society; these differences may be further activated by the presence of competing actors vying for authority.
In some cases, the community’s views of social order and moral consensus may be more in line with those espoused by the gang (if, for example, it is a local entity that is viewed as sharing the community’s values) than by the state (that may be viewed as distant, predatory, and/or repressive, particularly in “underserved” communities). In the past couple of decades, numerous scholars have attempted to grapple with the emergence of these non-state governance actors that not only offer rival services but also represent competing social orders. Much of this debate has centered on what this reality might mean for the concept of citizenship and the way community members interact with the state.
Davis (2010: 402), for example, has highlighted what she describes as the fragmentation of sovereignty, which is defined as the “proliferation of a wide range of competing and overlapping communities, with their own armed forces of protection and own allegiances, that leads to contested geographies of citizenship.” In general, work within this framework suggests that “state failure” (as well as voluntary state retrenchment through neoliberal policies) allow alternate actors (such as organized criminal enterprises) to emerge as replacement order-generating governance institutions. The literature tends to use terms such as “fractured” (Koonings & Kruijt 2007) or “disembedded” (Rodgers 2004) cities, implying the presence of “parallel polities” (Leeds 1996) and bringing to mind images of the urban as a patchwork of “island” neighborhoods each with its own borders, social order, public services, and imagined community (Sanchez-Jankowski 1991).
Figure 2 attempts to illustrate this separate and competing social order, by modifying our previous state capacity model to include the role of gangs:
Figure 2. The competing orders model
The model continues to assume that the police represent a direct extension of the state and, furthermore, assumes as before that the relationship between the police and the community is frayed. However, in contrast to the previous model, the gang is now present and strongly linked to that community. As Figure 2 indicates, the gang represents an alternate and competing relationship-path that is available to the community, offering a distinct services package and/or social order from that of the state. Based on what has been discussed previously, the relationship is assumed to be zero-sum, whereby as the line between the police and the community becomes more connected, it will then disrupt the line between the gang and the community.
To summarize, the principle assumptions of the model are that the gangs and the state/police represent inherently antagonistic forces, engaged in a zero-sum battle over the resources represented by the support of the community. What this implies for the community policing strategy is that not only may the police force have to provide services that are competitively “better” than those offered by the gang, but that they may have to employ tactics that attempt to alter the social and moral norms of the community (or have the state alter its norms to more closely align to those espoused by the community and the gang).
The main criticism that can be leveled at this model is that it ignores the deep and complicated ties between government institutions and sophisticated criminal organizations, and imposes a sort of “exclusion principle” whereby the community cannot utilize both the services of the gang and that of the state. Recent literature has shown that these assumptions can be easily refuted and that ties between the state and gangs can be quite extensive, so much so that it easily distorts the neat picture offered by this second theory. The following section further critiques the assumption that the gangs and the state represent independent and antagonistic forces in order to formally introduce the polycentric model.
Community policing under the polycentric model
The previous model relaxed the restriction on the number of actors involved in the governance of the community to include non-state actors (principally “the gang”). However it included the assumption of “parallel polities,” in which the gang represents an alternate governance structure with its own social code that does not interact with the state and competes for the same resources (in this case, authority over the community). In this section, I further question that assumption to propose a model where gangs form part of a larger governance eco-system, one in which the gangs and the state interact, at times in a competitive and at other times in a cooperative fashion.
The idea that states and gangs interact is not new, though serious attempts at examining the systemic patterns of that relationship have only recently come to the academic fore. Organized criminal enterprises have been shown to employ a variety of methods in dealing with the police and other state institutions. Bailey and Taylor (2009) divide these strategies into three categories: “evade, corrupt, or confront”. Criminal networks and the state develop “equilibrium relationships” at various levels, such as at the neighborhood level, where “fluid organizations involved in extortion, drug distribution, gambling, or prostitution work out stable relationships with individual officers, groups of police, or with whole precincts” (Bailey & Taylor 2009: 8). At higher levels, public officials may make the calculation that “a criminal organization that is able to achieve its goals without disturbing the public order may be valuable … especially in contrast to more fragmented criminal enterprises that compete for dominance and thus may generate politically inconvenient levels of violence” (Bailey & Taylor 2009: 9).
The institutionalization of such relationships and the state’s harnessing of the role of the criminal networks to advance its interests led William Chambliss (1988) to describe such dynamics as “state-organized crime.” This consists of “acts defined by law as criminal and committed by state officials in the pursuit of their job as representatives of the state” (Chambliss 1988: 184). The theory is in line with the work of authors like Loïc Wacquant who, in his critique of the “neoliberal state,” views the influence of criminal networks in marginalized communities not as a response to state weakness but as the inevitable consequence of purposeful government policies (see, for example, Wacquant 2008).
Other scholars have argued that the level of collaboration can be so extensive that gangs can be viewed as an extension of the state or, at the very least, as so inextricably enmeshed and intertwined with the state that an analytic separation of the concepts may be disingenuous. This leads to the idea of a “hybrid state” in which “the various unstable yet enduring coalitions between governmental officials and criminal organizations make it difficult to separate formal state governance and the rule of criminal [actors]” (Jaffe 2013: 738). Under this view, official government efforts to delegitimize and counteract gang rule, in reality, mask a symbiotic relationship that is both mutually beneficial and systemically engrained.
Under such circumstances, a useful framework for understanding the diversity of actors and the varied and complex relationships between them may be offered by Ostrom, Tiebout, and Warren (1961) in their classic article on the governance of metropolitan areas. In the paper they introduce the concept of “polycentric political systems” which they describe as:
“Polycentric” connotes many centers of decision making that are formally independent of each other. ... To the extent that they take each other into account in competitive relationships, enter into various contractual and cooperative undertakings or have recourse to central mechanisms to resolve conflicts, the various political jurisdictions in a metropolitan area may function in a coherent manner with consistent and predictable patterns of interacting behavior. To the extent that this is so, they may be said to function as a “system” (Ostrom, Tiebout, and Warren 1961; pg 831).
The polycentric systems approach posits that individuals within a system will band together to resolve collective action problems that are not addressed or resolved by larger governance authorities. This generates a new “decision making” center, which then interacts in some manner with other centers, establishing both hierarchical and horizontal relationships. The approach is therefore considered to be “bottom-up” in the sense that individuals are able to resolve local problems by banding together and forming their own governance structures, while simultaneously forming part of larger networks.
The polycentric model, therefore, suggests that, in order to address specific demands of the community, gangs may emerge as centers that are independent of other actors but that also form part of the larger overall system. Instead of constituting a separate and exclusive regime that cannot coexist with the state, as the competing orders model contends, the gang represents an additional actor within a preexisting overarching system that includes a variety of complementary and competing actors, such as the state and the police (McGinnis & Ostrom 2012). Crucially, these different centers, according to Ostrom, Tiebout, and Warren, do not form part of strict hierarchical structure, but operate on shared principles and rules that dictate their interactions. The power balances between the centers are contingent on the claims being made and may vary from horizontal to vertical relations depending on the issue.
The polycentric approach has only recently resurfaced with regards to the case of community policing, having largely laid dormant following Ostrom’s seminal work on the subject in the 1970s.2 Boettke, Lemke, and Palagashvili’s (2016) recently
published a paper that argues that the framework provides insights into why the community policing strategy has had such ambiguous results in the United States. Boettke et al. (2016) suggest that separate governance centers in the federal
government have impaired the work of decentralized policing units, diverting their attention from local problems and pushing them (through the dangling of federal funds) towards the objectives and interests of the national authority, which has tended to promote more militarization and anti-terrorism initiatives.
In this paper, I propose extending the polycentric approach to apply to urban contexts that feature “extra-legal” authorities, an area of study not traditionally covered by academic literature on polycentricity. Previous work that applied the framework tended to focus on informal governance arrangements in rural communities dealing with issues of common pool resource management, such as fisheries and forests.3 Yet, this framing complements Enrique Desmond Arias’s
(2006) work on the networks between trafficker-dominated favelas and the broader political and social system in Rio de Janeiro. As he emphasizes, “criminals have emerged as political actors who must operate in the political system through other state and social actors” (Arias 2006: 324).
With this literature in mind, Figure 3 illustrates the polycentric framework as an alternative to the existing models identified in previous sections:
Figure 3. The polycentric network model
In this case, all actors are connected to each other, but in contrast to the previous models, these connections are not completely subordinate and hierarchical but contingent and diffuse (which is represented by the dotted line). Figure 3 shows that the gangs not only interact with the state but also with the police (for example,
3 An exception is Skarbek and Marcum (2012), which applies the polycentric framework to analyze prison gang networks in the US. The study, however, focuses on inter-gang ties and, for the most part, ignores the role of the formal authorities (such as guards and wardens) within the system.
by tacitly collaborating to delineate territories of authority or by exchanging information in order to capture or punish rule violators). Two additional centers are introduced, one representing state actors (or the various agencies that conform state)—which, it should be noted, no longer subsume the police—and the other representing a conjunction of non-state actors (such as businesses or civic organizations).
In contrast to previous models, the polycentric model assumes a much more organic ecosystem in which each actor reinforces the other, as opposed to a rigid structure where power emanates specifically from the top (the state) and flows towards the bottom (the community). The relationship between the actors is not necessarily horizontal or vertical (it can be either/or depending on the situation). Though not made explicit in Figure 3, the model also assumes that the actors are less unitary, since there is always the possibility for new decision-making centers to emerge within the centers that are outlined. For example, the community may represent a variety of independent decision-making centers, but for the sake of the analysis we must settle on one contingent stylized actor.
Under this model, community policing efforts set forth by the police to combat the influence of the gangs and reestablish the primacy of the police over matters of security in the community would be ineffective without coordinating with the rest of the actors. The polycentric model suggests that focusing exclusively on “winning the hearts and minds” of the residents may fail to disempower the gangs if, for example, their sources of power go beyond the allegiance offered by the community to include the backing of other actors within the system (such as businesses, civic organizations, the police, or the state). Furthermore, the strategy could even produce adverse effects by weakening the overall governance structure, especially if the gang is overthrown without the resources that it channels to the community (through its networks to the state, the police, and other actors) being adequately supplanted.
Conclusion
This chapter attempted to unpack the assumptions underlying the community policing strategy by first, examining the US context in which the strategy emerged and then connecting the strategy’s assumptions to academic debates on urban governance. By combining the implicit assumptions of community policing and recent theories on modern urban governance dynamics, I developed three governance models, each with a different combination of urban actors and a different set of assumptions. The models range from the most basic state capacity model, where a unitary state imposes a well-defined and universally accepted order through the police; to the slightly more complex competing orders model, which introduces the gang as an extra-legal actor; to the final polycentric model, in which a multiplicity of legal and extra-legal actors interact in competitive and complementary ways.
I base these models both on the logic used by policymakers in the United States and on urban governance theories generated by academics. I argued that the first two models apply to US-based discussions on community policing, and went on to question policymakers’ assumptions of an independent state, uninfluenced by “extra legal” actors. Therefore, I proposed a third model that relaxed that assumption, and introduced new actors and ties.
In the following chapters I explore the manner in which these policies and their assumptions are exported by foreign aid organizations (such as USAID) and then interpreted by street-level bureaucrats (such as police officers). These models serve to frame the assumptions I identify in my interviews with Jamaican police officers as well as local residents. I argue that the manner in which the assumptions of the policymakers (including foreign aid organizations and the top brass of the JCF), the street-level bureaucrats (beat police officers), and the community residents complement or contradict one another has direct implications for the success or failure of the community policing strategy.
However, before turning to the assumptions of the JCF and the residents of inner-city communities, I first discuss the process by which I obtained the data that forms the basis of my empirical analysis. The sensitivity of the topic of discussion along with the violent urban context I was studying had important implications for how one should interpret my analysis.
Chapter 3
Navigating the Violent Urban Context
Introduction
Before analyzing the logic, implementation, and effects of the community policing strategy in Jamaica, I begin by exploring the ethical and methodological implications of the field. This investigation centers on data obtained from five months of fieldwork in two inner-city communities of Kingston, Jamaica, and utilizes research methods that include semi-structured interviews, participant observation, and quantitative and qualitative secondary sources. The sensitivity of the research topic along with the violent context of the field requires a careful and transparent reflection of the research process.
As the chapter highlights, my fieldwork took place in a conflict-prone urban environment where disputes among criminal factions and repressive state-authorities produced high levels of violence. I discuss some of the literature on doing fieldwork in violent settings, and describe the considerations taken to ensure both my own safety and that of my research participants. Specifically, I look at how the violence directly affected the selection of my case studies, my sampling process, and my freedom to visit and move around the case study communities. This first part of the chapter centers primarily on steps taken to ensure my own safety and analyzes the implications of my entering and studying a violent urban context.
