1

Introduction

Policing relies heavily on voluntary compliance with the law. While historically policing is associated with utilitarian theories of deterrence based on fear of punishment, no society can afford to employ enough police to rely solely on threats of sanction to keep the peace. A broad social consensus of support for police and law is associated with lower levels of crime, at least cross-nationally (Adler, 1983; Eisner and Nivette, 2013). Moreover, the effectiveness of deterrence-based compliance mechanisms often depends on the level of community support that the police enjoy (Bottoms, 2019; Sherman, 1993). In recent decades, theorists have examined and parsed this consensus into the conceptual framework of police legitimacy and trust.

This report explores how the police can develop practices to build community trust and legitimacy (as well as to minimize loss of legitimacy). This report is one of a series of five reports completed at the request of the U.S. Department of State, Bureau of International Narcotics and Law Enforcement Affairs (INL). INL is part of an extensive network of international and regional organizations, bilateral donors, international financial institutions, and civil society organizations that work with governments in several countries to provide foreign assistance and support capacity building for criminal justice systems and police organizations. In 2018, guided by the Foundations for Evidence-Based Policymaking Act, INL created the Office of Knowledge Management to assemble evidence from research to inform its work. To support its efforts to synthesize findings from scientific research, INL asked the Committee on Law and Justice of the National Academies of Sciences, Engineering, and Medicine (the National Academies) to convene an ad hoc consensus committee to review and assess

existing evidence on policing institutions, police practices and capacities, and police legitimacy in the international context.

THE COMMITTEE’S CHARGE

The National Academies of Sciences, Engineering, and Medicine assembled the Committee on Evidence to Advance Reform in the Global Security and Justice Sectors (“the committee”) to review the available research evidence on how police reform can promote the rule of law (including human rights) and protect the public. See Box 1-1 for the committee’s perspective on rule of law and protection of the population.

The committee includes experts in criminology, economics, international and organized crime, law, policing, and political science. It brings knowledge and local experience from a portfolio of work that spans five continents. Such experience includes conducting research as well as advising governments on police policy in several countries, including but not limited to Afghanistan, Australia, Brazil, Colombia, Denmark, Ghana, India, Japan, New Zealand, South Korea, Sweden, Uganda, and the United States (see the Appendix for more details).

The committee was charged with producing a series of five reports, each addressing areas of interest to INL (see Box 1-2). To assist with this assignment, the committee developed a series of five public information-gathering sessions to bring together researchers and practitioners with experience in each subtopic to be examined.

This report is the fourth of the series, addressing the fourth question in the committee’s charge: What policing practices build community trust and legitimacy in countries with low-to-moderate criminal justice sector capacity?¹

Countries with low-to-moderate criminal justice sector capacity were understood to refer to INL’s range of partner countries. INL partners with countries in Sub-Saharan Africa (Ghana, Kenya, Nigeria etc.), East Asia and the Pacific (Indonesia, Malaysia, Philippines, etc.), Europe and Eurasia (Georgia, Serbia, Ukraine, etc.), Middle East and North Africa (Egypt, Iraq, Libya, etc.), South and Central Asia (India, Nepal, Pakistan, etc.), and the Western Hemisphere (Colombia, El Salvador, Mexico, etc.).²

___________________

¹ Each consensus report in the series of five reports is being released as a PDF as it is completed.

² More information on INL partner countries is available on its website: https://www.state.gov/bureau-of-international-narcotics-and-law-enforcement-affairs-work-by-country/.

APPROACH TO THIS STUDY

Like the others in the series, this report reflects the development of the committee’s consensus on advice to address the questions in its charge. The committee was tasked to carry out the entire study within a year and a half and release each of the five reports separately and sequentially during this time period. In forming its advice, the committee draws specifically on information from a prepared paper and a single workshop on the topic of the fourth question as well as its members’ years of experience investigating policing policies and practices.

The public workshop, entitled Police and Community Trust and Legitimacy, was held virtually on January 25 and 26, 2022. Workshop participants included committee members, representatives from INL, and international researchers. An effort was made to assemble a diverse set of participants who work with or study policing in several different country contexts such as Brazil, Colombia, India, Indonesia, Liberia, Mexico, Nigeria, Pakistan, the Philippines, Uganda, and the United States. The questions posed to participants included:

• How have trust and legitimacy been operationalized in research studies?
• Why is it important to care about improving trust and legitimacy?
• How do motivations and definitions of trust and legitimacy vary between countries?
• Has improvement of trust and legitimacy been the primary goal or secondary benefit for recent interventions?
• Which police practices seem most promising for building community trust and legitimacy?
• What is known about the influence of contextual factors on the effectiveness of such police practices?

The workshop was framed around a commissioned paper prepared by Robert Blair (Brown University) titled Policing Practices to Build Community Trust and Legitimacy in the Global South.³ Blair (2022b) summarizes high-quality research on citizens’ perception of police regarding six policing practices: (1) community policing; (2) saturation policing; (3) body-worn cameras; (4) procedural justice and soft skills; (5) integration and descriptive representation; and (6) militarization or constabularization. We draw on that review and notable findings from the workshop throughout this report.

___________________

³ The author, as does the committee, uses the term “Global South” to refer to low to middle income countries most likely to receive assistance from INL. “Global North” refers to developed countries, notably the United States and the United Kingdom, and includes Australia despite its geographic location.

It is important to note that while the science of policing and theoretical work on conceptualizing police legitimacy and community trust in police has grown in recent years, empirical studies are limited in context. Due to the research available and studies presented to the committee, examples in this report tend to draw primarily on studies conducted in the United States, United Kingdom, India, Colombia, Mexico, and, to a lesser extent, Liberia and Turkey. This does not mean that issues of police legitimacy are limited to these regions, nor are solutions and opportunities limited to these contexts.

Interventions to improve police legitimacy, and assessments of those interventions, have been dominated by the procedural justice perspective as popularized by Tyler (2006). However, some scholars have raised questions about the complexity and implications of context on the relationship between procedural justice and perceptions of legitimacy (Bottoms and Tankebe, 2012, 2020; Graham and Cullen, 2021; Nagin and Telep, 2017, 2020; Terpstra and van Wijck, 2021). While a procedural justice viewpoint to policing continues to demonstrate value, an exclusive focus on procedural justice limits knowledge about means to increase legitimacy and community trust (Bell, 2017; Bottoms and Tankebe, 2017; Loader and Sparks, 2013).

CONCEPTUALIZING LEGITIMACY AND TRUST

Police have the power to arrest and interrogate suspects and to use deadly force against civilians. These powers inevitably raise questions about their justified use. Legitimacy is a complex concept that is anchored in these questions. One potentially fruitful first step toward conceptualizing legitimacy is differentiating between different types of power relations. Joseph Raz (2006) specifies three of these power relationships:

1. A naked power relation in which powerholders make no claim to have the right to exercise power, but instead, rely on their ability to exert brute force or to coerce the population into obedience. Hence, it is a power relation without social consensus;
2. A de facto power relation in which powerholders are conscious of the importance and necessity of cultivating legitimacy, but public acceptance is limited or lacking; and
3. A legitimate power relation, in which claims to rightful power are made and these are recognized as valid (Bottoms and Tankebe 2012) without the need to resort to coercion and irrespective of the material implications of the execution of power (Baldassarri and Grossman, 2011). In other words, to say power is legitimate is to say it is “acknowledged as rightful by relevant agents, who

include power holders and their staff, those subject to the power and third parties whose support or recognition may help confirm it” (Beetham, 2013, p. 19, emphasis in original).

There are different perspectives of police legitimacy. Scholarly arguments by Tyler and colleagues (Sunshine and Tyler, 2003; Tyler, 2006; Tyler and Fagan, 2008; Tyler and Huo, 2002) advance legitimacy as a core element of policing. This body of work leans toward defining legitimacy as “a property of an authority or institution that leads people to feel that that authority or institution is entitled to be deferred to and obeyed” (Sunshine and Tyler, 2003, p. 514). Following from this, legitimacy is often measured by asking people the extent to which they feel obliged to obey the police (see discussion below). This is combined commonly with measures of support for the police (Tyler, 2006) or confidence in the police (Sunshine and Tyler, 2003). Various empirical studies within this perspective show that a key correlate of police legitimacy is perception of police procedural justice—how fairly people feel the police treat them during everyday interactions with them (Tyler, 2003). However, while useful as a starting point, this approach to legitimacy has been criticized for failing to differentiate between true legitimacy and naked power situations, with the latter often rooted in a sense of powerlessness, fear, and fatalism (Bottoms and Tankebe, 2012; Tankebe, 2013). It is an important condition for true legitimacy that the recognition of the validity of police power be voluntary rather than coerced.

A second perspective to police legitimacy builds on the former approach. In attempt to address criticisms, Jackson and colleagues incorporate into the conceptualization of legitimacy what they describe as “moral alignment” (Jackson et al., 2012). It is the extent to which people believe “authority as morally aligned with their own values and thus have a sense that they have a duty to obey that authority” (Hough, 2019, p. 301). This is an important improvement since legitimacy presupposes the existence of shared values between power-holders and those subject to such power (Beetham, 1991). However, in practice the authors often combine moral alignment with obligation to obey the police (Hough, Jackson, and Bradford, 2013), and therefore are at times unable to completely separate coerced attributions of legitimacy and true legitimacy.

A third perspective comes from Bottoms and Tankebe (2012), who emphasize the situational and relational nature of legitimacy. In their article, they argue that legitimacy is the recognition of power as rightful, and that this recognition arises from the dialogues between power-holders and those subject to power. As they put it:

They argue that audiences approach this metaphorical dialogue with power-holders with certain expectations; the fulfilment of which generates recognition of power as legitimate. Bottoms and Tankebe (2017) describe these expectations as “basic legitimation expectations” (see Chapter 2 for discussion). Further, they argue that the acknowledgment required for police power to be legitimate comes from two sources: civilians and the police themselves. The civilian dimension has been described by Bottoms and Tankebe (2012) as audience legitimacy; that is, an acknowledgment by civilians that policing is rightful. Building and sustaining legitimacy requires that the police also recognize their entitlement to exercise power. Bottoms and Tankebe (2012) refer to this dimension as self-legitimacy: the self-belief that police officers hold the right to occupy the roles they do in society and to exercise the police powers vested in them, “a self-belief that then underpins the claims to [public] legitimacy” (p. 150). The available correlational evidence shows that organizational justice and perceived public support affect the confidence officers have in their self-legitimacy (Nix and Wolfe, 2017; Tankebe, 2022). Recent research has focused on police officers’ self-perception of legitimacy and recognition by their supervisors, peers, and the public (Bradford and Quinton, 2014; Jonathan-Zamir and Harpaz, 2014; Tankebe, 2019, 2022; Trinkner, Tyler, and Goff, 2016). While the effects of self-legitimacy are not yet fully known, self-legitimacy appears to be associated with an increased commitment to community engagement (Wolfe and Nix, 2016), greater support for the use of procedural justice in interactions with citizens (Bradford and Quinton, 2014; Jonathan-Zamir and Harpaz, 2018 ; Tankebe, 2019), and lowered use of force (Noppe, 2016; Tankebe and Mesko, 2015). These findings suggest that self-legitimacy might be important for behaviors that support ROL. Therefore, it may be necessary for practitioners to think about how to build both police self-legitimacy and public perceptions of police legitimacy.

Common to the varied perspectives is a recognition of the complexity of police legitimacy. This complexity lies partly in the frequent disconnect between the police’s normative authority, police behavior, and public perceptions. The police might be perceived as illegitimate even though their conduct complies with established laws and norms (and vice versa). Conversely, police may be considered legitimate by segments of the public, even though their conduct contravenes formal laws and established norms. In an illustration of these tensions, Waddington and colleagues (2015) showed a video of an actual police-civilian interaction to multiple focus groups in a major city in the United Kingdom. Even though the officer in the video

behaved in ways consistent with “good” policing practices, different discussants perceived and assessed the officer’s actions differently. As the authors conclude, this complicates the relationship between action and perception, and demonstrates that “there is no simple recipe for winning legitimacy” (Waddington et al., 2015, p. 212).

The concept of trust is often considered to be separate from, although correlated with, legitimacy (Bottoms and Tankebe, 2012; Kaina, 2008). Community trust for institutions has been defined in various ways in the literature regarding policing and legitimacy (see Box 1-3 for some examples). A perception that the police are legitimate can lead citizens to trust them to do specific things, such as responding to calls for service, or investigating allegations of criminal victimization. Citizens’ acknowledged trust in the police has been linked to a greater willingness to obey the law, report crimes to the police, and cooperate in criminal investigations (Blair, 2022b; Bolger and Walters, 2019; Walters and Bolger, 2018). However, citizens can trust that the police will do specific things without necessarily viewing them as legitimate (Bell, 2016). Research has demonstrated that mistrust in state institutions does not necessarily reduce citizens’ likelihood of engaging with state security institutions. Studies find that compliance with the law and police can be driven by emotional, cultural, and/or ideational factors beyond sense of trust (Gallagher, 2006; Hilbink et al., 2021; Lake, Mukatha, and Walker, 2016). In sum, the array of findings suggests that the relationship between trust, legitimacy, and citizen behavior is complex.

With regard to the committee’s charge, an ultimate aim is for citizens to trust police to promote the rule of law and protect the population. As

such, we focus this report on the concept of legitimacy and ways of building legitimacy to foster this kind of trust and expectations. The committee proposes this as a working definition of legitimacy: the acceptance by the police, citizens, and third parties of the validity of claims to rightful power without the need to resort to coercion and irrespective of the material implications of the execution of power (Baldassarri and Grossman, 2011; Beetham, 1991; Bottoms and Tankebe, 2012).

MEASURING LEGITIMACY

Given the centrality of legitimacy to police authority, consistent and accurate measurement of public perceptions of legitimacy are important to track over time. The four pillars of legitimacy discussed later in this report can be measured through triangulation of various data sources, including public opinion surveys, citizen complaint data, police administrative data, interviews, focus groups, and ethnographic accounts. The qualitative methods of ethnography, focus groups, and in-depth interviews with key informants are all useful in understanding, for example, potential reasons behind citizens’ views of police legitimacy and the social, cultural, and institutional barriers to reform.

Many existing studies of legitimacy employ surveys using samples of the population to capture how people feel about the authority of police officers (Jackson et al., 2012; Tyler, 2006) either with crime victims (Murphy and Barkworth, 2014) or with criminal offenders (Augustyn, 2015; Tyler and Fagan, 2008). The mode and content of survey administration have varied between studies (see Eisner and Nivette, 2013). Some studies measure people’s normative alignment with police behavior, asking for a level of agreement on whether police “…usually act in ways that are consistent with [one’s] own ideas of right and wrong” (Bradford, 2022, p. 4). Some studies combine a measure of people’s perceived obligation to obey the law with a measure of their trust or confidence in the police (Sunshine and Tyler, 2003), or else with “moral alignment” (Jackson et al., 2012, p. 1067). Jackson and colleagues (2011, p. 1066) note that to measure obligation, respondents are typically asked to indicate the extent to which they feel a duty to:

1. Do what the police tell you to do, even if you don’t like how they treat you?
2. Accept the decisions made by the police even when you disagree with them?
3. Do what the police tell you even if you don’t understand or agree with the reasons?

None of these statements differentiate between legitimacy and duty to obey that might be coerced. Therefore, caution is required in interpreting measures tapping feelings of obligation to obey the police. There is still a need to develop research measures that capture the perceived rightful authority of the police. One approach advocated by some researchers is to develop survey instruments that capture four pillars of legitimacy: police effectiveness, lawfulness, procedural justice, and distributive justice (see discussion in Chapter 2) (Tankebe, 2013; Tankebe, Reisig, and Wang, 2016). More recently, Kearns, Ashooh, and Lowrey-Kinberg (2020, p. 201) asked a sample of U.S. residents an open-ended question: “When thinking about the police, what does “legitimacy” mean to you?” Coding of the results showed legitimacy meant the police “follows the law,” acted with “honesty,” demonstrated “fairness,” had a recognized “right to govern,” were “effective,” and offered “protection” to residents (Kearns, Ashooh, and Lowrey-Kinberg 2020, p. 198).

A distinctive feature of these survey-based measures is that they tap general perceptions of the police. A more fruitful approach could also measure the legitimacy of specific policing units, such as counterterrorism vs. community policing or specific actions, such as use of deadly force, home raids, and stop-and-search. Further, Bell (2016) shows that people differentiate between the police as an institution and individual neighborhood officers; as such, the public might express deep cynicism about the police institution as a whole but express trust in individual beat officers.

CONTEXTUAL CONSIDERATIONS

Measures, meanings, and interventions related to police legitimacy and trust will likely vary across social and political contexts. Globally, some approaches to establishing police legitimacy may be affected negatively by the underlying historical context, or by regional and racial divisions. As discussed in Blair (2022b), long legacies of community fear, suspicion, and resentment can hinder attempts at building police legitimacy. These dynamics can arise in any country, but may be more entrenched in places with histories of colonialism, authoritarianism, economic instability, civil war, or low-to-moderate criminal justice capacities (the committee’s focus).

Moreover, in many countries, the criminal justice system is just one of many potential venues for adjudicating crimes and resolving disputes (Baker and Scheye, 2007; Hills, 2012). In these settings, the police must compete for citizens’ loyalties with myriad alternative providers of security and justice, including gangs, warlords, and traditional, customary, or religious authorities (Baker, 2007; Blair, Karim, and Morse, 2019; Isser, 2011). Citizens who do not trust the police or perceive them as illegitimate are more likely to seek redress for even the most serious grievances through

one of these alternatives (Isser, 2011). For example, experimental and quasi-experimental studies from Brazil, Liberia, and Papua New Guinea suggest that while community policing can occasionally improve perceptions of police and increase crime reporting (Blair, Karim, and Morse, 2019; Cooper, 2019; Magaloni, Díaz-Cayeros, and Ruiz Euler, 2019), the effects of community policing seem to vary depending on contextual factors. These factors include not only the resources the police can devote to community policing (and to address issues that came up during community consultations), but also whether non-state actors already provide security and whether citizens already have access to non-state security providers, such as traditional chiefs or gangs (Blair, 2022b).

Legitimacy also generally takes long periods of time to build, a factor that studies rarely can measure adequately. Nearly all studies evaluate programs only to determine their short-term effects on people’s perceptions of the police. To the extent that countries vary in their capacity to adjust (or reform) justice organizations, they may also vary in the length of time it takes for police legitimacy to change. Where there has been long-standing historical and social discrimination, a resulting lack of legitimacy cannot be expected to be undone through a single, quick reform program, even if it is well executed. Large-scale, longer-term studies would provide a more complete picture of why interventions to improve trust and legitimacy may or may not be successful.

Finally, it is worth noting that in some locations (e.g., Central African Republic and the Democratic Republic of the Congo), the public reports a comparatively higher level of trust in the police than compared to other security actors, the government, and the media (Balthazard and Vinck, 2022). An interesting aspect is that trust is not necessarily based on effectiveness or fair treatment so much as on the predictability of the police’s actions. Citizens seem to appreciate police operations in their countries; they may not like petty corruption, frivolous arrests, or harassment, but they tolerate those things as part of the expected behavior of police. However, they may have higher expectations of other security actors.

Such contextual factors lend to a complex dynamic around police legitimacy. A “dialogic” approach, as discussed above, can aid in efforts to be sensitive to country contexts and dynamic expectations as interventions to build legitimacy are considered (see Bottoms and Tankebe, 2012). Four basic legitimation expectations are offered in the next chapter.

REPORT ORGANIZATION

Chapter 2 examines four pillars of legitimacy, which includes procedural justice in combination with effectiveness, lawfulness, and distributive justice. Chapter 3 presents the state of empirical evidence on police

legitimacy in consideration of these four pillars. Drawing on the policing strategies presented at the workshop and findings from the committee’s first three reports (NASEM, 2021, 2022a,b), the chapter highlights the state of knowledge on crime reduction, community policing, procedural justice, and distributive justice. It also summarizes the committee’s advice for foreign assistance donors. The Appendix provides biographical sketches of committee members and study staff.

This page intentionally left blank.