Reconciling Police and Communities with Apologies,Acknowledgements,or Both: A Controlled Experiment

Reconciliation as a Process to Build Consensus

If law enforcement were beginning a new relationship with communities devoid of historical context, demonstrating procedural justice would likely be sufficient to earn public trust. As many in law enforcement know, however, legal authorities in America are not operating outside of a historical context. There is a historical context in North America and other parts of the world that makes clear why many communities would not view criminal justice authorities as a source of security and help (Kendi 2016). The lessons of this history are reinforced through extreme incidents of abusive authority that affect the mental health of African Americans through vicarious experience (Bor et al. 2018). Additionally, that history is reinforced through less dramatic but highly consequential personal experiences with authorities (Tyler, Fagan, and Geller 2014) and nonovert lessons about the status of African Americans within the state (Justice and Meares 2014). Both laws and broader inequality have enforced norms that subjugate individuals of darker skin tones living in America, with particular subjugation of blacks and African Americans. Despite attempts at progress through law, racial inequality has plagued the United States and its criminal justice system (Alexander 2010).

For authorities to work toward a consensual relationship with all people that the criminal justice system is charged with serving, it must begin to address this history. Research suggests that the community-level actions authorities take to build trust with communities are related to more legitimacy and cooperation when people perceive them to be sincere in their intention to help the community (O’Brien, Tyler, and Meares 2019). Experimental research among African Americans in particular suggests that inviting community participation in how authorities reform is crucial for people to see those initiatives as sincerely intended to help the community (O’Brien and Tyler 2020).

The research discussed here builds on psychological studies of reconciliation, which demonstrate that empowerment promotes willingness to reconcile among groups who have been disempowered through conflict (Shnabel and Nadler 2015). The police-community reconciliation research is consistent with evaluations of community policing efforts (Gill et al. 2014; Skogan 2006), as well as with research on participation as a form of empowerment in government (Fung 2009; Fung and Wright 2003). Reconciliation is not, however, a process that occurs in isolation; it occurs as part of a process that begins with reforms acknowledging injustice (O’Brien and Tyler 2019).

Sincerity is central to the process of reconciliation. Communities must perceive that initiatives to build trust are sincere; otherwise they may backfire and reduce legitimacy (O’Brien, Tyler, and Meares 2019). Part of what shapes perceptions of sincerity is people’s preexisting beliefs about whether authorities are procedurally just. The way in which authorities go about undertaking initiatives to improve their relationship with communities is also important to shaping these perceptions (O’Brien and Tyler 2020).

One strategy for reaching out to communities in the reconciliation process is for authorities to offer public apologies. Police have used public apologies, in a variety of contexts, as a strategy to reach out to communities and address racial discrimination in policing. A police chief in Alabama, for example, apologized in person and on camera to a congressman who had been part of a civil rights protest group that the police force had failed to protect a half century earlier (Giammona 2013). Such statements have the advantage of addressing the history and current grievances of an entire community at one time. The use of public apologies, however, may or may not fully address the reasons for distrusting police. If they do not, such apologies may undermine reconciliation rather than enhance it. One example of this risk dates from 2016, when the head of a major policing organization issued a statement in which he said his organization must

acknowledge and apologize for the actions of the past and the role that our profession has played in society’s historical mistreatment of communities of color. . . . At the same time, those who denounce the police must also acknowledge that today’s officers are not to blame for the injustices of the past.

There are two key messages here. The first is conciliatory: an apology for “the past.” The second is defensive: as he refers to “the past,” the chief separates the admitted injustice of the past from the present and defends the modern institution of policing. Later in the speech, the chief makes an argument for removing responsibility from law enforcement of “the past” by emphasizing that they were doing their job to enforce laws that they did not create (see International Association of Chiefs of Police 2016). Thus, despite the conciliatory language in the apology, the message as a whole could be interpreted as largely defensive.

Political pressure may make intergroup apologies in the real world partial (see Dhami 2017), in that the apology does not acknowledge the scope of the relevant injustice or acknowledge responsibility for it. Perhaps this is because apologies that acknowledge specific faults might admit culpability that could lead to criminal charges or lawsuits (see Robbennolt 2003). Moreover, an admission of guilt may also end up angering those who feel unjustly blamed, bringing political heat on the public official. These dynamics lead to a common pattern in which government officials express regret without acknowledging responsibility: an apology that avoids responsibility, or what we call here a “partial apology,” to be distinguished from a case in which an official neither apologies nor acknowledges responsibility by doing nothing.

Our Research

The current research examines whether such “partial apologies” can backfire among those who are less likely to trust the police, the very people for whom these apologies are most relevant. As police know well, how members of the public interpret what police do and say involves both the actual behavior of police and people’s views of police. One factor is race: on average, black and African Americans trust police less than other groups in the United States (Gallup 2017); they view them as less fair and respectful (Tyler 2005), reflecting the historical role of law enforcement as well as modern disparities. Despite these factors, there is of course variation in how black and African Americans view police, which is why we split our sample according to those participants who view police as fair and respectful (procedurally just) and those who do not view police as fair and respectful (procedurally unjust). We do this so that readers can see how people’s prior views of the police, and in particular how the people least likely to trust them (black and African Americans who view police as unfair and disrespectful), will react to two kinds of apologies—apologies with acknowledgment of responsibility for distrust in the police and those without.

This article summarizes analyses of an experiment we undertook that tested the hypothesis with 678 blacks and African Americans. We also conducted an experiment among a general U.S. sample not limited to blacks and African Americans. The results of this other experiment with 421 participants are available in the appendix, which is available online, and they support the results we present here. We focus here on the results with blacks and African Americans because of the particular challenge that law enforcement have building trust with this group—a group that is disproportionately affected by police violence and that, given the historical context of the United States and ongoing disparities within its criminal justice system, trust police less than other groups in the United States.

We focus our analyses on how apology and acknowledgment impact people’s willingness to reach out to police as a source of help and offer police help in solving crime. People’s willingness to reach out and report crime or talk with police when they are not legally compelled to do so indicates that they view police as a source of help rather than harm. Past research has also examined how beliefs about procedural justice impact legitimacy and the willingness to voluntarily cooperate with the police (Tyler and Fagan 2008; Tyler and Jackson 2014). Since research has already established that people who believe that police are fair and respectful will cooperate with police more, the research presented here is focused on understanding how apology and acknowledgment will impact cooperation. Our particular focus is on those people who do not view the police as fair. These are the people who do not trust police. They should be the central focus of efforts at trust building.

Research Methods

The experiments for this article were conducted using participants recruited and paid through a company called “TurkPrime” through their paneling service. TurkPrime is an online platform through which researchers can recruit participants specifying specific parameters (e.g., nationality, race, ethnicity; Litman, Robinson, and Abberbock 2017). TurkPrime allows anonymous participants to fill out surveys on an internet webpage, for compensation, and in response to various stimuli. In these experiments, we recruited participants, instructed them to read one of four scenarios about a police chief making a public statement, and then asked them to answer structured questions after they read the scenario.

Materials and procedure

Following their consent, participants were presented with information about the context and existing levels of distrust between the police and the community (complete materials appear in the online appendix). The survey instructed participants to imagine “that this was happening in your community” and to reflect on this in writing. The next block showed one of four scenarios. Participants were randomly assigned to one of the four conditions.

In all scenarios, the chief “recognize(d) distrust” between police and the community and called for initiatives to build trust. In the scenario described on the far right of Figure 1, that is all the police chief did; this is to say that the statement acknowledged distrust but neither apologized for it nor acknowledged agency responsibility in its creation. In the second scenario from the far right, the chief acknowledged agency responsibility for community distrust of police but did not apologize. In the third scenario from the far right, the chief apologized for the distrust but did not acknowledge police responsibility in creating the distrust. The apology without acknowledgment focused on “the distrust in the relationship between police and members of many communities, especially black and African American communities, from the past.” The intention of this statement was to make clear that the chief was apologizing but not taking responsibility for the cause of distrust. In the fourth and final scenario (far left), the chief both acknowledged police responsibility for public distrust of the force and apologized for that distrust.

OPEN IN VIEWER

Figure 1

Differential Impact of Acknowledgment and Apology on Cooperation

Results

We focus our discussion on the results for those who do not view police as procedurally just. The presentation of our results here illustrates the differences in impact of the different kinds of gestures police make to the public when those gestures are perceived by people who have low trust in the police compared to people who have relatively high trust. ² These results are supported by a more detailed statistical analysis, which, for purposes of brevity, we do not report in this article but which makes us more confident in our results (see the online appendix).

As Figure 1 shows, the people with the least positive views of police as procedurally just (presented in black bars) generally show less cooperation than those with strongly positive views of police. People with the least positive views of police also respond differently than those with strong positive views to variations in the chief’s public statement regarding community distrust: the most positive response is among those who received both an apology and an acknowledgment, and the most negative response is among those who received a kind of partial apology—an apology for the distrust itself but no acknowledgment of the role of police. The other kind of partial apology we tested—acknowledgment without an apology—was slightly more positive but not as positive as the acknowledgment combined with an apology.

Among people who already have strong positive views of procedural justice, there is little variation between the condition with both apology and acknowledgment and the acknowledgment only condition. People are least positive in the apology without acknowledgment condition and the condition where police neither apologize nor acknowledge.

The analysis suggests two findings. First, among those with a negative view of the police, the responses vary substantially across the four ways in which a police chief makes a statement. Second, it is only the combination of apology and acknowledgement that motivates a higher level of cooperation than doing nothing. On the contrary, partial efforts are less helpful than doing nothing. This finding is consistent with prior evidence that these types of efforts at reconciliation will enhance cooperation (O’Brien and Tyler 2020; O’Brien, Tyler, and Meares 2019). The research design advances our understanding of reconciliation because it allows us to make inferences about the differences between apologies that include or do not include an apology and acknowledgments that include or do not include an apology.

Discussion

The results reported in this study demonstrate the potential for apologies without acknowledgment or acknowledgment without an apology to backfire among the parts of the population with whom police most need to build trust. The experiment demonstrated this effect among those with less positive beliefs about procedural justice represented by a sample of online participants identifying as black and African American. The results of this experiment and the one reported in the online appendix demonstrate that half-hearted attempts at reconciliation through partial apologies to the public may backfire among those for whom they are most important: members of the population that are least likely to trust police.

The results do not suggest that police apologies are negative; rather, they make clearer the kinds of public statements that agency leaders ought to offer. When agency leaders attempt reconciliation efforts with the community by issuing public apologies, those apologies should be paired with an acknowledgment of responsibility. The results emphasize the point that efforts to build a consensual strategy to be effective must not adopt alternative motives, such as appeasing those who may insist that police have nothing to apologize for. The results suggest, especially in the analyses among people perceiving low levels of procedural justice, that denial of the problem further distances communities from authorities, making it more difficult for them to benefit from community cooperation.

It may seem as though partial apologies would never happen. However, in reality they are common. Once officials decide they will make some kind of overture that addresses reasons for distrust, there are political motivations to avoid full acceptance of responsibility, which is why these “partial apologies” are ubiquitous in the real word (Dhami 2017). Officials balance the interest of reaching out to communities with the interest of appeasing those who want the institution to avoid any blame.

This refusal to acknowledge institutional responsibility can always devolve into a conversation about specific incidents and whether they are indicative of bias. Yet the impact of racial inequality in the United States and its criminal justice system transcends conversations about individual incidents. Racial inequality in the criminal justice system goes far beyond the question of whether an individual officer demonstrates racial bias; it goes back at least as far as the historical factors that have left African Americans as a group at large disadvantage both in terms of economic, educational, and housing opportunities and in terms of disproportionate contact with the criminal justice system (Munger and Seron 2017). Individual actions do have strong impact on the relationship between authorities and communities; the brutal beating of Frank Jude, for example, was linked to a drop of 911 calls to police in Milwaukee, particularly in African American neighborhoods (Desmond, Papachristos, and Kirk 2016). Of course, this example and others like it are not isolated incidents, they are incidents that happen in the context of a history of inequality and pervasive discrimination both at the level of law and policy and in individuals’ behaviors toward blacks and African Americans. Beyond the question of whether they as individuals are biased against particular groups, employees of the criminal justice system represent an institution that has demonstrated procedural injustice. That historical context of injustice has created estrangement between authorities and communities (Bell 2016).

Acknowledging responsibility for community distrust on behalf of the police can be politically difficult. Officers, unions, and political activists may react with anger, feeling that innocent officers are being blamed for the actions of a few, or for historical reasons for which they are not responsible. Indeed, at one funeral of New York City Police Department officers, officers turned their backs to Mayor Bill de Blasio, apparently in retaliation for remarks he had made on a separate occasion about conversations he has had with his own son (who has an African American mother and a white American father) about “dangers that he may face” (Altman 2014).

We argue that the response from leaders should not be dictated by the potential for internal opposition. Instead, if policing is fundamentally about serving communities, leaders should seek to promote consciousness of the factors that shape beliefs about policing. Instead of fearing that some will dislike acknowledgment of injustice, police leaders should promote a commitment among their ranks to building trust with communities, which includes recognizing why many distrust police. Irrespective of what officials ought to do, these findings suggest that unless they are willing to both apologize and acknowledge responsibility they are unlikely to achieve the gains they hope for in trust and cooperation from the public.

Although the specific study we presented on apologies does not offer a clear solution for how police leaders can best issue an apology beyond acknowledging responsibility, our other research (O’Brien, Tyler, and Meares 2019; O’Brien and Tyler 2020) suggests that taking proactive measures to address distrust helps to build legitimacy and cooperation. These initiatives are part of a process, not a one-off event that dispels the distrust that is built both through specific incidents and through the larger context of our national history. Taken together with our other research, we recommend following an apology with inviting community participation in initiatives to build trust with the community, whereby the community can have an active role in shaping how police manage crime (O’Brien, Tyler, and Meares 2019; O’Brien and Tyler 2020).

Conclusion

Police need to build trust with communities for their work to be sustainable, and that is not an easy process particularly in contexts of distrust. The results presented here emphasize that reconciliation is not an easy process that can be achieved with a speech. Reconciliation between police and communities will require that people see real improvements in how police treat members of the public. This means that people will need to see both that individual officers treat members of the community with respect when dealing with them one-on-one and that the department as a whole demonstrates respect to the community as a whole. Seeing these changes as the “new normal” will necessarily take time (O’Brien and Tyler 2019). Notably, the effects that the experiment demonstrated were on the decision to cooperate. We measured but did not find effects on beliefs about legitimacy beyond its relationship to beliefs about procedural justice. This has been consistent in experimental research that we conduct (e.g., O’Brien and Tyler 2020): short-term knowledge impacts the decision about whether to cooperate in reporting crime but does not change personal beliefs about authorities. This is also consistent with other psychological research suggesting that behaviors can change before personal beliefs (Paluck 2009). It is an important point because it suggests that people might first decide to cooperate with police, and then begin seeing police as more legitimate once they have procedurally just experiences as they are cooperating. In other words, the willingness to cooperate creates contact with the police, which provides an opportunity for the police to act in ways that further promote trust. The process of reconciliation may be cyclical, with the development of trust hinging on community members experiencing procedural justice, following initiatives that acknowledge the barriers to trust.

Following through with the necessary steps for promoting trust will require considerable investment from police departments. However, if authorities succeed in transforming the relationship between authorities and communities to one of consensual authority, the benefits would far outweigh costs. Such a transformation would help to create self-sustaining governance through legitimacy and informal social controls (Meares 2009; Tyler 2016). Research demonstrates that gestures people perceive as sincere may promote both legitimacy and cooperation (O’Brien et al. 2019; O’Brien and Tyler 2019). With higher legitimacy in the eyes of the public, officers will less often find themselves in negative interactions, much less in situations that call for the use of force.

If American police departments are to succeed in transforming their relationship with communities to a consensual style of authority, they must undertake a long-term strategy that involves understanding what gestures communities will perceive as sincere. The data presented in this article suggest that gestures that do not make deep investments are not a viable route to achieving these gains. Authorities will need to take the route of acknowledging responsibility for why communities, in particular black and African American communities, may distrust police. Understanding precisely how to make such gestures may require community participation to better understand their views, which may in itself promote a more trusting relationship (Fung 2009; O’Brien and Tyler 2020; O’Brien, Tyler, and Meares 2019).

Supplemental Material