Principles for Just and Rational Policing

Police reform has long been a topic of heated debate in the United States. But it assumed new urgency and political significance during the past decade, as national news has carried story after story about the killing of unarmed Black men and women at the hands of law enforcement officials. In 2015, not long after the death of Michael Brown in Ferguson, Mo., the American Law Institute (ALI) launched its Principles of the Law, Policing project to address pressing questions about law enforcement failures and to provide a written framework for building just and rational policing laws, policies, and practices — a framework that police agencies and police reform advocates alike might agree on.

ALI Principles of the Law projects aim to offer best practices for issues that have significant legal underpinnings. Drawing on a variety of sources, including existing policies and practices in various jurisdictions, social science research, and constitutional norms, the publications are primarily addressed to legislatures, administrative agencies, and private groups. The audience for the Policing Principles project is broad, including lawmakers, police agencies, bodies that regulate or conduct oversight on policing, the public, and also, in some instances, the courts. For judges who regularly interact with law enforcement on cases in which police are involved, the principles offer new insights into policing practices.

Now nearing final publication, Principles of the Law, Policing is organized into 14 chapters: General Principles of Sound Policing; General Principles of Searches, Seizures, and Information Gathering; Policing with Individualized Suspicion; Police Encounters; Policing in the Absence of Individualized Suspicion; Policing Databases; Use of Force; General Principles for Collecting and Preserving Reliable Evidence for the Adjudicative Process; Forensic-Evidence Gathering; Eyewitness Identifications; Police Questioning; Informants and Undercover Agents; Agency and Officer Role in Promoting Sound Policing; and Role of Other Actors in Promoting Sound Policing.

Earlier this year, David F. Levi, ALI president and former director of the Bolch Judicial Institute, interviewed the Policing Principles project’s reporters for Judicature. Those reporters — the scholars and practitioners responsible for developing the project and preparing and presenting drafts to a large and diverse list of project advisers and ALI members for feedback — are Barry Friedman, the Jacob D. Fuchsberg Professor of Law and Affiliated Professor of Politics at NYU Law, who served as lead reporter on the project, and associate reporters Brandon L. Garrett, the L. Neil Williams Jr. Professor of Law at Duke Law School; Rachel Harmon, the Harrison Robertson Professor of Law at the University of Virginia School of Law; Tracey L. Meares, the Walton Hale Hamilton Professor of Law at Yale Law School; Maria Ponomarenko, an associate professor at the University of Texas at Austin School of Law; and Christopher Slobogin, the Milton R. Underwood Chair in Law and director of the Criminal Justice Program at Vanderbilt Law School. Their discussion, conducted via email, follows.

More information on the project and the full list of advisers can be found at www.policingprinciples.org.

---

David F. Levi: Barry, let’s start with you. What was the impetus of Principles of the Law, Policing, and why did you decide to get involved?

Barry Friedman: Ricky Revesz, then the director of ALI, came to me in the fall of 2014, following the protests in Ferguson after the shooting of Michael Brown. The entire country had watched the stunning footage of a very militarized policing agency facing down unarmed protesters. I initially said no, as did every associate reporter I asked — and one reason we all declined was the same: In the 1960s and 1970s, ALI had put a decade of work into a Model Code of Pre-Arraignment Procedure, and after all that work, it received very little takeup. But ultimately, we concluded that the issue of sound policing was too vital to the country, and that this might be the moment to have an impact.

As the project proceeded, the country witnessed many more shootings and great controversy over policing. And yet our group of advisers worked together on all of this with great civility and a remarkable degree of agreement. It’s worth looking at who those advisers were, because it was an extremely diverse group over many dimensions. We had numerous policing leaders in the same room as the Movement for Black Lives, the ACLU, and the NAACP Legal Defense Fund, as well as the CATO Institute, the Institute for Justice, and the Heritage Foundation. And, of course, judges, prosecutors, defense lawyers, and much more.

Levi: You have said that, unlike our regulation of most other government agencies, regulation of policing has traditionally been done on the back end, and that this is a mistake. Can you explain this thought?

Friedman: Sure. The concept of “accountability” cashes out very differently around policing than it does elsewhere in government. In most cases, the norm is that there are statutes, regulations, rules, and guidelines that tell public officials what is expected of them. Call that “front-end accountability.” In policing, though, there is too much of a vacuum on the front end, and we try to rely instead on a host of back-end measures, which kick in only after things go wrong: criminal prosecutions of officers or civil rights suits, oversight by civilian review boards, federal pattern and practice investigations, and even body cameras — all are designed to figure out what went wrong and who should be held responsible.

As we all know, these back-end measures don’t work very well. One of the reasons is that there is not enough guidance on the front end about what is expected. We hope Principles of the Law, Policing can begin to fill that gap. Principle 1.06, which we refer to again and again, calls for written policies to guide officer behavior, formulated with community input where possible. Since the project was concluded, several organizations affiliated with the reporters have begun to write model legislation based on it. And since the protests following the murder of George Floyd, more and more states are passing such legislation. Similarly, we hope to spark the writing of model department policies based on the principles. The Knowledge Lab — a joint product of the Department of Justice’s Bureau of Justice Assistance and the National Policing Institute — recently allocated some funding to develop toolkits to help policing agencies put the principles into action.

Levi: You have noted that policing has changed to incorporate more predictive or proactive policing in addition to the reactive policing that we are most aware of. Can you expand on this?

Friedman: Back in the day, police investigated “suspects,” which is to say their work was driven by the fact that a crime had been committed, or was imminent, and the goal was to identify the person or persons responsible. Police of course did more, including what might be called “order maintenance” (which often involved policing marginalized or unpopular groups), but the point is that when an investigation was occurring, it was of a suspected crime. Today, policing increasingly is “suspicionless,” which is to say it involves investigating people not suspected of anything. Think here of airport security or sobriety roadblocks — the goals are deterrence as much as anything.

One of the most significant parts of the Policing Principles project is Chapter 5, directed at such suspicionless policing. The courts have struggled with what to do with this sort of programmatic policing, but, drawing on existing precedents, we provide a structure. The reporters believe it is entirely workable — and we got very little pushback at any stage of the drafting process on this, likely because it is so logical and based in precedents.

Levi: Thank you. Tracey, given that you were doing this work during the period when George Floyd was killed, one might expect that the project would in some way be itself controversial. But that wasn’t the experience here, even though the advisers to this project included the entire spectrum of those whom one might view as “stakeholders,” ranging from defense lawyers and civil liberties groups to police chiefs and prosecutors. Would you agree that the project was not controversial? And why do you think this was?

Tracey L. Meares: We began our work with a firm belief that the project of articulating principles to improve policing should not be and is not controversial. We grounded the specific chapters — directed at issues such as encounters with police, use of force, evidence gathering, police questioning, and so on — within Chapter 1’s set of guiding principles. These “General Principles of Sound Policing” were written to capture a set of critical points that even groups that do not always agree on specific strategies and tactics can agree upon. Those principles include the importance of ensuring that policing is constitutional, legitimate, and focused on reducing the harm it imposes, and that it pays special attention to the concerns of vulnerable populations.

Given the centrality of Chapter 1 throughout the entire project, I think we found many opportunities not only for individuals with differing perspectives — but also, even more important, opportunities for those working across different institutions of criminal legal processing — to find common ground.

Levi: ALI principles projects tend to propose best practices and are addressed primarily to legislatures, administrative agencies, and private actors — unlike ALI Restatements of the Law, which are primarily addressed to courts. Maria, can you talk about who the intended audience was?

Maria Ponomarenko: These principles offer guidance to anyone whose role includes influencing police conduct. That clearly means policing agencies that dictate the policies, supervision, training, and incentives for officers; the legislatures that write laws to govern their behavior; and the judges who interpret those laws and see officers in court. But it also means the state administrative agencies, prosecutors, defense lawyers, civil litigators, consent decree monitors, state attorneys general, government officials, community leaders, civil rights groups, and insurers who seek to shape what officers do in the context of their work. Some principles offer guidance to anyone interested in assessing police policies and practices. But Chapter 14 is somewhat unique in that it has principles directed at the specific responsibilities of nonpolice actors — legislators, judges, and prosecutors, as well as the federal government, private and philanthropic entities, researchers, and more.

Levi: Many of our readers are judges, so they would be particularly interested in whether the principles or any part of them are addressed to the courts. The final chapter sets out the role of nonpolice actors in promoting sound policing. In it, you say that each branch of government should “take steps” to ensure that policing is lawful and to advance the principles detailed in the project. What responsibilities do judges have to encourage sound policing, and how should they go about promoting these ends without jeopardizing their judicial role or violating other principles concerning the separation of powers and constitutional and other limitations on jurisdiction?

Christopher Slobogin: Many scholars and policymakers tend to view the exclusionary rule as the judiciary’s primary tool for regulating the police. But suppression of evidence obtained through violation of the Constitution affects only a fraction of policing activity, and in any event often has only a very indirect impact on the police. Assessing damages against individual officers or departments is another way judges can influence police behavior, but current law imposes significant impediments for plaintiffs seeking relief under 42 U.S.C. § 1983 and related statutes.

In recognition of these facts, Section 14.04 of the Policing Principles project provides guidance on a host of very specific things judges are already doing, and can and should continue to do in the ordinary course, to promote sound policing. One of these is the critical role of judges in finding the facts around police conduct. Section 14.04(a) provides that, when judges consider police assertions in warrant applications at suppression hearings and at trial, they should do so “without a presumption of credibility.” To aid in this fact-finding, the principles in various chapters call for the recording whenever possible of police-citizen interactions, whether on the street, the interrogation room, during identification procedures, or in the forensic lab. Chapter 14 also calls on judges to enforce disclosure and documentation requirements directed at the police and prosecutors.

On a systemic level, Section 14.04(b) points out that the judiciary can foster sound policing by establishing case-assignment procedures to eliminate or minimize judge-shopping, including in connection with the warrant docket; fostering transparency in the adjudicative process, including providing to the public data relevant to sound policing; and minimizing nondisclosure orders regarding policing practices. The principles also recommend that the judiciary develop mechanisms and protocols that enable individual judges to take note of, and report, instances of unlawful conduct by officers in the performance of official duties.

In carrying out these tasks, judges can also be helped immensely by greater input from legislatures and from police agencies themselves. That is why the principles throughout emphasize the need for statutes, regulations, and policies that detail the role of law enforcement in carrying out investigations and how these rules can be enforced. For example, Section 14.03 provides that “legislative bodies should ensure that immunities from liability do not vitiate remedial goals,” and the commentary suggests that one way of implementing that objective is “to make agencies and jurisdictions liable in respondeat superior for the actions of their officers.”

Levi: Rachel, do you think that judges have fallen short of whatever duty they have to supervise law enforcement agencies? In what ways?

Rachel Harmon: Judges face significant obstacles as regulators of policing. First, judges rarely have the opportunity to evaluate the conduct of agencies because constitutional and other doctrines make it hard for plaintiffs to get such claims into court. That means judges do not have a chance to review the departmental policies, training, supervision, and accountability systems that drive officer conduct. It also makes it harder for judges to understand policing.

Instead, judges review the conduct of individual officers, most often in the context of the exclusionary rule. That can be a problem. By definition, motions to suppress evidence only come about when the defendant has been caught with the goods. As a consequence, judges see a biased sample of what happens in the world; they do not see the vast majority of police encounters in which police are either not looking for criminal evidence or do not find it. Nor do they see the cases in which the government would rather drop criminal charges than bring officer conduct into court. Even in nonexclusionary rule cases, such as those involving damages claims against officers who use violence, officer-friendly immunity doctrines press judges to view and treat officer conduct favorably. As a result of these forces, many judges tend to defer to the police version of encounters with citizens. Some judges also appear to assume that police, prosecutors, and other law enforcement personnel always or almost always act in a competent, fair, and nondiscriminatory manner, an assumption that can lead them astray.

The principles that guide police conduct — the bulk of this project — can help judges by providing insight into sound police practices. The reporters’ notes (which are published with the principles) offer citations to actual policies of law enforcement agencies that spell out rules for how things can be done.

Judges can rely on these policies knowing that the reporters did not draft these principles out of whole cloth or their own notion of ideal policing. They received input and direction from a group of advisers that included many judges, police, and prosecutors, as well as those who have advocated for reform. It again is worth noting how little disagreement there was over many of the principles.

Levi: One of the key principles that you identify is that the police should operate subject to written rules, policies, and procedures. It is somewhat surprising to learn that this is an aspirational principle and that many police departments lack written policies on many aspects of policing such as the use of photo lineups. Can you discuss this phenomenon?

Harmon and Ponomarenko: Part of the problem is that there are nearly 18,000 law enforcement agencies (and that is an approximate number because the federal government does not reliably count these agencies, unlike their tracking of, for example, schools) at the state and local level — the vast majority of which are small agencies that lack the resources to develop policies on all of the issues that their officers are likely to encounter. But the reality is that even larger agencies often provide insufficient guidance on key policing issues. Justice Department investigations of major city police departments have routinely pointed to significant gaps in police department policies on everything from use of force to the use of confidential informants. Sometimes agencies provide guidance to officers through training or agency bulletins. But these materials rarely are visible to the public — and often are not binding in ways that make it possible to then hold officers to account. We have chapters on these issues. The first chapter we completed, which became Chapter 7, is on the use of force. The reporters since have relied upon it to draft a model statute governing the use of force as well as a model department policy.

Levi: The project also contains a set of general principles for searches, seizures, and information gathering and addresses police access to data, records, or physical evidence held by third parties. Can you discuss the concerns here and what is recommended?

Slobogin: In Chapters 2, 3, and 5, the Policing Principles project set forth a number of guidelines that should help courts, legislatures, and law enforcement agencies navigate the complicated terrain of what the principles call “information gathering.” Chapter 2, which sets out general principles governing searches and seizures, explains in Section 2.02 that this concept is meant to encompass not only traditional searches and seizures governed by the Fourth Amendment but also technological surveillance techniques, access to third-party databases, and undercover operations that currently are not governed by the case law construing that amendment. Even if these information-gathering techniques are not searches for Fourth Amendment purposes, they can involve intrusive law enforcement actions that compromise privacy, speech, and other important interests. Thus, the principles take the position that they should be regulated by law.

The principles divide information gathering into two categories: suspicion based and suspicionless. Suspicion based policing occurs when police want to investigate a particular person. In this situation, Chapter 3 provides that, in addition to or, in cases involving lesser intrusions, instead of the constitutionally recognized warrant, probable cause, and reasonable suspicion requirements, other regulatory mechanisms may be appropriate, including court orders based on less than probable cause, authorization by police supervisors, limitations on the types of law enforcement officials who may carry out the specific information-gathering technique at issue (see Section 3.02(a)), and notice to individuals about their right to refuse cooperation (see Section 3.04). Under Section 3.02(b), the decision about which of these regulatory options applies and in what combination will depend upon the intrusiveness of the information-gathering technique; its potential for exacerbating racial disparities, chilling constitutional rights, and facilitating other forms of abuse; the feasibility of obtaining advance approval and the likelihood that abuse can be identified through after-the-fact review; and the degree to which a particular safeguard would impose undue burdens on the agency. The principles also call for documentation of information-gathering activities and require notice about the use of particularly invasive techniques to any affected individual who is not prosecuted or otherwise not informed of their use through normal discovery procedures.

Chapter 4 applies the same objective of minimizing unregulated policing to what it calls “encounters,” which are defined to include not only Fourth Amendment seizures but also most other confrontations between police and citizens. One implication of this approach is the recommendation in Section 4.06 that police be prohibited from asking for consent to search a person or effects unless they have reasonable suspicion that the individual is involved in criminal activity. Another, found in Section 4.07, is that searches incident to arrest should be based either on reasonable suspicion or on a written policy that specifies the scope of the search and is applied “evenhandedly.” A third implication, set out in Section 4.05(a), is that custodial arrests be reserved for those situations where they “directly advance[] the goal of public safety”; otherwise, citations should be preferred.

In contrast to Chapters 3 and 4, Chapter 5 deals with information gathering and encounters that are suspicionless, defined in Section 5.01 to include actions that are “conducted in the absence of any cause to believe the particular individual, place, or item . . . is involved in prohibited conduct.” In these situations — which can encompass a wide range of common policing techniques, including checkpoints, inspection regimes, drug-testing programs, and surveillance systems — an individualized suspicion requirement does not work. Courts, including the Supreme Court, have had great difficulty analyzing these types of policing, which are often treated as “special needs” situations that call for different, and usually very relaxed, regulation. The principles provide for more robust regulation of these programmatic information-gathering techniques, which can often affect tens of thousands of individuals, virtually all of whom are innocent of wrongdoing.

In brief, the principles provide that suspicionless programs should occur only when there is a strong rational basis for the program after considering its impact on collective and individual interests; policies are developed that explicitly identify the program’s purpose and scope; and the program is implemented in a neutral, evenhanded fashion (see Sections 5.02, 5.03, 5.05). The rationale for these conditions is straightforward. As the commentary states, “In the absence of warrants and individualized suspicion, it is essential that there be alternative mechanisms in place to ensure that search and seizures and other policing activities are justified, are not directed at individuals or groups in an arbitrary or discriminatory fashion, and are limited in scope consistent with their justification.”

Chapter 6, which deals with policing databases, likewise requires, in Section 6.01(c), written policies that specify “the purpose of the data collection, including the criteria for inclusion in the database; the scope of data to be collected, including the types of individuals, locations or records that will be the focus of the database; and the limits on data retention, the procedures for ensuring the accuracy and security of the data, the circumstances under which the data can be accessed, and mechanisms for ensuring compliance with these rules.” Other principles in Chapter 6 spell out the implications of this language. The courts have rarely weighed in on these types of issues. Yet given the amount of data now collected by law enforcement agencies, either directly or by paying private companies, it is crucial that they be addressed by statute or regulation. This is one of the fastest growing areas of practice, especially in an era of AI, and it needs rapid attention.

The principles also confront the important controversies raised by “pretextual” policing, which involve an investigation of a minor crime (often a traffic violation) that would not occur but for a desire on the part of the police, often based on a hunch or nothing at all, to discover evidence of a more serious crime. Chapter 2 states that police “should not use pretextual policing as a general strategy to address unlawful or undesirable conduct” and further states that, if used at all, pretexts should only be permitted when officers are investigating “a specific serious offense” that they have an “articulable belief” the targeted individual has committed. Chapter 5 similarly provides that suspicionless actions should not exceed the scope of their authorization. This concern about pretextual policing is bottomed on the reality that police officers often exploit the ubiquity of minor offenses and programmatic searches and seizures in biased and discriminatory ways that can exacerbate tensions with the community, undermine their own legitimacy, and occasionally erupt into serious physical altercations.

Levi: The project also addresses the use by police of profiles and predictive models. These models can be used in a variety of settings to identify likely criminals or victims or patterns of activity that might suggest criminal conduct. There is concern that some of these models may incorporate racial or other kinds of stereotyping or bias. Can you explain? What recommendations do you make and how did you balance the competing considerations?

Meares: Of late, suspicionless policing occurs frequently around data. Police agencies increasingly rely on a range of data to inform their decision-making, often by creating profiles or engaging in prediction. This can, in theory, make their work more informed. In practice, however, agencies have often implemented “black box” or proprietary technology, where its usefulness is unknown. It may rely on data that is inappropriate for policing because it is biased, including racially biased, and simply error prone. And these systems may lead to poor public safety outcomes, as well as harm people’s rights, in ways that are difficult for law enforcement or the public to assess. Section 2.06 provides guidelines for how algorithms can be used in a transparent and sound way. As technology advances, it is essential that principles like this one be followed. Similarly, we have an entire chapter on policing databases and how they should be managed.

Levi: The ALI decided to release the chapter on use of force in 2020 before completion of the project because of the killing of George Floyd. The chapter was distributed to police agencies throughout the country. How helpful do you think this chapter is and what was the reaction to it from police departments and police chiefs?

Friedman: As we note above, this chapter formed the basis for a model statute and a model department policy. We have received a number of requests for the model policy. And we have worked with many states on use of force legislation that relies at least in part on that model statute. However, the fact remains that most states lack adequate regulation of the use of force, and this really needs to change. The use-of-force chapter, Chapter 7, and the model state statute should provide a guide.

Levi: You also address best practices for forensics labs. This seems like a very important topic. Can you explain the context and the recommended best practices and how they differ from current practice?

Brandon L. Garrett: Collecting evidence accurately, including forensic evidence, is crucial to the crime-solving mission of police agencies, as we describe in Section 9.01 of the Policing Principles project. While police agencies have long had patrol guides that have been the subject of legislation and constitutional criminal procedure, that kind of law is largely absent when it comes to forensic evidence. Many assume that it would be the opposite; after all, clinical laboratories have long been regulated by detailed federal legislation.

Crime laboratories have not been so regulated. As we highlight in Section 9.01: “Basic scientific standards should apply to the use of forensic evidence, whether it is scientific or technical evidence, or a combination.” And yet the lack of scientific standards has been a longstanding challenge in forensics, a field largely not developed by scientific researchers. The result has been high-profile quality-control failures around the country, where labs have been shut down, thousands of cases have been reopened, and serious wrongful convictions have come to light. As we note in Section 9.04, “If the process for supervising forensic staff does not include checks on the quality of their casework, then there is no way to know how well they are performing.”

Judges can play a crucial role around forensic evidence. In December 2023, Federal Rule of Evidence 702 was amended for the first time since 2000, highlighting the special need for judges to carefully examine the reliability of expert evidence and emphasizing that need in criminal cases. [See The Amendment to Federal Rule 702 in this edition.] That amendment also underscored the need to evaluate the opinions that an expert reaches, and not just the underlying methods, when conducting that analysis. Police agencies should do the same. They need to critically assess what forensic tools they are using, whether they are supported by adequate research and for the tasks for which they are used. As we describe, “[a]gencies, as consumers of forensic science, should insist through policy and practice that only scientifically valid methods are used, and that any limitations of a method be included as part of any conclusions reported.” In short, agencies need to adhere to scientific standards when collecting and analyzing forensic evidence — and all evidence.

One important way to accomplish this structurally is through the independence of crime laboratories. As we describe in Section 9.01: “Leading scientific organizations, such as the National Academy of Sciences, have called for the independence from law enforcement of crime laboratories and forensic-evidence-related work.” That type of independence ultimately powerfully benefits law enforcement by safeguarding the accuracy and integrity of forensic evidence.

Levi: The reporters and all who participated in Principles of the Law, Policing deserve our congratulations for doing such terrific work and reaching consensus on so many difficult issues. I think judges will be particularly interested in the project, and not just the recommendations but also the discussions and summaries of the literature on current police administration. I hope that the project is widely read, considered, and adopted in whole or in part by police departments and law enforcement agencies.

In light of that hope, which I am sure you share, once the project is published, what can we all do to promote the dissemination and adoption of these principles and what can others do to help?

Friedman: We welcome all the help we can get. As we were writing our answers to these questions, we also were meeting to discuss how ALI can launch these principles successfully. We expect the official text of the principles to be available by early 2025. With the help of the project’s advisers and reporters, we hope to get the principles into the hands of the organizations that will benefit from guidance on any of the topics addressed. We expect to host a full slate of events and webinars and to make them available in a variety of formats to make them as widely accessible and useful as possible.

---

Barry Friedman is the Jacob D. Fuchsberg Professor of Law and Affiliated Professor of Politics at NYU Law. He was lead reporter on the policing project.

Brandon L. Garrett is the L. Neil Williams Jr. Professor of Law at Duke Law School.

Rachel Harmon is the Harrison Robertson Professor of Law at the University of Virginia School of Law. 

David F. Levi is president of the American Law Institute and director emeritus of the Bolch Judicial Institute at Duke Law.

Tracey L. Meares is the Walton Hale Hamilton Professor of Law at Yale Law School.

Maria Ponomarenko is an associate professor at the University of Texas at Austin School of Law.

Christopher Slobogin is the Milton R. Underwood Chair in Law and director of the Criminal Justice Program at Vanderbilt Law School.