3

Law and Legality

This chapter examines the relationship between proactive policing and the law. Supporters of proactive policing strategies that are intended to build community relationships, such as community-oriented policing and procedural justice policing, suggest that these strategies will help protect legal values and lead to less law-breaking by police. Critics sometimes argue that proactive policing—through strategies such as hot spots policing; stop, question, and frisk (SQF); and broken windows policing—lead police officers and departments to violate the law (see, e.g., Rosenbaum, 2006; Kochel, 2011). In either case, law is a critical constraint on policing; however effective a policing practice may be in reducing or preventing crime, it is impermissible if it violates the law.

Proactive policing, as defined in Chapter 1, is rarely forbidden by law. The proactive policing practices discussed in this report generally are law enforcement strategies or tactics, and occasionally, higher-level philosophies of policing.¹ Law primarily regulates individual acts by officers and the decisions and policies set by municipalities and departments that guide these acts; the law neither encourages nor discourages particular strategies or philosophies. Nevertheless, since some proactive policing strategies are implemented through common sets of policies and acts, and those policies and acts are governed by federal, state, and local law, the law governs proactive policing strategies indirectly. Since different kinds of proactive strate-

___________________

¹ The logic models discussed in Chapter 2 for the four proactive policing approaches, including the associated primary objective and key ways to accomplish the objective shown in Table 2-1, are examples of philosophies of policing.

gies may be implemented with similar police action, and proactive strategies are often implemented in a variety of ways, proactive strategies based on the same logic model may raise disparate concerns, and strategies based on different logic models may raise similar issues. Moreover, proactive strategies may vary in how much they raise legal concerns, depending on what activities are used to implement them. In light of these considerations, this chapter highlights proactive strategies with significant legal implications rather than considering each proactive strategy by its logic model.

The committee considered several ways that law and proactive policing might interact. First, since constitutional and statutory law regulates police activities that might be used to implement a proactive strategy, a strategy could cause violations of law by increasing the probability that police action falls outside the boundaries of existing legal constraints. Second, legal rules concerning permissible conduct, or legal consequences for violations such as those arising from civil suits and criminal prosecutions of officers, could shape departmental and officer decision making about whether and how to conduct proactive policing. Third, even if police action pursuant to proactive policing does not violate the law, it may undermine legal values and principles such as privacy, bodily integrity, autonomy, or accountability, or it may foster inequality in ways that generate public concern. This public concern could in turn be the basis for changing the law to expand regulation of proactive policing. Fourth, some proactive strategies could reduce opportunities for lawbreaking by the police or increase incentives for police compliance with the law.

There are other ways that proactive policing and law interact that were not central to the committee’s charge and therefore were not considered by the committee. Most notably, law sometimes promotes particular proactive policies. The U.S. Department of Justice (DOJ), for example, promotes community policing through litigation by its Civil Rights Division against police departments for patterns and practices of constitutional violations, leading to enforceable settlements that mandate implementing community policing. The Office of Community Oriented Policing Services (COPS) in DOJ also awards grants promoting community-oriented policing and procedural justice policing pursuant to federal legislation.

This chapter comprises several parts. Since the most important legal constraints on proactive policing are the Fourth Amendment and the Equal Protection Clause of the U.S. Constitution, the first two sections of the chapter describe ways proactive policing interacts with these constitutional rights and related statutory provisions. This discussion of legal rights and proactive policing is largely based on court decisions, federal investigations, and non-empirical legal scholarship because the limited existing empirical research does not permit strong conclusions about whether proactive strategies lead to constitutional violations. The third section considers the

implications of the major remedial mechanisms the law uses to induce police compliance with constitutional rights, in order to consider the effects legal consequences might have on the means by which proactive policing strategies are implemented. In addition to Fourth Amendment and Equal Protection law, proactive strategies must comply with a diverse array of other federal, state, and local law that regulates the police. The fourth section therefore considers some of these other rules and, more broadly, discusses ways that proactive policing strategies may violate legal values even when they are implemented in ways that comply with the law. The fifth section discusses the relationship between law and community-based proactive policing strategies, namely, community-oriented policing and procedural justice policing, which raise different issues than do other proactive policing strategies.

FOURTH AMENDMENT

Legal Overview

The Fourth Amendment to the U.S. Constitution states: “The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no warrants shall issue, but upon probable cause, supported by oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.” Under the Fourth Amendment, a police officer seizes a person when he restricts his liberty, either by a show of government authority to which the individual submits or by physical force (Terry v. Ohio, 392 U.S. 1 [1968]; Hodari D. v. California, 499 U.S. 621[1991]). Thus, arrests, pedestrian stops, traffic stops, and all uses of force by the police constitute seizures within the meaning of the Fourth Amendment. Fourth Amendment searches occur when the government intrudes upon an individual’s reasonable expectation of privacy (Katz v. United States, 389 U.S. 347 [1967]) or when it physically trespasses onto a person’s property for the purpose of gathering information (United States v. Jones, 565 U.S. __ [2010]; Florida v. Jardines, 569 U.S. 1 [2013]). Searches include both physical searches, such as looking in a car trunk or frisking a suspect, and electronic searches, such as listening in on a phone call or placing and monitoring a GPS [global positioning system] unit on a suspect’s car.

Under the terms of the Fourth Amendment, a government search or seizure must be reasonable. Fourth Amendment reasonableness often requires that the police possess a quantum of evidence about an individual’s involvement in a criminal offense before initiating a search or seizure. In some cases, the amount of suspicion required to engage in a stop or search is tied to the intrusiveness of the activity. Thus, a stop, which is a brief detention

of a person short of a full arrest, requires that an officer have evidence rising to a “reasonable suspicion” that the person stopped is currently involved in criminal activity or has just committed or is about to commit an offense.² In order to lawfully conduct a frisk, which involves patting down a person’s body outside his or her clothes for weapons, the officer must reasonably suspect that a person with whom an officer is interacting is armed and dangerous (Terry v. Ohio, 1968; Adams v. Williams, 407 U.S. 143 [1972]). Since arrests are more intrusive than stops, the U.S. Supreme Court requires that an officer have “probable cause” to believe that a person has committed a crime, a higher level of justification than “reasonable suspicion” (Draper v. United States, 258 U.S. 307 [1959]; Atwater v. City of Lago Vista, 432 U.S. 318 [2001]). Many types of searches other than frisks, such as searches of homes, also require probable cause to believe that a suspect or evidence of a crime will be found in the location searched, and these searches sometimes require a warrant, which ensures that a police officer establishes probable cause to a neutral magistrate before the search takes place.

For some searches and seizures, including some that may be used in proactive policing, it does not make sense to measure reasonableness by whether there is individualized suspicion because the police actions in these instances are not carried out primarily because someone is suspected of a crime. These actions include, for example, uses of force, DNA sampling of arrestees, and immigration checkpoints. Courts evaluate whether these activities are reasonable by balancing the severity of the intrusion on the individual against the interests of the government (Graham v. Connor, 490 U.S. 386 [1989]; Indianapolis v. Edmond, 531 U.S. 32 [2000]; Maryland v. King, 569 U.S. ___ [2013]).

While stops, searches, and arrests are all regulated by the Fourth Amendment, the Fourth Amendment case law defining what constitutes a search or seizure also puts many common policing activities used in proactive policing strategies beyond the scope of the Fourth Amendment’s restrictions. Most critically, the doctrine governing consensual encounters, the third party doctrine, and the doctrine concerning movements in public permit police to gather information and monitor individual action in several ways without engaging in a search or seizure within the meaning of the Fourth Amendment. First, unlike encounters that would communicate “to a reasonable person that the person was not free to decline the officers’

___________________

² As noted in Chapter 2, the U.S. Supreme Court has not ruled as to whether Terry can be used to investigate a completed misdemeanor, and it has suggested that it might not be permissible. However, Terry can be used as the legal justification for police to investigate a completed felony (United States v. Hensley (469 U.S. 221 [1985]); see also Navarette v. California (572 U.S. ___ [2014]).

requests or otherwise terminate the encounter” (Florida v. Bostick, 501 U.S. 429 [1991]), consensual encounters between police officers and pedestrians do not constitute a seizure. Similarly, searches to which a subject voluntarily consents—even if the action is a search within the meaning of the Fourth Amendment—are considered reasonable (Schneckloth v. Bustamonte, 412 U.S. 218 [1973]). Second, under the third party doctrine, police may obtain, without probable cause or a warrant, information an individual has revealed to a third party, though the police would have had to comply with these requirements if the information had not been previously disclosed. This is true even if the information was disclosed on a limited basis or for a limited purpose, such as to one’s bank through bank transactions (United States v. Miller, 425 U.S. 435 [1976]; Smith v. Maryland, 442 U.S. 735 [1979]). Third, the police may also watch a person’s movements in public, including through technological means, unless they engage in a physical trespass to do so, without triggering the Fourth Amendment (United States v. Knotts, 460 U.S. 276 [1983]; United States v. Jones, 2012). Finally, the decision to investigate is not itself an activity regulated by the Fourth Amendment, though it can lead to searches and seizures that are regulated. When a police activity does not constitute either a search or a seizure within the scope of the Amendment, it need not be reasonable and does not require probable cause or a warrant under the Fourth Amendment, though it may still be subject to other law.

Deterrence-Oriented Proactive Strategies

As Chapter 2 suggests, several proactive policing strategies work to maximize the perceived consequences of criminal activity to potential criminals as a means to discourage that activity. One way some departments pursue this aim is to engage in frequent searches and seizures to deter criminal activity. Thus, SQF promotes stopping and frisking pedestrians as a means of discovering weapons and drugs and deterring people from carrying them. Similarly, hot spots policing often involves intensive patrols, including stops, frisks, and arrests within the microgeographical high-crime locations, and zero tolerance policing includes frequent stops, searches, and arrests, often for minor offenses (Mastrofski, Worden, and Snipes, 1995). Although both reactive and proactive stops, frisks, and arrests are subject to the same legal standards, deterrence-oriented proactive strategies interact with the Fourth Amendment in distinctive ways. Specifically, proactive practices often take significant advantage of Fourth Amendment discretion generated by the U.S. Supreme Court in reactive contexts, and there is some indication that in doing so these proactive practices may produce constitutional violations.

The Court frequently crafts Fourth Amendment rules that are simpler

and more permissive than a determination of government need and individual interests in individual cases might otherwise warrant, in order to ensure that law enforcement has guidance and yet adequate flexibility to address the myriad, and sometimes unpredictable, circumstances that officers face (Atwater v. Lago Vista, 532 U.S. 318 [2001]) (Harmon, 2012b). More specifically, the rules governing stops, frisks, and arrests permit officers generous discretion. Thus, lower courts following constitutional case law permit officers to stop a suspect on the street based on reasonable suspicion of criminal activity rather than probable cause; to make a frisk based on reasonable suspicion that a suspect that has been stopped is armed and dangerous (Terry v. Ohio, 1968); and to make a warrantless custodial arrest, even for a very minor offense, such as a seat belt violation, that is punishable only by a fine (Atwater v. Lago Vista, 2001).

In justifying giving officers clear rules and flexibility, the U.S. Supreme Court has reasoned in part that officers usually have weak incentives to use intrusive means to address minor or equivocal conduct. As a result, officers are most likely to use the full zone of flexibility permitted by Fourth Amendment doctrine only when circumstances most warrant it (Atwater v. Lago Vista, 2001; cf. Hudson v. Michigan, 547 U.S. 586 [2006]). Notably, this reasoning assumes conventional policing: traditionally, stops, frisks, and arrests are tools police use reactively as a means to address a particular crime they witness or have reported to them or to investigate specific suspicious behavior. In this context, harmless or ambiguous individual conduct often will not justify the resources that would be necessary to address it, and officers are assumed to leave such conduct unaddressed rather than intrude on individuals.³ By contrast, in proactive policing, departments often employ coercion more expansively to promote forward-looking, preventative ends rather than merely to investigate or enforce criminal law. Thus, proactive policing may encourage legal stops, frisks, and arrests even for equivocal or minor individual conduct.

This strategic use of Fourth Amendment doctrine for proactive policing is legal: the U.S. Supreme Court has repeatedly resisted considering subjective officer motives in evaluating searches and seizures for reasonableness, and it has permitted the pretextual use of legal authority to engage in searches and seizures (Whren v. United States, 517 U.S. 806 [1996]). Nevertheless, deterrence-oriented proactive strategies that rely on stops, frisks,

___________________

³ This assumption about the frequency with which police do not take formal enforcement action even when there is sufficient evidence to do so (“leniency”) is generally supported by empirical research (National Research Council, 2004, pp. 115–116). Legal factors (e.g., strength of evidence) are among the more powerful predictors of police use of formal enforcement, but they are hardly determinative. However, this literature does not compare police practices under high and low levels of proactivity, and many of the studies were conducted at times when proactive practices were not strategically promoted.

and arrests generate incentives for officers to conduct more frequent and intrusive, and therefore liberty reducing, searches and seizures than reactive policing would generate, and those strategies are aided by the legal rules developed for reactive policing.⁴ Moreover, some scholars and critics argue that using these tools proactively potentially affects the legality of the police activities that result because departments encourage stops, frisks, and arrests for reasons other than the individual suspicion that provides the legal basis for the activities (Meares, 2015). Departments need to employ strong incentives for officers to engage in only those searches and seizures that satisfy the demands of the Fourth Amendment. Otherwise, encouraging stops, frisks, and arrests could easily result in searches and seizures that do not comport with constitutional standards. Without a strategy to ensure that officers comply with the Fourth Amendment, when departments encourage aggressive and frequent use of stops, summonses, and arrests pursuant to proactive strategies, they also increase the frequency of illegal stops, summonses, and arrests both in absolute numbers (because they conduct more) and in relative terms (because more of the additional stops, summonses, and arrests conducted are illegal).

The litigation against the New York City Police Department’s (NYPD’s) SQF program illustrates some of these issues. For many years, the NYPD claimed that its SQF and broken windows policing policies encouraged—except for occasional mistakes—only stops, frisks, and arrests that satisfied the Fourth Amendment’s reasonable suspicion requirement. Plaintiffs contended, by contrast, that the program resulted in many stops and frisks without adequate suspicion. In Floyd v. City of New York (959 F. Supp 2d 540 [2013]), the district court declared the program unconstitutional in part because it agreed with the plaintiffs, finding that many of the stops pursuant to the program violated the Fourth Amendment. According to the Floyd decision, the pressure to conduct stops as part of the program, when combined with inadequate training about the constitutional standard, led officers to engage in a practice of routine, unconstitutional stops that violated both the Fourth and Fourteenth Amendments.

DOJ has similarly contended that proactive policing that utilizes widespread stops and arrests for minor crimes causes constitutional violations. In its investigation of the New Orleans Police Department, for example, DOJ found that an organizational focus on arrests and statistical measures of productivity, in combination with poor training and policies, contributed to illegal stops, pat downs, and arrests (U.S. Department of Justice, 2011).

More recently, in its investigation into the Baltimore Police Department, DOJ found that the police department emphasized “an aggressive,

___________________

⁴ Proactive strategies that emphasize narrowly focused deterrence are unlikely to have this effect.

‘zero tolerance’ strategy that prioritized making large numbers of stops, searches, and arrests—often for misdemeanor street offenses like loitering and disorderly conduct.” This strategy was conducted “with minimal training and insufficient oversight from supervisors or through other accountability structures” (U.S. Department of Justice, 2016, p. 17). According to DOJ, the consequences were “repeated violations of [] constitutional and statutory rights, further eroding the community’s trust in the police” (U.S. Department of Justice, 2016, p. 5). Moreover, according to DOJ, the strategy had long-term effects. Even though Baltimore no longer formally uses a zero tolerance policing strategy, zero tolerance “continues to drive policing in certain Baltimore neighborhoods and leads to unconstitutional stops, searches, and arrests” (U.S. Department of Justice, 2016, p. 5).

Several scholars have argued that it is unlikely that any programmatic use of stops, frisks, and arrests could be effective in preventing crime and still survive proper constitutional scrutiny (Bellin, 2014; Meares, 2015). If so, then no department should adopt these strategies. That said, courts have not forced many departments to give up SQF, broken windows, or zero tolerance policing. And other scholars assume that a legal version of these strategies is possible, if departments aggressively use the legal authority to conduct stops or arrests (by encouraging officers to make all possible legal stops and arrests) and still train and supervise officers to avoid unconstitutional conduct (Harmon, 2012b). Either way, legal scholars conclude that deterrence-based strategies that employ aggressive stops, frisks, and arrests raise the prospect of increased constitutional violations, and the litigation surrounding these programs supports that conclusion.

The committee identified little systematic empirical research documenting either exactly how large the problem of unconstitutional behavior resulting from programmatic action is or exactly why it occurs. In finding the NYPD’s use of SQF unconstitutional, the court was strongly influenced by an expert report by Fagan and Macdonald (2012), which found that many of the stops apparently violated the Fourth Amendment. Analyzing the reasons officers provided for stops and frisks in the reports they were required to make when they conducted stops, the authors found that at least 7 percent of the stops conducted by the NYPD during the program lacked legal justification and another 24 percent lacked sufficiently detailed documentation to support a conclusion that the stop was legal.

Fagan and MacDonald’s report offers a rare window into the justifications for police action on a large scale. Still, it only provides limited evidence either about how many illegal stops occurred or, more importantly, whether any individual policy or menu of department policies caused them to take place. The findings by the court in Floyd and by DOJ are grounded in legal evidence, rather than social science evidence that satisfies the standards for attributing causation as used by this committee. Fagan and McDonald

(2012) do not undertake the mathematical exercise of statistically evaluating whether or not the 7 percent rate of illegal stops was larger, smaller, or indistinguishable from the rate of illegal stops that would have occurred in the absence of SQF.⁵ Similar problems arise with the evidence discussed in other court decisions and legal commentary. Whether there is “evidence” that a particular policy is associated with constitutional violations from a legal perspective is not the same issue as whether there is “evidence” that the policy causes constitutional violations in the sense of statistical causation. In acknowledging this distinction, the committee is not giving priority to either the legal or the social science definition of evidence. Rather, the purpose of credibly testing a previously defined null hypothesis against an alternative hypothesis (the social science assessment of causal connection) is, quite plainly, different from the goal of establishing a legal finding that an unconstitutional act occurred.

Further, even as a basis for describing how common illegal stops were during the period they studied, there are important limitations to the kind of data available to Fagan and McDonald (2012), namely, the self-reports of stops generated by NYPD officers. As the court pointed out in its decision, the reports on which the authors relied likely overestimated the legality of the stops conducted because officers may overstate the legal grounds for stops and may fail to document illegal stops more often than legal ones. In addition, the study intentionally estimated the legal sufficiency of the reports generously. As a result, many more of the NYPD’s stops under SQF could have been illegal. However, the court did not mention an alternative way in which the reports could understate legal stops. Fagan and McDonald (2012) considered merely whether the stops are “apparently unjustified,” a standard designed to capture those stops for which the reports indicated inadequate grounds for the purposes of the litigation. However, an officer’s conduct is legal if an objective basis for the stop exists, regardless of whether he or she provides adequate documentation of that basis. Since the criminal code is vast, and reasonable suspicion requires only “a minimal level of objective justification for making the stop” (Illinois v. Wardlow, 528 U.S. 119 [2000]), it is possible that some proportion of the stops found to be “apparently unjustified” by Fagan and McDonald (2012) could have had a legal basis that the officer had not stated in the documentation (Bellin, 2014). Given the weak scrutiny the NYPD gave to the reports, officers might have had little incentive to take care to include all of the grounds that justified the stops. Thus, it is difficult to know to what degree Fagan and McDonald (2012) estimates overstate or understate the proportion of SQF stops that were in fact illegal.

___________________

⁵ See the Chapter 1 section on “Assessing the Evidence” for additional discussion of the points made here about assessing evidence of causal relationships.

Yet Fagan and McDonald (2012) provide far more information about proactive policing and legality in the NYPD context than exists with respect to proactive policing efforts in most other cities.⁶ It is not easy, using existing data sources, for empirical researchers to count constitutional violations or develop meaningful proxies for them, and quantitative and qualitative criminological research often does not evaluate policing in terms that align with legal categories (Harmon, 2017). As a result, the limited empirical research about how proactive strategies change the frequency of constitutional violations does not provide a basis for concluding that proactive strategies either increase or reduce constitutional violations, according to the standards of causality used by the committee. The empirical evidence on whether SQF policies affect crime rates does not further clarify the issue (Meares, 2014).

Place-Based Strategies

Place-based strategies focus resources on locations where crime is concentrated in order to prevent and to respond more effectively to crime. To a substantial degree, the Fourth Amendment implications of a high-crime-area strategy depend on the kind of efforts police departments take to deter crime in the identified areas. If, for example, a department uses closed circuit television to deter crime at a particular street intersection or in a public park, it may do so without triggering Fourth Amendment scrutiny because that policing practice monitors individual movements only in public places and therefore does not constitute a search within the meaning of the Fourth Amendment. By contrast, if predictive policing or hot spots policing leads a department to engage in intensive stops, frisks, and arrests in a limited geographic area, these strategies will raise many of the same concerns as do the deterrence-based strategies discussed above. However, in addition to the Fourth Amendment issues raised by policing practices within specified areas, place-based strategies raise a distinctive set of Fourth Amendment issues by identifying specific microgeographic areas as locations of intensive recent or likely future criminal activity.

In Illinois v. Wardlow (2000), the U.S. Supreme Court held that unprovoked flight in a “high-crime area” can constitute reasonable suspicion justifying a pedestrian stop. Although lower courts have been slow to refine what constitutes a high crime area, a police department’s designation that a location is a hot spot is relevant to the legal analysis in which courts

___________________

⁶ The spreading use of body-worn cameras may provide the opportunity to study whether or not stops are constitutional, an approach that may yield better data on the proportion of SQFs that are illegal.

engage in making that determination.⁷ As a result, hot spots policing can have consequences for the legal rights of those who interact with the police in a hot spot location.

Under Illinois v. Wardlow, courts consider the fact that a suspect’s actions occurred in a “high-crime area” in evaluating whether the officer’s suspicion was sufficient to warrant a stop by the officer. Courts may permit stops in high crime areas on the basis of weaker suspicious behavioral cues by individual suspects than would be permissible in other areas because those cues can be taken to have additional meaning in a neighborhood with higher levels of criminal activity. As an extension of this logic, some scholars have suggested that courts in the future could include conclusions drawn from predictive policing technologies in assessing whether adequate suspicion exists to justify a traffic or pedestrian stop (Ferguson, 2012, p. 263). Thus, by lowering the amount of evidence of criminal activity (other than a department’s designation or prediction) necessary to make an officer’s intrusion constitutional, the department’s implementation of the policing strategy can now, and might further in the future, affect the scope of the rights of citizens to act free from interference. In doing so, all other things being equal, the strategy will also reduce the likelihood that an officer’s actions in conducting a stop will violate the Constitution because it is not justified by adequate suspicion.

A department’s characterization of an area as one of high crime can be consequential even when it is wrong. First, courts are unlikely to uncover or reveal a conflict between police assertions about an area and crime rates in that area. The vast majority of stops are never challenged legally because they result in no criminal charge, and a motion to suppress evidence in a criminal case is the primary mechanism by which the constitutionality of stops is contested. Moreover, in the absence of a clear legal standard about what constitutes a high-crime area, even when a stop is challenged, courts often defer to police assessments of the status of a neighborhood, sometimes without requiring specific evidence to support the designation. (Ferguson, 2011; Harris, 1998; see also, e.g., United States v. Smith, 594 F. 3d 530 [6th Cir. 2010]; United States v. Ruidiaz, 529 F. 3d 25 [1st Cir. 2008]). If such a designation is made without adequate basis, then the inferences a court draws about whether adequate suspicion exists within that area could be similarly unfounded.

Analogously, if predictive policing strategies that generate conclusions about the area are unreliable or nontransparent, they may produce predictions that are either unjustified or unfair and similarly lead to unsupported

___________________

⁷ Hot spots are often very small geographically, as small as a single intersection. Although courts have not clarified the size of a high-crime area within the meaning of Illinois v. Wardlow, cases seem to suggest that it may be substantially larger than a hot spot might be.

judgments that stops and frisks defended on the basis of those predictions are constitutional as a result. The discretion awarded to departments in designating hot spots may itself raise Fourteenth Amendment issues. Although not legally or empirically tested, ethnographic research has argued that the race and nationality of local residents and business people can play a role in labeling an area as “high crime” (Brunson and Miller, 2006; Chesluk, 2004; Muniz, 2012, 2014; Quillian and Pager, 2001; Sampson and Raudenbush, 2004).

Even if a court scrutinizes a department’s designation of a high-crime area and eventually concludes that the department erred in classifying the area as one of high crime at the time an officer conducted a stop, the department’s designation would make it reasonable for an individual officer to believe that it was a high crime area and therefore to believe that he had a greater basis for suspicion then he had in fact.⁸ Even if an officer lacks reasonable suspicion, making the stop unconstitutional, his reasonable mistake would change the consequences of his illegal act. Several remedies for constitutional violations, including the exclusionary rule and civil suits for damages under § 1983, are mostly unavailable against officer conduct that is unconstitutional but based on an officer’s reasonable mistake about the legal status of his actions (Herring v. United States, 555 U.S. 135 [2009]; Harlow v. Fitzgerald, 457 U.S. 800 [1982]; Mullenix v. Luna, 577 U.S. ___ [2015]). Assuming that the likelihood of civil damages or evidentiary exclusion shapes an officer’s incentives to ensure that reasonable suspicion exists before engaging in a stop, a proactive policing strategy in which high-crime areas are sometimes erroneously designated could cause additional, albeit unknowing, constitutional violations by officers.

The law governing high crime areas also has implications for the deterrence-oriented policing strategies discussed above. In place-based proactive policing, hot spots are designated in advance by departments. But individual police officers may equally use a history of crime in a location as part of the circumstances that justify a stop under Illinois v. Wardlow even when an agency has not previously labeled the area. Officers encouraged to engage in aggressive enforcement pursuant to deterrence-oriented proactive strategies need legal reasons to justify their activities, and the history of crime in the area often provides one (Fagan and Geller, 2015). Thus, for example, Fagan and Geller (2015) found in a study of 4.7 million stops by NYPD officers that police officers asserted that more than one-half of the stops took place in an area with a high incidence of crime. Weisburd, Telep, and Lawton (2014) showed that stops were indeed concentrated in specific loca-

___________________

⁸ As discussed in Chapter 7 of this report, social psychological processes of implicit bias and discrimination may affect policing in minority neighborhoods (see also Sampson, 2012, and Sampson and Raudenbush, 2004).

tions and that those high-SQF locations were strongly correlated with crime hot spots. Used in this way, deterrence strategies, combined with Illinois v. Wardlow, can have significant distributional consequences, exposing individuals to additional scrutiny because of perceived or actual neighborhood characteristics, which often correlate with race and economic status. The committee did not find causal empirical research to date that adequately engaged with this question, in spite of the psychological, ethnographic, and correlational social science literature documenting this phenomenon.

Third Party Policing

Third party policing leverages the actions of third parties in deterring and reducing the opportunities for targeted offenders or criminal conduct. For example, as described in Box 2-3 (see Chapter 2), as a means to indirectly control drug and disorder problems the Oakland, California, Beat Health Program focused on civil remedies for addressing conditions of physical decay and property management problems of specific commercial establishments, private residences, and rental properties. As this program illustrates, departments often take advantage of existing civil laws and regulations in implementing third party policing because these laws provide much of the leverage to demand third-party participation in crime prevention and control. Nevertheless, departments can also utilize third parties to prevent or reveal crime in another way, one that three aspects of Fourth Amendment doctrine facilitate: officers may use information obtained through third parties that would otherwise be unavailable without establishing individualized suspicion or obtaining a search warrant.

First, the Fourth Amendment does not apply to information that a person voluntarily provides to a third party when the third party makes that information available to the government (United States v. White, 401 U.S. 745 [1971]; United States v. Miller, 425 U.S. 435 [1976]; Smith v. Maryland, 442 U.S. 735 [1979]). Thus, when police officers secure information about individuals from third parties, their conduct is not subject to the Fourth Amendment protection, whereas efforts to obtain the information directly from the suspect may involve protected searches and seizures.

Second, when a third party shares or reasonably appears to share common authority over a location or over property, he or she may consent to a search by government actors (United States v. Matlock, 415 U.S. 164 [1974]). Although that consent is not valid against an objecting co-occupant who remains present during the search (Illinois v. Rodriguez, 497 U.S. 177 [1990]; Georgia v. Randolph, 547 U.S. 103 [2006]), it otherwise has the potential to permit police access to locations unavailable without the cooperation of the third party.

Third, the Fourth Amendment applies only to government conduct, and

any exposure of private information usually negates the argument that an individual has a reasonable expectation of privacy against the government. This means that private searches by third parties can limit the applicability of the Fourth Amendment to subsequent searches made by law enforcement of the same locations or the same information. This third avenue is illustrated by the U.S. Supreme Court decision in United States v. Jacobsen (466 U.S. 109 [1984]). In that case, Federal Express employees examined a package damaged during transport and discovered a white powdery substance they suspected was contraband. They reassembled the package and called the Drug Enforcement Administration (DEA). When the DEA agents arrived, they reopened the package and subjected the powder to a field chemical test that indicated the substance was cocaine. In upholding the government’s use of the cocaine in a criminal case against the package’s recipients, the U.S. Supreme Court held that since private actors had already opened the package, the government’s re-inspection of the contents uncovered nothing new and therefore did not constitute a search within the meaning of the Fourth Amendment. In its decision, the court noted that even an illegal private search can undermine the reasonableness of an expectation of privacy with respect to the information discovered (United States v. Jacobsen, 1984).

There are limitations on the use of private searches by the government. Most notably, if a private actor is an agent of the state or if government actors are deeply entangled in private searches, the search he or she conducts may be a public rather than private one, and therefore fall within the scope of the Fourth Amendment (Coolidge v. New Hampshire, 403 U.S. 443 [1971]). Similarly, if a police officer or department compels, encourages, endorses, or participates in a search or seizure by a third party, the action may be subject to Fourth Amendment protections (Skinner v. Railway Lab. Execs. Ass’n, 489 U.S. 602 [1989]). Thus, while a proactive strategy that takes advantage of third party access to private information would likely permit officers to gather evidence without triggering Fourth Amendment scrutiny for that evidence gathering, a proactive strategy that induces searches by private parties may be subject to constitutional regulation.

Even with these limitations on private searches, it might be said that in each of the circumstances described above—voluntary disclosure, consent by a third party, and involuntary exposure—proactive policing that leverages the cooperation of private third parties may narrow the applicability of Fourth Amendment protection to police efforts to obtain information. Ceteris paribus, officers who are able to obtain information from third parties (and thus without searches and seizures), are likely to conduct fewer searches and seizures and therefore have less opportunity to violate the Fourth Amendment. In this way, third party policing may reduce constitutional violations. At the same time, if third party policing gives police offi-

cers an incentive to strongly encourage private searches, it may lead to more frequent violations of the rules limiting the use of private searches by the government. In addition, to the degree that proactive policing encourages information gathering outside the scope of the Fourth Amendment, it may increase intrusions on privacy that are unregulated by Fourth Amendment law in ways that raise concerns about private invasion, even if the intrusions comply with constitutional law. The committee knows of no empirical literature assessing these risks.

As the Beat Health example suggests, police departments can also leverage searches designed to enforce civil regulatory laws, such as health and safety codes, building codes, and environmental regulations. Although administrative searches are governed by the Fourth Amendment, the Supreme Court has not usually demanded individualized suspicion or warrants for them (Camara v. Municipal Court, 387 U.S. 523 [1967]). Instead, the Court’s doctrines permit civil government inspections, such as housing code inspections, so long as they are reasonable, which often requires nothing more than that reasonable legislative or administrative standards govern them. This is therefore another mechanism by which proactive policing may allow police to avoid standards governing individualized suspicion that might otherwise limit access to the information in the absence of an administrative search regime. As with third-party searches, although such a strategy might be construed to limit protection for privacy, it also reduces the opportunities for constitutional violations against the same individuals. Under existing law, police officers may attend, or use information discovered during, such searches when they are carried out by other government officers, or they may conduct administrative searches themselves, consistent with the Fourth Amendment (New York v. Burger, 482 U.S. 691 [1987]), so long as the primary motive for the search is not to uncover ordinary criminal wrongdoing (Indianapolis v Edmond, 531 U.S. 32 [2000]).

EQUAL PROTECTION AND STATUTES PROHIBITING DISCRIMINATION

Legal Overview

Unlike the Fourth Amendment, the Equal Protection Clause of the Fourteenth Amendment applies to all police activities, including policy decisions by departments to investigate suspects or to search or seize them. It guarantees equal and impartial treatment by government actors under the law. A policy or police action may violate the Equal Protection Clause either because it expressly singles out individuals for disfavored treatment on the basis of their race or other impermissible classification or because, though facially neutral, the policy is selectively enforced against members of one

race or other impermissible classification in an intentionally discriminatory manner.

However, not all policies involving racial classifications or creating racial disparities in investigation or enforcement violate the Equal Protection Clause. Laws or policies that draw express racial or ethnic classifications among citizens do not violate the Equal Protection Clause if they are narrowly tailored to serve a compelling state interest (Wayte v. United States, 470 U.S. 598 [1985]). This test, known as “strict scrutiny,” is difficult to pass. Facially neutral laws and policies that are selectively enforced in a discriminatory manner violate the Equal Protection Clause only if they are also motivated by a discriminatory purpose (Washington v. Davis, 426 U.S. 229 [1976]).

Proving discriminatory effect requires establishing that an individual received less favorable treatment because of his race or other classification. Plaintiffs often establish this disfavored treatment with statistical evidence. Chapter 7 considers further the difficulties of establishing unfavorable treatment, including the difficulties of establishing the proper comparison populations. As with the Fourth Amendment, however, the legal concept of causation in Equal Protection law does not necessarily satisfy the criteria social scientists use to identify causal relationships. For instance, federal courts are divided as to whether plaintiffs claiming that police officers selectively enforced the law against them because of their race must demonstrate that “similarly situated individuals of a different race” did not have the law enforced against them in order to demonstrate discriminatory effect. This standard, which is always required for plaintiffs attempting to establish selective prosecution (United States v. Armstrong, 517 U.S. 456 [1996]; United States v. Davis, 793 F.3d 712 [7th Cir. 2015]; United States v. Mason, 774 F.3d 824 [4th Cir. 2014]), makes selective enforcement by the police exceptionally difficult to establish (United States v. Whitfield, 649 F. App’x 192 [3d Cir. 2016]).

In contrast to some of the historical practices discussed in Chapter 7 of this report, most policing policies today do not expressly target racial or ethnic groups, so most Equal Protection challenges require proving discriminatory purpose as well as discriminatory effect. The concept of discriminatory intent in Equal Protection law is distinct from the concepts of racial bias used in the psychological literature and discussed in Chapter 7 of this report. Proving discriminatory purpose requires showing (1) that the government intended to treat an individual unequally because of his or her classification, and (2) that it acted because of the harmful effect on a chosen group, not merely in spite of that effect. In other contexts, Equal Protection strictly scrutinizes government conduct even if the plaintiff does not prove that the desire to treat a group unequally was the only purpose guiding an activity, so long as it is demonstrated to be one motivating factor behind the

harm. However, some lower courts have refused to apply law enforcement on race unless the decision was based solely on race (e.g., United States v. Travis, 62 F. 3d 170, 174 [6th Cir. 1995].⁹ Discriminatory intent can be proved through direct evidence, such as admissions by a policy maker or officer, or circumstantially, using statistical evidence of discrimination to show that discriminatory intent likely exists (Washington v. Davis, 1976), including the kind of statistical evidence discussed in Chapter 7.

Though the legal concept of discriminatory intent is distinct, efforts to prove that intent in lawsuits are plagued by many of the same evidentiary challenges, discussed in Chapter 7, that affect social scientists’ efforts to establish the reasons for racial disparities. In addition, assessing the legal adequacy of evidence of discriminatory intent is complicated both by the social and historical context in which law enforcement operates also discussed in that chapter and by the subtle and nonobvious ways racial bias and animus may operate in society. For instance, symbolic racism, as defined in Sears (1988), involves the belief that prejudice against Black people is no longer a problem in U.S. society today, that the overrepresentation of Black Americans in low-income, low-educated, and high-crime groups is primarily due to their own personal shortcomings, and that Black people in general demand too much from society at large and have also “gotten more than they deserve.” A core part of symbolic racism, as described by Sears (1988), is therefore the belief that if a Black person received less favorable treatment, it was likely because they objectively deserved less favorable treatment. Holding such a view would presumably influence whether one believed that indirect evidence established the discriminatory purpose necessary to prove an Equal Protection violation.

In addition to the Equal Protection Clause, federal statutes, including Title VI of the Civil Rights Act of 1964 and the Omnibus Crime Control and Safe Street Act of 1968, also prohibit discrimination by police departments that receive federal funding. These statutes provide protection against discrimination that significantly overlaps with Equal Protection law, but they also sometimes permit liability for unintentional discrimination when Equal Protection does not (28 CFR § 42.104(b)(2); 28 CFR § 42.203).

___________________

⁹ This view finds some support in the Supreme Court’s Fourth Amendment jurisprudence. The Court has suggested that seizures in the context of an immigration checkpoint based solely on ethnicity are arbitrary and therefore unreasonable under the Fourth Amendment (United States v. Martinez-Fuerte, 428 U.S. 543, 554 [1976]), but that seizures largely on the basis of ethnicity may be permissible at least where ethnicity is relevant to the law enforcement interest at stake (United States v. Brignoni-Ponce, 422 U.S. 873 [1975]). Nevertheless, the Supreme Court has also indicated that “the constitutional basis for objecting to intentionally discriminatory application of law is the Equal Protection Clause not the Fourth Amendment” (Whren v. United States, 517 U.S. 806, 813 [1996]), raising questions about the relevance of this analysis to the Equal Protection context.

They also allow federal agencies to address noncompliance by terminating federal financial assistance to the offending agency, a remedy unavailable under Equal Protection law.

Deterrence-Oriented Proactive Strategies

Proactive policing strategies that use frequent stops, frisks, and arrests to prevent future crime often raise Equal Protection concerns as well as Fourth Amendment issues. Many critics have argued that such strategies cause unwarranted racial disparities, and both the district court’s decision in Floyd and DOJ’s analyses in its pattern-and-practice investigations in New Orleans and Baltimore found that the proactive policing strategies at issue caused discriminatory policing in violation of the Equal Protection Clause.

For example, in Floyd, Judge Scheindlin found that, in carrying out SQF, the NYPD violated the Equal Protection Clause by disproportionately and discriminatorily stopping non-Whites. Specifically, she noted that officers likely targeted Blacks for stops based on a lesser degree of objectively founded suspicion than they applied in stopping Whites, and officers subjected them to different treatment during stops, including more frequent use of force, despite the fact that Whites who were stopped were more likely to be found with weapons or contraband (Floyd v. City of New York, 2013). She also found that the NYPD had an unwritten policy of targeting “the right people” in carrying out SQF, which encouraged subjecting young Black and Latino men to heightened police enforcement on the basis of their race, and that the department had shown deliberate indifference in the face of evidence that the program was carried out in a discriminatory manner.

In Baltimore, DOJ linked the Baltimore Police Department’s (BPD’s) zero tolerance policy—which was implemented in the early 2000s and included frequent stops, searches, and arrests—to “overwhelming statistical evidence of racial disparities in BPD’s stops, searches, and arrests,” in violation of Title VI and the Safe Streets Act (U.S. Department of Justice, 2016, p. 48). DOJ concluded that the evidence was sufficient to establish discriminatory impact under the Equal Protection Clause. DOJ also found evidence suggesting that the discrimination against Blacks was intentional because of the magnitude of the statistical relationship between race and stops, because the proactive strategy focused on Blacks and Black neighborhoods, because of statements by officers and supervisors indicating that the program was being carried out in a discriminatory fashion, and because of the department’s failure to act in the face of evidence of discrimination. For example, one supervisor allegedly instructed officers to carry out the zero tolerance strategy by arresting “all the Black hoodies” in a neighborhood (U.S. Department of Justice, 2016, p. 66). In the course of DOJ’s investi-

gation, at least some top BPD officials shared the view that its proactive policing strategy had discriminatory effects. One told DOJ that “stop and frisk killed the hopes and dreams of entire communities” (U.S. Department of Justice, p. 63). DOJ contended that even after zero tolerance was no longer the formal policy of the police department, supervisors within the department continued to implement this form of proactive policing, with its discriminatory and other consequences.¹⁰ Other DOJ and private civil suits resulting in settlements have alleged that the frequent use of stops, frisks, and arrests in other cities has also violated the Equal Protection Clause but have drawn less express connection between the enforcement practices and proactive policing strategies.

More broadly, concerns about discrimination in proactive policing are often framed as concerns about racial profiling. Racial profiling usually refers to police decisions to engage in vehicle or pedestrian stops, searches, or arrests or to take other law enforcement actions based at least in part on an individual’s race, ethnicity, religion, or national origin, outside of the context in which officers target an individual because he satisfies a specific description of a criminal suspect or other person of interest. For instance, officers implementing a deterrence-oriented proactive strategy might use race as a factor in choosing which people to stop, frisk, or arrest because they believe that the targeted race is overrepresented in the criminal population the strategy is intended to deter, and they would thereby engage in racial profiling. Even if their belief were accurate and hit rates or deterrence could be improved using race as a criterion, this use of race may not pass constitutional scrutiny. The overwhelming number of people selected would still likely be innocent in the sense of needing no deterrence from the targeted conduct; those selected on the basis of their race would suffer additional harm from being selected for this reason; and courts applying strict scrutiny would be unlikely to find this use of race “narrowly tailored” to serve a “compelling state interest.”

Although legal claims about unwarranted racial disparities have focused on stops, frisks, and arrests, other kinds of intensive enforcement resulting from proactive policing may also raise questions about disparate impacts, including third party enforcement of civil regulatory codes, specifically “nuisance violations.” These violations, which are filed against landlords whose tenants contact 911 frequently, require the landlords to take steps to reduce the frequency of these calls. In practice, the steps taken frequently involve evicting tenants who request police assistance by calling 911. Desmond and Valdez (2013) documented a positive correlation between the use of third-party enforcement and the fraction of neighborhood

___________________

¹⁰ Legal claims that proactive policing led to discrimination often remain unadjudicated either because procedural barriers bar suit or the parties settle, making a court ruling unnecessary.

residents who are Black. Similar to Fagan and MacDonald’s (2012) analysis of the geographic incidence of SQF in New York, documenting such a pattern may constitute legal evidence of Fourteenth Amendment violations in the use of third-party enforcement.¹¹

The difficulties of assessing and understanding racial disparities and racial bias are discussed further in Chapter 7. It remains an open question whether any tendency that proactive policing strategies have to cause Fourth and Fourteenth Amendment violations are linked, though some theorists suggest that such linkage is likely (Meares, 2015; Bellin, 2014). To be clear, this is not due to mixed or null conclusions of credible evaluations of the causal impact of proactive policing strategies on the incidence of Fourth or Fourteenth Amendment violations; it is because the empirical social science literature that could establish such causal effects has not adequately engaged with the question.

Predictive Policing Strategies

As noted in Chapter 2, predictive policing strategies seek to anticipate, prevent, and respond more effectively to crime by collecting information and identifying patterns in aggregated data about past crime and other information. To the degree that these predictions focus on individuals or groups who may commit or fall victim to crime, rather than to places where crimes may be committed, they could raise Equal Protection concerns. First, predictive strategies or the law enforcement interventions based on the resulting predictions may be implemented by departments with discriminatory effect and intent. Doing so would violate the Equal Protection Clause, just as implementing SQF or broken windows policing with discriminatory effect and intent violates the law. Second, these strategies are sometimes directed intentionally at members of a particular religion or national origin and therefore contain an express classification that singles out members for unfavorable treatment. This raises a distinctive kind of Equal Protection claim, one only touched upon above.

When predictive policing is targeted at members of a religion or national origin, they are likely to be subject to heightened scrutiny, requiring that the government prove a strong justification between the governmental interest and the means used to achieve it. In Hassan v. City of New York (2015), for example, plaintiffs alleged that the NYPD adopted a long-term

___________________

¹¹Desmond and Valdez (2013) do not do a counterfactual analysis of whether or not the rate at which Black residents were denied emergency response service or evicted changed as a result of the adoption of third-party enforcement. As a result, their study does not address the question of whether third party policing in this instance exacerbated racial disparities in victimization or simply relabeled an existing phenomenon.

program of extensive surveillance and investigation of Muslim individuals, businesses, and institutions after the terrorist attacks of September 11, 2001. Among other claims, the plaintiffs contended that this selective investigation violated Equal Protection law. Though this claim has not yet been fully litigated, the U.S. Court of Appeals for the Third Circuit permitted the case to go forward for discovery and trial because the allegations, if true, could establish a constitutional violation, even if the NYPD was motivated by a legitimate law enforcement purpose in establishing the program. Specifically, the Third Circuit panel ruled that allegations of religious discrimination are subject to heightened Equal Protection scrutiny, even if the program containing them was motivated by national security and public safety concerns.

Although the program challenged in Hassan would not fall within the bounds of proactive policing as described in this report because it sought to uncover rather than prevent criminal activity, it raises the same legal concerns as would a proactive strategy that is similarly directed at members of a particular religion or national origin and is thus illustrative. Similar legal analysis might have applied, for example, to the Los Angeles Police Department’s 2007 plan to identify and map Muslim communities in Los Angeles to help them avoid the influence of extremist elements that might lead to terrorism had criticism not led the department to abandon the plan (Roush, 2012).

EMPIRICAL EVIDENCE ON PROACTIVE POLICING AND ILLEGAL POLICE BEHAVIOR

There are relatively few empirical studies that credibly examine whether or not proactive policing is causally related either to police behavior that is likely to raise constitutional challenges or to legal findings of constitutional violations. Two challenges make such research exceptionally difficult. First, researchers have limited data about the kinds of police conduct that often raise constitutional challenges. Unlike serious crime and arrest rates, there is little nationwide data collection on many kinds of police behaviors, including stops, searches, and uses of force, that may trigger a constitutional challenge. Individual agencies often have different standards for how police conduct is reported internally, including, for example, different standards for definitions of what constitutes force (Alpert, 2016), and for whether data concerning police conduct is available for research. To the extent that many proactive policies may alter the legality of police behavior and that there is value in social science evaluation of this possibility, systematic and standardized collection of data on relevant police outcomes is necessary.

Second, even with such data, constitutional violations are difficult for researchers to define and to measure. Such violations require fact-specific

analysis and legal judgments, and different observers are likely to come to differing conclusions about whether a violation occurred (Harmon, 2017). Unless and until a court has given a final judgment on the question, there is no authoritative basis for concluding that a researcher’s determination about whether a constitutional violation occurred is accurate. Nor do easy proxies for legality exist. For example, citizen complaint rates might vary for reasons independent of the constitutional violations that might spur them, including agency-specific methods of taking (or resisting) complaints. Lawsuit rates might vary with the strength of the local bar and with settlement practices (Harmon, 2017). And, in the extreme, it is possible that the majority of residents could be very satisfied with a department that regularly violated the constitutional rights of a small minority of the population, making community satisfaction surveys a similarly weak measure.

The committee conducted a systematic search of peer-reviewed publications examining the relationship between proactive policing and the legality of police officer actions. The committee found notably less research on the impact of proactive policing strategies on legality than it found on the implications of proactive policing for crime control or community satisfaction. The few studies that were found generally assessed satisfaction with the police or perceptions of police legitimacy; this literature is reviewed systematically in Chapters 5 and 6 of this report as part of the committee’s examination of community impacts.

Fagan and colleagues (2010) used a modified pre-post design to attempt to identify the impact of broken windows policing on officer SQF behavior in New York City. To the extent that this deterrence-oriented strategy led to unequal treatment of people of different races or ethnicities, this could be interpreted as evidence that SQF led to an increase in violations of the Equal Protection Clause. The authors found a sharp increase in stops of Blacks and Latinos in the “late” period of broken windows policing relative to the early period, from 27 and 15, respectively, per 1,000 people to 131 and 64 per 1,000 people. For White people, the comparable change was from 4 per 1,000 in the early period to 18 per 1,000 in the late period. These findings are consistent with, but by no means evidence of, the proactive SQF policy causing a large increase in illegal racial targeting by the NYPD. That said, comparing officer behavior in New York City to stop behavior in a different city, or making a comparison with a “pre” period that is not defined by the low level of stops, would make this evidence more convincing.

In essence, the calculation by Fagan and colleagues (2010) assumes that, in the absence of the broken windows policing policy in New York City, the rate at which Blacks or Hispanics would be stopped by the NYPD would have been constant over time. Potentially alternate explanations include demographic change, variation in the taste of residents and police

officers, or changes in recording practices. Fagan and colleagues, (2010) further demonstrated that the percentage of neighborhood residents who are Black was a strong predictor of the number of stops, conditional on crime rates, but they did not explore whether the increased use of broken windows policing had changed the relationship between the racial composition of a neighborhood and the frequency with which police make (potentially illegal) stops.

LEGAL MECHANISMS FOR CHALLENGING PROACTIVE POLICING

Police departments and officers have myriad complex reasons for following the law, including the costs and consequences of litigation challenging the constitutionality of police conduct. As a general matter, departments may, in part, determine if and how proactive strategies are employed in response to their perceptions about this litigation and the remedies likely to be imposed as a result. However, given the substantial limitations on constitutional remedies for police misconduct in the context of proactive policing and the limited information departments collect about lawsuits and their connection to police practices, these legal consequences may provide only limited incentives for departments and officers with respect to proactive strategies. To the degree this occurs, the law may not substantially discourage even those proactive strategies that result in provable constitutional violations.

Several kinds of legal actions can be brought against police conduct. Individuals whose rights have been violated by the police can bring civil suits under federal and state law for damages, for a declaration of the rights of the parties, or for a command to adopt particular reforms. The federal government (and occasionally states) can also bring civil suits against police departments who have engaged in a “pattern or practice” of rights violations, seeking reform.¹² In addition, criminal defendants whose rights have been violated can challenge police conduct by moving to exclude illegally obtained evidence from criminal trials in which the government would

___________________

¹² The use of the term “pattern” by DOJ also diverges from the social science meaning of the term. Identifying a pattern in, say, use of force, in social science research would imply identifying some measure (e.g., time, officer, or place) that was correlated with that variable. Claiming to have identified a correlation would require statistically distinguishing the estimated correlation from zero, which involves mathematical calculations. However, with regard to legal findings of a pattern, the U.S. Court of Appeals for the Fifth Circuit has found that “The number of [violations]...is not determinative. . . . In any event, no mathematical formula is workable, nor was any intended” (United States v. Peachtree Tenth Corp., 437 F.2d 221, 227 [5th Cir. 1971], cited in June 28, 2013, DOJ Findings Letter regarding the Investigation of the Los Angeles County Sheriff’s Department Stations in Antelope Valley).

introduce it, and federal and state prosecutors can bring criminal charges against police officers for their actions.

Title 42 U.S.C. § 1983 was passed in its original form as part of the Civil Rights Act of 1871. It permits a civil suit against any person, agency, or municipality that, while acting under color of law, deprives another of his or her constitutional rights; and it is frequently used to challenge police practices (Monell v. Dept. of Soc. Serv., 436 U.S. 658 [1978]). When successful, these suits typically result in settlements or other judgments against individuals and municipalities for monetary damages, though they can also lead to equitable relief in the form of a court declaration that a policy or act is unconstitutional or a command to an agency either not to engage in some conduct or to carry out particular reforms to prevent future constitutional violations.

Under 42 U.S.C. § 1983, governments act by making policies or decisions or by permitting practices that are so persistent and widespread that they function as policy or law. A municipality or police department can only be sued under section 1983 if a departmental policy, custom, or practice causes—in the sense of being the moving force behind—a constitutional violation by an officer (Monell v. Dept. of Soc. Serv., 1978). Thus, a city will only be directly liable for harms associated with a proactive policing strategy if the policies, decisions, or practices that implement that strategy cause constitutional injury (Monell v. Dept. of Soc. Serv., 1978). For example, a federal district court found New York City liable for the NYPD’s program of aggressively stopping, questioning, and frisking suspects because the program resulted in a widespread practice, amounting to a policy, of conducting unconstitutional stops and frisks and targeting racially defined groups in a disproportionate and discriminatory manner (Floyd v. City of New York, 2013). The court did not bar the proactive goal of deterring weapons possession, nor the practice of using stops and frisks aggressively to achieve it, so long as the policy as implemented did not cause constitutional violations or otherwise violate the Equal Protection Clause.

Plaintiffs can also bring civil suits against individual officers for violating clearly established constitutional rights while acting under color of law (42 U.S.C. § 1983). Where prior law makes clear that an officer’s conduct under the specific circumstances violates the Constitution, the officer can be liable for the injuries that result. If an officer violates a right that is not clearly established under existing law, he is entitled to qualified immunity, which protects him against being sued or held liable for his actions (Pearson v. Callahan, 555 U.S. 223 [2009]). A right is not clearly established unless preexisting court decisions squarely govern the question, such that every reasonable officer would have understood that the particular conduct violated the law. For example, in one recent U.S. Supreme Court case, Mullenix v. Luna (2015), the court held that existing precedent had not put “beyond

debate” the conclusion that an officer who shot “a reportedly intoxicated fugitive, set on avoiding capture through high-speed vehicular flight, who twice during his flight had threatened to shoot police officers, and who was moments away from encountering an officer” had acted unreasonably (Mullenix v. Luna, 2015). Thus, it found the officer entitled to qualified immunity, shielding him from liability and suit.

While suits against officers might seem less relevant to influencing departmental decision making than suits against municipalities or police departments, they may have similar effects on policy. Even when an individual officer(s) is named as a defendant and not the department or municipality, municipalities almost inevitably indemnify officers, meaning that they pay the costs of damages actions against them. Thus, municipalities bear the financial burden for judgments for damages in section 1983 suits even when the judgments operate formally only against individual officers (Schwartz, 2014). Given indemnification, civil judgments could, at least in theory, deter cities from adopting policies that give rise to unconstitutional conduct that might lead to liability, and cities that pay frequent civil judgments might be encouraged to reform strategies that tend to produce constitutional violations. However, municipalities only infrequently collect and analyze information about civil suits or the police practices that give rise to them. In departments that do not use the information provided by civil suits to manage their liability risk, damages actions may have limited effect on decision making about continuing proactive strategies that lead to such suits (Schwartz, 2010).

Sometimes monetary damages are inadequate to repair an injury to a plaintiff. In those circumstances, private plaintiffs may seek equitable relief instead. Equitable relief can include a judicial order to do something, an order not to do something, or a declaration about the rights of the parties, among other remedies. Though equitable relief is less common than damages, it can operate powerfully on the government agency against which it is levied. If damage actions incentivize reform, it is by making reform a cost-effective alternative to costly future judgments. By contrast, equitable relief can mandate immediate policy change and imposes stark legal and reputational consequences for those who refuse to comply.

While private suits for equitable relief have played an historic role in efforts to promote civil rights in many other arenas, including housing, school desegregation, and prison conditions, the U.S. Supreme Court has established notable obstacles to civil lawsuits for equitable relief against police departments, mostly importantly in the form of limits on standing (City of Los Angeles v. Lyons, 461 U.S. 95 [1983]; Rizzo v. Goode, 423 U.S. 362 [1976]). In general, unlike a plaintiff seeking only damages, a plaintiff asking for forward-looking relief must demonstrate that there is a “real and immediate threat” of future injury. In City of Los Angeles v. Lyons (1983),

the U.S. Supreme Court applied this standard with special vigor to plaintiffs seeking injunctive relief against a police department, holding that it cannot be satisfied by demonstrating a past injury by the police or by speculation that the police might injure the same plaintiff. Thus, Adolph Lyons, who had sought to challenge the chokehold policy of the Los Angeles Police Department after he had been choked to unconsciousness during a traffic stop, did not have standing because he could not show that he would likely be stopped again, and then either that he would illegally resist, resulting in a chokehold, or that officers would subject him to a chokehold without provocation. Although the “real and immediate threat” standard applies to all plaintiffs seeking injunctive relief, given the vagaries of police–citizen interactions, the standard has proven to be an especially high bar for plaintiffs challenging police policies.

Though Lyons has stymied many suits against departments, plaintiffs challenging proactive policing may have a somewhat easier time bringing equitable relief claims than plaintiffs challenging traditional policing methods. The same qualities that make preventative policing policies proactive—their forward looking, strategic focus—can make the threat of future injury more “real and immediate.” For instance, courts are more likely to find standing for equitable challenges under Lyons when a policy targets relatively innocent or common conduct—as proactive policing sometimes does when it encourages stops based on minimal suspicion or arrests for very minor offenses—because the risk to the plaintiff of being targeted under such a policy is less dependent on his own future wrongdoing and therefore less speculative (United States v. Chang, Civ. Action No 02-2010, Memorandum Op., D.D.C. [Sep. 9, 2010]; National Congress for Puerto Rican Rights v. City of New York, F. Supp. 2d 154 [S.D.N.Y. 1999]). Similarly, plaintiffs are likely to have an easier time showing that they are likely to be injured in the future when a department engages in the challenged conduct frequently or when the policy targets a subpopulation of which they are a part (United States v. Chang, 2010; National Congress for Puerto Rican Rights v. City of New York, 1999). Strategies that depend on widespread use of stops, frisks, and arrests, like SQF, broken windows, and zero tolerance, often encourage a large volume of police-citizen encounters and are often accused of disproportionately focusing police action against particular racial or ethnic groups. They therefore may make it more likely that the burdens of the policy will fall on a particular plaintiff attempting to establish standing (Floyd v. City of New York, 2013; U.S. Department of Justice, 2016). Strategies that expressly concentrate resources on identifiable activities, places, or defendants, such as problem-oriented, hot spots, and focused deterrence policing, are similarly more likely than are general patrol strategies to create a realistic risk that plaintiffs who fall within those parameters will be subject to the allegedly unconstitutional police intervention.

Section 1983 suits are the primary method for challenging the consequences of proactive strategies in court, but they are not the only one. Title 42 U.S.C. § 14141 permits DOJ to bring suit for equitable relief against police departments that engage in a pattern or practice of constitutional violations. DOJ is not subject to the standing requirements of City of Los Angeles v. Lyons and therefore can bring cases seeking forward-looking remedies that could not be brought by private individuals.

In most of the early efforts to pursue pattern and practice suits against police departments, DOJ focused on policing acts rather than on strategies. However, in some recent suits DOJ has expressly linked proactive policing strategies to constitutional violations. Most recently, as noted above, DOJ found that zero tolerance policing as implemented by the BPD caused a pattern of constitutional violations (U.S. Department of Justice, 2016). “Pattern-and-practice” suits are usually settled through consent decrees, in which the city and DOJ agree to reforms the department will adopt in order to prevent future constitutional violations. In these decrees, DOJ sometimes expressly promotes one proactive strategy, community-oriented policing, as well as other mechanisms for encouraging transparency, accountability, and community participation in determining policing policy. DOJ can similarly demand that departments not engage in proactive strategies it views as linked to violations. To the extent that police departments look to prior consent decrees for information on what activities might get them sued, this linkage could discourage some departments from adopting zero tolerance policing or similar proactive strategies that DOJ has previously described as facilitating constitutional violations.

Other legal remedies for police misconduct, such as the exclusionary rule, are much less likely to affect police department use of proactive policing strategies. The exclusionary rule prohibits the use in any criminal trial of evidence obtained unconstitutionally, and it is often labeled the primary remedy for deterring Fourth Amendment violations (Utah v. Strieff, 579 U.S. ___ [2016]). However, the exclusionary rule cannot deter constitutional violations that do not produce evidence or do not result in a criminal prosecution of the individual whose rights were violated (Terry v. Ohio, 1968; Rakas v. Illinois, 439 U.S. 128 [1978]). Many proactive strategies do not emphasize prosecuting criminal conduct, or if they do, they focus on minor crimes that may not involve physical evidence or extensive motions practice. Even beyond these limitations, U.S. Supreme Court cases have notably limited the circumstances in which the exclusionary rule applies (Utah v. Strieff, 2016; Herring v. United States, 2009). Thus, even if a proactive strategy leads to illegally obtaining evidence and introducing it in criminal trials, the expected value of the strategy is unlikely to be undermined significantly by the increasingly remote threat of evidentiary exclusion. Finally, as with civil rights lawsuits, departments often do not

gather sufficient information about evidentiary suppression to effectively internalize the expected costs of exclusion for policies that might trigger the exclusionary rule.

Criminal prosecutions of police officers are similarly unlikely to notably affect proactive policing, both because such prosecutions are relatively rare and because the costs of those prosecutions are borne heavily by the individual officers who are prosecuted, so are far less likely to be internalized by departments in a manner that prompts reform (Harmon, 2012a).

OTHER LEGAL STANDARDS AND VALUES

In addition to the Fourth and Fourteenth Amendments, a wide variety of federal, state, and local statutes constrain proactive policing. The federal Electronic Communications Privacy Act, for example, which incorporates the federal Wiretap Act, Stored Communications Act, and Pen Register Act, restricts how police may gather private information and how they collect information from third parties, such as Internet or cell service providers. State constitutions and statutes, along with local charters and ordinances, determine how police executives are hired and fired and how budgets are formulated. They determine when police are disciplined and what kinds of judicial or administrative review disciplinary mechanisms receive. And they determine what kinds of information about police activities are collected by departments and made available to the public. As these examples suggest, the entirety of law that could influence proactive strategies is extensive and diverse and cannot be easily summarized.

Even when proactive policing does not violate constitutional law or this array of additional legal constraints, or does so in unenforceable ways, proactive strategies sometimes violate deeply held legal values, such as privacy, bodily integrity, equality, autonomy, accountability, and transparency. Threats to these values may subject policing strategies to political responses that can, in turn, push municipalities and states to more aggressively impose additional regulation on policing. For instance, the Maryland State Police engaged in an extensive and intrusive undercover operation to investigate political activists in 2005 and 2006, which led to public outrage when it was revealed in 2008. In response to the public reaction and an investigation of the surveillance program, the Maryland General Assembly passed the Freedom of Association and Assembly Protection Act of 2009. The law sets additional controls over police surveillance activities, even when those activities comply with the Constitution, and requires local law enforcement agencies in Maryland to adopt policies implementing those controls (Roush, 2012). Similarly, after concerns about privacy and accountability were raised about the city’s use of drones and video cameras, Seattle passed

an ordinance setting up new political checks on law enforcement acquisition and use of surveillance equipment.¹³ In light of potential legislative responses to concerns that proactive policing strategies violate traditional legal values, even when the strategies comply with existing law, some of these concerns are considered here.

For example, even assuming that SQF, broken windows, and zero tolerance policing can comply with Equal Protection and antidiscrimination law, many have argued that these strategies undermine equality and have unfair distributional consequences (Sekhon, 2011; Colb, 2001). Critics also contend that the practices used in these strategies invade bodily integrity and privacy in ways Fourth Amendment law cannot fully address (Harmon, 2012b). To address these concerns, legal scholars often advocate changing constitutional doctrine to forbid the strategies (Stuntz, 2002; Colb, 2001; Capers, 2010), but they might as easily argue that departments should give up the strategies preemptively or that other legal avenues be used to prohibit them.

Similarly, though focused deterrence (a person-focused strategy) and place-based strategies often comply with constitutional law, when departments identify chronic offenders or high-crime neighborhoods, they do so based on criminal histories and crime data. Blacks are likely to be overrepresented in criminal history data (Snyder, 2011; Kaeble, Maruschak, and Bonczar, 2015; Raphael and Stoll, 2013) and to live in neighborhoods in which crime is more likely to take place (Lofstrom and Raphael, 2016). To the degree that the data reflect earlier discriminatory criminal justice policy or historical housing discrimination, proactive strategies that seem neutral and may survive legal challenge can nevertheless have the effect of compounding earlier discrimination. In this way, proactive strategies can, in effect, “launder” racial disparities that result from prior government decision making: they can make the disparities appear to be driven by reasonable and legitimate policy goals rather than preexisting discrimination.

Similar concerns are often raised about using predictive policing methods, including the power of “big data” and crime analytics techniques, to isolate patterns among past criminal incidents. These methods can replicate discrimination and provide it with a superficially neutral justification. Such concerns are often aggravated by the absence of transparency and accountability for the algorithms used to identify patterns and predict future crime incidence (see, e.g., Joh, 2017). One of the most commonly used recidivism risk assessment tools, the Correctional Offender Management Profiling

___________________

¹³ Seattle, WA, Municipal Code 14.18.20 (2013); Seattle, WA., Ordinance 124142 (2013).

for Alternative Sanctions, or COMPAS, is based on calculations that are considered proprietary by its creator, Northpointe (Angwin et al., 2016).¹⁴

More generally, concentrating policing on particular problems or neighborhoods is likely to have important distributional consequences, including focusing the costs of police and prosecutorial power more heavily on places where specific groups are overrepresented. For instance, a decision to interrupt open-air drug markets, rather than targeting doctors who run prescription drug mills, will mean that some culpable offenders are more likely to suffer criminal justice consequences than others. David Weisburd (2016) argued that such focusing of policing can reduce overall harm. A focused policing approach, for example, at crime hot spots will not lead to large-scale police intrusion in a neighborhood overall. But such focusing can have negative consequences in the form of reduced liberty for some when people who live in identified hot spots suffer additional police stops or arrests.

Beyond distributional effects, although some types of focused policing may reduce overall harm, other proactive strategies may increase individual and aggregate negative consequences of policing. Even when legal, and even when effective in preventing crime, each additional stop and arrest imposed constitutes a significant intrusion on individual interests in liberty, autonomy, bodily integrity, and privacy and potentially constitutes an erosion in perceptions of the police, at least among some in the community. The negative consequences can be both financial, in the form of lost income, and intangible, such as the dignity harms of being frisked in public. Yet these various harms are sometimes overlooked in existing assessments of policing policies (Harmon, 2015).

Some scholars have suggested that the risk of unfair policing that many proactive strategies entail indicates that those proactive strategies should be replaced with a “newer policing” that focuses on changing public perceptions of the police (Tyler, Goff, and MacCoun, 2015). Others have argued that the negative consequences resulting from some proactive strategies can be mitigated by programs designed and implemented with an emphasis on public participation, legitimacy, and fairness (Braga and Weisburd, 2010). For instance, an ongoing test program in Brooklyn Park, Minnesota, proactively focuses additional policing on hot spots but also seeks to establish effective and trusting relationships between police and residents of the hot spots and shared expectations for the program (D. Weisburd, 2016).

___________________

¹⁴ At the same time, increased availability of administrative data on police activity may allow police departments to prevent, and others to better assess, potential Fourth or Fourteenth Amendment violations, as demonstrated by Goel, Rao, and Shroff (2016).

COMMUNITY-BASED POLICING

The heterogeneity of policing programs under the rubric of community policing makes it difficult to assess credibly the relationship between such programs and legal constraints and values. Activities associated with the approach, this report calls “community-based policing” (see Chapter 2), such as engaging in foot patrols or attending community meetings, have no significant legal implications. They are not governed by the Fourth Amendment, the Equal Protection Clause, or by federal or state statute. Nevertheless, in addition to encouraging officers to engage in particular activities, community-based policing also changes the allocation of discretion and responsibility within police departments and alters the mechanisms by which the department hears the concerns of the community. Whatever the positive benefits for legitimacy, community satisfaction, and crime control, these organizational changes can also limit traditional pathways of accountability in policing.

Traditionally, elected mayors and city councils and appointed city managers influence policing through police chiefs and other top commanders, whom they often hire and fire.¹⁵ Police executives make and implement policy and priorities through a hierarchical command staff that oversees street-level officers. Thus, chiefs operate at the fulcrum of an external accountability mechanism by which voters, through elected officials and more directly, influence police executives and an internal accountability mechanism in which chiefs operate through a hierarchical command staff to shape officer action through rules governing officer conduct, professional rewards for good behavior, and sanctions for noncompliance. State and municipal law often draws the outer boundaries of this system of accountability in multiple ways: (1) through laws determining the form of the local government and the local electoral process, (2) by requiring departments to collect and disclose some kinds of information to the public, (3) by setting qualifications and powers for police executives, and (4) by regulating administrative investigation and discipline of officers.

Though community-based policing strategies are unlikely to violate the structural parameters set by state law, a community-based approach nevertheless changes the nature of both internal and external accountability in police departments. First, with respect to internal accountability, community-based policing often includes devolving authority down the organizational hierarchy to frontline officers, whose patrol assignments are geographic areas (Skogan, 2006c). Communities are encouraged to provide input directly to street-level officers. Those officers in turn are given discre-

___________________

¹⁵ Most sheriffs are elected, meaning that unlike police chiefs, they are directly accountable to voters.

tion to allocate policing resources and shape problem solving, pursuant to that input, without as much command approval as is often required in traditional policing. These direct lines of communication and additional discretion can enable officers to act quickly and reactively to community input and can enable officers to develop valuable problem-solving skills (Weisburd, McElroy, and Hardyman, 1988; Bittner, 1983). However, the process concomitantly weakens the traditional power that command staff has over policy and officer action. For example, in an early study of a community-oriented policing strategy, Weisburd, McElroy, and Hardyman (1988) found that pilot community-policing units in the NYPD engaged in aggressive patrol tactics against low-level drug dealers, activities that were otherwise discouraged for patrol officers (as contrasted with specialized drug enforcement units) because of corruption hazards.

In concept, the additional officer discretion generated by the community-oriented policing strategy could permit additional violations of law and policy by individual officers. The potential problems here are highlighted in systematic social observations of police departments, which found higher rates of illegal searches among officers who embraced community-oriented policing than among those who did not (Gould and Mastrofski, 2004).

However, departments sometimes develop alternative means of supervising officers to replace traditional rules, monitoring, and sanctions. For example, in one study, supervisors of patrol officers engaged in community-based policing developed alternative metrics for productivity, such as assessing whether the officers made progress on priority problems in the neighborhood, rather than looking at arrests or response times. They also used supervisor approval for patrol strategies, careful selection of officers, and positive reinforcement of values to encourage law-abiding conduct by officers while they were out on patrol (Weisburd, McElroy, and Hardyman, 1988). These alternative supervision mechanisms may mitigate or eliminate effects on legal compliance by individual officers. For instance, a positive relationship between documentation and legality was noted by Gould and Mastrofski (2004), who observed the constitutionality of more than 100 searches in a single agency. In addition, whether decentralizing discretion results in a net increase or decrease in legal violations depends on several additional factors, including whether the counterfactual, more hierarchical, structure effectively promotes legal compliance. There is little empirical research about the comparative effectiveness of these alternative supervisory strategies, and therefore no way exists at present to assess the net accountability effects of community-oriented policing or similar strategies.

Second, with respect to external accountability, proactive policing strategies frequently emphasize informal community involvement in identifying, prioritizing, and solving problems through neighborhood meetings or through collaboration with business, religious, and neighborhood leaders,

rather than either formal processes of aggregating community will, such as elections, or individual methods for providing input into police priorities, such as 911 calls (Skogan, 2006c). Replacing traditional means for collective input allows departments to respond more precisely (e.g., at the neighborhood or street corner level) and more thoroughly to more local concerns, and it allows voices that may get drowned out in the political process to be heard. But it may also make departmental choices less representative of broader community values. Moreover, since neighborhoods often lack elected leaders designated to represent their specific areas, officers have less structured and clear guidance about how to balance competing views. In addition, when departments replace other traditional, individual, forms of input, such as citizen calls, to set priorities, they move departments away from the classic account of policing by Reiss and Bordua (1967), which holds that “[i]n a democratic society, the major volume of police work derives from an external source, the citizen complaint, rather than from an internal organizational source.” In this way, the community-based policing approach can change the basis for the legitimacy of police departments.

Like community-oriented policing, procedural justice policing operates both as a philosophy and as a strategy in police departments. As described in Chapter 2, in procedural justice policing, police officers give citizens voice, make decisions neutrally, treat people with dignity and politeness, and convey concern and benevolence, in order to promote perceptions of police legitimacy and thus achieve greater public cooperation with and deference to the police and increased compliance with the law.

Though the four pillars of procedural justice—giving voice, acting neutrally, treating citizens with dignity and respect, and conveying trustworthy motives—could reduce constitutional violations, procedural justice strategies may nevertheless sometimes exist in tension with other legal values. For example, one important principle in liberal legal regimes is that citizens should be able to limit their cooperation with law enforcement to no more than what is legally required of them. To that end, the law’s commands should be clearly defined in advance and ascertainable to those subject to them, a principle known in some contexts as legality.

Outside of the context of Miranda warnings (384 U.S. 436 [1966]), police officers are not usually required by constitutional law to tell citizens that they may refuse consent in order for their consent to be found knowing and voluntary (Schneckloth v. Bustamonte, 412 U.S. 218 [1973]; United States v. Drayton, 536 U.S. 194 [2002]). Procedural justice practices often seek to facilitate compliance by having officers request cooperation, both in circumstances when the officers might have no power to compel cooperation and in circumstances where they could issue an order enforceable either by force or by the threat of an arrest. Given that an invitation to cooperate is ambiguous, procedural justice practices can comply with the law while

making it harder for people to distinguish requests from commands that they are legally obliged to follow. Doing so can thereby make it harder for citizens to enforce fully their legal rights, if they wish to comply with the law but do not wish to cooperate with police requests that are not legally obligatory. The empirical literature studying the effects of procedural justice policing largely fails to distinguish cooperation with optional requests from compliance with legally mandatory commands, which makes it harder to assess the effects of procedural justice practices on populations with different preferences about cooperation versus compliance. There is some empirical evidence that suggests that police officers themselves may not fully understand the difference between a citizen’s failure to comply with an optional request and resistance to a lawful order. While not examining the distinction between requests and orders per se, Heffernan and Lovely (1991) presented police officers, lawyers, and lay people with hypothetical search and seizure scenarios and found that, on average, officers were better at identifying constitutional violations than lay people but worse than lawyers.

More broadly, the logic model underlying procedural justice emphasizes the centrality of citizen feelings about policing and deemphasizes the significance of the legal or normative status of police conduct (Meares, 2013). This logic model emphasizes the importance of community satisfaction with the police and the benefits that may accrue from the perception that the police are trustworthy and legitimate. Thus, procedural justice scholars define terms such as legitimacy and fairness differently than legal and political philosophers do. In these latter perspectives, procedural justice is a virtue of the decision-making process, not a quality of how that process is perceived (Solum, 2004). Similarly, legitimacy is a quality of political institutions, not of perceptions of those institutions. In legal and political philosophy, perceptions of how an institution functions may be considered in deciding whether it lives up to the normative demands of procedural justice and political legitimacy, but those demands have content independent of how the institution is perceived.

Criteria for police action based on perceptions that were developed in accordance with the procedural justice logic model often align closely with criteria based on deep legal values such as fairness and accountability. But there could be some distance between the normative standards by which policing might be meaningfully assessed from an objective perspective and standards based on subjective perceptions—the yardsticks by which police departments are encouraged to measure themselves under this logic model. By contrast, to the degree that procedural justice policing operates as intended, it may make violations of the law less likely. For example, procedural justice policing tries to induce citizens to comply voluntarily with officers. If an officer invites a person to talk to police on the street,

and the person cooperates, then the officer may avoid a seizure that triggers Fourth Amendment scrutiny (or custody that triggers the need for Miranda warnings and a waiver by the person in custody before asking questions). Similarly, if an officer invites a citizen to turn out his pockets, and the individual voluntarily complies, the Fourth Amendment requires no individualized suspicion for the search. Fewer rules for officers to follow in carrying out their duties could, mechanically, mean fewer legal violations by police. Less directly, if procedural justice policing increases citizen compliance and reduces conflict between citizens and officers, it may limit the situational factors that can lead to escalation, such as arrests and use of force, and therefore reduce the opportunities for making an arrest illegally or using excessive force (Owens et al., 2016).

In addition, procedural justice may include changes within the police department, namely, the application of procedural justice principles internally to how officers are treated by their organization and those who oversee it. Thus, for example, a department might seek to give officers voice, treat them neutrally and with dignity, and display trustworthy motives before imposing administrative discipline. Or it might solicit input for policies and priorities that affect an officer’s work. If adopting procedural justice policing increases the legitimacy of internal rules to officers, and thereby increases their compliance with departmental policies regarding treatment of civilians, then procedural justice policing could decrease officers’ legal violations, including Fourth Amendment violations.

This argument was made by Wolfe and Piquero (2011) and by Tyler and colleagues (2007), who surveyed groups of law enforcement officers about their perceptions of procedural justice in their agency and their willingness to follow their supervisor’s orders. Both studies found that perceptions of fairness and procedural justice were positively correlated with various measures of rule compliance by officers. Wolfe and Piquero (2011) found that officers who felt that they were treated fairly within the Philadelphia Police Department were less likely to engage in misconduct on the job and were also less likely to be the subject of an internal affairs investigation. Since all officers work for the same organization, in the absence of further information on the supervisory strategies to which each officer was subject, it is difficult to attribute this finding to changes in procedural justice. For example, it seems highly plausible that officers developed poor opinions of their employers because they were subject to investigation. Similar concerns apply to the findings by Tyler, Callahan, and Frost (2007), who surveyed officers in multiple agencies and estimated the correlation between the officer’s perceptions of legitimacy and procedural justice with their self-reported propensity to violate department rules. Without a better understanding of why, exactly, individual officers vary in their perceptions of legitimacy, it is difficult to draw causal conclusions from these studies

about the impact of introducing procedural justice–oriented policies on the legality of officer actions. In short, given their research design, the existing literature does not provide evidence supporting or refuting the hypothesis that procedural justice principles applied internally and that officers’ perceptions of the legitimacy of the police organization will increase the likelihood that officers follow department rules.

CONCLUSION

However effective a policing practice may be in preventing crime, it is impermissible if it violates the law. The most important legal constraints on proactive policing are the Fourth Amendment and the Equal Protection Clause (of the Fourteenth Amendment) of the Constitution, along with related statutory provisions. Several proactive practices are made possible by particular aspects of contemporary Fourth Amendment doctrine: SQF, broken windows, and hot spots policing strategies take advantage of the low level of individualized suspicion required for stops and frisks. Closed circuit television depends on the doctrine that puts most movements in public beyond the scope of the Fourth Amendment. Third party policing sometimes uses doctrine that permissively allows police to use information gathered from third parties.

Empirical evidence on the relationship between particular policing strategies and constitutional violations is insufficient to draw any significant conclusions about the likelihood that particular proactive strategies increase or decrease constitutional violations. Research about whether proactive policing leads to constitutional violations is hampered by inadequate data on police conduct that raises constitutional concerns, including stops, searches, and uses of force; the absence of accurate objective measures of constitutionality or proxies for constitutional violations; and studies that do not adequately engage in counterfactual analysis. Nevertheless, there are case-specific evidence and ethnographic and theoretical arguments consistent with the hypothesis that proactive strategies that use aggressive stops, searches, and arrests to deter criminal activity may decrease liberty and increase Fourth Amendment and Equal Protection violations. In addition, proactive policing strategies can affect the Fourth Amendment status of policing conduct.

Community-oriented policing and procedural justice policing strategies differ from other proactive policing strategies in that there are plausible mechanisms by which they may decrease constitutional violations rather than increase them. However, there is insufficient empirical evidence to support the existence of these effects (especially given the heterogeneity of these approaches and the activities used to pursue them), and both community-oriented policing and procedural justice policing sometimes may disrupt

traditional mechanisms of accountability by changing how departments make decisions or how demands and requests are conveyed to individuals with whom the police interact.

Civil lawsuits for damages and equitable relief are likely to be both the most common and most successful legal mechanisms for enforcing constitutional rules when police departments engage in proactive policing. While civil lawsuits for equitable relief have more direct effect, such suits face practical and legal obstacles that sometimes make them difficult to bring successfully. Civil lawsuits for damages face different obstacles and are unlikely to encourage constitutional compliance unless departments collect information about the number and kinds of lawsuits they face, enabling the departments to identify and mitigate sources of constitutional violations within them. DOJ has also sought to limit some kinds of proactive policing, such as zero tolerance policing, and encourage other kinds of proactive policing, such as community-oriented policing, in its pattern and practice lawsuits against departments.

Even when proactive policing does not violate or encourage violations of the law, it may implicate important legal values such as privacy, equality, and accountability that are of substantial public concern. In doing so, proactive policing strategies can raise substantial distributional and equality concerns and can sometimes spur local and state law changes, adding to existing regulation of the police.

Compared to the other outcomes examined in this report, there is relatively less empirical evidence on the impact of proactive policing policies on the legality of officer actions. This is at least in part due to the nature of legality itself, which is intrinsically determined in an ex post, individual manner relative to evolving case law, rather than a more objective, a priori, standard such as the standards for determining assault, racial disparities, or community satisfaction. The committee drew the following overarching conclusions regarding law, legality, and proactive policing:

CONCLUSION 3-2 Even when proactive strategies do not violate or encourage constitutional violations, they may undermine legal values, such as privacy, equality, and accountability. Empirical studies to date have not assessed these implications.