Argued April 18, 2023; decided May 18, 2023

People v Johnson, 206 AD3d 1702, reversed.

In People v De Bour (40 NY2d 210, 216 [1976]), we established a four-tiered framework for evaluating the propriety of police-initiated encounters with civilians (see People v Hollman, 79 NY2d 181 [1992]). In this case, Tyquan Johnson was stopped and frisked after he exited a parked car and walked down the street. He unsuccessfully moved to suppress the contraband found on him. On appeal, he contends that the police did not have the requisite level of suspicion to justify an intrusion under any level of De Bour. We agree that the police lacked reasonable suspicion to believe Mr. Johnson had committed a crime or was in possession of a weapon. Accordingly, we hold that the circumstances did not warrant a level three stop and frisk under De Bour and that the evidence seized as a result of the frisk must be suppressed.

On an early evening in April 2015, Officers Bradley Pike and Darrel Schultz were patrolling an area in Rochester that had recently experienced a reported rise in violent crime. Both officers were in uniform and were riding in a marked vehicle.^[FN*] As they drove up Harvest Street, Officer Pike noticed a parked Ford Explorer about 50 feet ahead. There is no indication that the car was parked improperly or that it was otherwise violating any vehicle or traffic laws. Officer Pike observed the vehicle's only occupant, Mr. Johnson, move from the driver's seat to the passenger seat. As Officer Pike approached the car, he saw Mr. Johnson momentarily move his upper body back toward the driver's seat. Officer Pike stopped his patrol car behind Mr. Johnson's car and turned on the overhead lights (not the emergency lights), so he could better see into Mr. Johnson's vehicle. Nothing in the record suggests that Mr. Johnson was aware of the presence of the police when the police car stopped or when Officer Pike turned on the overhead lights.

Officer Pike and Mr. Johnson exited their respective vehicles and Officer Pike noticed that Mr. Johnson's pants were unbuttoned, his belt undone, and that he was trying to pull his pants up as he walked down the street. Officer Pike asked Mr. Johnson to hold up, but Mr. Johnson continued to walk away.{**40 NY3d at 175} When Officer Pike caught up to Mr. Johnson, he asked whether Mr. Johnson was nervous; Mr. Johnson replied that he was not. Officer Pike asked whether Mr. Johnson had any weapons on him, to which Mr. Johnson replied, "Nothing." Officer Pike then frisked Mr. Johnson—finding no weapon. During the frisk, Officer Pike felt an object in Mr. Johnson's pocket he thought might be a bag of drugs. He asked Mr. Johnson what was in his pockets and Mr. Johnson replied, "Nothing." According to Officer Pike, Mr. Johnson began emptying his pockets, throwing two bags of marijuana on the ground. He also noticed that Mr. Johnson was holding a clear bag in his fist containing what appeared to be heroin. Officer Pike placed Mr. Johnson under arrest.

Mr. Johnson moved to suppress the drugs found on his person as the fruits of an illegal search and seizure. At the suppression hearing, Officer Pike testified that he thought it was not "common" for someone to move from the driver to the passenger seat of a car and that Mr. Johnson's moving his upper torso back toward the driver's seat meant there was "potential" that Mr. Johnson could be trying to stash or retrieve a weapon. He also considered it suspicious that Mr. Johnson was pulling up his pants and attempting to buckle his belt because suspects commonly hide weapons in their waistband. The court denied the motion to suppress, and the case proceeded to a bench trial. Mr. Johnson was convicted of two counts of criminal possession of a controlled substance in the third degree and was sentenced to five years on each count, to run concurrently. On appeal, Mr. Johnson renewed his arguments that Officer Pike's initial request to stop violated level one of De Bour; that the Officer's questioning violated level two, and that the stop and frisk violated level three. The Appellate Division affirmed, summarily holding that "the action taken by [Officer Pike] was justified in its inception and at every subsequent stage of the encounter leading to [Mr. Johnson]'s arrest" (206 AD3d 1702, 1702-1703 [4th Dept 2022]). We now reverse.

There is no need for us to consider whether Officer Pike's initial approach and questioning violated levels one and two of De Bour because his frisk of Mr. Johnson clearly runs afoul of level three. To [*3]conduct a stop and frisk under De Bour level three, the police must at a minimum have "reasonable suspicion that the particular person has committed or is about{**40 NY3d at 176} to commit a crime" (People v Benjamin, 51 NY2d 267, 270 [1980]) or that the person is "armed or dangerous" (People v Carney, 58 NY2d 51, 52 [1982]; see People v Brannon, 16 NY3d 596, 602 [2011] [reasonable suspicion requires "specific and articulable facts which, along with any logical deductions, reasonably prompted the intrusion" (internal quotation marks and alterations omitted)]).

Here, Mr. Johnson's actions, as observed by Officer Pike, do not meet the minimum standard required to justify a stop and frisk under De Bour. Prior to the frisk, Officer Pike observed Mr. Johnson: (1) move from the driver's seat to the passenger seat of his parked car; (2) move his upper torso back toward the driver's seat; (3) pull up his pants and attempt to buckle his belt; and (4) appear nervous while being questioned. These circumstances do not support a reasonable view that Mr. Johnson was armed or that he had committed or was about to commit a crime. These actions "constituted [nothing] other than 'innocuous behavior,' sole reliance on which would impermissibly reduce the foundation for [this] intrusion to nothing but 'whim or caprice' " (People v Carrasquillo, 54 NY2d 248, 252 [1981], quoting De Bour, 40 NY2d at 216-217; see also People v Sierra, 83 NY2d 928, 930 [1994] [no reasonable suspicion where defendant "grabbed at his waistband and then fled"]; People v Milaski, 62 NY2d 147, 156 [1984] [nervousness in response to questioning does not justify further detention]; People v Howard, 50 NY2d 583, 590 [1980] [presence in area of "frequent burglaries" did not support reasonable suspicion and furtive movements were "at best ambiguous"]). Because Officer Pike lacked reasonable suspicion to justify the stop and frisk of Mr. Johnson, the evidence should have been suppressed.

Accordingly, the order of the Appellate Division should be reversed and the indictment dismissed.

Rivera, J. (concurring).

A person's "right to be left alone is 'too precious to entrust to the discretion of those whose job is the detection of crime' " (People v De Bour, 40 NY2d 210, 219 [1976], quoting McDonald v United States, 335 US 451, 455 [1948]). And yet, police interactions with the public for criminal investigative purposes are necessary to ensure public safety. This is risky business, as shown by media accounts of police encounters prompted by innocent conduct that quickly escalate to violent altercations. These dangers are not new, but their continued prevalence requires us to reassess whether the legal principles set out in People v De Bour nearly half a{**40 NY3d at 177} century ago can protect individuals today from "subconstitutional abuses" (40 NY2d 210, 220 [1976]).

Police officers protect our neighborhoods and the judiciary is constitutionally tasked with determining the legality of the officers' interactions with the public. I agree that, in this case, there is no record support for the officer's stop and frisk. But, unlike the majority, I believe it pertinent to the resolution of this appeal and vital to the sound development of our police-encounters jurisprudence that we also evaluate whether the officer's initial approach and inquiries were justified and not merely, as the majority opines, cumulative factors in assessing the propriety of defendant's eventual detention and frisk. After all, these initial observations eventually contributed to his arrest.

Lawfully parking one's car on the street and moving from the driver's seat to the front passenger's seat is not suggestive of criminality. Nor is reaching from the passenger seat back to the driver's side or exiting through the closest door. Therefore, the officer's actions were unjustified at [*4]their inception and at every subsequent stage of the encounter until defendant's arrest. Any conclusion to the contrary invites the government to cast a wide net that ensnares the innocent in the hopes that a fishing expedition will uncover incriminating evidence. Apart from the danger to people on the street and the officers who patrol our neighborhoods, these aggressive investigative tactics strain relations with communities, and distract from good policing practices that focus on actual criminal behavior. Such aggressive encounters are inconsistent with the socially-accepted goals of police engagement: public safety and security.

Moreover, although we have reached the right conclusion in this case—but have taken different paths—I conclude that our street-encounters jurisprudence has veered in the wrong direction and that the De Bour legal framework in practice does not serve the ends of justice. The way to properly safeguard the right to be left alone and the safety of officers and individuals alike is a rule that requires reasonable suspicion of criminality for all police-initiated encounters.

I first discuss why reversal is required here. Defendant Tyquan Johnson argues that, under De Bour's four-level framework, the police were unjustified in approaching, questioning,{**40 NY3d at 178} and detaining him before his arrest (40 NY2d at 223). Based on the evidence before the suppression court, defendant is correct that the officer's initial approach and every subsequent intrusion failed to comport with De Bour.

Undeniably, De Bour is one of our Court's most significant decisions, having set the course for police encounters since it was first decided in 1976. De Bour rejected the Federal Constitution's "all or nothing approach" to seizures of persons by "emphasiz[ing] the primacy of the right to be free from aggressive governmental interference" and more broadly defining the term "seizure" to mean "a significant interruption with an individual's liberty of movement" (40 NY2d at 216-217). However, the Court also rejected Mr. De Bour's proposal for a "blanket prohibition of all police-citizen encounters conducted in the absence of probable cause or reasonable suspicion based on concrete observations" (40 NY2d at 216). Instead, the Court concluded that police-initiated street encounters with the public, short of a federal constitutional seizure under Terry v Ohio (392 US 1 [1968]) and its progeny, were permissible but nonetheless subject to judicial scrutiny (40 NY2d at 217-218). The Court then drew a distinction between "public service functions, not related to criminal law enforcement," and criminal investigations, concluding that the latter are "viewed and measured by an entirely different standard of reasonableness" (De Bour, 40 NY2d at 218-219). Thus, De Bour's legal standards were informed by and responsive to the "multiplicity and complexity of tasks assumed by the police" as public servants (id. at 218). Nevertheless, as crime prevention "is highly susceptible to subconstitutional abuses," the Court reasoned, police interactions with the public "will be subject to the greatest scrutiny; for whereas a police[ officer's] badge may well be a symbol of the community's trust, it should never be considered a license to oppress" (id. at 220). With those principles in mind, the Court adopted its four-level framework, explaining that "[i]n evaluating the police action we must consider whether or not it was justified in its inception and whether or not it was reasonably related in scope to the circumstances which rendered its initiation permissible" (id. at 222-223).

As the Court explained years later in People v Hollman:

[*5]"De Bour represents the culmination of a number {**40 NY3d at 179}of State common-law cases that provided a framework for the evaluation of police-civilian encounters . . . . In De Bour, constitutional law and common law both played a part in the articulation of the four-part test. Although we stated that 'constitutional considerations do not disappear' when police encounters fall below the level of a seizure (People v De Bour . . . at 217), we did not rest our analysis squarely upon the language of either the Federal or State Constitution. Rather, we noted that '[t]he basic purpose of the constitutional protections against unlawful searches and seizures is to safeguard the privacy and security of each and every person against all arbitrary intrusions by government. Therefore, any time an intrusion on the security and privacy of the individual is undertaken with intent to harass or is based upon mere whim, caprice or idle curiosity, the spirit of the Constitution has been violated' (id. at 217). To some extent, then, our holding in De Bour was not compelled by the specific language of either the State or the Federal Constitution. Rather, it reflected our judgment that encounters that fall short of Fourth Amendment seizures still implicate the privacy interests of all citizens and that the spirit underlying those words required the adoption of a State common-law method to protect the individual from arbitrary or intimidating police conduct" (79 NY2d 181, 195 [1992]).

Under this framework, the first step governs an officer's least invasive engagement with a private individual, and "[e]ach progressive level" of De Bour "authorizes a separate degree of police interference with the liberty of the person approached and consequently requires escalating suspicion on the part of the investigating officer" (id. at 185). The justification for each escalation is based on the totality of the circumstances at the moment the escalation occurs, building on the officer's prior observations and actions of both the officer and the private individual (see People v Moore, 6 NY3d 496, 500-501 [2006]).

At the first level, law enforcement may engage in minimally-intrusive questioning to request information "when there is some objective credible reason for that interference not necessarily indicative of criminality" (De Bour, 40 NY2d at 223). The second level—the common-law right of inquiry—permits officers {**40 NY3d at 180}"to gain explanatory information, . . . short of a forcible seizure," upon a "founded suspicion that criminal activity is afoot" (id.). The third level, "a forcible stop and detention," requires the "officer entertain[ ] a reasonable suspicion that a particular person has committed, is committing or is about to commit a felony or misdemeanor," and "[a] corollary of the statutory right to temporarily detain for questioning is the authority to frisk if the officer reasonably suspects that [they are] in danger of physical injury by virtue of the detainee being armed" (De Bour, 40 NY2d at 223 [citation omitted]; see also CPL 140.50 [1]). "Finally[,] a police officer may arrest and take into custody a person when [the officer] has probable cause to believe that person has committed a crime, or offense in [the officer's] presence" (40 NY2d at 223 [citation omitted]). This framework applies to police encounters with both pedestrians and individuals in motor vehicles, whether moving (People v Hinshaw, 35 NY3d 427, 431 [2020]; People v Garcia, 20 NY3d 317, 324 [2012]) or parked (see People v Harrison, 57 NY2d 470, 475-476 [1982]).

A full analysis of each De Bour level implicated in a defendant's case is not merely a thought exercise. Rather, addressing each implicated level of the framework is critical to the development of our law in this area and our continuing constitutional obligation "to provide clear guidance for police officers seeking to act lawfully in what may be fast-moving street encounters and a cohesive [*6]framework for courts reviewing the propriety of police conduct in these situations" (Moore, 6 NY3d at 499). When, as here, the Court avoids discussion of levels one and two and starts its analysis at the higher De Bour levels, it implicitly legitimizes the information police gather as a result of those initial, lower-level intrusions, which is then later used to justify greater intrusions (see e.g. id. at 497-499 [beginning analysis with the "gunpoint stop" where officers acting on an anonymous tip had arrived at the scene and approached the defendant "who began to walk away"]; People v Townes, 41 NY2d 97, 99-101 [1976] [holding that, given the prosecution's concession, the gunpoint command to "freeze" was not "justified in its inception" when officers had "followed the defendant and his companion for approximately one hour and . . . observed them doing nothing more than walking about and looking at people and stores" (internal quotation marks and alterations omitted)]; see also People v Perez, 31 NY3d 964, 966 [2018] [concluding, without specifying the De Bour level at issue, that{**40 NY3d at 181} record evidence "provided the requisite level of support to satisfy the applicable level of intrusion"]; People v Thorne, 207 AD3d 73, 75-77 [1st Dept 2022] [officers lacked reasonable suspicion for a level three encounter where officers first stopped defendant on the street and asked if they could speak with him, to which he asked "why they were stopping him"]; Matter of Darryl C., 98 AD3d 69, 71-74 [1st Dept 2012] [invalidating stop and frisk under level three even though officers approached and questioned juvenile based on his being on a street corner during school hours holding a nondescript "object"]).^[FN1] I therefore turn to a level-by-level analysis of the police encounter that led to defendant's arrest.

According to the only police officer who testified at the hearing, he and another officer were in uniform and patrolling a neighborhood that had recently experienced increased violence when the testifying officer saw a Ford Explorer parked on the street 50 feet in front of him. He observed defendant—the sole occupant—move or jump from the driver to the passenger seat and then move his upper body back towards the driver's seat. Believing there was a "potential" that defendant might have been trying to stash or retrieve a weapon, the officer stopped his vehicle behind the Explorer, turned on his overhead lights to look inside, exited his marked vehicle, and walked towards the Explorer.

None of the officer's observations justified this initial approach. As we have said, police must have an objective, credible reason for approaching a defendant, equally so when a defendant is seated in a parked car (see Harrison, 57 NY2d at 475-476; De Bour, 40 NY2d at 216-217). Although the officer's reason for conducting a level one request for information need not be indicative of criminality, it cannot be based on conduct that is otherwise innocent (De Bour, 40 NY2d at 216) and the approach "must be predicated on more than a hunch, whim, caprice or idle curiosity" (People v Ocasio, 85 NY2d 982, 985 [1995]; De Bour at 217). Here, the officer did not testify that{**40 NY3d at 182} defendant's vehicle was illegally parked or that defendant violated a rule of the road. Instead, the officer approached based on having observed defendant changing front seats. Moving from one seat to another in a parked car, however, is not "unusual" and defendant's doing so therefore supplied no "objective credible reason" for the level one approach (Hollman, 79 NY2d at 190-191). Nor was it "unusual" when defendant reached back toward the recently-vacated driver's seat (id. at 190-191). It was the officer's pure speculation that the defendant's movements evinced a specific attempt to hide or retrieve a weapon. Therefore, because there was no "objective credible reason"—nothing more than a mere "hunch"—to approach defendant's vehicle (De Bour, 40 NY2d at 223), the officer's observations at this point did not justify a level one encounter.

Nor did the officer's observations of defendant exiting through the passenger-side door to the street justify either a level one or level two interaction because, as with defendant's seat change and reach toward the driver's seat, this action was also not "unusual" (Hollman, 79 NY2d at 191) but was "innocuous behavior" that "alone [did] not generate a founded . . . suspicion" (De Bour, 40 NY2d at 216). For example, traffic conditions may have made it dangerous for defendant to exit on the driver's side. Or perhaps the driver's-side door was malfunctioning and could not be opened from the inside. And the fact that the car was parked in a neighborhood with a reported increase in crime does not provide the requisite basis to approach since presence in a high-crime neighborhood does not justify a police encounter with private individuals (see People v McIntosh, 96 NY2d 521, 526-527 [2001]). Otherwise, residents of an entire community would be subject to intrusive police encounters merely because of their home addresses.

After defendant exited the Explorer, the officer first observed that defendant's pants and belt were undone, and then saw defendant pull up his pants and buckle his belt while walking away. These observations did not support the officer's stated suspicion that defendant might have been hiding a weapon. Indeed, the officer's conclusion defies common sense, as defendant would have needed to button up and close his belt first in order to be able to hide a weapon.

Since none of these observations supported the officer's initial, level one approach—and because individuals have the right to walk away without responding to such a request (People v Howard, 50 NY2d 583, 586 [1980])—these observations also {**40 NY3d at 183}did not provide justification for the officer to follow defendant, ask him to "hold up," or walk alongside him. Nor did the officer's subsequent belief that defendant was nervous based on his chest moving up and down support a founded suspicion of possible criminal activity to permit a level two inquiry about whether defendant had any weapons. As the Court has noted, nervousness does not indicate criminality (see People v Garcia, 20 NY3d 317, 323 [2012]), as "[i]t is certainly unsettling to be approached by a police officer" requesting information and "a reasonable person would . . . be taken aback by such a request" (Hollman, 79 NY2d at 192).

When defendant answered that he had "nothing," the officer also had no basis to detain defendant to further question him about whether defendant was nervous (see Moore, 6 NY3d at 501; see also CPL 140.50 [3]). Nor did the quantum of knowledge the officer possessed up to that point provide a reasonable suspicion that defendant had committed, was committing or was about to commit a felony or misdemeanor to support his subsequent frisk (Terry, 392 US at 30; People v Batista, 88 NY2d 650, 654 [1996]). For example, the officer did not testify that he saw a bulge or any other physical outline of a weapon before he conducted the frisk (see De Bour, 40 NY2d at 221). The officer frisked defendant based on pure supposition—which ultimately proved unfounded since defendant in fact had no weapon.

During this physical intrusion the officer did not feel any weapon but thought he felt drugs in defendant's pocket. He then asked defendant what was in the pocket and defendant again answered "nothing." According to the officer, even though defendant responded "nothing" to every question regarding what defendant had on his person, defendant spontaneously emptied his pocket [*7]and threw what appeared to be two bags of marijuana and some dollar bills to the ground. The officer also observed a baggie in defendant's hand which the officer suspected contained heroin. Based on these final observations of the drugs, the officer believed he had probable cause for defendant's arrest. However, because "[d]efendant's later conduct cannot validate an encounter that was not justified at its inception," and the officer's actions leading to the arrest—at levels one, two and three—exceeded the scope of permissible police interference with defendant, the contraband should have been suppressed (see Moore, 6 NY3d at 498).{**40 NY3d at 184}

I have engaged in a straightforward De Bour analysis, working through every level and describing why the police interaction here fails under each level of De Bour's framework. However, this atomized analysis and our experience with policing under the shadow of De Bour illustrate why police encounters with private individuals should be further cabined. Though driven by a goal of safeguarding the "right to be left alone," De Bour, as originally framed, renders it illusory in practice (40 NY2d at 219).

The Court has previously reconsidered De Bour. Fifteen years after the Court announced the four-tiered framework, the prosecution in Hollman asked the Court to overrule De Bour and adopt the federal standard, arguing that Terry and its progeny "made it increasingly clear that police-initiated encounters falling short of actual seizures do not implicate the Fourth Amendment" (79 NY2d at 194). That argument was in part historical. The De Bour Court acknowledged in 1976 that, at the time, there was "scant appellate authority" on the constitutionality of "an investigative confrontation"—i.e. the authority of police to approach private individuals (40 NY2d at 219). The prosecution in Hollman noted that, during the decade and a half since the Court had decided De Bour, the United States Supreme Court had concluded that a police approach short of a full-blown seizure does not implicate the Fourth Amendment (Hollman, 79 NY2d at 194-195). Under the Federal Constitution, the police have not seized an individual under Terry only when " 'a reasonable person would feel free "to disregard the police and go about [their] business" ' " (id. at 195, quoting Florida v Bostick, 501 US 429, 434 [1991]). In other words, "[i]f the civilian would feel free to go, 'the encounter is consensual and . . . will not trigger Fourth Amendment scrutiny unless it loses its consensual nature' " (id., quoting Bostick, 501 US at 434).

The Hollman Court rejected this invitation to discard the De Bour framework and clarified that "[t]he continued vitality of De Bour" was "not contingent upon the interpretation that the Supreme Court gives the Fourth Amendment, because De Bour is largely based upon considerations of reasonableness and sound State policy" (79 NY2d at 195). The Court further explained that police encounters that are not Fourth Amendment seizures should continue to be evaluated under De Bour because judicial scrutiny of those encounters best serves "[t]he {**40 NY3d at 185}aims of predictability and precision in judicial review of search and seizure cases and the protection of the individual rights" of the public (id. at 196 [internal quotation marks omitted]).

Thirty years have now passed since Hollman reaffirmed De Bour and almost 50 years since the Court decided De Bour. Given De Bour's analytic premise and our troubling historical experience with police-initiated encounters, we should again reconsider De Bour's "continued vitality" (Hollman, 79 NY2d at 195). In De Bour, the Court permitted intrusions under levels one and two, in part, based on "the practical necessities of law enforcement and the obvious fact that any person in our society may approach any other person and attempt to strike up a conversation" (40 NY2d at 219). At the same time, the Court purported to create a check by subjecting these sub-Terry encounters to [*8]judicial scrutiny with the objective of safeguarding the primacy of "the right to be left alone" (id.). However, equating intrusions by police officers with those of "any person in our society" ignores the simple reality that everyday people on the street understand: a police officer is not just "any person," but is an agent of the State, cloaked with governmental authority and specially authorized to use lethal force (id.). Thus, while a private individual might not think twice about ignoring a stranger, there is, as the De Bour Court described it, "the tendency to submit to the badge" (id.).

Empirical evidence confirms the obvious, as "studies over the last several decades on the social psychology of compliance, conformity, social influence, and politeness have all converged on a single conclusion: the extent to which people feel free to refuse to comply is extremely limited under situationally induced pressures" (Janice Nadler, No Need to Shout: Bus Sweeps and the Psychology of Coercion, 2002 Sup Ct Rev 153, 155 [2002] [collecting and summarizing studies]). Here in New York, one psychologist who submitted findings in connection with the New York Stop-and-Frisk Joint Remedial Process's 2018 final report and recommendations found that "[t]he intrusive commanding-presence style of policing embodied in stop and frisk interventions at unexpected and unprovoked times, combined with a history of use of excessive or deadly force by the police, has generated considerable fear" (New York City Joint Remedial Process, Final Report and Recommendations, Appendix A, White Paper of Michael Britton, Ed.D., Dec. 10, 2016 at 7 [2018], available at https://www.jamsadr.com/files/uploads/documents/articles/belen-new{**40 NY3d at 186}-york-city-joint-remedial-process-may-2018.pdf [last accessed May 15, 2023]).

A New York Civil Liberties Union survey conducted that same year similarly found pervasive fear among community residents exposed to aggressive policing:

"More than two-thirds (67 percent) of respondents in heavily policed communities feared having a friend or family member killed by police (a surprising 15 percent of respondents in lightly policed communities felt the same way). Slightly fewer (64 percent versus 10 percent) feared that they themselves could be killed by police. And almost half (43 percent) of the respondents in heavily policed neighborhoods feared they could be sexually assaulted by police compared to six percent in lightly policed communities" (New York Civil Liberties Union, Shattered: The Continuing, Damaging, and Disparate Legacy of Broken Windows Policing in New York City at 12 [2018], available at https://www.nyclu.org/sites/default/files/field_documents/nyclu_20180919_shattered_web.pdf [last accessed May 15, 2023]).^[FN2]

In addition, the survey found that 64% of respondents in "heavily policed communities reported that police at times made them feel scared," 71% "unsafe," and 74% "nervous" (id.). Significantly, police officers' shouting commands contributed to individuals' apprehension of the police in more heavily-policed communities:

[*9]"Sixty-one percent of survey respondents in heavily policed communities reported at least one negative verbal police encounter in 2016, compared to 15 percent in less policed communities. One in four people in heavily policed communities said they were shouted at by police, (25 percent versus five percent), cursed at (26 percent versus four percent) or threatened with arrest (33 percent versus three percent)" (id. at 18).

The fear is not unfounded, because, as I discuss infra, for members of certain racial, ethnic and religious communities, ignoring a police officer can carry dire consequences.{**40 NY3d at 187}

The De Bour Court aimed to shield "the right to be free from aggressive governmental interference" and "the right to be left alone" (40 NY2d at 216, 219), but experience has shown that the means it supplied for doing so—the regulation of sub-Terry encounters on less than reasonable suspicion—has not only proved inadequate, but, at times, undermined those very rights by promoting heightened forms of governmental intrusion triggered by noncriminal behavior. The resulting escalation of police-initiated encounters jeopardizes the safety of the public and police officers, and frustrates community policing efforts.

One commentator observed in 1991 that the De Bour framework's atomized levels are susceptible to conflation with one another in a manner that spurs the police to initiate and escalate encounters with members of the public (see Emily J. Sack, Police Approaches and Inquiries on the Streets of New York: The Aftermath of People v. De Bour, 66 NYU L Rev 512, 520, 548-553 [1991]). People v Reyes (83 NY2d 945 [1994]), which the prosecution cites in favor of affirming the denial of suppression here, illustrates the point. In Reyes, two officers positioned themselves on both sides of the defendant and "approached him with their hands on their holstered guns" while commanding " 'Hey stop, excuse me' or 'Stop, hey, stop, police,' or words to that effect" after they observed him "clutch" with his hands in his pants pockets and "walk briskly away from a group of men" in a " 'drug-prone' area" (id. at 946). When the defendant complied, a brick-like object fell from his armpit to the ground, which turned out to be a kilogram of cocaine (id.). It is difficult to imagine how Mr. Reyes could have regarded the officers' conduct as "nonthreatening" (Hollman, 79 NY2d at 191). Indeed, given the gravity of the intrusion—officers issuing commands with their hands on their firearms—it is difficult to imagine that a person in Mr. Reyes's position would not have felt "suspected of . . . wrongdoing," or even free to leave, thus qualifying the encounter as (at least) a level two (id. at 185, 191-192; see also United States v Mendenhall, 446 US 544, 553 [1980]). Nonetheless, the Appellate Division held that the actions of the police constituted a level one encounter and reasoned that commanding a person to stop in such a manner "is a necessary preliminary to a request for information when a person is ahead of the officer, walking away from him, and—for all that appears—unaware that the officer wished to inquire" (People v Reyes, 199 AD2d 153, 155 [1st Dept 1993]).{**40 NY3d at 188} This Court affirmed, agreeing with the Appellate Division that the encounter amounted to a level one "permissible request for information based on some objective credible reason" (Reyes, 83 NY2d at 946).

Reyes is a prime example of how De Bour's legal standard for limiting police power during sub-Terry encounters (see 40 NY2d at 219-220) can produce the opposite effect (see Sack, 66 NYU L Rev at 550). On one hand, the Court reiterated in Hollman that "[e]ach progressive level" of De Bour "authorizes a separate degree of police interference with the liberty of the person [*10]approached and consequently requires escalating suspicion on the part of the investigating officer" (79 NY2d at 185; see also Moore, 6 NY3d at 500). But Reyes inverts this graduated approach. Under its logic, an initial, unjustified intrusion can be transformed into permissible police conduct when it facilitates a justified lower-level intrusion. For example, applying Reyes's rationale to a hypothetical variation of the facts in defendant's appeal, the officers would have acted lawfully in stopping defendant's car if it were moving and immediately asking him if he was carrying a weapon on the ground that this level three seizure was a "necessary preliminary" to the level two inquiry (Reyes, 199 AD2d at 155, affd 83 NY2d 945 [1994]).

Concerns about the Court's atomization of police action and the potential for reverse engineering a lawful basis for an unjustified, escalating intrusion were identified in a strongly worded dissent by Judge Fuchsberg in People v Samuels (50 NY2d 1035 [1980]). In his view, "the mushrooming lexicographical distinctions into which the reasonableness of street stops and seizures are being categorized present a disturbing problem" (id. at 1039-1040 [Fuchsberg, J., dissenting]). "Since the circumstances of no two stops are ever precisely the same," he continued, "the semantics necessarily employed to effect such compartmentalization, however well intended, in the end make it all the easier to substitute labels for liberties" (id. at 1040). He welcomed what he regarded as "a trend away from the artificial and 'unworkable' hierarchy of police conduct erected in People v De Bour" (id., quoting Note, People v De Bour: The Power of the Police to Stop and Frisk Citizens, 30 Syracuse L Rev 893 [1979]).^[FN3]

In Samuels, an officer followed the defendant after he saw the defendant purchase a holster from a novelty shop and{**40 NY3d at 189} asked the defendant why he had made the purchase (50 NY2d 1035, 1036-1037 [1980]). The defendant did not respond, and instead put his hand in his coat pocket (id. at 1037). The officer commanded the defendant to remove his hand from his pocket; defendant did not comply (id.). The officer then grabbed the defendant's hand through his coat pocket and discovered a gun (id.).^[FN4]In upholding what was essentially a frisk, the Court acknowledged that the defendant's failure to respond to the officer's question would not have justified further action by the police, but that placing his hand in his pocket and refusing to withdraw it elevated the officer's "interest in his own safety" to a "sufficient basis for his grabbing the [defendant's] hand through the coat" (id.). And yet, the arresting officer testified that police indiscriminately confronted patrons who examined holsters (id. at 1039 n 1 [Fuchsberg, J., dissenting]).

Of course, one can sympathize with this officer's reaction. After all, who wouldn't feel threatened by somebody refusing to remove their hand from their pocket after just having purchased a holster? But as Judge Fuchsberg explained, the officer's safety would not have been at issue as a result of this rapidly-escalated confrontation had he simply refrained from approaching the defendant (see id. at 1039):

[*11]"The tone of life and spontaneity of spirit that characterizes a free society could not long survive if the police were allowed to exploit any unusual circumstance, rationalized with 20-20 hindsight as giving rise to suspicion, as a basis for an escalating intrusion into the privacy of anyone who insists on his right to be left alone . . . . This factor is crucial here, for without the initial baseless encounter, the officer could not conceivably claim that he was placed in a position where concern for his own safety compelled him to frisk defendant (see Terry v Ohio, 392 US 1, 27)" (id. at 1039 [citation and footnote omitted]).

Even if the officer's hunch turned out to be correct in Samuels, such fails to account for the cases featuring similar{**40 NY3d at 190} encounters where no weapon is recovered and which, consequently, never find their way into our courts (see Terry, 392 US at 13-14 [discussing the powerlessness of the exclusionary rule "to deter invasions of constitutionally guaranteed rights where the police either have no interest in prosecuting or are willing to forgo successful prosecution in the interest of serving some other goal"]). We know that there are numerous such cases.

Police intrusions that qualify as level one and two encounters under De Bour are quite common (see Lynn Langton & Matthew Durose, Police Behavior during Traffic and Street Stops, 2011, United States Department of Justice, Office of Justice Programs, Bureau of Justice Statistics at 11-12 [Sept. 2013], available at https://bjs.ojp.gov/content/pub/pdf/pbtss11.pdf [last accessed May 15, 2023]) and, as I have discussed, the De Bour model catalyzes their rapid spiral toward escalation and violence (see Sack, 66 NYU L Rev at 520, 548-553). But as history, social science, and demographic research has shown, the right to walk away is illusory. In Floyd v City of New York, the federal district court in the Southern District of New York found, based on expert testimony, that 83% of individuals stopped by the NYPD were black or Latino, even though those two groups made up only 52% of the city's population (959 F Supp 2d 540, 559-560 [SD NY 2013]; see also Dasha Kabakova, The Lack of Accountability for the New York Police Department's Investigative Stops, 10 Cardozo Pub L Poly & Ethics J 539, 562-563 [2012] [observing that, "in precincts in which blacks are less than 10% of the population, blacks remained over two times (2.17) more likely to be 'stopped' on suspicion of committing a violent crime than whites . . . (and) these differences are evident after controlling for race- and crime-specific arrest rates (within each precinct)"]).

The Floyd court also found that the NYPD had engaged in racial profiling and "the odds of a stop resulting in any further enforcement action were 8% lower if the person stopped was black than if the person stopped was white" (959 F Supp 2d at 560). Blacks and Latinos also bore the brunt of violent policing, as "blacks who were stopped were about 14% more likely" and Latinos "9% more likely—than whites to be subjected to the use of force" (id.). Hence, in addition to its numerous other deficiencies, De Bour has also failed to tackle racial profiling (see People v Epton, 19 NY2d 496, 501 [1967] [commenting that "tensions in Harlem" had "heightened over the issue of police brutality"]) and, if anything, has entrenched it (see Floyd, 959 F Supp 2d at 560).{**40 NY3d at 191}

As one legal scholar has argued, racial profiling as an investigative tool becomes a self-fulfilling prophecy under which successful police actions against one discrete group results in greater focus upon that group that permits sometimes-more-frequent crime by other groups to go undetected (see Barry Friedman, Unwarranted: Policing without Permission 197 [2017]). As he explained:

[*12]"Think about it this way. You like to fish, so you ask people what's a good spot. They say Trout Pond. You go there, and sure enough you catch some fish. Occasionally you go somewhere else, and you catch some fish there, too, but you've been told Trout Pond is a surefire bet, and you've seen some evidence of that, so you keep coming back to it. Now, if you'd done a careful study, you'd have learned that Trout Pond was no better than any other spot, and in fact might have been less good. There were more fish in other places, like Town Wharf or Towd Point. But having been told of Trout Pond and had your information confirmed, that is where you went" (id.).

The choice not to engage with police is no choice in fact for many people of color. Indeed, as De Bour itself acknowledged, it is difficult to understand how any person is not intimidated or otherwise believes that they cannot leave when approached by, as in this case, a uniformed officer or someone identifying themselves as a law enforcement official who tells them to stop (see De Bour, 40 NY2d at 219). And although the Court has said the right to walk away exists—elevating it to almost sacrosanct status—it has simultaneously recognized that turning and walking or running away from an officer may be a factor supporting level two, three, and four intrusions on the ground that "[f]light, combined with other specific circumstances indicating that the suspect may be engaged in criminal activity, could provide the predicate necessary to justify pursuit" (People v Holmes, 81 NY2d 1056, 1058 [1993]; see also People v Sierra, 83 NY2d 928, 930 [1994] [holding that the defendant's walking away from officers who, from a police cruiser, requested information under level one in a drug-prone area furnished officers with "reasonable suspicion that he was committing or was about to commit a drug-related crime"]).

Assume that, here, defendant had not merely said that he had nothing on him but also then tried to walk around the officer{**40 NY3d at 192} or back towards his vehicle. Would the officer have let him walk away? That is unlikely since the officer testified that, before either he or defendant had even exited their vehicles, he believed—despite a glaring lack of evidence—that defendant possessed a weapon. More likely, the officer would have stopped defendant and conducted the frisk anyway. So much for the right to be left alone. I have previously identified this same dilemma, noting that upholding a stop and frisk based on a defendant's refusal to respond to a level one or level two inquiry "suggests that members of the public must affirmatively respond to officers or risk forfeiting their freedom, a proposition we have clearly and unequivocally rejected" (Perez, 31 NY3d at 978 [Rivera, J., dissenting]).

New York City acknowledged in Floyd that: (1) "[b]etween January 2004 and June 2012, the NYPD conducted over 4.4 million Terry stops"; (2) "52% of all stops were followed by a protective frisk for weapons"; and (3) "[a] weapon was found after 1.5% of these frisks," meaning that, "in 98.5% of the 2.3 million frisks, no weapon was found" (959 F Supp 2d at 558). The City also conceded that only "6% of all stops resulted in an arrest, and 6% resulted in a summons," while "[t]he remaining 88% of the 4.4 million stops resulted in no further law enforcement action" (id. at 558-559). So many resources expended, so much liberty infringed, and so many people endangered, yet so little successful law enforcement outcomes to show for it. The individuals stopped of course bear the brunt of this failed policing strategy.

New York police recognize the deleterious effects this policing approach has on community relations. Former Mayor Michael Bloomberg issued an apology for the City's stop-and-frisk policy which he had championed during his Mayoralty (see Carl Campanile, Craig McCarthy & Aaron [*13]Feis, Police Union, de Blasio Blast Bloomberg Over Stop-and-Frisk Apology, NY Post, Nov. 17, 2019, available at https://nypost.com/2019/11/17/police-union-de-blasio-blast-bloomberg-over-stop-and-frisk-apology/amp/ [last accessed May 15, 2023]). In response, the New York City Police Benevolent Association (PBA) President conveyed his apparent long-standing view that stop-and-frisk was a " 'misguided policy' " that " 'inspired an anti-police movement that has made cops the target of hatred and violence, and stripped away many of the tools we had used to keep New Yorkers safe' " (id.). He asserted that the PBA had told the Mayor " 'in the early 2000s that the quota-driven emphasis on{**40 NY3d at 193} street stops was polluting the relationship between cops and our communities' " (id.).

As is abundantly clear from numerous sources—the stop and frisk litigation in the Southern District of New York (Ligon v City of New York, 925 F Supp 2d 478 [SD NY 2013]; Floyd, 959 F Supp 2d at 540), national statistics (e.g. New York Civil Liberties Union, Shattered: The Continuing, Damaging, and Disparate Legacy of Broken Windows Policing in New York City [2018], supra), and all too frequent news reports—the risks attendant with escalation from levels one through four of De Bour fall heaviest on people of color and the mentally ill, sometimes with tragic consequences (e.g. NYPD Housing Officers Beaten with Baton in Brooklyn, Suspect Shot and Killed, CBS NY, Nov. 19, 2016, available at http://newyork.cbslocal.com/2016/11/19/officers-hurt-in-brooklyn-shooting/ [last accessed May 15, 2023] [officers responding to a 9-1-1 call about a "suspicious man" in a public housing complex approached the decedent and asked him for identification, after which a scuffle ensued, the decedent struck one of the officers with the officer's own baton, and the officers fatally shot the decedent—who had a history of depression and addiction]).

By encouraging more police-initiated interactions with private individuals that risk dangerous, sometimes fatal outcomes for both individuals and officers, De Bour's method of balancing public safety and the individual right to be left alone has proved counterproductive to both (e.g. John Dias et al., NYPD: Devin Spraggins, 22, Arrested in Shooting of Rookie Officer Brett Boller in Queens, CBS NY, Apr. 7, 2023, available at https://www.cbsnews.com/newyork/news/devin-spraggins-arrested-in-shooting-of-nypd-officer-brett-boller-in-jamaica-queens/ [last accessed May 15, 2023] [defendant shot police officer who boarded a bus and confronted him in response to a report of an altercation between passengers]; Suspect Shot and Killed after NYPD Pursuit Ends in Crash, Closes Saw Mill River Parkway, ABC7: Eyewitness News, Dec. 8, 2015, available at https://abc7ny.com/westchester-yonkers-crash-nypd/1114221/ [last accessed May 15, 2023] [following chase of an unarmed driver who did not stop when commanded, police shot and killed him]; NYPD 'Inadvertently' Shot Dead Unarmed Man in Brooklyn after Dispute, Guardian, Oct. 2, 2014, available at https://www.theguardian.com/us-news/2014/oct/02/nypd-inadvertently-killed-unarmed-man-brooklyn-shooting [last accessed May 15, 2023] [man who intervened in an intimate-{**40 NY3d at 194}partner incident involving a man with a knife killed in an ensuing "hail of police gunfire"]; Joseph Goldstein & Marc Santora, Staten Island Man Died from Chokehold during Arrest, Autopsy Finds, NY Times, Aug. 1, 2014, available at https://www.nytimes.com/2014/08/02/nyregion/staten-island-man-died-from-officers-chokehold-autopsy-finds.html [officers approached decedent and questioned him on suspicion of selling unlicensed cigarettes and, after the man denied the allegations and accused the officers of harassing him, an officer placed him in a chokehold which killed him]). Empirical studies confirm these anecdotes: police-initiated encounters with members of the public increase the risks for everyone involved in them (see Darrel W. Stephens, Officer Involved Shootings: Officers/Subjects Executive Summary at 3, National Police Foundation [Oct. 2019], available at [*14]https://www.policinginstitute.org/wp-content/uploads/2019/05/2.-OIS_off_sub_8.28.19.pdf [last accessed May 15, 2023] [study finding that, nationwide, officers suffered injuries in 17.4% of police shootings following police-initiated encounters, compared to 12.2% in those initiated by citizens]; National Police Foundation, Research Summary: Non-Fatal Injuries to Law Enforcement Officers [Mar. 2018], available at https://www.policinginstitute.org/wp-content/uploads/2018/07/Non-fatal-Injuries-to-LEOs-Research-Summary-2018.pdf [last accessed May 15, 2023] [summarizing an academic study showing that, between 2003 and 2014, the frequency of non-fatal injuries to police officers far outpaced that of American workers generally and had steadily increased for over a decade]).

Simply stated, De Bour's framework has proved itself to be "an unworkable" one that "creates more questions than it resolves" (People v Taylor, 9 NY3d 129, 149 [2007]), "no longer serves the ends of justice or withstands the cold light of logic and experience" (Policano v Herbert, 7 NY3d 588, 604 [2006] [internal quotation marks omitted]), and has simply been undermined by "the 'lessons of experience and the force of better reasoning' " (People v Bing, 76 NY2d 331, 338 [1990], quoting Burnet v Coronado Oil & Gas Co., 285 US 393, 407-408 [1932, Brandeis, J., dissenting]). Although the Court viewed De Bour's approach as sensible when adopting it in 1976, and when reaffirming it in 1992, its application during the many years since has diminished its vitality, to the point that the events of the present now overshadow the good intentions of the past. The De Bour framework has calcified into an "archaic {**40 NY3d at 195}and obsolete doctrine which has lost its touch with reality" (People v Hobson, 39 NY2d 479, 487 [1976]). Our experiences with De Bour supply a "compelling justification" for adopting an approach to police-initiated encounters that better safeguards the right to be let alone, public safety and officer security (People v Peque, 22 NY3d 168, 194 [2013]).

To better ensure public safety and protect the rights of individuals and diverse communities from abusive and ineffective policing, we should require all police-initiated encounters to be justified on reasonable suspicion and actual signs of criminality—i.e., reasonable suspicion of an ongoing or completed felony or misdemeanor. Absent such cause for a criminal investigation, the law should not permit officers to approach private individuals to request information or engage in what De Bour termed a "common-law right to inquire" (40 NY2d at 223).

This wiser approach values autonomy and avoids "substitut[ing] labels for liberties" (Samuels, 50 NY2d at 1040 [Fuchsberg, J., dissenting]). It also leaves ample room for police action at times of heightened concern for public safety. For example, if an officer receives a reliable tip from an anonymous source or a confidential informant that a person has committed, is about to commit or planning a felony or misdemeanor, the officer would remain able to approach and inquire (see Florida v J.L., 529 US 266, 270-271 [2000]; People v Argyris, 24 NY3d 1138, 1140 [2014]). Or if a person approaches an officer and identifies someone on the street as involved in a crime, the officer may approach and inquire (People v Moore, 32 NY2d 67, 69 [1973]). And, of course, if an officer personally observes conduct that supplies reasonable suspicion that crime is afoot—not merely based on an incredulous view of innocent conduct, as was the case here—an officer may approach (id.). The Court has found such police interactions to be lawful under levels three and four of De Bour.

This is also not to say that such a rule would prohibit an officer from approaching to seek assistance in gaining information unrelated to whether the person approached is engaged in [*15]criminality. For example, an officer seeking information about a missing person would still be able to approach individuals to ask if they had seen or heard of the missing individual. That is good policing and in furtherance of public safety goals.{**40 NY3d at 196}

The federal approach is not a viable replacement for De Bour. As the Court recognized in Hollman, the Terry standard would be inconsistent with "our judgment that encounters that fall short of Fourth Amendment seizures still implicate the privacy interests of all citizens and that the spirit underlying those words required the adoption of a State common-law method to protect the individual from arbitrary or intimidating police conduct" (79 NY2d at 195). In other words, the federal approach does not reduce the risks that rapidly-escalating, violent encounters pose for individuals and officers.

Justice Sotomayor has aptly described the flaws in the federal approach. In Utah v Strieff, Justice Sotomayor powerfully described how the United States Supreme Court "has allowed an officer to stop you for whatever reason [they] want[ ]—so long as [they] can point to a pretextual justification after the fact" and reminded us that, under Terry and its progeny, although the requisite "justification must provide specific reasons why the officer suspected you were breaking the law, . . . it may factor in your ethnicity, . . . where you live, . . . what you were wearing, . . . and how you behaved" (579 US 232, 252 [2016, Sotomayor, J., dissenting] [citations omitted]). Justice Sotomayor discussed the dreadful consequences that attend this conduct, which the Fourth Amendment tolerates, including intrusive frisks, possible handcuffing and jailing, fingerprinting and DNA collection, and—even if innocent—an arrest record (id. at 253). The Justice also observed that "[f]or generations, black and brown parents have given their children 'the talk'—instructing them never to run down the street; always keep your hands where they can be seen; do not even think of talking back to a stranger—all out of fear of how an officer with a gun will react to them" (id. at 254). The Supreme Court's post-Terry reading of the Fourth Amendment, Justice Sotomayor concluded, "implies that you are not a citizen of a democracy but the subject of a carceral state, just waiting to be cataloged" (id.)

Legal scholars have vigorously critiqued this evolution in the Supreme Court's Fourth Amendment jurisprudence. For example, one scholar and former federal prosecutor has colorfully described how the Supreme Court had "greased the wheels of . . . controversial" stop-and-frisk tactics for which Terry and its progeny left room (Julian A. Cook, III, Suspicionless Policing, 89 Geo Wash L Rev 1568, 1580 [2021]). Another scholar and court reform expert has suggested that, while Terry may{**40 NY3d at 197} "create a clearer standard" and "reflect actual police behavior more realistically" than De Bour, Terry "skew[s] further the already imperfect balance between law enforcement and individual privacy interests" and that "the history of federal constitutional law demonstrates that this model creates a constant pressure to expand the area of police behavior not subject to constitutional scrutiny" (Sack, 66 NYU L Rev at 553-554).

New Yorkers need not be left vulnerable to subconstitutional abusive police conduct. The way forward is to simply adopt the standard Mr. De Bour proposed decades ago—"reasonable suspicion based on concrete observations"—before initiating encounters with individuals (De Bour, 40 NY2d at 216). Time and experience have proved this to be the just and fair approach.

The De Bour Court adopted a tiered approach, intending for it to provide some minimal protections against aggressive sub-Terry police investigative encounters. The right to privacy and the interest in public security that animated De Bour's analysis are no less significant today than they were almost 50 years ago. However, despite the De Bour Court's best intentions and earnest effort in 1976, the framework it adopted has failed to generate the predictability and precision in judicial review for which it strove. Instead, experience has shown that, because the bar is set so low, courts often treat an officer's speculation or hunch as "some objective credible reason" for an approach under level one. Similarly, level two is based on a right to inquire that permits targeted questioning without a criminal activity nexus. Whether driven by animus or implicit bias, the results are the erosion of the right to be left alone and increased risks to the safety of officers and private individuals.

So, we come full circle. Drawing on new wisdom in the form of accounts from persons in heavily policed communities and years of research, we should hold that police-initiated encounters with individuals are unconstitutional absent reasonable suspicion of criminality—exactly the approach Mr. De Bour suggested in 1976.

Judges Garcia, Singas, Cannataro and Troutman concur. Judge Rivera concurs in result in an opinion. Judge Halligan took no part.

Order reversed and indictment dismissed.