Harvard Law Review Harvard Law Review Harvard Law Review

Fourth Amendment

State v. Kelly

Connecticut Supreme Court Upholds Suspicionless Street Stop of Suspect's Companion.

Under the Fourth Amendment, a search or seizure must be “reasonable.”1×1. U.S. Const. amend. IV (“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 . . . .”). Traditionally, it has been presumed that the police meet this requirement only if they act with a warrant based on probable cause.2×2. See Skinner v. Ry. Labor Execs.’ Ass’n, 489 U.S. 602, 619 (1989). In Terry v. Ohio,3×3. 392 U.S. 1 (1968). this warrant presumption yielded to a general reasonableness test under which the Supreme Court balanced the government and individual interests involved in the search and seizure.4×4. See id. at 20–21. The Court held that police officers may briefly detain and pat down a suspect on the street based only on a “reasonable suspicion”5×5. “Reasonable suspicion” was actually a post-Terry formulation that soon became the canonical Terry standard. See, e.g., Almeida-Sanchez v. United States, 413 U.S. 266, 268 (1973). that the suspect was armed and committing a crime.6×6. Terry, 392 U.S. at 30. The Court thought stop-and-frisks “as a practical matter could not be[] subjected to” a warrant requirement, id. at 20, but “emphatically reject[ed] th[e] notion” that stop-and-frisks lie “outside the purview of the Fourth Amendment,” id. at 16. Since Terry, courts have used the reasonableness balancing test to craft several more exceptions to the warrant requirement, widening what one scholar discussing an early post-Terry case called “the kind of very small hole . . . which customarily begins the process by which entire tapestries unravel.”7×7. Anthony G. Amsterdam, Perspectives on the Fourth Amendment, 58 Minn. L. Rev. 349, 374 (1974). For a lucid overview of the Supreme Court’s shift away from the warrant requirement and toward reasonableness balancing, see Nadine Strossen, The Fourth Amendment in the Balance: Accurately Setting the Scales Through the Least Intrusive Alternative Analysis, 63 N.Y.U. L. Rev. 1173, 1178–84 (1988).

Recently, in State v. Kelly,8×8. 95 A.3d 1081 (Conn. 2014). the Connecticut Supreme Court further unraveled the tapestry when it held that the state’s police officers may seize a legitimate suspect’s “companion” if the officers “reasonably believe that the suspect presents a threat to their safety.”9×9. Id. at 1100 (emphasis added). In other words, an officer may detain the companion even if the officer has no reasonable suspicion that the companion “himself was involved in criminal activity or was armed and dangerous.” Id. at 1102. Eclipsing Terry’s holding while embracing the balancing test that Terry inaugurated, the court carefully detailed the state’s “weighty interest in ensuring officer safety”10×10. Id. at 1095. but scarcely explained the assertion that police stops impose a mere “inconvenience[]”11×11. Id. at 1102 (quoting State v. Kelly, 19 A.3d 223, 232 (Conn. App. Ct. 2011)). and a “relatively limited intrusion”12×12. Id. at 1094. on their targets. The individual interests at stake are much greater than the court acknowledged, and in an era of Fourth Amendment balancing, courts must account more fully for the factors weighing on this side of the scale.

Driving in an unmarked car around 11 a.m., Hartford police officers William Rivera and Jose Angeles saw Rafael Burgos and Jeremy Kelly walking together near a gas station that was a “known location for drug dealing.”13×13. Kelly, 19 A.3d at 225. Thinking Burgos might be Pedro Gomez, a man with an arrest warrant out for violating his probation, Rivera stopped the car, displayed his badge, and told the men to come to the vehicle.14×14. Id. at 225–26. Kelly and Burgos kept walking.15×15. Id. at 226. When Angeles then repeated Rivera’s order, Burgos and Kelly ran.16×16. Id. The officers chased on foot, and Kelly dropped a clear plastic bag of cocaine.17×17. See id. Rivera eventually tackled and arrested Kelly; in the process, he found another bag of cocaine on Kelly’s person.18×18. See id.

Kelly was charged with various drug-related crimes.19×19. Kelly, 95 A.3d at 1086. At trial, he moved to suppress the cocaine, claiming his seizure was illegal because the officers initially had no reasonable suspicion that he had committed a crime.20×20. Id. The trial court denied the motion: because the officers had reasonably stopped Burgos, thinking he was the wanted and potentially armed Gomez,21×21. The trial court erroneously found that Gomez was wanted for felony possession of a firearm; he was actually wanted for an unspecified probation violation. Id. at 1086 n.2. The Connecticut Supreme Court held this error “harmless,” id., because the trial court had confirmed that it had nevertheless “credited” testimony about the true nature of Gomez’s warrant and about an informant’s tip to the officers that Gomez likely had a gun — rendering the point “moot,” see id. at 1101–02. they were justified on safety grounds in briefly stopping Kelly too.22×22. Id. at 1086. The stop was therefore permissible under the U.S. and Connecticut Constitutions.23×23. See id. at 1086–87. The state and federal questions in Kelly largely overlapped but diverged on the question of when Kelly was “seized.” Under Connecticut law, Rivera officially seized Kelly when he showed his badge and ordered Kelly to the car. Kelly, 19 A.3d at 227–28. Under federal law, Kelly was seized only when Rivera tackled him. Id. at 228 n.6; see also California v. Hodari D., 499 U.S. 621, 626 (1991). But once it is established that there was a seizure, Connecticut courts look to the same Fourth Amendment doctrine that federal courts do to determine whether that seizure was reasonable. See State v. Oquendo, 613 A.2d 1300, 1311 (Conn. 1992). So most of the analysis in Kelly turned on federal Fourth Amendment precedent. See Kelly, 95 A.3d at 1091. (This body of doctrine did not help Kelly with his federal claims because Rivera clearly had probable cause to seize Kelly once he saw Kelly drop a plastic bag. Kelly, 19 A.3d at 228 n.6.) Kelly pleaded nolo contendere conditioned on his right to appeal the denial; he was ordered to serve three and a half years of a nine-year prison sentence and three years of probation.24×24. Kelly, 19 A.3d at 226.

The Connecticut Appellate Court affirmed. Acknowledging that the U.S. Supreme Court had yet to sanction a suspicionless street stop,25×25. See id. at 229 (calling Kelly “a matter of first impression for appellate review”). it analogized to precedent in which the Court, putting heavy weight on officer safety in applying the reasonableness balancing test, had found permissible the suspicionless seizures of certain car passengers26×26. See id. at 229, 231 (citing Maryland v. Wilson, 519 U.S. 408, 413–15 (1997)). and occupants of houses.27×27. See id. at 229–32 (citing Maryland v. Buie, 494 U.S. 325, 333 (1990); Michigan v. Summers, 452 U.S. 692, 702–03, 705 (1981)). Taken together, these decisions convinced the Appellate Court that “the interest in the officers’ safety during the investigatory stop of Burgos outweighed [Kelly’s] personal liberty interest in not being inconvenienced.”28×28. Id. at 232.

The Connecticut Supreme Court affirmed. Writing for the majority, Justice Palmer29×29. Chief Justice Rogers and Justices Norcott, Zarella, and Vertefeuille joined the opinion. acknowledged the traditional presumption in favor of requiring a warrant but cited Terry as an exemplar of the proposition that “reasonableness” is the ultimate Fourth Amendment criterion.30×30. See Kelly, 95 A.3d at 1091–94. The court was therefore not bound to apply any particular threshold standard: “[I]ndividualized suspicion is not an absolute prerequisite for every constitutional search or seizure.”31×31. Id. at 1094 (quoting United States v. Lewis, 674 F.3d 1298, 1305 (11th Cir. 2012)) (internal quotation marks omitted). Rather, the court was free to resolve Kelly’s case by conducting its own balancing analysis.32×32. See id.

Embarking on this test, the court asserted that Kelly’s seizure “represent[ed] a relatively limited intrusion”33×33. Id. into his personal liberty, while “the state has a weighty interest in ensuring officer safety.”34×34. Id. at 1095. Echoing the Appellate Court, Justice Palmer highlighted the same U.S. Supreme Court decisions permitting suspicionless seizures during traffic stops and during the execution of search warrants in houses — all in the interest of officer safety.35×35. See id. at 1093, 1095–99 (discussing Maryland v. Wilson, 519 U.S. 408 (1997); Maryland v. Buie, 494 U.S. 325 (1990); and Michigan v. Summers, 452 U.S. 692 (1981)). The court also cited numerous decisions from other state courts in support of its holding. Id. at 1099. These cases “strongly support[ed] the conclusion” that the police may, “as a reasonable safety measure,”36×36. Id. at 1098. stop the companion of a lawfully detained suspect when they “reasonably believe that the suspect is armed and dangerous.”37×37. Id. at 1099. The court also noted that its decision “permits only the brief detention of the defendant and does not entail or authorize a search of the defendant.” Id. at 1098 n.20.

The court also entertained and quickly dismissed the “sociological and policy” concern that allowing stops like Kelly’s would “lead to abuse by overzealous officers.”38×38. Id. at 1099. The court did “not believe that any such risk outweighs the state’s strong interest in officer safety,”39×39. Id. and it considered the state’s courts “well equipped to address any claim of police impropriety” during a Kelly-type seizure.40×40. Id. at 1099–100. All in all, Justice Palmer concluded, the balance fell in favor of “permitting the protective stop of a suspect’s companion when the officer has reason to be concerned for his safety.”41×41. Id. at 1100. Unwilling to go further than this, however, Justice Palmer declined the state’s invitation to adopt a bright-line rule categorically allowing “an officer to detain a suspect’s companion . . . regardless of whether the officer is reasonably concerned for his safety.” Id.

Justice Eveleigh42×42. Justice McDonald joined Justice Eveleigh in dissent. began his spirited dissent by quoting Benjamin Franklin’s admonition that “[t]hose who would give up essential [l]iberty, to purchase a little temporary [s]afety, deserve neither [l]iberty nor [s]afety.”43×43. Kelly, 95 A.3d at 1103 (Eveleigh, J., dissenting) (alterations in original) (quoting Pennsylvania Assembly: Reply to the Governor (Nov. 11, 1755), in 6 The Papers of Benjamin Franklin 238, 242 (Leonard W. Labaree ed., 1963)). He disputed the majority’s handling of precedent, contending that Ybarra v. Illinois44×44. 444 U.S. 85 (1979). — which held unlawful the suspicionless search of a bar patron during the execution of a search warrant aimed at the bar and its bartender45×45. See id. at 88, 96. — best applied to Kelly’s situation.46×46. See Kelly, 95 A.3d at 1104–06 (Eveleigh, J., dissenting). He then faulted the majority for relying on cases that involved searches or seizures in either “a car or a house,” both “confined area[s]”47×47. Id. at 1111. where the police act to protect their own safety, need to prevent the removal of evidence, and see clear indications of the detainees’ association with the suspects.48×48. See id. at 1109–11. Kelly, by contrast, was detained in an open area where he could easily have been allowed to “proceed on his way,” was in no position to remove evidence, and betrayed no hint of his relationship to Burgos “beyond the fact that they were walking together for a brief period of time.”49×49. Id. at 1111. Justice Eveleigh also distinguished the majority’s state precedents, see id. at 1112–13, arguing that the police in those cases had invariably been “able to point to particularized facts about the companion that amounted to a reasonable fear for their safety,” id. at 1113. Under Kelly’s holding, Justice Eveleigh warned, the police can presumably seize anyone they deem a suspect’s “companion,” including everyone in a high-crime neighborhood.50×50. Id. at 1113. The majority responded that “the police must reasonably believe that the [detained] person is accompanying the suspect.” Id. at 1097 n.19 (majority opinion). He would, he concluded, “require more than mere ‘guilt by association.’”51×51. Id. at 1113 (Eveleigh, J., dissenting).

The Connecticut Supreme Court’s approach in Kelly appears to reflect a growing judicial tendency to look past the Fourth Amendment’s warrant requirement — and the “jealously and carefully drawn” exceptions to that rule52×52. Jones v. United States, 357 U.S. 493, 499 (1958). — and to apply instead the reasonableness balancing test.53×53. See Strossen, supra note 7, at 1178–84; see also United States v. Place, 462 U.S. 696, 721 (1983) (Blackmun, J., concurring in the judgment) (lamenting “an emerging tendency . . . to convert the Terry decision into a general statement that the Fourth Amendment requires only that any seizure be reasonable”); Tracey Maclin, The Central Meaning of the Fourth Amendment, 35 Wm. & Mary L. Rev. 197, 202–07 (1993) (documenting the Court’s shift from a “warrant preference,” id. at 203 (internal quotation marks omitted), to a “rational basis standard,” id. at 206). If courts are indeed irreversibly set on adopting this indeterminate approach, then they should assume a corresponding duty to more rigorously examine the individual interests at stake.

By many accounts, reasonableness balancing has largely proved a boon to the state at the expense of the individual. For one scholar, it is a “freewheeling” method of Fourth Amendment analysis, which may lead judges to ignore the obvious dangers of expanded police discretion.54×54. See Carol S. Steiker, Terry Unbound, 82 Miss. L.J. 329, 348 (2013); see also id. at 342 (discussing Chief Justice Rehnquist’s preference for reasonableness balancing and his “dismissive attitude” toward allegations of intrusive police conduct). Another examines how “the Court’s balancing holdings have restricted rather than expanded individual rights.”55×55. Strossen, supra note 7, at 1194 n.98; see id. at 1194–207; see also Shima Baradaran, Rebalancing the Fourth Amendment, 102 Geo. L.J. 1, 14–29 (2013) (providing a detailed analysis of balancing in Supreme Court criminal procedure cases and concluding that “government interests typically trump individual rights,” id. at 15). And a dissenting Supreme Court Justice has voiced concern that in the Court’s use of balancing, “the judicial thumb apparently will be planted firmly on the law enforcement side of the scales.”56×56. United States v. Sharpe, 470 U.S. 675, 720 (1985) (Brennan, J., dissenting).

At the same time, other accounts suggest that a reasonableness test is, at least potentially, capacious enough to accommodate the moral and imaginative thinking that would illuminate the individual interests under threat. For instance, one scholar argues that an emphasis on Fourth Amendment reasonableness invites courts to “honestly and openly” discuss an array of constitutional concerns, including questions about racism in policing.57×57. Akhil Reed Amar, Fourth Amendment First Principles, 107 Harv. L. Rev. 757, 808 (1994); see id. at 804–11; see also Sherry F. Colb, Innocence, Privacy, and Targeting in Fourth Amendment Jurisprudence, 96 Colum. L. Rev. 1456, 1464–65 (1996) (noting that the Fourth Amendment’s “open-ended term ‘unreasonable’” invites “moral inquiry [to] assume[] a central role,” id. at 1465). Another scholar, though less sanguine about the Reasonableness Clause, argues that it “require[s] a fairly high level of abstraction of purpose” and “positively invites constructions that change with changing circumstances” — the growing imperative to address racially biased policing being perhaps the most prominent of those circumstances.58×58. See Carol S. Steiker, Response, Second Thoughts About First Principles, 107 Harv. L. Rev. 820, 824 (1994); cf. T. Alexander Aleinikoff, Constitutional Law in the Age of Balancing, 96 Yale L.J. 943, 984 (1987) (“A common objection to balancing as a method of constitutional adjudication is that it appears to replicate the job that a democratic society demands of its legislature.”).

If balancing is here to stay, then those calling for greater focus on individual interests in Fourth Amendment cases must learn to speak the language of balancing. Professor Shima Baradaran provides a model: in a recent article, she proposes that judges consider more evidence from the social sciences so as to avoid “blind balancing,” “the process of decision making based simply on common sense and a gut assessment of risk, without consideration of data, evidence, or empirical studies.”59×59. Baradaran, supra note 55, at 3; see id. at 3–8. Such creative ideas should be further explored. In the meantime, courts should take this moment to recognize their duty to give serious attention to the “imponderable weights” that sit in both of the Fourth Amendment’s “pans of judgment.”60×60. Amsterdam, supra note 7, at 354. In the context of street stops especially, courts should “abandon a naive theory of the Fourth Amendment, and consider the real world that exists on the street”61×61. Tracey Maclin, “Black and Blue Encounters” — Some Preliminary Thoughts About Fourth Amendment Seizures: Should Race Matter?, 26 Val. U. L. Rev. 243, 279 (1991). when assessing the individual interests in question.

A call for greater focus on individual interests would at a minimum ask courts to fully examine the indignities caused and resentments aroused by nearly any police stop62×62. See, e.g., Terry v. Ohio, 392 U.S. 1, 16–17 (1968) (calling it “simply fantastic,” id. at 16, to think a public pat search anything less than “a serious intrusion . . . which may inflict great indignity and arouse strong resentment,” id. at 17); Brinegar v. United States, 338 U.S. 160, 180–81 (1949) (Jackson, J., dissenting) (identifying “unheralded search and seizure,” id. at 181, as among the best tools for “cowing a population, crushing the spirit of the individual and putting terror in every heart,” id. at 180); Charles A. Reich, Police Questioning of Law Abiding Citizens, 75 Yale L.J. 1161, 1161 (1966) (describing a Yale Law professor’s indignation when two officers “summoned [him] off the sidewalk without even getting out of their patrol car,” as Officers Rivera and Angeles did Kelly). instead of insisting, like the Kelly court, that the typical suspicionless seizure is “a relatively limited intrusion”63×63. Kelly, 95 A.3d at 1094. and a mere “inconvenience[].”64×64. Id. at 1102 (quoting State v. Kelly, 19 A.3d 223, 232 (Conn. App. Ct. 2011)). More pressingly, this approach would involve a serious examination of the grave risks that police stops often pose in minority communities. For instance, recent, well-publicized data from 2004 to 2012 suggest that New York City police officers who were encouraged to increase their discretionary stops singled out and harassed blacks and Hispanics.65×65. See Floyd v. City of New York, 959 F. Supp. 2d 540, 562, 573, 574 (S.D.N.Y. 2013) (citing evidence that blacks and Hispanics accounted for 52% of the city’s population but 83% of the city’s 4.4 million stop-and-frisks from January 2004 to June 2012, id. at 573, 574, and that “blacks and Hispanics are more likely to be subjected to the use of force than whites, despite the fact that whites are more likely to be found with weapons or contraband,” id. at 562). And courts will find no shortage of seemingly routine stops that ended in physical harm or even death.66×66. See, e.g., Jesse McKinley, In California, Protests After Man Dies at Hands of Transit Police, N.Y. Times, Jan. 8, 2009, http://www.nytimes.com/2009/01/09/us/09oakland.html. See generally Jaeah Lee, Exactly How Often Do Police Shoot Unarmed Black Men?, Mother Jones (Aug. 15, 2014, 6:00 AM), http://www.motherjones.com/politics/2014/08/police-shootings-michael-brown-ferguson-black-men [http://perma.cc/9NVA-FL4V]. Notably, last year’s police killing of Michael Brown had just started roiling Ferguson, Missouri, and sparking a nationwide discussion about race and policing, when the Kelly decision came down. See Frances Robles & Michael S. Schmidt, Shooting Accounts Differ as Holder Schedules Visit to Ferguson, N.Y. Times, Aug. 29, 2014, http://www.nytimes.com/2014/08/20/us/shooting-accounts-differ-as-holder-schedules-visit.html.

A commitment to more rigorous balancing would also take into account the possibility of community-level harms, which burden individuals less directly but no less severely. Repeated stops by aggressive officers can, for example, stigmatize individuals and reverberate to crack the social bonds of whole neighborhoods.67×67. See Alice Goffman, On the Run 5–7 (2014) (reporting how a black neighborhood “where police circle overhead,” id. at 5, tags its residents “clean” or “dirty,” id. at 6 (emphasis omitted), depending on whether the person is likely to survive a police stop without being arrested, thereby deepening social divides). And most judges will be familiar with the notion that aggressive policing can cause entire communities to fear the social servants who are supposed to protect them.68×68. See, e.g., Ta-Nehisi Coates, Reparations for Ferguson, The Atlantic (Aug. 18, 2014, 4:29 PM), http://www.theatlantic.com/national/archive/2014/08/Reparations-For-Ferguson/376098 [http://perma.cc/WFQ7-3638] (“Black people . . . call[] the police . . . fully understanding that we are introducing an element that is unaccountable to us. We introduce the police into our communities, the way you might introduce a predator into the food chain.”). Each resident of a police-cowed community is stripped of “[w]hat the [F]ourth [A]mendment protects above all[:] the conduct of ordinary lives.”69×69. Lloyd L. Weinreb, Generalities of the Fourth Amendment, 42 U. Chi. L. Rev. 47, 85 (1974).

As one concrete step forward, the same courts that regularly conceive imaginative hypotheticals to spotlight the state’s legitimate interest in protecting police officers70×70. See Baradaran, supra note 55, at 21–22 & n.119 (collecting examples of “unfounded hypothetical[s]” supporting decisions to uphold searches, including one court’s comment that “[a]n ordinary pen or pencil, when plunged into the neck of the police officer . . . would have been as lethal as any hand gun,” id. at 22 (first alteration in second quotation in original) (quoting State v. Lombardi, 727 A.2d 670, 674 (R.I. 1999)) (internal quotation marks omitted)). should look also at readily available examples of individual interests under threat. In Kelly, the Connecticut Supreme Court raised the specter of a companion opening fire on police officers71×71. See Kelly, 95 A.3d at 1087; see also id. at 1093 (pointing to the U.S. Supreme Court’s vision of armed house occupants ambushing officers executing a search warrant (citing Maryland v. Buie, 494 U.S. 325, 333–36 (1990))). but assessed the potential risk to the companion only in abstract terms. Why should the court not have consulted, for instance, a dashboard recording72×72. Patrol Car Video, Pensacola News J., http://archive.pnj.com/section/videonetwork/?bctid=68642876001 (last visited Nov. 23, 2014). that shows an overzealous officer chasing teenager Victor Steen on his bicycle — because, said the officer, Steen had no light on his bike — and, after firing his Taser from his cruiser window, accidentally running over Steen and killing him?73×73. See Meg Laughlin, Death of Teen on Bike Shows Risks of Expanded Use of Tasers, Tampa Bay Times (July 30, 2010, 3:25 PM), http://www.tampabay.com/news/publicsafety/death-of-teen-on-bike-shows-risks-of-expanded-use-of-tasers/1112106 [http://perma.cc/5DXH-GPPN]. What if expanding discretion to stop people on the street emboldens officers to make such trivial stops more frequently, and young men who may harbor a “reasonable suspicion” of the police decide to keep moving when ordered to stop,74×74. See Illinois v. Wardlow, 528 U.S. 119, 132 (2000) (Stevens, J., concurring in part and dissenting in part) (noting that in some circumstances, even innocent people may flee, “believ[ing] that contact with the police can itself be dangerous”). and more Victor Steens end up harassed or hurt or dead? Would considering such worst-case (but real-life) scenarios inspire courts to tilt the scales a little further toward the individual than the Connecticut Supreme Court did?

Connecticut’s courts will likely have several opportunities to more evenly load the scales, for the Kelly holding was narrow enough to leave several questions unanswered. For instance, can an officer constitutionally search the companion as well as seize him? The decision ostensibly “permits only the brief detention of the defendant.”75×75. Kelly, 95 A.3d at 1098 n.20. But if an officer fears for her safety, then surely she will want to frisk the companion for weapons too. In the future, when Connecticut’s judges consider this question and others, and they balance the state and individual interests at stake, they must hold up for full inspection the well-documented risks that individual Americans rely on the Fourth Amendment to curb.