United States v. Massenburg, 654 F.3d 480 (4th Cir. 2011)

Vacated, reversed, and remanded by-published opinion. Judge DAVIS wrote the opinion, in which Judge MOTZ and Judge KEENAN joined.

OPINION

DAVIS, Circuit Judge:

In this appeal from a judgment of sentence, we are once again called on to determine whether evidence seized during a street encounter between law enforcement and citizens was properly admitted into evidence during a subsequent criminal prosecution. We conclude that the seizure of the evidence did not comport with settled Fourth Amendment principles, and we therefore reverse the district court’s denial of appellant’s motion to suppress and remand for further proceedings.

Responding one night to an anonymous tip that shots were fired in a high-crime neighborhood, Richmond police encountered four young men, including appellant Tyerail Massenburg, four blocks from the reported gunfire. When an officer approached them in a marked police car, the men were not evasive; they continued walking forward, toward the car, and voluntarily paused to speak with the officer upon the officer’s request. In fact, they were cooperative: one of the men reported that he had heard shots fired from a passing car two blocks away and handed over his identification when asked; and at least two of the men consented to voluntary pat-downs. Appellant Massenburg stopped with his friends, but he refused to consent to a frisk. As the officer interacting with Massenburg testified, he first thought Massenburg nervous when he began asking him to consent to a pat-down and Massenburg was “real reluctant to give consent.” J.A. 48. Based on the fact that appellant stood a foot or two away from the other men, who were shoulder-to-shoulder, and did not make eye contact as the officer renewed his requests for a consensual search, the officer undertook a nonconsensual search. The search produced a firearm and some marijuana, the subjects of the suppression motion at issue here.

Charged with one count of possession of a firearm by a drug user under 18 U.S.C. § 922(g)(3) and one count of possession of marijuana under 21 U.S.C. § 844, Massenburg moved to suppress the gun and drugs on the ground that the officer’s frisk was unlawful. The district court denied that motion, and Massenburg entered a conditional guilty plea, reserving his right to appeal the suppression ruling.

Before an officer can stop and frisk a citizen, she must have “reasonable and articulable suspicion that the person seized is engaged in criminal activity.” Reid v. Georgia, 448 U.S. 438, 440, 100 S.Ct. 2752, 65 L.Ed.2d 890 (1980). We recently warned against the Government’s proffering “whatever facts are present, no matter how innocent, as indicia of suspicious activity” and noted that we were “deeply troubled by the way in which the Government attempts to spin ... mundane acts into a web of deception.” United States v. Foster, 634 F.3d 243, 248 (4th Cir.2011). This concern is only heightened when the “mundane acts” emerge from the refusal to consent to a voluntary search. If the important limitations on the “stop and frisk” regime crafted by Terry v. Ohio, 392 U.S. 1, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968), are not to become dead letters, refusing to consent to a search cannot itself justify a nonconsensual search.

I.

A.

On the night of March 28, 2009, at 10:33 p.m., Richmond City Police received an anonymous tip that shots had just been

fired. The caller reported eight shots fired “possibly” two blocks south of 14th and Hull Streets, a high-crime area in which “drug activity as well as random gunfire” were “usual[ ].” J.A. 46, 77. The caller said nothing more; in particular, he or she included no description of a suspect.

Officers Stephen Gaines and Eric Fries responded to the call and arrived at 14th and Hull at 10:48 p.m. They split up and patrolled the area in marked police cars. Fries soon saw four young black men, including appellant Massenburg, walking north at the corner of East 17th Street and Stockton Street, four blocks west and two south of the intersection of 14th and Hull and thus four blocks from the alleged origin of the shots. They were walking in the direction of Fries’s marked car and did not stop or change course when they saw it.

Fries approached in his vehicle and asked, “hey guys, can you stop for a second?” J.A. 31. The men stopped to talk with him. Fries asked if they had heard gunfire, and one man reported hearing shots fired from a vehicle on Maury Street, two blocks south of their present location. Gaines arrived, the two officers exited their vehicles, and they began taking the men’s names. Fries then asked if they had weapons on them and if they would consent to a pat-down. The four men were now “all basically lined up in a row on the sidewalk,” with the man who reported hearing gunfire on Maury Street on the left end of the line and Massenburg on the right. J.A. 32. According to Gaines, the three left-most men were “pretty much shoulder-to-shoulder, and [Massenburg] was kind of offset from the group” by a “foot or two,” “give or take.” J.A. 57.

The man on the left consented to Fries’s request for a pat-down, as did the man nearest him. Gaines began at the other end of the line, asking Massenburg if he would consent to a frisk. Gaines testified that Massenburg, in reply to the request, “was kind of hesitant and stand-offish, and kind of real reluctant to give consent to a pat down or a search of his person.” J.A. 48. Instead, “[h]e stated he did not have anything. You don’t need to check me. Stood back and kind of air-patted himself down, stating, trying to show he didn’t have anything.” J.A. 48. At this point Gaines insisted and patted Massenburg down without his consent.

Officer Fries testified that he had seen “a small bulge in the left jacket pocket of Mr. Massenburg” prior to Officer Gaines’s frisk, but he “didn’t alert” Gaines to it. J.A. 32, 42. Officer Gaines, asked multiple times about the basis for his suspicion of Massenburg, never indicated in his testimony that he saw a sign or signal from Fries.

During the frisk of Massenburg, Gaines felt the handle of a firearm on Massenburg’s waist band (not in the jacket), and Massenburg fled before Gaines could grab it. Gaines pursued and directed him to drop the firearm, which Massenburg did, dropping it on the grass. Massenburg ran another 250 feet before Gaines caught up and arrested him. In addition to the firearm, police recovered a small amount of marijuana on Massenburg’s person.

B.

Massenburg was charged with one count of possession of a firearm by a drug user, in violation of 18 U.S.C. § 922(g)(3), and one count of possession of marijuana, in violation of 21 U.S.C. § 844. He filed a motion to suppress the firearm and marijuana, arguing that Gaines lacked the reasonable, particularized suspicion that he was engaged in criminal activity necessary to authorize a nonconsensual frisk under the Fourth and Fourteenth Amendments.

At the suppression hearing, the Government presented few objective bases for particularized suspicion of Massenburg. It was only able to point to the following: (1) Massenburg and his three friends were walking four blocks from the location of the shots reported by the tipster, the only people the responding officers encountered in the vicinity; and (2) several observations made by Gaines of Massenburg’s allegedly “nervous behavior.” In particular: (a) Massenburg was standing a foot or two from the other three men, who were “shoulder-to-shoulder,” J.A. 57; (b) he did not make eye contact with Gaines as Gaines asked him to consent to a frisk; and (c) he did not consent. Gaines’s testimony on these points is instructive.

Officer Gaines testified that “it wasn’t until actually I made contact with him that I noticed nervous behavior from him.” J.A. 48. He elaborated:

J.A. 57-59.

The district court denied Massenburg’s suppression motion, holding that the search was lawful. It found that reasonable suspicion existed on the basis of six factors: (1) “a vague report of shots fired”; (2) the four men were encountered “roughly two blocks from the location of the reported shooting incident”1 and were the

only people in the area; (3) this was a “high-drug, high-crime area”; (4) Massenburg was “acting nervously, looked down and refused to make eye contact and stood off from the group”; (5) Massenburg “continued to act strangely by making a series of two furtive movements” — that is, he “took a step back away from Officer Gaines, and he then began pantomiming a self pat-down search”; (6) Gaines’s actions were informed by a “year’s worth of practical experience serving as a law enforcement officer.” J.A. 73-75.

After the denial of his suppression motion, Massenburg entered a conditional guilty plea, reserving his right to appeal the court’s ruling. Judgment was entered and he was sentenced to 18 months in prison. He brought this timely appeal challenging the suppression ruling. We exercise jurisdiction pursuant to 28 U.S.C. § 1291.

II.

We review the district court’s legal conclusions de novo and its factual findings for clear error. See United States v. Day, 591 F.3d 679, 682 (4th Cir.2010).

To comport with the Fourth Amendment, even a “brief’ investigatory detention “must be supported at least by a reasonable and articulable suspicion that the person seized is engaged in criminal activity.” Reid v. Georgia, 448 U.S. at 440, 100 S.Ct. 2752; see United States v. Foster, 634 F.3d 243, 246 (4th Cir.2011). Considering the totality of the circumstances, we are to determine whether there was a sufficient objective, particularized basis for suspecting the person seized of criminal activity. United States v. Arvizu, 534 U.S. 266, 273, 122 S.Ct. 744, 151 L.Ed.2d 740 (2002). Evidence that would support only “a mere ‘hunch’ is insufficient,” though a reasonable basis need not establish probable cause and may well “fall[ ] considerably short of satisfying a preponderance of the evidence standard.” Id. at 274, 122 S.Ct. 744 (quoting Terry, 392 U.S. at 27, 88 S.Ct. 1868); of. United States v. Digiovanni, 650 F.3d 498, 511 (4th Cir.2011) (“The reasonable suspicion standard is an objective one, so we examine the facts within the knowledge of [the officer] to determine the presence or nonexistence of reasonable suspicion.”).

This quantum of suspicion is likewise required prior to a frisk when the officer’s initial encounter with the citizen is voluntary. See United States v. Burton, 228 F.3d 524, 528 (4th Cir.2000) (“[D]uring [initially consensual] police-citizen encounters, an officer is not entitled, without additional justification, to conduct a protective search. To conduct such a protective search, an officer must first have reasonable suspicion supported by articulable facts that criminal activity may be afoot.”); see also Terry, 392 U.S. at 32-33, 88 S.Ct. 1868 (Harlan, J., concurring) (“[I]f the frisk is justified in order to protect the officer during an encounter with a citizen, the officer must first have constitutional grounds to insist on an encounter, to make a forcible stop.... If and when a policeman has a right ... to disarm such a person for his own protection, he must first have a right not to avoid him but to be in his presence. That right must be more than the liberty ... to address questions to other persons, for ordinarily the person addressed has an equal right to ignore his interrogator and walk away; he certainly need not submit to a frisk for the questioner’s protection.”) (emphasis add

ed). Thus we can assume without deciding that Officer Fries’s initial conversation with Massenburg and his companions was consensual and that the Fourth Amendment was first implicated by Officer Gaines’s frisk of Massenburg.

We emphasize that the Constitution requires “a particularized and objective basis for suspecting the particular person stopped of criminal activity.” United States v. Griffin, 589 F.3d 148, 152 (4th Cir.2009) (quoting United States v. Cortez, 449 U.S. 411, 417-18, 101 S.Ct. 690, 66 L.Ed.2d 621 (1981)) (emphases added). As the Supreme Court noted in Cortez, “Chief Justice Warren, speaking for the Court in Terry v. Ohio, said that, “[t]his demand for specificity in the information upon which police action is predicated is the central teaching of this Court’s Fourth Amendment jurisprudence.” ” Cortez, 449 U.S. at 418, 101 S.Ct. 690 (quoting Terry, 392 U.S. at 21 n. 18, 88 S.Ct. 1868 (emphasis added by Cortez)). Indeed, as our late friend and colleague Judge Michael reminded us in the 2010 Madison Lecture at New York University, “The Fourth Amendment owes its existence to furious opposition in the American colonies to British search and seizure practices.... Th[e] controversy [over the use of general warrants] left citizens of the new American states with a deep-dyed fear of discretionary searches permitted by general warrants and writs of assistance.” The Honorable M. Blane Michael, Reading the Fourth Amendment: Guidance from the Mischief that Gave it Birth, 85 N.Y.U. L. Rev. 905, 907, 911-12 (2010). Cf. Arizona v. Gant, 556 U.S. 332,-, 129 S.Ct. 1710, 1720, 173 L.Ed.2d 485 (2009) (noting “the central concern underlying the Fourth Amendment” is “the concern about giving police officers unbridled discretion to rummage at will among a person’s private effects”); Delaware v. Prouse, 440 U.S. 648, 661, 99 S.Ct. 1391, 59 L.Ed.2d 660 (1979) (holding unconstitutional a discretionary, suspicionless stop for a “spot check” of a motorist’s license and registration, emphasizing that “[t]his kind of standardless and unconstrained discretion is the evil the Court has discerned when in previous cases it has insisted that the discretion of the official in the field be circumscribed”).

III.

On the facts of this ease, there is precious little to sustain the district court’s holding that Officer Gaines had reasonable, particularized suspicion of Massenburg such that a nonconsensual frisk was lawful under the Fourth Amendment. Among the six factors the district court cited in support of its ruling is Officer Gaines’s one “year’s worth of practical experience serving as a law enforcement officer,” which of course is wholly unrelated to appellant. J.A. 74. The first three factors it listed — that the officers were responding to a “vague report of shots fired,” J.A. 73, that Massenburg was found in the general vicinity (four blocks) of the reported site of the gunfire, and that this encounter occurred in a high-crime area — also do little to create particularized suspicion.

A.

As the district court noted, the officers were responding to “a vague report of shots fired.” J.A. 73. This report was not only “vague” — indicating only that eight shots were “possibly” fired two blocks south of a certain intersection, J.A. 77 — it was also anonymous. Reliance on an anonymous tip may be reasonable where, “suitably corroborated, [it]' exhibits sufficient indicia of reliability.” Florida v. J.L., 529 U.S. 266, 270, 120 S.Ct. 1375, 146 L.Ed.2d 254 (2000). Yet here corroboration did not exist until one of Massen

burg’s companions reported hearing shots fired — which cannot be said to increase reasonable suspicion of the companion’s own party, especially since he also reported that the shots were fired from a moving car (by unknown parties) several blocks away. Like the tip of illegal gun possession held unreliable in J.L., the tip here “provided no predictive information and therefore left the police without means to test the informant’s knowledge or credibility.” Id. at 271, 120 S.Ct. 1375. The tipster here disclosed her basis of knowledge — she heard the shots herself— but little else. Though that disclosure “enhance[s] the tip’s reliability,” United States v. Perkins, 363 F.3d 317, 322 (4th Cir.2004), we have held that even a “nearly contemporaneous report” of a drug transaction the tipster reportedly saw was unreliable in the absence of “[s]ome corroboration,” since “a fraudulent tipster can fabricate her basis of knowledge,” United States v. Reaves, 512 F.3d 123, 127-28 (4th Cir.2008). Cf. Perkins 363 F.3d at 322, 327-28 (anonymous tip held sufficiently reliable where contemporaneous viewing was corroborated by presence of a known drug user in front of a known drug house and where tipster, though she did not explicitly identify herself, was reasonably assumed to be a known, reliable informant).2

Furthermore, the poor match between the vague tip and the individuals encountered substantially undermines reliance on the tip for reasonable particularized suspicion of Massenburg. The tip contained no physical description of the perpetrators or any other outward identifying features; the only link between the tip and Massenburg’s group was the group’s rough proximity to the alleged site of the gunfire. The tipster reported hearing shots two blocks south of the intersection of Hull and 14th Streets; Massenburg and his friends were encountered four blocks west of that intersection.

Thus, while the district court appears to have heavily relied on the fact that Massenburg and his companions were the only people encountered as Officers Fries and Gaines responded to the tip, this provides little basis for reasonable, particularized suspicion of Massenburg. As J.L. and its progeny indicate, when a tip lacks sufficient indicia of reliability, presence in the area identified by the tip does not generate reasonable suspicion. Here, Massenburg was not even present at the site of the alleged gunfire—he was encoun tered four blocks away. Cf. United States

v. Moore, 817 F.2d 1105, 1106 (4th Cir. 1987) (finding reasonable suspicion where only individual in the vicinity was found “30 to 40 yards” from the entrance to a building burglarized two to three minutes before, “moving away from the scene of the crime”). To the extent that the tip, together with Massenburg’s location, did identify his group with particularity, J.L. and Reaves teach that an anonymous tip, absent some corroboration or sufficient other indicia of reliability, is not itself a reasonable basis for suspicion justifying a nonconsensual frisk.

The fact that this was a “high-drug, high-crime area” adds little to the anonymous tip. J.A. 74. This counts among the totality of the circumstances we consider, but it does little to support the claimed particularized suspicion as to Massenburg. “An individual’s presence in an area of expected criminal activity, standing alone, is not enough to support a reasonable, particularized suspicion that the person is committing a crime.” Illinois v. Wardlow, 528 U.S. 119, 124, 120 S.Ct. 673, 145 L.Ed.2d 570 (2000); see Brown v. Texas, 443 U.S. 47, 52, 99 S.Ct. 2637, 61 L.Ed.2d 357 (1979). This is true because “presence in a high crime neighborhood is a fact too generic and susceptible to innocent explanation to satisfy the reasonable suspicion inquiry.” Wardlow, 528 U.S. at 139, 120 S.Ct. 673 (Stevens, J., concurring in part and dissenting in part).

As the officers testified, the city police “usually get complaints ... [for] random gunfire” in this area. J.A. 46. That such incidents are common may make it more reasonable for otherwise innocent behavior to appear suspicious to officers on the beat; but where a tip has already indicated that shots were fired, the level of such crime in the neighborhood does not provide an additional reasonable basis for suspicion of particular individuals. That the tip concerned a common incident in a high-crime neighborhood does little to bolster its reliability and less to create particularized suspicion. While we appreciate the danger posed by firearms in our cities, the Supreme Court has rejected “an automatic firearm exception to our established reliability analysis.” J.L., 529 U.S. at 272, 120 S.Ct. 1375. Like any other anonymous tip, a tip concerning firearms must present certain indicia of reliability before it can provide a basis for reasonable, particularized suspicion.

To hold otherwise would be to authorize general searches of persons on the street not unlike those conducted of old by the crown against the colonists. Allowing officers to stop and frisk any individuals in the neighborhood after even the most generic of anonymous tips would be tantamount to permitting a regime of general searches of virtually any individual residing in or found in high-crime neighborhoods, where “complaints” of “random gunfire” in the night are all too “usual[ ].” J.A. 46. James Otis famously decried general searches as “instruments of slavery ... and villainy,” which “placet ] the liberty of every man in the hands of every petty officer,” warning against abuses by “[e]very man prompted by revenge, ill humor, or wantonness.” Timothy Lynch, In Defense of the Exclusionary Rule, 23 Harv. J.L. & Pub. P. 711, 722 (2000) (quoting James Otis, Speech on the Writs of Assistance (1761)). The Fourth Amendment, and the courts’ Fourth Amendment jurisprudence, is aimed at this evil. Without reasonable particularized suspicion of wrongdoing, such searches and seizures offend the Constitution.

B.

Reasonable suspicion determinations are made according to the totality of the circumstances, and in light of the above—

Massenburg’s presence in a high-crime neighborhood shortly after an (unreliable) tip concerning random gunfire in the general vicinity — we give Officer Gaines a good deal of leeway in his interpretation of Massenburg’s behavior. Yet, as we recently reminded the Government in Foster, it cannot simply proffer “whatever facts are present, no matter how innocent, as indicia of suspicious activity.” 634 F.3d at 248. We expressed serious concerns there about “the way in which the Government attempts to spin ... mundane acts into a web of deception,” id.; these concerns are amplified when these “mundane acts” are incident to the refusal to consent to a voluntary search.

Officer Gaines made clear in his testimony that “it wasn’t until actually I made contact with [Massenburg] that I noticed nervous behavior from him.” J.A. 48. His “blow-by-blow” account of the encounter— which is not contradicted by Fries or any other evidence — indicates that this “nervous behavior” was his characterization of Massenburg’s repeated refusal to consent to a voluntary pat-down: “We questioned if anybody had any weapons on them. The individuals besides Massenburg stated, we don’t have anything, you can check us. And Tyerail [Massenburg] was kind of hesitant and stand-offish, and kind of real reluctant to give consent to a pat down or a search of his person.” J.A. 48. Gaines reiterated this when asked a second time to describe Massenburg’s nervous behavior:

J.A. 49-50 (emphases added). On cross-examination, Gaines again explained that Massenburg “was the only one to be reluctant” and admitted, when asked if it was true he had “no new information to know [Massenburg] was armed and dangerous but for the fact he didn’t want to be patted down,” that there was “[n]othing” except Massenburg’s “statements” (he “kept on saying, I don’t need to be patted down”) and “the area of the call.” J.A. 57-59.

The evidence Gaines cites for Massenburg’s nervousness is slight: Massenburg was standing a foot or two from the other three, who were lined up shoulder-to-shoulder, and “[l]ooked down” or failed to make eye contact as Gaines repeatedly asked him if he would consent to a search. The district court accepted the Government’s characterization and deemed Massenburg’s lack of eye contact “nervous behavior,” yet as Judge Gregory noted in United States v. Foreman, the Government often argues just the reverse: that it is suspicious when “an individual looks or stares back at [officers].” 369 F.3d 776, 787 n. 1 (4th Cir.2004) (Gregory, J., concurring in part and dissenting in part) (collecting cases); see also United States v. McFarley, 991 F.2d 1188, 1192 (4th Cir. 1993) (noting, in support of reasonable suspicion, that appellant and his companion “each canvassed the terminal area, obtaining eye contact with Officer Faulkenberry”). Given the complex reality of citizen-police relationships in many cities, a young man’s keeping his eyes down during a police encounter seems just as likely to be a show of respect and an attempt to avoid confrontation. Cf. State v. Scott, 412 So.2d 988, 989 (La.1982) (“Nervousness on the part of a black laborer when confronted by an armed uniformed police officer does not seem so unusual as to indicate guilt or criminal proclivity.”)

It is, of course, highly relevant when suspects “engage[ ] in evasive behavior or act[ ] nervously.” United States v. Mayo, 361 F.3d 802, 806 (4th Cir.2004). Yet Massenburg did not attempt to evade the officers — in fact, he and his companions stopped to speak with Officer Fries, and one volunteered information about recent gunfire. And looking down as an officer persists in requesting consent to a search is a far cry from the “unusually nervous behavior” we cited in United States v. Mayo, which included “shaking hands, heavy breathing, and providing inconsistent answers.” 361 F.3d at 806 (citing to United States v. McFarley, 991 F.2d 1188, 1192 (4th Cir.1993)). As the Tenth Circuit explained in United States v. Salzano,

158 F.3d 1107, 1113 (10th Cir.1998) (internal quotation marks and citations omitted). See also State v. Lee, 265 Neb. 663, 658 N.W.2d 669, 678-79 (2003) (“[Nervousness is of limited value” to reasonable suspicion analyses as “it is common knowledge that most citizens whether innocent or guilty, when confronted by a law enforcement officer who asks them potentially incriminating questions are likely to exhibit some signs of nervousness.”).

Indeed, the Supreme Court of Wyoming has applied this commonsense principle to a situation much like this one, where an officer was asking a motorist for consent to search his ear and, upon the motorist’s refusal, continued to ask him “whether there was some reason he did not want the officer looking in the vehicle.” Damato v. State, 64 P.3d 700, 709 (Wyo.2003). Reasoning that “Realistically, few citizens would not have become uncomfortable to some degree with these questions,” the court discounted as a “factor of no significance” far more extreme signs of nervousness, including the motorist’s “sweating heavily although it was a chilly day, his carotid artery pulsating hard and fast, and an inability to keep eye contact.” Id.

And as a reasonable response to continued police questioning, looking down is a good deal more innocent than the defendant’s actions in United States v. Sprinkle, where the defendant “put his head down and his hand up to his face as if to avoid recognition” as an officer passed the car and then “drove away as soon as the officers walked by.” 106 F.3d 613, 617 (4th Cir.1997). In Sprinkle we found no reasonable suspicion existed, even though the officers knew the defendant to have been recently released from prison following narcotics convictions, defendant was in a neighborhood known for drug trafficking, and his evasive behavior was preceded by someone else’s entering the car and making gestures consistent with a covert exchange (“huddling” with the two men’s hands “closet ] together” as if to pass something). Id. at 615-16. When we have held that behavior far more extreme, by a known narcotics dealer, in a high-crime area does not create reasonable suspicion, it is difficult to imagine that Massenburg’s keeping his eyes down as he is asked repeatedly to consent to a voluntary search would suffice.

Indeed, we are especially conscious here of the fact that Massenburg’s looking down was incident to his repeated refusal to consent to a voluntary search. It cannot be doubted that “a refusal to cooperate [with a police request to conduct a voluntary search], without more, does not furnish the minimal level of objective justifi

cation needed for a detention or seizure.” Florida v. Bostick, 501 U.S. 429, 437, 111 S.Ct. 2382, 115 L.Ed.2d 389 (1991); see also Mayo, 361 F.3d at 806 (“A suspect’s refusal to cooperate with police, without more, does not satisfy Terry stop requirements.”). If the ordinary response of the innocent upon being asked to consent to a search — some mild nervousness — sufficed to create reasonable suspicion, then Terry ’s reasonable suspicion requirement would become meaningless: officers could ask a citizen for permission to conduct a voluntary search, and, if denied, they could use the citizen’s denial as evidence of criminal activity and perform the search anyway. Though, as an analytic matter, nervousness can be separated from the denial of consent itself,3 to attempt to extricate the very mildest indicators of nervousness — such as a failure to maintain eye contact during the refusal, as the officer becomes “more persistent,” J.A. 50 — from the denial itself is too nice a matter. Virtually any denial will be accompanied by these mild reactions to the request, and thus virtually any denial would go much of the way toward authorizing a nonconsensual search. This cannot be the case.

As for the district court’s characterization of Massenburg’s “self-pat down” as “[f]urtive movements,” J.A. 74, it recalls the Government’s suggestion in Foster that a man’s “sudden[ly]” “pop[ping] up” in a car with “his arms going haywire” was suspicious. Foster, 634 F.3d at 247. There we warned against “using whatever facts are present, no matter how innocent, as indicia of suspicious activity,” and reminded the Government that it “must do more than simply label a behavior as ‘suspicious’ to make it so”: “The Government must also be able to either articulate why a particular behavior is suspicious or logically demonstrate, given the surrounding circumstances, that the behavior is likely to be indicative of some more sinister activity than may appear at first glance.” Id. at 248. No such demonstration has been forthcoming. Massenburg’s “self-pat down” was interpreted as such by Officer Gaines, and as an obvious attempt to satisfy him without consenting to a frisk, it provided little basis, if any, as a matter of constitutional analysis, for a reasonable suspicion of wrongdoing.

Genuinely suspicious behavior, occurring in a high-crime neighborhood after a tip concerning gunfire, would certainly justify a Terry stop and almost certainly a frisk of the detainee. Where that tip is unreliable, the question becomes closer. But where the accompanying behavior — the only substantial basis for particularized suspicion — is simply a mild reaction to repeated requests to relinquish one’s constitutional right to be free from unreasonable searches, it is clear that reasonable, particularized suspicion of criminal activity does not exist.

rv.

The Government suggests that under the collective-knowledge doctrine (also called the “fellow officer” rule) Officer Fries’s observation of a bulge in Massenburg’s jacket pocket should be imputed to Officer Gaines, though, as the Government concedes, Fries never “inform[ed]” Gaines about it. Br. of Appellee, at 16 n.1.4 Be

cause this novel application of the doctrine would stretch it well beyond its purpose, we decline to do so.

The collective-knowledge doctrine, as enunciated by the Supreme Court, holds that when an officer acts on an instruction from another officer, the act is justified if the instructing officer had sufficient information to justify taking such action herself; in this very limited sense, the instructing officer’s knowledge is imputed to the acting officer. In Whiteley v. Warden, the Supreme Court recognized in dicta that “officers called upon to aid other officers in executing arrest warrants are entitled to assume that the officers requesting aid” had probable cause to support the issue of the warrant. 401 U.S. 560, 568, 91 S.Ct. 1081, 28 L.Ed.2d 306 (1971). The Court applied this principle in United States v. Hensley, holding that where officers stopped defendant “in objective reliance” on a flyer from another department that explained that defendant was wanted in connection with an aggravated robbery and requested that other police departments “pick up and hold” him, the stop was justified if and only if the officers who issued the request had reasonable, particularized suspicion sufficient to justify their own stop:

469 U.S. 221, 223, 232, 105 S.Ct. 675, 83 L.Ed.2d 604 (1985) (internal citations omitted).

We have applied the collective-knowledge doctrine often, both before and after Whiteley and Hensley, and our case law likewise establishes that the doctrine has a limited domain: officers acting on the information and instructions of other officers. In United States v. Pitt, federal police agent Wurms learned through an informant that a large quantity of heroin was being driven from New, York City to Washington, D.C. 382 F.2d 322 (4th Cir. 1967). Agent Wurms informed fellow agents, including Agent Worden, and instructed Baltimore City police to intercept the car. Pitt was arrested by Agent Worden, with the assistance of city police. Rejecting Pitt’s claim that Worden lacked personal knowledge of the facts constituting probable cause, we noted that “[p]rob-able cause ... can rest upon the collective knowledge of the police, rather than solely on that of the officer who actually makes the arrest.” Id. at 324. Though this shorthand reference to the collective-knowledge doctrine might be misleading out of context, we went on in the next sentence to explain that “[i]t was enough that Agent Wurms reported to Agent Worden the substance of his telephone conversation with the informant.” Id.

In our discussion of the doctrine in United States v. Laughman, we made its limitations explicit:

The law seems to be clear that so long as the officer who orders an arrest or search has knowledge of facts establishing probable cause, it is not necessary for the officers actually making the arrest or conducting the search to be personally aware of those facts.

618 F.2d 1067, 1072-73 & n. 3 (4th Cir. 1980) (emphasis added). Again, the collective-knowledge doctrine simply directs us to substitute the knowledge of the instructing officer or officers for the knowledge of the acting officer; it does not permit us to aggregate bits and pieces of information from among myriad officers, nor does it apply outside the context of communicated alerts or instructions. See 2 Wayne R. LaFave, Search and Seizure § 3.5(b) (4th ed. 2004) (“[U]nder the Whiteley rule (or, as it is sometimes termed, the ‘fellow officer’ rule) police are in a limited sense ‘entitled to act’ upon the strength of a communication through official channels directing or requesting that an arrest be made.”); cf. United States v. Wells, 98 F.3d 808, 810 (4th Cir.1996) (“[Although the agent who actually seized the weapon pursuant to the supervising agent’s instructions had no personal knowledge that Wells was a convicted felon, it is sufficient that the agents collectively had probable cause to believe the weapon was evidence of a crime at the time of the seizure.”) (emphasis added); United States v. Gaither, 527 F.2d 456, 458 (4th Cir.1975) (quoting Pitt to support application of collective-knowledge doctrine where arresting officer was “acting on” a “ ‘flash’ bulletin” issued by FBI agents who had just observed a bank robbery).

The Government would have us recognize a far more expansive rule, which would look to the aggregated knowledge of all officers involved to determine if reasonable suspicion or probable cause existed. Under this proposed rule, it would be irrelevant that no officer had sufficient information to justify a search or seizure. It would be irrelevant that no officer believed any other officer had pertinent information, and thus that the acting officer undertook a search or seizure she should have believed to be illegal. Indeed, as this aggregation rule is only required when the information at issue has not been communicated to other officers (as the “aggregation” it concerns is judicial, after-the-fact aggregation, not an acting officer’s reliance on instructions or information conveyed by another officer), this would be the paradigmatic case. Were we to adopt this rule, the legality of the search would depend solely on whether, after the fact, it turns out that the disparate pieces of information held by different officers added up to reasonable suspicion or probable cause.

The Tenth Circuit has helpfully distinguished “ ‘vertical’ collective knowledge relationships in which [one] officer’s conclusion [i]s conveyed” to others who effect the seizure from a “ ‘horizontal’ collective knowledge relationship in which the knowledge of several officers must be aggregated to create probable cause.” United States v. Rodriguez-Rodriguez, 550 F.3d 1223, 1228 n. 5 (10th Cir.2008). No case from the Supreme Court or from our own court has ever expanded the collective-knowledge doctrine beyond the context of

information or instructions communicated (“vertically”) to acting officers. Some of our sister courts have authorized “horizontal” aggregation of uncommunicated information. See United States v. Ramirez, 473 F.3d 1026, 1032-33 (9th Cir.2007) (collecting cases). Because we believe that this expansive aggregation rule strays from the purposes of the collective-knowledge doctrine recognized by the Supreme Court and promotes none of the proper ends of law enforcement, we decline to follow them.

The rationale. behind the Supreme Court’s collective-knowledge doctrine is, as the Court noted in Hensley, “a matter of common sense: [the rule] minimizes the volume of information concerning suspects that must be transmitted to other jurisdictions [or officers] and enables police ... to act promptly in reliance on information from another jurisdiction [or officer].” Hensley, 469 U.S. at 231, 105 S.Ct. 675. Thus, law enforcement efficiency and responsiveness would be increased: Police department search-and-seizure training would soon reflect Hensley’s rule, and officers would learn that they need not relay the information justifying an alert when issuing one nor wait for such information upon hearing one.

The Government’s proposed aggregation rule serves no such ends. Because it jettisons the present requirement of communication between an instructing and an acting officer, officers would have no way of knowing before a search or seizure whether the aggregation rule would make it legal, or even how likely that is. The officer deciding whether or not to perform a given search will simply know that she lacks cause; in ordinary circumstances, she will have no way of estimating the likelihood that her fellow officers hold enough uncommunicated information to justify the search. And as an officer will never know ex ante when the aggregation rule might apply, the rule does not allow for useful shortcuts when an officer knows an action to be legal, as Hensley did. Perhaps an officer who knows she lacks cause for a search will be more likely to roll the dice and conduct the search anyway, in the hopes that uncommunicated information existed. But as this would only create an incentive for officers to conduct searches and seizures they believe are likely illegal, it would be directly contrary to the purposes of longstanding Fourth Amendment jurisprudence.

As the Supreme Court recently reaffirmed in Davis v. United States, the exclusionary rule’s “sole purpose ... is to deter future Fourth Amendment violations.” — U.S. -, -, 131 S.Ct. 2419, 2426, 180 L.Ed.2d 285 (2011). It targets police action that “exhibits] deliberate, reckless, or grossly negligent disregard for Fourth Amendment rights” — in these cases the “deterrent value of exclusion is strong and tends to outweigh the resulting costs.” Id. at 2427 (internal quotation marks omitted). As the Government’s proposed aggregation rule would do nothing but redeem searches or seizures that the acting officers should have believed at the time to be unlawful, it would serve only to erode that deterrence. The Davis Court further broadened the “good-faith” exception to the exclusionary rule, recognizing that “when the police act with an objectively reasonable good-faith belief that their conduct is lawful ... the deterrence rationale loses much of its force.” Id. at 2427-28. The Government’s proposed aggregation rule would perversely reward officers acting in bad faith according to the result of an after-the-fact aggregation inquiry that is simply academic.

Though we have studied our sister circuits’ cases adopting an aggregation rule,

we can find no convincing defense of it.5 Most courts to have adopted the rule appear to have done so simply on the grounds that officers working closely together are “a team,” United States v. Terry, 400 F.3d 575, 581 (8th Cir.2005); United States v. Edwards, 885 F.2d 377, 383 (7th Cir.1989), or, as one court put it, “on the theory that officers working closely together during a stop or an arrest can be treated as a single organism,” United States v. Shareef, 100 F.3d 1491, 1504 & n. 6 (10th Cir.1996) (considering this rationale after rejecting a general aggregation rule). But why? We must frame the question in terms of deterrence, and for the purposes of deterrence we look to each individual officer’s decision-making process as she considers executing a search or effecting a seizure. Where officers working closely together have not communicated pertinent information, the acting officer weighs the costs and benefits of performing the search in total ignorance of the existence of that information — it is not known to her, so it cannot enter into the calculus. Therefore, for purposes of the exclusionary rule, that additional information must be irrelevant.6

Furthermore, if the “team” or “single organism” theory should apply when the information at issue is incriminating, should it not apply when the information is exculpatory? Yet, we held in United States v. Holmes that the collective-knowledge doctrine does not impute uncommunicated exculpatory knowledge to fellow officers in similar circumstances. 376 F.3d 270, 277 n. 3 (4th Cir.2004). Likewise, though most courts to allow aggregation have required “some degree of communication” among the officers, Terry, 400 F.3d at 581, see also Ramirez, 473 F.3d at 1032-33, it is not clear why. If the Fourth Amendment is satisfied when, unbeknownst to the officer conducting a search, a fellow officer on the scene has the information necessary to justify it, why should the analysis change when the other officer is not on the scene? Yet we recently held in United States v. Blauvelt that information held by others in the “law enforcement community at large” is not imputed to members of a particular investigative team. 638 F.3d 281, 289 (4th Cir.2011). Cf. People v. Hazelhurst, 662 P.2d 1081, 1087 (Colo.1983) (“The fellow officer rule, however, is not a means of creating probable cause by using post hoc combinations of information available to the police. The rule does not permit the police to cull its archives in hopes of justifying an arrest which is not supported by probable cause.”)

Because we believe the aggregation rule runs contrary to the Supreme Court’s Fourth Amendment jurisprudence, would seriously erode the efficacy of the exclusionary rule’s deterrent purposes, and serves none of the legitimate ends of law enforcement, we reject it. We do not impute Officer Fries’s observation of a “bulge” in Massenburg’s jacket pocket to Officer Gaines, and thus, for the reasons

stated above, we hold that Gaines lacked the reasonable suspicion needed to conduct a lawful nonconsensual frisk. Accordingly, the district court erred when it failed to suppress the fruits of that unlawful search.

V.

For the reasons set forth herein, the judgment is vacated, the district court’s order denying the motion to suppress is reversed, and the case is remanded for further proceedings consistent with this opinion.