ON REHEARING EN BANC
PUBLISHED
UNITED STATES COURT OF APPEALS
FOR THE FOURTH CIRCUIT

No. 18-4233

UNITED STATES OF AMERICA,
Plaintiff – Appellant,
v.
BILLY CURRY, JR.,
Defendant – Appellee.

Appeal from the United States District Court for the Eastern District of Virginia, at
Richmond. M. Hannah Lauck, District Judge. (3:17-cr-00130-MHL-1)

Argued: January 30, 2020

Decided: July 15, 2020
Amended: July 16, 2020

Before GREGORY, Chief Judge, and WILKINSON, NIEMEYER, MOTZ, KING, AGEE,
KEENAN, WYNN, DIAZ, FLOYD, THACKER, HARRIS, RICHARDSON,
QUATTLEBAUM, and RUSHING, Circuit Judges.

Affirmed by published opinion. Judge Floyd wrote the majority opinion, in which Chief
Judge Gregory, Judge Motz, Judge King, Judge Keenan, Judge Wynn, Judge Diaz, Judge
Thacker, and Judge Harris joined. Chief Judge Gregory wrote a concurring opinion. Judge
Wynn wrote a concurring opinion. Judge Diaz wrote a concurring opinion, in which Judge
Harris joined. Judge Thacker wrote a concurring opinion, in which Judge Keenan joined.
Judge Wilkinson wrote a dissenting opinion. Judge Richardson wrote a dissenting opinion,
in which Judge Wilkinson, Judge Niemeyer, Judge Agee, Judge Quattlebaum, and Judge
Rushing joined.

 ARGUED: Richard Daniel Cooke, OFFICE OF THE UNITED STATES ATTORNEY,
Richmond, Virginia, for Appellant. Caroline Swift Platt, OFFICE OF THE FEDERAL
PUBLIC DEFENDER, Alexandria, Virginia, for Appellee. ON BRIEF: G. Zachary
Terwilliger, United States Attorney, Alexandria, Virginia, Holli R. Wood, Special
Assistant United States Attorney, Michael A. Jagels, Special Assistant United States
Attorney, OFFICE OF THE UNITED STATES ATTORNEY, Richmond, Virginia, for
Appellant. Geremy C. Kamens, Federal Public Defender, Alexandria, Virginia; Paul G.
Gill, Assistant Federal Public Defender, Laura J. Koenig, Assistant Federal Public
Defender, OFFICE OF THE FEDERAL PUBLIC DEFENDER, Richmond, Virginia, for
Appellee.

2

 FLOYD, Circuit Judge:
This appeal presents the question of whether the Fourth Amendment’s exigent
circumstances doctrine justified the suspicionless seizure of Defendant-Appellee Billy
Curry, Jr. The police seized Curry after responding to several gunshots that were fired in
or near an apartment complex less than a minute earlier. When the police arrived, they
encountered five to eight men—including Curry—calmly and separately walking in a
public area behind the complex, away from the general vicinity of where the officers
believed the shots originated; several other people, likely visitors or residents, standing
around closer to the apartments; and another man walking toward the rear of the officers’
patrol car, who appeared to be favoring one of his arms.
The district court held that exigent circumstances did not justify the suspicionless,
investigatory stop of Curry, and so it granted his motion to suppress a firearm and other
evidence based on the unreasonableness of the seizure that led to its discovery. We agree
with the district court’s conclusion. To hold otherwise would create a sweeping exception
to Terry v. Ohio, 392 U.S. 1 (1968). The exigent circumstances doctrine typically involves
emergencies justifying a warrantless search of a home, not an investigatory stop of a
person, and the few cases that have applied the doctrine in the investigatory seizure context
are materially distinguishable. In those cases, the government isolated a discrete area or
group of people and engaged in minimally intrusive suspicionless searches in an effort to
search for a suspect implicated in a known crime in the immediate aftermath of that crime.
Requiring such suspicionless seizures to be narrowly targeted based on specific
information of a known crime and a controlled geographic area ensures that the exigency
3

 exception does not swallow Terry whole. Because these limiting principles were wholly
absent from Curry’s stop, we hold that the stop was not justified by exigent circumstances
and thus was not reasonable under the Fourth Amendment. Therefore, we affirm.

I.
Any Fourth Amendment analysis turns on the totality of the circumstances and thus
must be grounded on an accurate understanding of the facts. Because neither party disputes
the district court’s findings of fact—let alone challenges them as clearly erroneous, see
United States v. Kehoe, 893 F.3d 232, 237 (4th Cir. 2018)—we recite the operative facts
from the district court’s opinion. See generally United States v. Curry, No. 3:17-cr-130,
2018 WL 1384298 (E.D. Va. Mar. 19, 2018). 1

A.
On the night of September 8, 2017, four uniformed officers of the Richmond Police
Department (RPD)—Officers Gaines, Fitzpatrick, Janowski, and O’Brien—were
patrolling a densely populated area in Richmond, Virginia, when they heard what they
thought were five to six gunshots. The officers—all in the same marked car—were on
patrol duty in the Creighton Court neighborhood as members of the Focus Mission Team,
a division of the RPD dedicated to violent crime and drug suppression. Home to hundreds

1

The district court based its factual findings on testimony from Curry’s suppression
hearing and on video footage captured by police body cameras.
4

 of families, Creighton Court is a public housing community in the East End of Richmond.
The officers were assigned to patrol Creighton Court because there had been six shootings
and two homicides within the past three months in that area, with the most recent homicide
occurring just eleven days before the incident in question.
It was about 9:00 p.m. when the officers heard the gunshots. Upon hearing the
shots, the officers, who were then patrolling in a grass field near the 2100 block of
Creighton Court, made a U-turn and began driving toward a housing complex called
Walcott Place, where they thought the shots had originated. It took the officers only about
thirty-five seconds to arrive behind Walcott Place, which was two-to-three blocks away.
Behind Walcott Place was an open and poorly lit field flanked on two sides by
apartment buildings. Approximately five to eight men were walking in and around the
field, heading away from where the officers believed the shots were fired. Several people
were standing near the apartment buildings. None of the men shown in the body camera
footage taken that night were walking alongside one another or talking to one another as
they moved away from the complex. Officer Gaines testified that as the patrol car came to
a stop, the officers saw a man in a red shirt walking toward the rear of the car, who
“appeared to be maybe favoring one of his arms.” Id. at *2. Therefore, the officers were
“uncertain if he had been shot or not.” Id. 2

2

Before leaving the car, one of the officers remarked, “[h]e’s holding his arm.
What’s he doing?” Id. at *2 n.4 (alteration in original). Officer Fitzpatrick later spoke
with that man, and no call for medical assistance followed. Id.
5

 As the officers exited the patrol car, they received dispatch calls corroborating that
“random gunfire” had come from Walcott Place. See id. at *2. But they did not receive a
suspect description.

The officers then fanned out and began approaching different

individuals in the field. The officers stopped some of the individuals walking away from
the complex and illuminated their waistbands and hands.
Officer Gaines approached Curry and another man wearing a blue jacket, both of
whom were separately walking away from the officers. According to Gaines, he was not
looking for Curry, specifically, but was looking at everyone’s hands to ensure that they did
not have a firearm. When he first saw Curry, Gaines testified, Curry had a cell phone in
his left hand and was walking without putting his hands in his pockets or in his waistband.
Curry made no furtive gestures, nor did he walk at an accelerated pace indicative of flight.
Id. at *3.
Gaines instructed the two men to put their hands up, and they complied. Curry stood
completely still for about four seconds before pointing toward where the shots had come
from and telling the officers that he was looking for his nephew. Gaines instructed Curry
to pull his shirt up. Gaines claims he could not get a full view of Curry’s waistband and
asked him to lift his shirt again. See id. (“Officer Gaines characterized Curry as complying
with this command in a ‘lackadaisica[l]’ manner, ‘nonchalantly pick[ing up] the left side
[of his shirt] where [Gaines] could not get a full view of his . . . waistband.’ Officer Gaines
then ordered, ‘Pick your shirt up,’ and Curry responded, ‘I’m liftin’ it up.’” (first three
alterations in original) (citation omitted)). Gaines testified that Curry did not comply, but
instead turned away.
6

 Unable to visually check for a bulge because of what he deemed noncompliance,
Gaines called to Officer O’Brien to help him pat Curry down. Officers stated that Curry
remained evasive and prevented them from reaching his right side. Gaines and O’Brien
restrained Curry’s arms and began to pat him down. Gaines testified that he felt a hard
object like the butt of a handgun on Curry’s person, and the body camera video shows that
Gaines notified the other officers that “It’s on him,” referring to a firearm. An apparent
struggle ensued, and Gaines was never able to fully retrieve the object that he felt before
Curry was taken to the ground.
The officers handcuffed Curry and took him into custody. As they handcuffed him,
the officers told Curry that they had found his gun. Gaines testified that he found the
flashlight that he had dropped during the struggle approximately one to one-and-a-half feet
from the location where Curry had been taken to the ground. Next to Gaines’s flashlight
was a silver revolver.

B.
On October 3, 2017, a grand jury indicted Curry on one count of possession of a
firearm by a convicted felon. See 18 U.S.C. § 922(g)(1). Curry moved to suppress
evidence of the revolver, as well as statements he made after his seizure while in custody.
See generally Taylor v. Alabama, 457 U.S. 687 (1982); Wong Sun v. United States, 371
U.S. 471 (1963). Initially, the government relied solely on the theory that Curry’s seizure
was a lawful Terry stop and that the officers were justified in later patting down Curry to

7

 search for weapons. Later, in requested supplemental briefing, the government argued in
the alternative that exigent circumstances allowed for Curry’s seizure.
Following an evidentiary hearing, the district court issued a detailed and lengthy
opinion granting Curry’s motion to suppress. As a preliminary matter, the district court
concluded that Curry was seized when he halted and raised his hands upon Gaines’s initial
command. Curry, 2018 WL 1384298, at *5–9; see also United Sates v. Mendenhall, 446
U.S. 544, 553–54 (1980). Next, it held that the seizure was not a lawful Terry stop, as
Gaines possessed no reasonable, articulable suspicion of criminal activity particularized to
Curry at the time of Curry’s seizure. Curry, 2018 WL 1384298, at *10–12; see also Terry,
392 U.S. at 21; United States v. Black, 707 F.3d 531, 539 (4th Cir. 2013). Finally, the
district court rejected the government’s argument that Curry’s seizure was justified by
exigent circumstances. Curry, 2018 WL 1384298, at *12–14. Because the district court
held that the initial stop was unlawful, it granted Curry’s suppression motion without
considering whether the officers were justified in frisking Curry. See generally United
States v. Robinson, 846 F.3d 694, 700 (4th Cir. 2017) (en banc).
The government filed an interlocutory appeal, in which it effectively abandoned its
Terry justification by conceding that Curry’s stop was suspicionless. The government only
pursues its claim of reasonableness based on exigent circumstances, and that is the sole
justification for Curry’s seizure that is before us on appeal.
A split panel of this Court reversed the district court’s suppression ruling. See
United States v. Curry, 937 F.3d 363 (4th Cir. 2019). In doing so, the panel majority relied
not only on the exigent circumstances exception but also on the “special needs”
8

 exception—a separate Fourth Amendment doctrine never invoked by the government, see
infra note 4, which applies to programmatic searches such as vehicular checkpoints,
random drug tests, and administrative searches that are motivated by “special needs” that
go “beyond the normal need for law enforcement[ and] make the warrant and probablecause requirement impracticable.” Griffin v. Wisconsin, 483 U.S. 868, 873 (1987) (quoting
New Jersey v. T.L.O., 469 U.S. 325, 351 (1985) (Blackmun, J., concurring in the
judgment)). 3 We granted Curry’s petition for rehearing en banc, thereby vacating the
panel’s opinion. See United States v. Curry, 784 F. App’x 870 (4th Cir. 2019).

II.
“We review the factual findings underlying a district court’s ruling on a motion to
suppress for clear error and its legal conclusions de novo.” United States v. McGee, 736
F.3d 263, 269 (4th Cir. 2013). The parties here do not take issue with the district court’s
factual findings or its description of the situation confronted by the officers, and we have
reviewed the body camera footage corroborating the district court’s findings. Cf. Kehoe,

3

See, e.g., Bd. of Educ. of Indep. Sch. Dist. No. 92 of Pottawatomie Cty. v. Earls,
536 U.S. 822 (2002) (involving a program that subjected all students participating in
extracurricular activities to submit to random, suspicionless drug testing); Ferguson v. City
of Charleston, 532 U.S. 67 (2001) (involving a public hospital’s non-consensual drug
testing of maternity patients); Michigan Dep’t of State Police v. Sitz, 496 U.S. 444 (1990)
(involving a highway checkpoint program); Skinner v. Ry. Labor Execs.’ Ass’n, 489 U.S.
602 (1989) (involving compulsory blood and urine tests of railroad employees involved in
certain train accidents); Camara v. Mun. Court of S.F., 387 U.S. 523 (1967) (involving
warrantless administrative inspection to ensure compliance with city housing
code); Griffin, 483 U.S. 868 (involving a probation officer’s search of a probationer’s
home).
9

 893 F.3d at 237 (citing Scott v. Harris, 550 U.S. 372, 378–81 (2007)). Thus, this case
presents a pure question of law that we review de novo: Did the situation here rise to the
level of exigent circumstances and do those circumstances justify a warrantless and
suspicionless seizure of a pedestrian? See United States v. Singleton, 441 F.3d 290, 293
(4th Cir. 2006) (reviewing de novo whether “exigent circumstances excused the police’s
failure to follow the knock-and-announce requirement” before executing a search warrant).
In answering that question, we view the evidence in the light most favorable to Curry, as
the prevailing party below, see Black, 707 F.3d at 534, and bear in mind that the
government has the “burden of proof in justifying a warrantless search or seizure,” McGee,
736 F.3d at 269; see also United States v. Willis, 443 F. App’x 806, 807 (4th Cir. 2011)
(“The Government bears the burden of demonstrating exigent circumstances.” (citing
Welsh v. Wisconsin, 466 U.S. 740, 749–50 (1984))).

III.
The Fourth Amendment guards the “right of the people to be secure in their
persons . . . against unreasonable searches and seizures” and provides that “no Warrants
shall issue, but upon probable cause.” U.S. Const. amend. IV. The protection against
unreasonable seizures includes “brief investigatory stops.” Kehoe, 893 F.3d at 237; accord
Reid v. Georgia, 448 U.S. 438, 440 (1980) (stating that the Fourth Amendment applies to
all seizures, “including seizures that involve only a brief detention short of traditional
arrest” (quoting United States v. Brignoni-Ponce, 422 U.S. 873, 878 (1975))). These stops,

10

 also known as Terry stops, are not subject to the Fourth Amendment’s probable cause and
warrant requirements. See United States v. Holmes, 376 F.3d 270, 275 (4th Cir. 2004).
In Terry, the Supreme Court addressed the “narrow question” of “whether it is
always unreasonable for a policeman to seize a person and subject him to a limited search
for weapons unless there is probable cause for an arrest.” 392 U.S. at 15. Had the police
conduct there been subject to the Warrant Clause of the Fourth Amendment, the Court
“would have [had] to ascertain whether ‘probable cause’ existed to justify the search and
seizure which took place.” Id. at 20. But that was not the case. The “stop and frisk” in
Terry dealt “with an entire rubric of police conduct—necessarily swift action predicated
upon the on-the-spot observations of the officer on the beat—which historically has not
been, and as a practical matter could not be, subjected to the warrant procedure.” Id. at 16,
20. As a result, the Supreme Court held that the conduct involved in that case had to be
“tested by the Fourth Amendment’s general proscription against unreasonable searches and
seizures.” Id. at 20. And it proceeded to find that the search and seizure in Terry was
reasonable, providing us with the now-famous Terry rule. See id. at 20–31. Under Terry,
an officer may conduct a brief investigatory stop based on reasonable, articulable suspicion
of criminal activity. Id. at 20–21, 30; see Holmes, 376 F.3d at 275. If the person is “validly
stopped,” then the officer may conduct a protective “frisk,” a pat-down of the person’s
outer clothing for weapons, so long as “the officer reasonably believes that the person is
‘armed and dangerous.’” Robinson, 846 F.3d at 700 (quoting Terry, 392 U.S. at 27).
Notably, in this case, the government does not challenge the district court’s holding
that the officers lacked individualized suspicion that Curry was involved in criminal
11

 activity at the time of his seizure. In other words, the government does not claim that
Curry’s stop was a valid Terry stop. Instead, it claims that the seizure was justified by the
so-called exigent circumstances exception.

Therefore, we must decide whether the

suspicionless seizure of Curry—which was not a legal Terry stop—was nevertheless lawful
due to exigent circumstances. If it was not, then the revolver was properly suppressed.
Because the officers lacked reasonable suspicion to stop Curry in the first place,
exigent circumstances must have justified the initial seizure; we do not consider Curry’s
actions after he was seized—that is, after he halted and put his hands up at Officer Gaines’s
command. If the initial stop was unsupported by exigent circumstances, then Curry’s
subsequent actions cannot make the stop legal. See United States v. Kennedy, 32 F.3d 876,
882 (4th Cir. 1994) (determining whether failure to comply with knock-and-announce
requirements was excused by exigent circumstances that existed “at the time of the entry”);
cf. United States v. Simmons, 560 F.3d 98, 107 (2d Cir. 2009) (“The events that occurred
after [the defendant’s seizure] do not factor into the analysis of reasonable suspicion for
the initial stop.”); Black, 707 F.3d at 539 n.5 (same).
For the reasons discussed below, we hold that exigent circumstances did not justify
Curry’s suspicionless seizure and thus affirm the district court’s well-reasoned opinion. 4

4

At oral argument, the government conceded that exigent circumstances and special
needs are distinct—though, in its view, related—doctrines, and that the government has
always relied solely on exigent circumstances to justify Curry’s stop. See Oral Arg. 3:20–
4:13, 32:05–34:25. These are wise concessions. See Curry, 937 F.3d at 378 (Floyd, J.,
dissenting) (“[T]he majority’s analysis blurs the lines that have heretofore defined the
exigent circumstances exception, conflating exigent circumstances with the special needs
exception.”); see also McGee, 736 F.3d at 269 (government bears the burden of justifying
12

 A.
“[B]ecause the ultimate touchstone of the Fourth Amendment is ‘reasonableness,’”
the Supreme Court has articulated certain exceptions to the Fourth Amendment’s warrant
requirement. Brigham City v. Stuart, 547 U.S. 398, 403 (2006) (citing Flippo v. West
Virginia, 528 U.S. 11, 13 (1999) (per curiam)). One such “reasonableness” exception is
exigent circumstances. United States v. Yengel, 711 F.3d 392, 396 (4th Cir. 2013). Thus,
although “warrants are generally required to search a person’s home or his person,” Mincey
v. Arizona, 437 U.S. 385, 393–94 (1978), or to “enter a home to make an arrest,” Collins
v. Virginia, 138 S. Ct. 1663, 1672 (2018) (citing Payton v. New York, 445 U.S. 573, 587–
90 (1980)), the Supreme Court has held that “the exigencies of the situation” may render
“the needs of law enforcement so compelling that [a] warrantless search is objectively
reasonable under the Fourth Amendment,” Mincey, 437 U.S. at 393–94; see also Payton,

a warrantless search or seizure); United States v. Archibald, 589 F.3d 289, 296 (6th Cir.
2009) (waiver principles generally preclude the government from raising Fourth
Amendment arguments for the first time on appeal (citing Steagald v. United States, 451
U.S. 204, 208–09 (1981); Giordenello v. United States, 357 U.S. 480, 488 (1958))).
Special needs and exigent circumstances are separate doctrines of Fourth Amendment
jurisprudence that are animated by very different concerns and applied in different
contexts. The lynchpin of the court’s inquiry in all special needs cases is whether it is
impracticable to require a warrant in light of the primary purpose of a programmatic search.
In the context of an investigatory seizure like the one at issue here, such an inquiry is
untenable. See generally Curry, 937 F.3d at 379–80 (Floyd, J., dissenting). Moreover,
special needs cases all involve a critical feature not present here: programmatic safeguards
designed to protect against a law enforcement officer’s arbitrary use of unfettered
discretion. Id. at 381 (citing cases); see also id. (“Applying the special needs doctrine [in]
this case is especially dangerous considering that the officers selectively stopped and
frisked certain individuals while not seizing others who were closer to the suspected scene
of the shooting.”).
13

 445 U.S. at 590. The rationale underlying the exigent circumstances exception is a
“compelling need for official action and no time to secure a [search] warrant.” Missouri v.
McNeely, 569 U.S. 141, 149 (2013) (quoting Michigan v. Tyler, 436 U.S. 499, 509–10
(1978)); see also Birchfield v. North Dakota, 136 S. Ct. 2160, 2173 (2016) (“[T]he exigent
circumstances exception allows a warrantless search when an emergency leaves police
insufficient time to seek a warrant.”).
But like all exceptions to the warrant requirement, the exigent circumstances
exception is a “narrow” one that must be “well-delineated in order to retain [its]
constitutional character.” Yengel, 711 F.3d at 396 (citing Flippo, 528 U.S. at 13); see also
Mincey, 437 U.S. at 390 (“The Fourth Amendment proscribes all unreasonable searches
and seizures, and it is a cardinal principle that ‘searches conducted outside the judicial
process, without prior approval by judge or magistrate, are per se unreasonable under the
Fourth Amendment—subject only to a few specifically established and well-delineated
exceptions.’” (quoting Katz v. United States, 389 U.S. 347, 357 (1967))). Thus, while
application of the exigency doctrine often depends on the “totality of the circumstances,”
see Mitchell v. Wisconsin, 139 S. Ct. 2525, 2535 n.3 (2019) (plurality opinion) (quoting
McNeely, 569 U.S. at 150), the Supreme Court, to date, has identified only “a
few . . . emergency conditions” that rise to the level of exigent circumstances, Welsh, 466
U.S. at 749–50. They include: (1) the need to “pursue a fleeing suspect”; (2) the need to
“protect individuals who are threatened with imminent harm”; and (3) the need to “prevent
the imminent destruction of evidence.” Carpenter v. United States, 138 S. Ct. 2206, 2223

14

 (2018) (citing Kentucky v. King, 563 U.S. 452, 460 & n.3 (2011)); accord Birchfield, 136
S. Ct. at 2173; see Welsh, 466 U.S. at 750.
Here, the government relies upon the second type of exigency recognized by the
Supreme Court—the need to “protect individuals who are threatened with imminent harm.”
Carpenter, 138 S. Ct. at 2223. This exigency encompasses what is often referred to as “the
‘emergency aid’ exception,” King, 563 U.S. at 460 (quoting Brigham City, 547 U.S. at
403), 5 or, in this Circuit, as the more general “emergency-as-exigency” exception, Yengel,
711 F.3d at 397 (similarly relying on Brigham City); cf. United States v. Taylor, 624 F.3d
626, 631 (4th Cir. 2010) (“[A] warrant is not required to break down a door to enter a
burning home to rescue occupants or extinguish a fire, to prevent a shooting or to
bring emergency aid to an injured person. The need to protect or preserve life or avoid
serious injury is justification for what would be otherwise illegal absent an exigency or
emergency.” (alteration in original) (quoting Wayne v. United States, 318 F.2d 205, 212
(D.C. Cir. 1963) (Burger, J.))). Whatever the terminology used, under this approach, to
determine whether a given emergency justifies officers bypassing a Fourth Amendment
protection, courts look to officers’ “objectively reasonable belief . . . based on specific
articulable facts and reasonable inferences that could have been drawn therefrom.” Yengel,
711 F.3d at 397; see Michigan v. Fisher, 558 U.S. 45, 47 (2009) (citing Brigham City, 547
U.S. at 406).

5

As the Seventh Circuit observed in Sutterfield v. City of Milwaukee, 751 F.3d 542,
558 (7th Cir. 2014), the Supreme Court effectively deemed the emergency aid doctrine a
“subset” of the exigent circumstances exception in Brigham City.
15

 Though the “emergency-as-exigency approach,” Yengel, 711 F.3d at 397, may
sound broad in name, it is subject to important limitations and thus is quite narrow in
application.

For example, the requirement that the circumstances present a true

“emergency” is strictly construed—that is, an emergency must be “enveloped by a
sufficient level of urgency.” Id. 6 Indeed, standing alone, even a “possible homicide” does
not present an “emergency situation” demanding “immediate [warrantless] action.”
Mincey, 437 U.S. at 392. 7

6

Yengel addressed the reasonableness of a warrantless entry into a locked closet in
the defendant’s home following his arrest for an armed domestic dispute. There, we found
three facts “highly persuasive” in concluding that no exigency existed: (1) “the stable
nature of the threat,” namely, the possible presence of a grenade; (2) its “immobile and
inaccessible location”; and (3) “the failure by police officers on site to view the threat as
serious enough to warrant evacuation of a nearby child.” Id. at 398–99.
7

Our dissenting colleagues denounce this characterization of Mincey as teetering
on the edge of “imprecision and error.” Richardson Dissenting Op. 83 n.6. By
acknowledging Mincey’s holding that a possible homicide does not necessarily present an
emergency situation demanding immediate action, however, we do not intend to suggest
that officers who arrive on the scene of a homicide may never lean on the exigent
circumstances exception to excuse the ordinary warrant requirement. As the principal
dissent correctly notes, the Supreme Court in Mincey explained that under such
circumstances, officers may conduct a “prompt warrantless search of the area to see if there
are other victims or if a killer is still on the premises.” 437 U.S. at 392. Therefore, “Mincey
teaches that a response to exigent circumstances must end roughly when the exigency
does.” Richardson Dissenting Op. 83 n.6; see also Mincey, 437 U.S. at 393 (holding that
the warrantless search of the crime scene in the defendant’s apartment was not “‘strictly
circumscribed by the exigencies which justif[ied] its initiation,’” as “[a]ll the persons in
[the] apartment had been located before the investigating officers arrived there and began
their search” (quoting Terry, 392 U.S. at 25–26)). But Mincey also teaches that the
seriousness of the offense under investigation cannot give rise to exigent circumstances,
see 437 U.S. at 392–94, and it is this teaching that we simply reiterate here in outlining the
contours of the exigent circumstances exception.
Moreover, although Mincey suggested that a warrantless, preliminary sweep of a
murder scene would be justified by exigent circumstances, we would be remiss if we did
16

 More

importantly,

exigent

circumstances—including

those

emergencies

contemplated by the Brigham City line of cases—typically only justify “the warrantless
entry and search of private property,” usually a home. Yengel, 711 F.3d at 396 (citing
Mincey, 437 U.S. at 392–94); see, e.g., Fisher, 558 U.S. 45 (exigent circumstances allowed
warrantless entry into home to provide emergency assistance); Brigham City, 547 U.S. 398
(same); United States v. Santana, 427 U.S. 38 (1976) (exigent circumstances allowed
warrantless entry into home when police were in “hot pursuit” of fleeing suspect); United
States v. Turner, 650 F.2d 526 (4th Cir. 1981) (exigent circumstances justified warrantless
entry into residence to protect against imminent destruction of evidence); see also Mora v.
City of Gaithersburg, 519 F.3d 216 (4th Cir. 2008) (applying doctrine to justify a
warrantless search of a suspect’s apartment as well as his luggage and van); cf. McNeely,

not point out the grave differences between the “[t]he threat presented here,” Richardson
Dissenting Op. 89, and the initial threat to the responding officers in Mincey, as described
by the Supreme Court. In Mincey, an undercover officer returned to the defendant’s
apartment with several plainclothes narcotics agents after allegedly arranging a controlled
buy earlier in the day. One of the three acquaintances of the defendant who was present at
his apartment that afternoon answered the door, and the undercover officer “slipped inside
and moved quickly into the bedroom.” Mincey, 437 U.S. at 387. When the acquaintance
attempted to slam the door to prevent the other officers from entering, they pushed him
back against the wall. As the other officers entered the apartment, they heard “a rapid
volley of shots” from the bedroom. Id. The undercover officer then emerged from the
bedroom and collapsed on the floor. “[T]hinking that other persons in the apartment might
have been injured,” the narcotics agents “looked about quickly for victims” and found
several who had been injured. Id. at 388. Those are the facts of Mincey. Thus, even if we
set aside the thorny doctrinal questions regarding the application of the exigent
circumstances exception outside the context of a warrantless search or entry of a home, see
infra, we note that, from a pure exigency standpoint, there are important distinctions
between the immediate emergency present in Mincey and the situation that the officers
faced here, see generally Wynn Concurring Op. 40 (describing operative facts in this case
as amounting to “gunshots in a high-crime area”).
17

 569 U.S. at 148 (explaining that because warrants “are ordinarily required for searches of
dwellings” absent an emergency, “no less [can] be required” for blood-alcohol-content
testing, which is an “intrusion[] into the human body” (quoting Schmerber v. California,
384 U.S. 757, 770 (1966))). Warrantless seizures of persons, by contrast, are governed by
the law of arrests as well as the Terry line of cases, which itself is an exception to the Fourth
Amendment’s warrant requirement. See Terry, 392 U.S. at 20; see also Mora, 519 F.3d at
222 (“[B]ecause the ultimate touchstone of the Fourth Amendment is reasonableness, the
warrant requirement is subject to certain exceptions—as when exigent circumstances
justify the warrantless search of a home, or when the need for on-the-spot response justifies
a search based on reasonable suspicion.” (emphasis added) (citations and internal quotation
marks omitted)).
Thus, applying Brigham City—or any exigency case, for that matter—proves
challenging here. While some of the abstract principles articulated in the home-entry cases
may be relevant to our inquiry, we have little guidance on when and how the exigent
circumstances exception may apply to a suspicionless, investigatory seizure. Nevertheless,
as we explain below, the few cases that have extended the exigent circumstances exception
to such seizures all involve specific and clear limiting principles that were absent in Curry’s
stop. As a result, we are confident that the exigent circumstances exception does not apply
here.

18

 B.
There are relatively few cases that purport to extend the exigent circumstances
exception to suspicionless, investigatory seizures of a person. But in each of these cases,
officers typically have searched for a suspect implicated in a known crime in the immediate
aftermath of that crime, and—per that objective—have isolated a geographic area with
clear boundaries or a discrete group of people to engage in minimally intrusive searches.
One line of cases purporting to apply the exigent circumstances exception to
suspicionless seizures involves law enforcement officers establishing vehicular
checkpoints along routes that they reasonably expect will be used by suspects leaving the
scene of a known crime. For example, in United States v. Harper, after officers observed
a docked boat holding illicit drugs, an agent stationed his car close to the “only paved road
with access to the [landing area].” 617 F.2d 35, 40 (4th Cir. 1980). In holding that a
suspicionless stop resulting from the checkpoint was constitutional, we noted that “[t]he
purpose of these stops was to arrest suspects for a known crime, not to discover evidence
of undetected crimes by happenstance of visual searches.” Id. Officers’ observation of the
crime, the suspects, and the likely escape route allowed them to narrowly target individuals
“known to be fleeing the scene along a route reasonably expected to be used for their
escape.” Id. at 40–41; see also United States v. Paetsch, 782 F.3d 1162, 1170–71 (10th
Cir. 2015) (holding that a police barricade at an intersection and the search of twenty cars
without individualized suspicion was “appropriately tailored” because a GPS tracker
showed that “the stolen money (and likely the armed criminal who stole it) sat in a car
idling at that very intersection”).
19

 Beyond the context of vehicular checkpoints, courts have similarly required that
officers have specific information about the crime and suspect before engaging in
suspicionless seizures. In Palacios v. Burge, the Second Circuit considered whether the
exigent circumstances exception applied to the detention of a group of people in a nightclub
without particularized suspicion as to any one of the individuals. 589 F.3d 556, 563 (2d
Cir. 2009). Officers surveilling the club had encountered a man who told them that his
brother had been stabbed and that the assailants had run into the club. Id. at 559. Moments
earlier, the officers had seen a group of men matching the description of the assailants enter
the club. Id. The officers sealed the exits to the club, detained all the individuals inside,
and lined up approximately 170 men who matched the description of the suspect. Id. The
Second Circuit held that this detention and show-up was not unreasonable because “police
knew that the perpetrators were within the finite group of men . . . inside or lined up outside
of the . . . [c]lub.” Id. at 563. Moreover, police feared that witnesses on the scene who
could identify the suspect would not be available at a later time. Id.
In both the vehicular checkpoint cases and in Palacios, officers possessed specific
information about the suspect and the alleged crime that allowed them to narrowly confine
their suspicionless seizures. The officers who stopped Curry had no such information.
Here, although the officers heard shots, moved toward them, and received corroboration
regarding the general location of the shots in or near an apartment complex, the officers
did not have any specific description of the scene of the shooting or the perpetrator. It is
also clear that the officers did not stop everyone close to the scene, akin to a checkpoint.

20

 Unlike the officer in Harper, who stationed his car along the only escape route from
the site of an observed crime, here the officers approached Curry in an open field, at one
of several possible escape routes, in an area that they only suspected to be near the scene
of an unknown crime. See 617 F.2d at 40. Likewise, the officers lacked a description of
the suspect’s appearance or, more importantly, any indication that the suspect was in the
vicinity—factors that were central to the court’s analysis in Paetsch. See 782 F.3d at 1168.
And unlike the officers in Palacios, who were virtually certain that those responsible for a
known crime were in an immediately discernible and discrete location, officers here had
no reason to believe that the men walking in the field had anything to do with the gunshots
they heard. See 589 F.3d at 563. Indeed, the fact that the officers stopped those walking
in the field but not those standing closer to the apartment complex—who were closest to
the reported location of the shots—illustrates the relatively unrestricted nature of the
search.
In addition to allowing the officers to narrowly confine their suspicionless seizures,
the facts known to the officers in the previously described cases enhanced the exigency in
a way that the facts known to the officers in this case could not have. Because the officers
there had specific information about the suspect and the alleged crime, it was clear that
there was a definitive and ongoing crime that required emergency intervention. The sound
of gunshots could not have given the officers that knowledge here, and, therefore, they also
did not give rise to a similar exigency.

21

 *

*

*

In sum, the exigent circumstances exception may permit suspicionless seizures
when officers can narrowly target the seizures based on specific information of a known
crime 8 and a controlled geographic area. This reading of the exception does not transform
it into individualized suspicion by another name. After all, the officers in Harper and
Palacios had no particularized suspicion as to any individual they seized. Nor does it
require that officers be virtually certain that one of the individuals they stop is the suspect.

8

To be clear, we agree with our dissenting colleagues that exigent circumstances do
not arise only in the context of criminal activity. See Richardson Dissenting Op. 95. But
as we explain above, we deal here with a different type of Fourth Amendment encounter—
a non-consensual pedestrian stop—than the types of encounters that are addressed in a
typical exigency case. In the context of a non-consensual pedestrian stop, it is difficult for
us to imagine an underlying exigency that is divorced from criminal activity and yet poses
a danger to public and officer safety. Perhaps a non-criminal emergency may one day arise
that would require officers to render aid or prevent imminent injury to the person being
stopped, as in the exigency-justified home-entry cases discussed by the dissent. See
Richardson Dissenting Op. 96 (citing Brigham City, 547 U.S. at 403–04 (injured civilians);
Fisher, 558 U.S. at 47–58 (same); Tyler, 436 U.S. at 509–10 (burning building); Taylor,
624 F.3d at 631–33 (wandering children)). But that is not this case, and we strain to think
of any realistic examples.
Moreover, by definition, the exigency cases cited elsewhere in the principal
dissent—which mention a safety-driven preliminary sweep of a crime scene, see Mincey,
437 U.S. at 392, or sanction a safety-driven temporary seizure at a crime scene, see Figg
v. Schroeder, 312 F.3d 625, 639 (4th. Cir. 2002)—relate to criminal activity in some way.
And even in the temporary seizure case, where we stated that exigent circumstances permit
a warrantless search or seizure without probable cause that a crime has been committed,
Figg, 312 F.3d at 639, there was at least some level of particularized suspicion relating to
safety that was tethered to the underlying crime under investigation. See id. at 640
(explaining that the officers who arrived on scene knew (1) that there had been a shooting,
but not that the shots had been fired by a fellow officer; (2) that the family members who
were detained on scene had a “reputation for violence and possession of assault weapons
[that] was well-known to many members of the police force”; and (3) that the officers’
presence was “strongly objectionable” to the detainees, “who were understandably angry
and distraught” that one of their family members had been shot).
22

 But officers must support their “objectively reasonable belief” that there is an emergency
with “specific articulable facts and reasonable inferences.” See Yengel, 711 F.3d at 397.
Allowing officers to bypass the individualized suspicion requirement based on the
information they had here—the sound of gunfire and the general location where it may
have originated—would completely cripple a fundamental Fourth Amendment protection
and create a dangerous precedent.

C.
In urging us to conclude otherwise, the government invites us to apply a different
analytical framework. Rather than asking whether the circumstances faced by the officers
here were “exigent” under existing precedent—and, if so, whether such precedent should
apply with equal force in the context of an investigatory seizure of a person (a category of
encounters ordinarily governed by Terry)—the government insists that we must simply ask
whether Curry’s stop was “reasonable.” See Gov’t Br. 15, 29, 31. And to answer that
question, the government says, we must balance “the gravity of the public concerns served
by the seizure, the degree to which the seizure advances the public interest, and the severity
of the interference with individual liberty.” Gov’t Br. 31–32 (alterations omitted) (quoting
Brown v. Texas, 443 U.S. 47, 50–51 (1979)).
To support this argument, the government principally relies on Illinois v. McArthur,
531 U.S. 326 (2001). See Oral Arg. 2:05–3:20, 1:26:40–1:28:52. In McArthur, the
Supreme Court held that the refusal to allow the defendant to enter his own dwelling
without a police officer for two hours while the police obtained a search warrant did not
23

 violate the defendant’s Fourth Amendment rights. Because the surrounding circumstances
in that case gave rise to a “plausible claim of . . . exigent circumstances,” namely, the need
to prevent the destruction of evidence—and because the restraint at issue was tailored to
that need and avoided significant intrusion to the home itself—the Court refused to impose
a per se rule of unreasonableness despite the absence of a warrant. See McArthur, 531 U.S.
at 331. Instead, it balanced the relevant “privacy-related and law-enforcement related
concerns” to determine if the temporary seizure was reasonable under the Fourth
Amendment. Id.
We decline the government’s invitation to apply a free-form balancing test to
determine the reasonableness of Curry’s stop here, and we do so for two reasons.
First, although McArthur might seem like a pure “exigent circumstances” case at
first blush, particularly given its explicit use of that term, a closer examination reveals that
it is not.
McArthur addressed the lawfulness of a two-hour seizure of the defendant’s home
and thus is better viewed as an extension of Supreme Court precedent regarding the
temporary seizure of property. See, e.g., United States v. Place, 462 U.S. 696, 702 (1983)
(extending “the principles of Terry” to permit temporary, investigatory seizures of luggage
on the basis of reasonable, articulable suspicion that “the luggage contains contraband or
evidence of a crime”); see also Riley v. California, 573 U.S. 373 (2014) (citing McArthur
for proposition that officers could have seized and secured respondents’ cell phones “to
prevent destruction of evidence while seeking a warrant”); Georgia v. Randolph, 547 U.S.
103, 116 n.6 (2006) (stating that “if [an] objecting tenant cannot be incapacitated from
24

 destroying easily disposable evidence during the time required to get a warrant,” per
McArthur, then “a fairly perceived need to act on the spot to preserve evidence may justify
entry and search” under the Court’s exigent circumstances precedent (emphasis added)).
Indeed, in referencing the possibility of exigent circumstances, the McArthur Court cited
not only to its destruction-of-evidence exigency cases but also to Place itself. McArthur,
531 U.S. at 331; see also id. at 332–33 (citing again to Place in discussing the limited
nature of the restraint, i.e., one of the four factors that led the Court to conclude that the
brief seizure of the premises was lawful); id. at 334 (citing to Place once more in pointing
to the “various other circumstances” in which the Court has “upheld temporary restraints
where needed to preserve evidence until police could obtain a warrant”).
Justice Souter highlighted this point in his concurrence:
Respondent McArthur’s location made the difference between the exigency
that justified temporarily barring him from his own dwelling and
circumstances that would have supported a greater interference with his
privacy and property. As long as he was inside his trailer, the police had
probable cause to believe that he had illegal drugs stashed as his wife had
reported and that with any sense he would flush them down the drain before
the police could get a warrant to enter and search. This probability of
destruction in anticipation of a warrant exemplifies the kind of present risk
that undergirds the accepted exigent circumstances exception to the general
warrant requirement. That risk would have justified the police in entering
McArthur’s trailer promptly to make a lawful, warrantless search. When
McArthur stepped outside and left the trailer uninhabited, the risk abated and
so did the reasonableness of entry by the police for as long as he was
outside. . . .
Since, however, McArthur wished to go back in, why was it reasonable to
keep him out when the police could perfectly well have let him do as he
chose, and then enjoyed the ensuing opportunity to follow him and make a
warrantless search justified by the renewed danger of destruction? The
answer is not that the law officiously insists on safeguarding a suspect’s
privacy from search, in preference to respecting the suspect’s liberty to enter
25

 his own dwelling. Instead, the legitimacy of the decision to impound the
dwelling follows from the law’s strong preference for warrants, which
underlies the rule that a search with a warrant has a stronger claim to
justification on later, judicial review than a search without one. The law can
hardly raise incentives to obtain a warrant without giving the police a fair
chance to take their probable cause to a magistrate and get one.
Id. at 337–38 (Souter, J., concurring) (citations omitted).
In short, then, McArthur is a case about when, with suspicion, police may
temporarily seize property to prevent the potential destruction of evidence. It does not
address when police may, without any suspicion at all, seize a person. Thus, McArthur
does not provide the governing framework here. 9

9

This conclusion makes further sense in light of two other related considerations.
First, although the Supreme Court in Place engaged in the same general balancing that it
applied in Terry—a move that itself sparked criticism from some Justices—the Court did
so prior to announcing a new rule, which, in that case, permitted Terry seizures of luggage
based on reasonable suspicion. See Place, 462 U.S. at 702–03; see also infra pp. 28–29
(discussing balancing in Terry). But see id. at 718 (Brennan, J., concurring) (characterizing
the majority’s reliance on the Terry line of cases as authority for its reasonableness
balancing as “inappropriate,” and cautioning that “balancing inquiries should not be
conducted except in the most limited circumstances”); id. at 721 (Blackmun, J., concurring
in the judgment) (“I cannot fault the Court’s desire to set guidelines for [lower courts] . . . .
I am concerned, however, with what appears to me to be an emerging tendency on the part
of the Court to convert the Terry decision into a general statement that the Fourth
Amendment requires only that any seizure be reasonable.”). McArthur arguably does the
same thing, only in a different context and in an area of doctrinal overlap. Second, the
Supreme Court does not typically apply a free-form balancing test in pure exigency cases;
instead, it focuses on the more specific question of whether “the exigencies of the situation
make the needs of law enforcement so compelling that [a] warrantless search is objectively
reasonable under the Fourth Amendment.” See King, 563 U.S. at 460 (alteration in
original) (quoting Mincey, 437 U.S. at 394); see also Mitchell, 139 S. Ct. at 2535 n.3
(plurality opinion) (explaining that the totality-of-the-circumstances test “has not
prevented [the Court] from spelling out a general rule for the police to follow,” and that
the Court’s exigency case law is “full of [such] general rules,” like the rule that police may
proceed without a warrant “when an occupant of a home requires ‘emergency assistance’”
(first quoting McNeely, 569 U.S. at 168 (Roberts, C.J., concurring in part and dissenting in
26

 Second, even were a balancing test appropriate here, McArthur would tell us little
about the appropriate balance to be struck in a case such as this one.
As even the government acknowledged at oral argument, McArthur is factually
distinguishable from this case in almost every respect, especially with regard to suspicion
(or lack thereof). In McArthur, police officers had probable cause to believe that the
defendant’s home contained unlawful drugs; here, the government concedes that the police
had no reasonable basis to suspect that Curry fired the gunshots.
Terry, on the other hand, has much to say about the appropriate balance of lawenforcement and privacy-related interests in the context of a suspicionless, investigatory
stop. Thus, even if we were inclined to apply a balancing test in this case—an approach
we expressly reject for the reasons already stated above—we would find Terry itself to be
more persuasive on the question of balancing than any of the cases highlighted by the
government on appeal. While the Supreme Court did not apply the exigent circumstances
exception in Terry, its ruling was premised on the same general type of “exigenc[y]” that
exists here—namely, the need to “discover[] . . . weapons which might be used to harm the
officer or others nearby.” See 392 U.S. at 26 (citing Warden v. Hayden, 387 U.S. 294, 310
(1967) (Fortas, J., concurring)).

part); then quoting Brigham City, 547 U.S. at 403)); cf. United States v. Colyer, 878 F.2d
469, 478 (D.C. Cir. 1989) (“[T]he Court has indicated that balancing is only appropriate
when warranted by special needs, beyond the normal need for law enforcement.” (internal
quotation marks omitted)).
27

 Recall that in Terry, an off-duty police officer stopped and frisked several men
without probable cause or a warrant. The officer did so after he became suspicious that the
men were about to rob in a store in downtown Cleveland in the middle of the afternoon and
feared that they might have a gun. See id. at 7. In assessing the constitutionality of the
stop and frisk, the Supreme Court noted at the outset that this type of police conduct was
not subject to the Fourth Amendment’s warrant requirement. See supra p. 11. Specifically,
the Court explained that although it did “not retreat from [its prior] holdings that the police
must, whenever practicable, obtain judicial approval of searches and seizures through the
warrant procedure, or that in most instances failure to comply with the warrant requirement
can only be excused by exigent circumstances,” Terry dealt with “an entire rubric of police
conduct” that “historically ha[d] not been, and as a practical matter could not be, subject to
the warrant procedure.” 392 U.S. at 20 (citations omitted). Accordingly, the Court
proceeded to evaluate the lawfulness of such police conduct under the Fourth
Amendment’s general proscription against unreasonable searches and seizures, applying a
similar balancing test to the one employed later in McArthur. See id. at 20–21.
Turning first to the government’s interest, the Court observed that a stop and frisk
implicates not only a general interest in effective crime prevention and detention—which
is furthered by an investigative stop—but also a “more immediate interest” in ensuring that
the person who has been stopped is not “armed with a weapon that could unexpectedly and
fatally be used” against the officer. Id. at 22–23. More broadly, the Court explained that
it could not “blind [itself] to the need for law enforcement officers to protect themselves
and other prospective victims of violence in situations where they may lack probable cause
28

 for an arrest.” Id. at 24. Thus, it stated that when an officer reasonably concludes that
criminal activity is afoot and that “the individual whose suspicious behavior he is
investigating at close range is armed and presently dangerous to others,” the government
has a clear interest in taking “necessary measures to determine whether the person is in fact
carrying a weapon and to neutralize the threat of physical harm.” See id. At the same time,
the Court made clear that a frisk, like any other search, must “be strictly circumscribed by
the exigencies which justify its initiation,” namely, the discovery of weapons. Id. at 25–
26 (citing Hayden, 387 U.S. at 310 (Fortas, J., concurring)). Therefore, after balancing the
government’s interest against the nature and quality of the intrusion on individual rights,
the Court concluded that the Fourth Amendment permits “narrowly drawn authority to
permit a reasonable search for weapons for the protection of the police officer, where he
has reason to believe that he is dealing with an armed and dangerous individual, regardless
of whether he has probable cause to arrest the individual for a crime.” Id. at 27.
Although the Terry Court only discussed the threat to officer and public safety in
the context of assessing the legality of the frisk, as opposed to the initial stop, see id. at 22–
28; see also Robinson, 846 F.3d at 700, we cannot ignore the significance of the balance
struck by the Supreme Court in that case: Faced with a concern about a particular person
having a gun that might be used to inflict harm against investigating officers or others, the
Supreme Court drew the line at reasonable, individualized suspicion.
To be sure, we do not suggest that the exigency doctrine may never apply in the
Terry context. See supra Part III.B. But we are most reluctant to recalibrate the “balance”
struck by the Supreme Court in that case when, as here, the officers lacked the type of
29

 specific information and tailored response that justified a narrow extension of the exigency
doctrine to suspicionless seizures in Harper, Paetsch, and Palacios. Like any investigatory
search or seizure conducted when police respond to a suspected crime, Curry’s stop was
conducted with the safety of both the officers and the public in mind. But if that alone
could invoke a Fourth Amendment exception and eliminate the individualized suspicion
requirement, then the requirement would be rendered moot. Cf. Florida v. J.L., 529 U.S.
266, 272 (2000) (“Firearms are dangerous, and extraordinary dangers sometimes justify
unusual precautions. Our decisions recognize the serious threat that armed criminals pose
to public safety; Terry’s rule, which permits protective police searches on the basis of
reasonable suspicion rather than demanding that officers meet the higher standard of
probable cause, responds to this very concern.”). 10

10

To erase any doubt as to why we find Terry informative here, we pause to address
the dissent’s accusation that our analysis “attempts to shoehorn the government’s exigency
arguments,” which disavow any reliance on Terry, “into the Terry framework.”
Richardson Dissenting Op. 93. With all due respect, we think this criticism is misplaced.
It is the government that has attempted to shoehorn what would otherwise be a
straightforward Terry case into the exigent circumstances exception, which courts have
rarely applied to justify a pedestrian stop and even still have done so only in limited
circumstances not present here. Our discussion of Terry merely highlights why the facts
of this case and the exigency argument that the government has pursued on appeal are a
mismatch doctrinally, and why we are thus unwilling to disregard the general rule
governing investigatory seizures and condone a novel application of the exigent
circumstances exception. Furthermore, to the extent our dissenting colleagues complain
that Curry’s stop was not an “investigatory” stop contemplated by Terry because it was
based instead on “public[ ]safety,” see Richardson Dissenting Op. 93–94, we do not think
these two aims can be so neatly divided here. In this case, the officers were clearly acting
to investigate the source of the gunfire, even if in the interest of public safety writ large.
30

 D.
A few final points bear mentioning here. First, our ruling today will not hinder the
police’s ability to forcefully respond to emergencies such as active-shooter situations. We
simply hold that the limited and vague information regarding a possible crime that the
officers possessed in this case was insufficient. See Mincey, 437 U.S. at 392–93 (rejecting
the notion that even a “possible homicide presents an emergency situation demanding
immediate [warrantless] action”). And we need not define an “active shooter” situation to
conclude that the situation faced by the officers in this case—as described by the district
court and viewed in the light most favorable to Curry—was not one. The officers here
relied on the sound of five to six gunshots in the distance near an apartment complex,
corroborated by two 911 calls, and the sight of several men (separately and calmly) walking
in a public area nearby less than a minute later, including an apparent victim who seemed
to be favoring one of his arms. Nothing in our opinion is intended to preclude officers
facing actual exigent circumstances (or officers who reasonably believe that they are facing
such circumstances) from taking appropriately tailored action in the future.
Second, we note that in attempting to frame the situation as an emergency that
allowed police to stop and frisk Curry, the government makes much of the fact that there
were several shootings in the Creighton Court area in the weeks leading up to the incident
in question. Although this fact is certainly relevant to our analysis, it did not provide the
officers with the type of specific information that would be necessary to justify a
suspicionless seizure, even when combined with the other pertinent facts. See supra Part
III.B. Moreover, for the same reason a person’s presence in a high-crime area cannot alone
31

 create reasonable suspicion to justify a Terry stop, see Illinois v. Wardlow, 528 U.S. 119,
124 (2000); Brown, 443 U.S. at 52, we refuse to give this fact any special weight in
determining whether the officers faced exigent circumstances. To do so would deem
residents of Creighton Court—or any other high-crime area—less worthy of Fourth
Amendment protection by making them more susceptible to search and seizure by virtue
of where they live. Such an approach “risk[s] treating members of our communities as
second-class citizens,” see Utah v. Strieff, 136 S. Ct. 2056, 2069 (2016) (Sotomayor, J.,
dissenting), and we emphatically reject it here. As we have said in the past:
In our present society, the demographics of those who reside in high crime
neighborhoods often consist of racial minorities and individuals
disadvantaged by their social and economic circumstances. To conclude that
mere presence in a high crime area at night is sufficient justification for
detention by law enforcement is to accept carte blanche the implicit assertion
that Fourth Amendment protections are reserved only for a certain race or
class of people. We denounce such an assertion.
Black, 707 F.3d at 542.

IV.
For the foregoing reasons, the district court’s order granting Curry’s motion to
suppress is
AFFIRMED.

32

 GREGORY, Chief Judge, concurring:
Our decision today affirms that a central tenet of law nearly as old as this country—
namely, “[t]he right of the people to be secure . . . against unreasonable searches and
seizures”—applies equally to all. U.S. Const. amend. IV. I join the majority Opinion in
its entirety. However, I must say a few words in response to Judge Wilkinson’s dissent.
When I read the first line of Judge Wilkinson’s dissent I was heartened by the
thought: well, at least he acknowledges that there are “two Americas.” But this glint of
enlightenment was to serve as a “soap box” for his charge against the majority’s decision.
It is understandable that such a pseudo-sociological platform was necessary as his
assertions are bereft of any jurisprudential reasoning. More to the point, his recognition of
a divided America is merely a preamble to the fallacy-laden exegesis of “predictive
policing” that follows. Through his opinion, my colleague contributes to the volumes of
work gifted by others who felt obliged to bear their burden to save minority or
disadvantaged communities from themselves.
Of course, the story of two Americas of which Judge Wilkinson speaks is an ancient
tale to some. See, e.g., Frederick Douglass, “What to the Slave is the Fourth of July?”
1852. There’s a long history of black and brown communities feeling unsafe in police
presence. See, e.g., James Baldwin, A Report from Occupied Territory, The Nation, July
11, 1966 (“[T]he police are simply the hired enemies of this population. . . . This is why
those pious calls to ‘respect the law,’ always to be heard from prominent citizens each time
the ghetto explodes, are so obscene.”). And at least “[s]ince Reconstruction, subordinated
communities have endeavored to harness the criminal justice system toward recognition
33

 that their lives have worth.” Deborah Tuerkheimer, Criminal Justice and the Mattering of
Lives, 116 Mich. L. Rev. 1145, 1146 (2018). Thus, just a few decades ago, laws designed
to decrease violence in these communities were considered “a civil rights triumph.” James
Forman, Locking Up our Own: Crime and Punishment in Black America 73 (2017). The
thought being that our government had finally “promised to provide police protection to a
community so long denied it.” Id. This increased protection, however, led to what has
been described as “a central paradox of the African American experience: the simultaneous
over- and under-policing of crime.” Id. at 35.
Judge Wilkinson chooses to focus largely on one dimension of this paradox,
ignoring the details of the familiar perils of over-policing. See, e.g., Marie Gottschalk,
Caught: The Prison State and the Lockdown of American Politics (2015); Michael Tonry,
Punishing Race: A Continuing American Dilemma (2011); Michelle Alexander, The New
Jim Crow (2010); Khalil Gibran Muhammad, The Condemnation of Blackness: Race,
Crime, and the Making of Modern Urban America (2010); Ruth Wilson Gilmore, Golden
Gulag (2007). Describing the hazard of “hot spot policing” as “the danger of overreaction,”
Wilkinson Dis. Op. at 68, Judge Wilkinson mitigates the concerns of some that any
encounter with an officer could turn fatal. See Utah v. Strieff, 136 S. Ct. 2056, 2070 (2016)
(Sotomayor, J., dissenting) (describing “the talk” that black and brown parents frequently
give to their children “all out of fear of how an officer with a gun will react to them”); see
also United States v. Black, 707 F.3d 531, 541 (4th Cir. 2013) (“In certain communities
that have been subject to overbearing or harassing police conduct, cautious parents may
counsel their children to be respective, compliant, and accommodating to police officers,
34

 to do everything officers instruct them to do.”). In so doing, my dissenting colleague in
turn presents a sordid view of under-policing, suggesting that our decision today will lead
to “an America where gated communities will be safe enough and dispossessed
communities will be left to fend increasingly for themselves.” Wilkinson Dis. Op. at 69.
But we know that many of our fellow citizens already feel insecure regardless of
their location. In a society where some are considered dangerous even when they are in
their living rooms eating ice cream, asleep in their beds, playing in the park, standing in
the pulpit of their church, birdwatching, exercising in public, or walking home from a trip
to the store to purchase a bag of Skittles, it is still within their own communities—even
those deemed “dispossessed” or “disadvantaged”—that they feel the most secure.
Permitting unconstitutional governmental intrusions into these communities in the name of
protecting them presents a false dichotomy. My colleague insists on a Hobson’s choice for
these communities:

decide between their constitutional rights against unwarranted

searches and seizures or forgo governmental protection that is readily afforded to other
communities. But those inclined to shrug their shoulders at citizens who wave their
Constitutions in the air during uncertainty must not forget “[h]istory teaches that grave
threats to liberty often come in times of urgency, when constitutional rights seem too
extravagant to endure.” Skinner v. Ry. Labor Executives’ Ass’n, 489 U.S. 602, 635 (1989)
(Marshall, J., dissenting); cf. Korematsu v. United States, 323 U.S. 214 (1944). Indeed, it
is in moments of insecurity that our constitutional bells ring the loudest.
Why even suppose that checking police power in these circumstances would lead to
some communities falling into a Hobbesian state of nature? It’s unclear. Judge Wilkinson
35

 supports this slippery slope argument in a couple of mutually incompatible and individually
questionable ways. He mentions Professor Rod K. Brunson’s work on policing to bolster
the view that our decision here will further entrench the perception that police fail to serve
those in disadvantaged communities. But Professor Brunson has long argued that this
perception is largely created by aggressive policing strategies and discourteous treatment
of members in their community. See, e.g., Rod K. Brunson, “Police Don’t Like Black People”:
African-American Young Men’s Accumulated Police Experiences, 6(1) Criminology &
Pub. Pol’y 71 (2007). Indeed, Professor Brunson has noted that “arrests and successful
prosecutions are unlikely without cooperating witnesses.” Rod K. Brunson, Protests focus
on Over-policing. But under-policing is also Deadly, Wash. Post, June 12, 2020. And
those from disadvantaged communities “want a different kind of policing than the
aggressive approaches they typically see—one that values their humanity.” Id.; see also
Estate of Jones v. City of Martinsburg, W. Va., –– F.3d ––, 2020 WL 3053386, at *7 (4th
Cir. 2020) (recognizing a “desperate need” for more and different police training).
From this perspective, the video of the present incident mimics the aggressive,
discourteous, and ineffective policing that concern many. As the officers approached the
scene seconds after gunshots rang out, the members of this community, including Curry,
pointed them in the direction in which the perpetrator was likely to be found. Because, as
Judge Diaz notes in his concurrence, it would have been difficult for the officers “to
determine whether any firearm (which, of course, are generally lawful to possess) seized
in the effort to identify the suspect was the source of the gunfire,” Judge Diaz Op. at 57,
one would think that the officers’ best hope for finding the shooter was to accept the
36

 guidance offered by community members. See Black, 707 F.3d at 540 (“Being a felon in
possession of a firearm is not the default status.”). That, of course, was not the case here.
Cf. Miranda Fricker, Epistemic Injustice 4 (2007) (describing the notion of “testimonial
injustice,” where a speaker suffers from deflated credibility owing to an identity prejudice
on the hearer’s part). The officers ignored the assistance and the shooter got away. Like
most citizens, it is likely that residents of the Creighton Court community do not want
police officers to be tough on crime, or weak on crime—they want them to be smart on
crime.
No doubt it is beyond the scope of our roles to explain to any institution what it
means to be smart on crime. I will leave that to our clever colleagues in the chambers of
City Council. But it is “emphatically the province and duty of the judicial department to
say what the law is.” Marbury v. Madison, 5 U.S. 137, 177 (1803). Thus, “[i]n some
circumstances . . . we must remind law enforcement that the Fourth Amendment protects
against unreasonable searches and seizures,” and that those protections extend to all people
in all communities. Black, 707 F.3d at 534. This is one of those circumstances.
Contrary to Judge Wilkinson’s suggestion, our decision today does not deliver “a
gut-punch to predictive policing.” Wilkinson Dis. Op. at 71. As Judge Wilkinson notes,
predictive policing programs “differ in their details,” but generally seek to use “smart
policies” to “affirmatively prevent crime from happening, rather than just solve it.” Id. at
65; see also Andrew Guthrie Ferguson, Predictive Policing and Reasonable Suspicion, 62
Emory L.J. 259, 265 (2012) (“In simple terms, predictive policing involves computer
models that predict areas of future crime locations from past crime statistics and other
37

 data.”). But see id. at 321 (“Predictive policing may well become an effective tool for law
enforcement. Yet, the technology will also create tension for police in defending Fourth
Amendment challenges by defendants.”); Andrew Guthrie Ferguson, Policing Predictive
Policing, 94 Wash. U. L. Rev. 1113, 1149 (2017) (“More bluntly, the initial predictive
policing projects have raised the question of whether this data-driven focus serves merely
to enable, or even justify, a high-tech version of racial profiling.”). But, as with all policies,
the devil is going to lie in those details. Nothing in the majority Opinion prevents the
police from using, in good faith with constitutional principles, smart policies to identify
where crimes may occur and accordingly dispatching officers to those neighborhoods. But
it is how they, upon arrival, engage with the people in those neighborhoods that is important
here. A suspicionless, investigatory stop was not warranted under the circumstances.
Affirming our long-standing rules is nothing novel. If merely preventing crime was enough
to pass constitutional muster, the authority of the Fourth Amendment would become moot.
Don’t get me wrong—I understand the frustrations and uncertainties that attend
most discussions of how to abate crime. As a country, we are in a moment of reckoning.
And the unpredictability of the future encourages us to want to hang on to those entities
that make us feel secure. Still, “[t]he facts of this case give us cause to pause and ponder
the slow systematic erosion of Fourth Amendment protections for a certain demographic.”
Black, 707 F.3d at 542. The “lifelines a fragile community retains against physical harm
and mental despair,” Wilkinson Dis. Op. at 70, must be the assurance that there truly is
equal protection under law. Thus, “[i]n the words of Dr. Martin Luther King Jr., we are
[once again] reminded that ‘we are tied together in a single garment of destiny, caught in
38

 an inescapable network of mutuality,’ [and] that our individual freedom is inextricably
bound to the freedom of others.” Black, 707 F.3d at 542. It is with these truths that I join
my colleagues in the majority in ensuring that “the Fourth Amendment rights of all individuals
are protected.” Id. (emphasis in original).

39

 WYNN, Circuit Judge, concurring:
I fully concur in the majority’s opinion and briefly write to address Judge
Wilkinson’s discussion of predictive policing, and Judge Richardson’s discussion of the
operative facts that supposedly created exigency in this case: gunshots in a high-crime area.
I.
Turning first to Judge Wilkinson’s laudatory treatment of predictive policing
metrics, I note that my colleague relies on statistical inferences derived from social science
studies on predictive policing, and concludes that it may be appropriate to trade more
complete information about criminal wrongdoing for faster responses by police in highcrime areas. That reliance is in tension with the Supreme Court’s expressed reluctance to
use sociological studies and statistical inferences—even those with a more straightforward
scientific or mathematical basis than is present here—in deciding constitutional issues.
Consider Gill v. Whitford, 138 S.Ct. 1916 (2018), a challenge to a computerized,
statistically-generated partisan gerrymander. The plaintiffs introduced a statistical metric
that measured the extent to which a legislative redistricting map was biased in favor of a
political party. That metric, known as the “efficiency gap,” was a calculation similar to a
percentage. It first required counting how many votes were “wasted” by each party in an
election: for the losing party, how many votes were cast, and for the winning party how
many votes were cast beyond the winning threshold of fifty percent. The efficiency gap
was then calculated by taking the difference of wasted votes between the two parties and
dividing by the total number of votes cast in the election. See Brief for RespondentAppellee at 12-13, Gill v. Whitford, 138 S.Ct. 1916 (2018) (No. 16-1161).
40

 Although that metric encapsulated a fulsome statistical dataset via arithmetic, the
Supreme Court was skeptical of its utility. See Transcript of Oral Argument at 40, Gill v.
Whitford, 138 S. Ct. 1916 (2017) (No. 16-1161) (Roberts, C.J., characterizing the metric
as “sociological gobbledygook”); see also id. at 43 (Alito, J., asking, “Has there been a
great body of scholarship that has tested this efficiency gap? It’s full of questions.”).
I see no reason why the sociological sources cited by Judge Wilkinson, which
simply note that predictive systems receive a variety of input data and process it to generate
a statistical map of potential crime hot spots, should receive any less skepticism in our legal
analysis. See Chandler Harris, Richmond, Virginia, Police Department Helps Lower Crime
Rates with Crime Prediction Software, Government Technology (Dec. 21, 2008) (stating
that the Richmond Police Department generates hot-spot maps by inputting “all the crimes
in the past, weather, times of day, day of week, and moon phases” into a computer system,
but declining to explain how such maps are generated, and instead concluding without
further explanation that the system helped the department reduce major crime by 21
percent).
Likewise for studies which utilize statistical modeling and analysis far more
complex than in Gill. For instance, Judge Wilkinson relies on an article noting that surveys
have shown that most hot-spot policing programs reduce crime. See Wilkinson Dis. Op. at
66 (citing David Weisburd, Does Hot Spot Policing Inevitably Lead to Unfair and Abusive
Police Practices, or Can We Maximize Both Fairness and Effectiveness in the New
Proactive Policing?, 2016 U. Chi. Legal F. 661, 670 (2016)). That article, in turn, relies on
a complex survey that describes the use of, inter alia, “standardized mean difference effect
41

 sizes for all outcomes,” “chi-square values comparing the difference in calls for service at
RECAP and control targets before and after the intervention,” “effect sizes from exact pvalues from the F tests used in the two-way analysis of variance calculations for calls for
service data,” “standardized mean effect sizes based on the t-test results reported for the
intervention variables’ effects on the outcome variables,” “odds ratios based on reported
pre-test and post-test (or intervention period) crime outcome counts,” and “standardized
difference in means between the treatment and control or comparison conditions (effect
size) with a 95 percent confidence interval plotted around them for all tests.” Anthony
Braga et al., Hot Spots Policing Effects on Crime 24-25 (David B. Wilson & Charlotte Gill
eds., Campbell Systematic Reviews 8th ed. 2012). That sounds just like what the Chief
Justice in Gill called “gobbledygook.” And although I believe the Supreme Court was
wrong to discount the scientific analysis in Gill, we are bound by that decision until the
Supreme Court decides otherwise or Congress acts.
While Judge Wilkinson agrees “that the phrase ‘high-crime area’ cannot serve as a
facile excuse for indiscriminate interventions,” I am puzzled that in the next sentence, he
advocates doing just that by stating, “neither can skepticism toward the preventive potential
of predictive policing in violent crime locales allow us to deny its benefits ab initio to
communities that might welcome them.” Wilkinson Dis. Op. at 67. Justifying predictive
policing on the policy basis that neighborhoods—whether termed “violent crime locales”
or “high-crime areas”—“might welcome [it]” still results in the citizens of those
communities being accorded fewer constitutional protections than citizens of other
communities, as the police accept Judge Wilkinson’s proposed tradeoff: “faster responses
42

 for fuller information.” Id. Such an outcome fails as a matter of law. We may not treat
citizens as second-class simply because they live in areas that my good colleague calls
“violent crime locales.” See, e.g., United States v. Black, 707 F.3d 531, 542 (4th Cir. 2013).
I am even more puzzled by Judge Wilkinson’s attempt to dismiss the Supreme
Court’s skeptical treatment of social science in Gill as concerning an “inapposite line of
cases.” Wilkinson Dis. Op. at 67. The studies he cites fall into the same class of statisticsbased sociological reasoning that our high court, noting that such topics are beyond judicial
expertise, hesitated to credit in Gill. See Transcript of Oral Argument at 40, Gill v.
Whitford, 138 S. Ct. 1916 (2017) (No. 16-1161). Following the Chief Justice’s reasoning
in Gill, we are no better equipped to evaluate the statistical social science in this case.
Rather, that policy expertise would rest with the legislative branch of the government, not
this Court.
I do not wish to decide the constitutional question before us solely on the relative
merits of sociological studies. Nor do I believe we should convert our Fourth Amendment
jurisprudence into a policy-based battle of the experts. But it is worth pointing out that
Judge Wilkinson’s reliance on sociological studies simply substitutes policy considerations
for legal analysis.
We must be mindful of the potential for new police technologies to erode
constitutional protections. See, e.g., Carpenter v. United States, 138 S. Ct. 2206, 2223
(2018). It follows that to make an informed constitutional determination, we must
understand how those technologies work, and how they shape the behavior of the police

43

 and the public. 1 At the same time, our analysis must stay rooted in constitutional principles,
rather than turn on naked policy judgments derived from our perception of the beneficial
effects of novel police techniques.
Such policy decisions are better left to the representative branches of our
government, which even now are addressing unprecedented national protests regarding the
devastating effects of over-policing on minority communities—protests sparked by the
very deaths my colleague decries as militating greater police presence in disadvantaged
areas. It is through that conversation that the appropriate policy balance will be struck.
II.
I turn now to the operative facts of this case. This matter arose when the police
responded to gunshots in the Creighton Court housing development in Richmond—a socalled “high-crime area.” From that response flow various facts that Judge Richardson’s

1

To that end, as Judge Thacker aptly explains, talismanic references to
technological terms such as “big data” and “machine learning” should not be used as a
screen for objectivity when analyzing the constitutional sufficiency of hot-spot policing
programs and the law enforcement responses they enable. Wilkinson Dis. Op. at 66. As
one scholar has noted:
By design, machine learning algorithms learn and reproduce the data they
are given. If the data police provide to these systems already reflects a variety
of priorities, filters, and decisions, then the results will too repeat those
choices. And as police rely upon these predictive policing results to deploy
their resources, they produce even more data that appear to confirm what the
algorithm has predicted. That feedback loop reproduces a pattern of future
policing, not future crime.
Elizabeth E. Joh, Feeding the Machine: Policing, Crime Data, & Algorithms, 26 Wm. &
Mary Bill Rts. J. 287, 300-01 (2017) (footnote omitted). So, predictive algorithms may
only potentially confirm (or reinforce) what the police already know (or believe) about
where crime is occurring.
44

 dissent relies upon to support its finding of exigency. But giving the two dispositive facts
in this case—gunshots in a high-crime area—their appropriate weight, that dissent dictates
one of two conclusions. Either people who happen to live in high crime areas are subject
to a lower degree of Fourth Amendment protection, or any time a gun is fired in a populated
area, the police may conduct suspicionless and wholly discretionary stops of all individuals
in the vicinity. The former conclusion is untenable, as we may not relegate individuals who
live in high-crime areas to second-class citizen status. The latter conclusion suggests that
Fourth Amendment protections are lessened for those who own firearms or happen to be
within earshot of gunfire.
A.
Whether exigent circumstances support police action is determined by considering
the totality of the circumstances. See Mitchell v. Wisconsin, 139 S. Ct. 2525, 2535 n.3
(2019). We must candidly consider the particular facts the government offers that
supposedly created exigency here. There are just two: (1) the presence of gunshots (2) in a
high crime area. When the police officers stopped Curry, they had no reliable information
regarding any person who was shot, no suspect description, and no concrete indication of
a crime beyond discharge of a firearm in a public area—a Class 1 misdemeanor in the
Commonwealth of Virginia. See Va. Code § 18.2-280(A) (2005). The officers had simply
heard five to six gunshots. Based on that information, they approached a handful (but not
all) of the individuals in an open field, which was only one of several “escape” routes
adjacent to an area they suspected to be the scene of a serious crime.

45

 Judge Richardson’s dissent adopts a host of additional factors conjured by the
government to bolster its analysis and find exigency: that the officers saw a man who
appeared to be favoring one arm walking calmly near the individuals stopped, allegedly
making the officers believe the man may have been shot (though the officers who stopped
Curry did not pause to evaluate whether that man was harmed); that they observed people
near the scene of the crime (though all the video record shows is the men calmly walking
in the vicinity of the area where the officers thought the gunshots originated); that the police
learned of 911 calls reporting gunshots as they exited their cruiser (though these calls
merely reported discharge of a firearm, which the officers had already heard); and that they
were near the gunshots and arrived on the scene shortly thereafter (though the mere fact
that the police responded quickly does not change what evidence of exigency they had at
their disposal).
Other than taking up space on the page, these recitations add nothing to the essential
facts of this case: gunshots in a high-crime area. Simply put, they are factors that would
arise in any instance in which the police respond to gunfire in a populated area. Calls to the
police are common when guns are fired. Similarly, individuals will tend to move away
from the vicinity of gunshots, while the police will rapidly converge. Nor does the presence
of someone who, in the recollection of the officer who stopped Curry, “appeared to be
maybe favoring one of his arms,” J.A. 257, and who was walking calmly in a field with the
individuals the officers stopped as potential shooters, suggest a victim of an active
shooting.

46

 At bottom, the police heard gunfire in a high-crime area. Those are the essential
facts contributing to exigency in this case, and our analysis must give them the appropriate
weight.
B.
Once this case is boiled down to its operative facts, Judge Richardson’s dissent
suggests that the police may conduct wholly discretionary stops of individuals merely
because they live in high crime areas or happen to be in the vicinity of gunshots.
As ably noted by the majority and Chief Judge Gregory’s concurrence,
consideration of the high crime area alone is anathema under our jurisprudence. Individuals
who happen to live in high crime areas are not second-class citizens. 2 See Black, 707 F.3d
at 542; United States v. Griffin, 589 F.3d 148, 158 (4th Cir. 2009) (Gregory, C.J.,

2

By “second-class citizens,” I refer to law-abiding individuals who happen to reside
in disadvantaged communities and who, unlike their counterparts in more privileged areas,
would suffer daily the indignities of the “wide-ranging exploratory searches the Framers
intended to prohibit” were we to accept the dissents’ proposed diminution of Fourth
Amendment protections. Maryland v. Garrison, 480 U.S. 79, 84 (1987). People attempting
to conduct daily business in communities like Creighton Court—a neighborhood where
“hundreds of families live”—would face wholly arbitrary stops of themselves and their
children by police officers primed for action in a “high-crime” area. J.A. 256. Indeed, the
video record in this case demonstrates the belligerent, humiliating, and capricious nature
of such stops. Nor is it clear the police would limit themselves to directing a person to raise
their hands and show their waist, as occurred here. If exigency justifies that conduct, it
would seem to readily extend to more intrusive searches of individuals’ persons and
property—at least if one lives in a “high-crime” area. Subjecting people in disadvantaged
areas to that “too permeating police surveillance” while declining to do the same for those
in wealthier communities relegates the less fortunate to second-class status in the eyes of
the law. United States v. Di Re, 332 U.S. 581, 595 (1948).
47

 dissenting); see also Utah v. Strieff, 136 S. Ct. 2056, 2069 (2016) (Sotomayor, J.,
dissenting); Illinois v. Wardlow, 528 U.S. 119, 123 (2000).
Once the high-crime-area factor is properly discounted, we are left with the fact of
gunshots. If that element alone sufficed for exigency, we would sanction a broad rule
indeed: any time the police hear gunshots in a populated area, they may conduct wholly
discretionary, suspicionless stops of anyone in the vicinity. That would include, as relevant
here, the multiple people who were outside near Walcott Place at 9:00 PM on a Friday in
September and the hundreds of people who lived within earshot of the gunfire. The
government acknowledges as much, framing such widespread seizures as “controlling the
scene.” Reply Br. at 13. However, Judge Richardson’s dissent declines to address the broad
scope of the rule it prescribes.
People of course have a right to own firearms. And sometimes firearms are
discharged near others, whether deliberately or accidentally. If anybody in the vicinity of
gunfire can be stopped without reference to any of the Fourth Amendment’s protections, it
follows that those protections apply with lesser force to those who own guns and discharge
them—and to those who are merely within earshot. And unless we wish to limit such a
diminution in constitutional safeguards to those in high crime areas—which we cannot,
and should not, do—Judge Richardson’s reasoning would apply with equal force to both
economically disadvantaged public housing communities like Creighton Court and
wealthy suburban gated complexes.
Judge Richardson’s dissent thus contributes to a line of jurisprudence in this Circuit
that lessens constitutional protections for those who choose to own and carry inherently
48

 dangerous instrumentalities such as firearms. 3 See United States v. Mitchell, No. 18-4654,
2020 WL 3525542, at *5-6 (4th Cir. June 30, 2020) (finding reasonable suspicion for a
police stop of an individual near a bar fight who was described by a bystander and identified
as carrying a gun); United States v. Robinson, 846 F.3d 694, 700 (4th Cir. 2017) (en banc)
(holding that if during a lawful stop an officer reasonably suspects that a person is “armed
and therefore dangerous,” the officer may frisk the person); cf. Florida v. J.L., 529 U.S.
266, 272 (2000) (“Firearms are dangerous, and extraordinary dangers sometimes justify
unusual precautions.”). I express no opinion as to the correctness of this approach, other
than to note that it is a consequence of the position set forth in Judge Richardson’s dissent.

3

Indeed, at oral argument, the government’s counsel conceded that this case would
be “a very different scenario” if there were simply reports of a fistfight or a stabbing. Oral
Argument at 47:33-47:40, https://www.ca4.uscourts.gov/OAarchive/mp3/18-423320200130.mp3.
49

 DIAZ, Circuit Judge, with whom Judge HARRIS joins, concurring:
I agree with my colleagues in the majority that exigent circumstances didn’t justify
the government’s suspicionless stop of Curry, such that the district court properly granted
his motion to suppress. But I write separately to emphasize that the criteria under which
exigent circumstances may justify a suspicionless stop—that is, under which exigent
circumstances may operate as a “special need”—are those articulated by the Supreme
Court in City of Indianapolis v. Edmond, 531 U.S. 32 (2000). Because those criteria are
not met here, the exigent circumstances on which the government relies didn’t authorize
the measures that the officers took to investigate the origin of the gunfire.
The majority states that the government has “never invoked” the special needs
doctrine, and that the panel opinion relied on it of its own accord. Majority Op. 8–9. But
the government has always maintained (both in the district court and before us) that the
exigencies faced by the officers justified their suspicionless stop of the defendant. That
argument necessarily implicates Edmond’s doctrine of exigent circumstances as special
needs—which, in my view, provides much of the “guidance” that the majority finds
wanting in this area of the law. See Majority Op. 18.
As the majority correctly observes, exigent circumstances and special needs are, at
least typically speaking, “distinct” doctrines “animated by very different concerns and
applied in different contexts.” Majority Op. 13 n.4. Exigent circumstances constitute an
“exception” to the warrant requirement that “allows a warrantless search when an
emergency leaves police insufficient time to seek a warrant.” Birchfield v. North Dakota,
136 S. Ct. 2160, 2173 (2016). The special needs doctrine, by contrast, excuses the
50

 requirement of “individualized suspicion of wrongdoing” altogether, under circumstances
where “special needs, beyond the normal need for law enforcement,” render it
impracticable for police to observe any of the conventional protections of the Fourth
Amendment. See Edmond, 531 U.S. at 37 (emphasis added) (cleaned up).
Edmond has much to say about the resolution of this case. There, the Supreme Court
recognized that under certain criteria, a special need justifying suspicionless stops may in
effect be provided by exigent circumstances, despite their intrinsic relation to normal law
enforcement needs. Specifically, while the Court held that the City’s drug-interdiction
checkpoint was unreasonable because its primary purpose was “indistinguishable from the
general interest in crime control,” it articulated (in dictum) the following caveat:
Of course, there are circumstances that may justify a law enforcement
checkpoint where the primary purpose would otherwise, but for some
emergency, relate to ordinary crime control. For example, . . . the Fourth
Amendment would almost certainly permit an appropriately tailored
roadblock set up to thwart an imminent terrorist attack or to catch a
dangerous criminal who is likely to flee by way of a particular route. The
exigencies created by these scenarios are far removed from the circumstances
under which authorities might simply stop cars as a matter of course to see if
there just happens to be a felon leaving the jurisdiction.
Id. at 44.
This dictum acknowledges that police would be justified in detaining a potential
suspect as to whom they lacked individualized suspicion as part of an effort to apprehend
a dangerous criminal at large—provided, however, that they use an appropriately tailored
roadblock (or comparable discretionless and systematic means) to do so. Because that’s
what the officers sought to do here, I would measure the reasonableness of their conduct
by the “limiting principles” set forth in Edmond. See Majority Op. 18. And while the
51

 majority’s analysis largely does that in substance, I would emphasize a few different
points—especially the foundation on which Edmond’s dictum is premised, that (to pass
constitutional muster) the officers must employ a discretionless and systematic method of
conducting the suspicionless stops.
As the government points out, a handful of our sister circuits—including in two of
the cases discussed by the majority in Part III.B, see Majority Op. 19–23—have relied on
Edmond’s dictum to uphold stops under the precise criteria set forth therein. See United
States v. Arnold, 835 F.3d 833, 838–40 (8th Cir. 2016) (roadblock to catch armed bank
robber); United States v. Paetsch, 782 F.3d 1162, 1169–75 (10th Cir. 2015) (same); United
States v. Whitehead, 567 F. App’x 758, 767–68 (11th Cir. 2014) (per curiam) (perimeter to
catch armed bank robber); United States v. Rodger, 521 F. App’x 824, 828–29 (11th Cir.
2013) (per curiam) (roadblock to catch armed bank robber); Palacios v. Burge, 589 F.3d
556, 562–63 (2d Cir. 2009) (perimeter around nightclub to catch murder suspects); United
States v. Abbott, 265 F. App’x 307, 309 (5th Cir. 2008) (per curiam) (roadblock to catch
armed bank robber); United States v. Rogers, 244 F. App’x 541, 542–43 (5th Cir. 2007)
(per curiam) (same).
Similarly, this court—in the other case discussed by the majority in Part III.B, see
Majority Op. 19–23—has upheld a stop in a case presenting analogous circumstances. See
United States v. Harper, 617 F.2d 35, 40–41 (4th Cir. 1980) (roadblock to catch drug
smugglers heading inland from North Carolina coast).

And while Harper precedes

Edmond by several decades, our decision there may be viewed as the ancestor to this whole
line of authority. See Edmond v. Goldsmith, 183 F.3d 659, 662–63 (7th Cir. 1999) (relying
52

 on Harper for the proposition that police may “establish a roadblock on a route that they
know or strongly suspect is being used by a dangerous criminal to escape”), aff’d sub nom.
City of Indianapolis v. Edmond, 531 U.S. 32; cf. Edmond, 531 U.S. at 44 (“[A]s the Court
of Appeals noted . . . .” (citing Goldsmith, 183 F.3d at 662–63)).
The problem for the government is that this case simply doesn’t satisfy Edmond’s
criteria—especially, as noted, the need for a roadblock, perimeter, or other discretionless
and systematic method of carrying out the stops, which the officers here plainly didn’t use.
Instead, the officers (however well-intentioned) merely drove their vehicle to the nearest
of several grassy areas flanking Walcott Place, “fanned out,” and individually proceeded
to approach whomever there happened to draw their attention. J.A. 59. In doing so, the
officers failed not only to see whether others were lingering in the wider area where the
suspect might have been, but even to round up everyone in the field where they focused
their efforts, including those closest to Walcott Place.
Judge Richardson’s dissent asserts that the requirement of discretionless and
systematic conduct “ignores the very nature of an exigency, which calls for urgent action.”
Richardson Dissent 89. But to the contrary, the very purpose for which Edmond allows
police to use “an appropriately tailored roadblock” to apprehend a dangerous criminal in
the absence of individualized suspicion is to respond to an “emergency” or exigency. See
531 U.S. 44. Thus, it comes as no surprise that the cases I discuss above—including
Palacios, Paetsch, and Harper—involved both exigent circumstances and appropriately
tailored roadblocks or perimeters, most of which appear to have been set up “within

53

 minutes.” See Whitehead, 567 F. App’x at 767. * By the same token, Judge Richardson’s
dissent fails to identify a single case that has upheld a suspicionless stop based on exigent
circumstances without using such a means of cabining their exercise of discretion.
Judge Richardson’s dissent similarly errs in positing that Edmond’s permission to
conduct a roadblock in response to an emergency like the one at hand extends equally
(despite the dearth of case law) to “nonsystematic” seizures. See Richardson Dissent 93–
94 n.12. Rather, throughout the special needs doctrine, the Supreme Court has invariably
insisted that suspicionless intrusions be conducted in such a way as to “minimize the
discretion of the officers on the scene.” Michigan Dep’t of State Police v. Sitz, 496 U.S.
444, 452 (1990); see, e.g., Illinois v. Lidster, 540 U.S. 419, 427–28 (2004) (upholding
roadblock where “police stopped all vehicles systematically”); Skinner v. Ry. Labor
Executives’ Ass’n, 489 U.S. 602, 624 (1989) (stressing “the assurances of certainty and
regularity” afforded by programmatic drug tests); see also Turner v. Dammon, 848 F.2d
440, 446–47 (4th Cir. 1988) (observing that “[t]he cases upholding warrantless
administrative searches clearly establish” the requirements of “certainty, regularity, and
neutrality in the conduct of the searches”), abrogated on other grounds by Johnson v.
Jones, 515 U.S. 304 (1995).

*

The urgency with which the stops were undertaken in each of these cases belies
the suggestion in Judge Richardson’s dissent that the officers would have needed “time to
study the problem, await further information, and formulate a discretionless programmatic
response (perhaps after forming a committee)” to have complied with the requirement of
discretionless and systematic conduct. See Richardson Dissent 89. Rather, what the law
doesn’t allow (and what the dissents would sanction) is a roving band of officers, left to
their own devices, and untethered by any obligation to respect the Fourth Amendment.
54

 Indeed, that principle makes perfect sense in light of the impracticality, in the
context of suspicionless intrusions, of asking police to obtain a warrant or have any level
of particularized suspicion before interfering with individual liberty.

Because these

conventional Fourth Amendment protections are a poor fit in this context, the Court has
substituted the protection of “standardized” and “discretionless” police conduct in carrying
out the intrusions. After all, if police have no reason to suspect any particular individual,
there’s no reason for them to employ their discretion in selecting whom to investigate. And
as Edmond acknowledges, that’s equally so whether the stops are carried out as part of a
checkpoint or to catch a dangerous criminal in flight.
The consequences of failing to use a discretionless and systematic method to carry
out stops unsupported by any measure of individualized suspicion is laid bare in Delaware
v. Prouse, 440 U.S. 648 (1979), a case bearing all too much resemblance to the manner in
which the stops were carried out here. In Prouse, the Court held unreasonable a regime of
“roving-patrol stops” to check for motorists’ license and registration, finding that the
“sporadic and random” nature of the spot checks, which gave the officers “unbridled
discretion” to select whom to detain, interfered too severely with individual liberty. See
id. at 657–61; accord Almeida-Sanchez v. United States, 413 U.S. 266, 270–75 (1973)
(same with respect to roving border-patrol stops conducted with “unfettered discretion”).
And yet the Court has repeatedly acknowledged that suspicionless stops carried out
for the same governmental purpose would be reasonable under the special needs doctrine
if conducted using a discretionless and systematic roadblock. See, e.g., Edmond, 531 U.S.
at 37–38. This fundamental dichotomy is illustrative here: Whereas the officers might
55

 have been justified in using a roadblock or, better yet, a perimeter to help locate the suspect,
their exercise of discretion in deciding whom amongst the numerous individuals in the
vicinity to stop—despite lacking suspicion as to any of them—suffers from the same
fundamental defect as the roving-patrol stops in Prouse.
As the dissents would have it, the exigency present in this case would justify the
officers’ suspicionless stop despite the absence of any of these Fourth Amendment
protections. In their view, the sound of gunfire made it reasonable for the officers to forgo
not only the requirement of individualized suspicion that typically applies to intrusions
based on exigent circumstances, but also that of discretionless and systematic conduct that
invariably applies to intrusions based on special needs. Such an unsupported view of the
law, however, disregards the compromise that Edmond struck in giving officers more—yet
not unfettered—leeway to pursue certain kind of exigencies.
An additional problem for the government in this case, albeit one not expressly
accounted for in Edmond, is the difficulty the officers would have faced in identifying the
suspect even if they had used a roadblock, perimeter, or comparable method. In each of
the cases upholding suspicionless stops on the basis of exigent circumstances, the officers
possessed some reliable method of distinguishing the suspect from the innocent people
detained in the process. See Arnold, 835 F.3d at 836 (description of vehicle and license
plate); Paetsch, 782 F.3d at 1165–66 (electronic tracking device hidden in stacks of stolen
cash); Whitehead, 567 F. App’x at 762 (description of suspect); Rodger, 521 F. App’x at
829 (same); Palacios, 589 F.3d at 559 (same, plus eyewitness on hand); Abbott, 265 F.
App’x at 309 (another electronic tracking device); Rogers, 244 F. App’x at 542–43 (same);
56

 see also Harper, 617 F.2d at 37–40 (roadblock was set up along “the only paved road with
access” to the vessel suspected of being “equipped” with narcotics).
Here, by contrast, the officers had no description of the suspect, no information
about his precise whereabouts, and no apparent way to pick him out of the crowd. And
since it would likely have been difficult to determine whether any firearm (which, of
course, are generally lawful to possess) seized in the effort to identify the suspect was the
source of the gunfire, it may well be that any attempt to do so on these peculiar facts, no
matter how discretionless and systematic, would have been futile and, thus, unreasonable.
See Sitz, 496 U.S. at 453–54 (discussing the need for “effective” intrusions).
These shortcomings, and especially the officers’ failure to cabin their exercise of
discretion in conducting the stops, may explain why the government hasn’t fully relied on
Edmond. But in my view, the government’s effort to invoke a patchwork of additional
arguments is unavailing because Edmond already teaches how police may carry out a
regime of suspicionless stops in an effort to catch a dangerous criminal. And because the
officers in this case didn’t abide by Edmond’s teaching, the district court properly granted
Curry’s motion to suppress.
*

*

*

With these observations, I join my colleagues in the majority.

57

 THACKER, Circuit Judge, with whom Judge KEENAN joins, concurring:
I concur in all respects with the majority opinion. I am compelled to write this
separate concurring opinion only in response to the dissenting opinion authored by Judge
Wilkinson, who writes today with a smooth pen and a tin ear. Judge Wilkinson’s dissenting
opinion accuses the majority members of our court of all but dismantling the rule of law
and of “overstep[ping our] proper role.” Wilkinson Dissenting Op., post at 63. I cannot
sit silent in the face of Judge Wilkinson’s dissent. In my view, the use of predictive
policing, which Judge Wilkinson endorses, is little more than racial profiling writ large.
Judge Wilkinson’s dissent hails predictive policing as an “innovation” in policing
and notes, “As relevant here, Richmond has relied in part on ‘hot spot policing’ -- the idea
that, by identifying specific areas where crimes are most likely to occur, and dispatching
police officers accordingly, police can reduce the level of crime that ultimately takes
place.” Wilkinson Dissenting Op., post at 67, 65.

But predictive policing is not the

panacea Judge Wilkinson professes -- far from it. Although of relatively recent vintage,
the “innovation” of preventive policing, which uses computer algorithms to predict high
crime areas, is no longer the shiny new object it may once have appeared to be, but instead
has revealed itself to be tarnished with racial bias. Predictive policing is merely a covert
effort to attempt to justify racial profiling.
Over time, predictive policing has been shown to be, at best, of questionable
effectiveness, and at worst, deeply flawed and infused with racial bias. Indeed, the Los
Angeles Police Department, an early adopter of the predictive policing experiment,
recently put an end to its predictive policing program. See The Sentencing Project, “Race
58

 and Justice News: Los Angeles Discontinues A Predictive Policing Program.”
https://www.sentencingproject.org/news/race-justice-news-los-angeles-discontinuespredictive-policing-program/ (May 11, 2020) (available as PDF attachment). While the
Los Angeles police chief opined that the program was discontinued due to financial
constraints, the program had been widely criticized as biased, inasmuch as it resulted in
heavier policing of communities of color. See id. at 1. Moreover, an inspector general
charged with studying the efficacy of the Los Angeles predictive policing program was
unable to determine whether the program was actually effective in reducing crime. See id.
Predictive policing, in particular “hot spot policing,” has been controversial since
its inception. On August 31, 2016, a coalition of 17 civil rights, racial injustice, and
technology organizations issued a statement of concern drawing attention to the racial bias
inherent in such programs, citing “deep flaws that lead to injustice, particularly for people
of color.” American Civil Liberties Union, “Statement of Concern about Predictive
Policing by ACLU and 16 Civil Rights Privacy, Racial Justice, and Technology
Organizations,” https://www.aclu.org/other/statement-concern-about-predictive-policingaclu-and-16-civil-rights-privacy-racial-justice#:~:text=Predictive%20Policing%20
Today%3A%20A%20Shared%20Statement%20of%20Civil%20Rights%20Concerns&te
xt=A%20growing%20number%20of%20police,or%20who%20will%20be%20involved

59

 (Aug. 31, 2016) (available as PDF attachment) (hereinafter, “Statement of Concern”). * Of
particular note to the case at hand, the Statement of Concern highlights:
Predictive policing systems threaten to undermine the
constitutional rights of individuals. The Fourth Amendment
forbids police from stopping someone without reasonable
suspicion -- a specific, individualized determination that is
more than just a hunch. Computer-driven hunches are no
exception to this rule, and a computer’s judgment is never a
further reason (beyond the articulable facts that intelligibly
caused that judgment) for a stop, search, or arrest. Similarly,
predictive policing must not be allowed to erode rights of due
process and equal protection. Systems that manufacture
unexplained “threat” assessments have no valid place in
constitutional policing.
Technology cannot override human flaws. It stands to reason that any computer
program or algorithm is only as good as the data that goes into it. In the computer science
arena, this is known as “GIGO” (garbage in, garbage out); flawed data input produces
nonsense output. In predictive policing, GIGO is a real concern because “hot spot
policing,” which utilizes historical crime data to predict future crime hot spots, can be
infected with years of racial bias. Because “historic crime data is biased through the
practice of racialized enforcement of law, predictive policing will inherently reinforce and
perpetrate this structural racism.” Stop LAPD Spying Coalition, “Predictive Policing:

*

The 17 organizations joining in the issuance of the Statement of Concern are: The
Leadership Conference on Civil and Human Rights, 18 Million Rising, American Civil
Liberties Union, Brennan Center for Justice, Center for Democracy & Technology, Center
for Media Justice, Color of Change, Data & Society Research Institute, Demand Progress,
Electronic Frontier Foundation, Free Press, Media Mobilizing Project, National
Association for the Advancement of Colored People, National Hispanic Media Coalition,
Open MIC (Open Media and Information Companies Initiative), Open Technology
Institute at New America, and Public Knowledge.
60

 Profit

Driven

Racist

Policing”

at

2,

https://stoplapdspying.org/wp-

content/uploads/2016/12/Statement-of-Concern-on-Upturn-Predictive-Policing-ReportDecember-2016.pdf (Dec. 7, 2016) (available as PDF attachment); see also Renata M.
O’Donnell, Challenging Racist Predictive Policing Algorithms Under the Equal Protection
Clause, 94 N.Y.U. L. Rev. 544 (2019).
In at least acknowledging that predictive policing is not free of problems, my
colleague’s dissent cites the “danger of overreaction” by law enforcement. Wilkinson
Dissenting Op., post at 68. In this regard, the Statement of Concern points out that
experience has demonstrated,
[P]olice misconduct follows consistent patterns, and . . .
offering further training and support to officers who are at risk
can help to avert problems. Police should be at least as eager
to pilot new, data-driven approaches in the search for
misconduct [within their own ranks] as they are in the search
for crime, particularly given that interventions designed to
reduce the chances of [police] misconduct do not themselves
pose risk to life and limb.
Statement of Concern at 4.
I note that Judge Wilkinson and I agree on one point. “If change is to occur, part of
the obligation must lie with police themselves, and the essential efforts they must daily
make to earn the trust of their communities and prove themselves responsible stewards of
their power.” Wilkinson Dissenting Op., post at 70–71 (citing Barry Friedman & Maria
Ponomarenko, Democratic Policing, 90 N.Y.U. L. Rev. 1827, 1881 (2015) (noting link
between effective policing and perceived legitimacy)). Thus, to achieve this result, we
must never lose sight that it is individual police officers, not a computer program, who
61

 abuse their authority by violating the constitutional rights of citizens such as Billy Curry,
based on the simple fact that they committed the offense of “walking while black.” No
fact in this case ever suggested Billy Curry’s involvement in the commission of the crime
under investigation. And the majority’s enforcement of his constitutional rights is the
shield that will protect him and others from the bleak future imagined by Judge Wilkinson.

62

 WILKINSON, Circuit Judge, dissenting:
We face again in this day of sad and unhappy truths the divide between what are
already two Americas. In one America, where citizens possess the means to hire private
security or move to safer neighborhoods, the impact of judicial barriers to effective law
enforcement may be minimal. In another America, though, people have no choice but to
endure the unintended consequences of our missteps, as crime moves to fill the vacuum
left by the progressive disablement of the law’s protections.

These repercussions,

moreover, serve as reminders of the fact that only a segment of this country, the least
fortunate among us, ends up shouldering the weightier burden when our branch of
government oversteps its proper role. *
The deaths of George Floyd, Eric Garner, and far too many others have been
heartbreaking. They are crimes not only against law but against humanity. There exist
grave concerns about unfair police treatment of minorities, and I respect the majority for
being sensitive to that. There is also the risk, for the reasons so ably described in the

*

I wish to thank the Chief Judge for his concurring opinion in this case. I wish also
to take the occasion to reiterate what I have said often, namely that he has provided real
leadership to our fine court in his capacity as Chief.
I welcome, as always, his perspective and believe it provides a good impetus for the
conversation that jurists and others need to have on the significant issues of equal justice
under law that come before the courts. I regret our disagreement on this issue, but I do not
regret at all the opportunity to share with my esteemed colleague our respective perceptions
and views.
The lead concurring opinion also laments the long line of those who have “felt
obliged to bear their burden to save minority or disadvantaged communities from
themselves.” See Gregory Concurring Op., ante at 33. This opinion is not tendered in that
spirit. For as divided as we now are, one hopes that we shall not always remain so, and
that the bonds of brotherhood may yet come to transcend the boundaries that the shameful
aspects of our past have bequeathed us as to race.
63

 principal dissent by Judge Richardson, that impenetrable and impractical legal regimes will
drive police officers from disadvantaged communities—a situation that comes with dire
consequences, not only for the wellbeing of our most vulnerable fellow citizens, but for
society as a whole.
We are in danger of making law enforcement in our dispossessed communities a
thankless task. Being an officer in a high-crime area often carries its own distinctive risks.
In such areas, the dangers to police officers exist alongside the dangers to residents. See
United States v. Travis Ball, No. 3:19-cr-128 (E.D. Va. 2019) (indictment in connection
with fatal shooting of officer blocks from Creighton Court). And if the police are rightly
condemned for their abuses, they are often wrongly scapegoated for that whole collection
of conditions in deprived communities that even our best efforts and intentions scarcely
seem to dent. Finally, if courts, as here, overturn perfectly professional efforts at law
enforcement, and steadily immobilize officers from reasonably protecting themselves and
the people they serve, then we risk making the notion of high-crime areas an even more
self-fulfilling designation.
Of course, the police do act beyond the strictures of the law, and courts exist to
check them. Here, however, the law enforcement officers deserve appreciation, not a
rebuke. What would be inappropriate in ordinary circumstances can become appropriate
when a serious emergency arises. Courts have the luxury of reflecting at leisure upon what
these officers had to assess in an instant. Recall that the police arrived on the scene of a
live shooting at Creighton Court within thirty-five seconds of the first shots being fired.
Their intrusions were limited to a stop and weapons frisk. In dangerous circumstances,
64

 this can be policework at its best. And it happened because Richmond’s Police Department
decided to take innovative steps to modernize its policing tactics—steps that I worry will
be made all-but-impossible under the unduly rigid interpretation of the Fourth Amendment
advanced by the majority today.
Officers responded so quickly to the Creighton Court shooting because Richmond,
like other municipalities throughout the country, has adopted certain “predictive policing”
strategies to make its streets safer. See Jennifer Bachner, Predictive Policing: Preventing
Crime with Data and Analytics, IBM Center for the Business of Government 29-30 (2013)
(describing Richmond Police Department use of predictive policing); see also Chandler
Harris, Richmond, Virginia, Police Department Helps Lower Crime Rates with Crime
Prediction Software, Government Technology (Dec. 21, 2008) (noting 21% reduction in
major crimes from 2005 to 2006). While predictive policing programs are subject to
modification and differ in their details, they share a common premise: through smart
policies, law enforcement can affirmatively prevent crime from happening, rather than just
solve it.
As relevant here, Richmond has relied in part on “hot spot policing”—the idea that,
by identifying specific areas where crimes are most likely to occur, and dispatching police
officers accordingly, police can reduce the level of crime that ultimately takes place.
Creighton Court was such an area. A recent murder, repeated shootings, and regular reports
of violent crime understandably led the police department to assign a prevention task force
there. Richmond, as noted, is not alone. See Steve Lohr, The Age of Big Data, N.Y. Times,
Feb. 11, 2012 (“Police departments across the country, led by New York’s, use
65

 computerized mapping and analysis of variables like historical arrest patterns, paydays,
sporting events, rainfall and holidays to try to predict likely crime ‘hot spots’ and deploy
officers there in advance.”). With the advent of big data and machine learning, moreover,
hot spot policing has allowed departments to evolve well beyond the days of red pushpins
on a corkboard, and has empowered officers to identify likely areas of crime with blockby-block precision. See Andrew Guthrie Ferguson, The Rise of Big Data Policing 3 (2017)
(“Police can identify the street corner most likely to see the next car theft or the people
most likely to be shot.”).
With emergencies such as a potential active shooter, predictive policing rests on a
tradeoff: as the facts of this case make plain, better-positioned officers will arrive on the
scene faster, but they might have less information than if responding to second-hand
reports. For many communities, putting a premium on response time has worked. David
Weisburd, Does Hot Spot Policing Inevitably Lead to Unfair and Abusive Police Practices,
or Can We Maximize Both Fairness and Effectiveness in the New Proactive Policing?,
2016 U. Chi. Legal F. 661, 670 (2016) (noting “twenty of twenty-five tests of hot spot[]
policing show positive statistically significant findings” of crime reduction). To that end,
hot spot policing—and predictive policing writ large—has proven itself a boon to
communities that have otherwise suffered under outdated modes of policing that were
unable to stop the worst criminal misconduct. See Philip B. Heymann, The New Policing,
28 Fordham Urb. L.J. 407, 417 (2000) (“What does make a difference, careful evaluations
show, is focusing patrol resources on places and times that have the most crime.”).

66

 It is hard to see how this innovation can continue under the majority’s conception
of the Fourth Amendment. Indeed, the sole practical takeaway from the majority opinion
is that police officers on the scene of an unfolding emergency must sit and wait for
identifying information, rather than use discretion and judgment to get control of a possibly
deadly event, lest the prevention of a homicide violate the Constitution. This injunction
entirely saps predictive policing of its potency, and effectively forecloses the tradeoff—
faster responses for fuller information—that innumerable cities have opted for in making
their streets safer. This is a mistake. While courts have been scrupulous to guard against
specific police excesses, we have not until now invited the evident programmatic
consequences that are sure to follow from this case, depriving cities of the flexibility to
choose for themselves what sort of policing works best for their citizens.
My colleague Judge Thacker raises the thoughtful concern that “predictive policing
. . . is little more than racial profiling writ large.” Thacker Concurring Op., ante at 58. But
the answer is not to approve or condemn predictive policing wholesale but to look to its
implementation in a given community to discern whether it is infected with odious racial
assumptions that society cannot and should not tolerate.
I thank Judge Wynn as well for his concurring opinion. He is surely right that the
phrase “high-crime area” cannot serve as a facile excuse for indiscriminate interventions.
But neither can skepticism toward the preventive potential of predictive policing in violent
crime locales allow us to deny its benefits ab initio to communities that might welcome
them. Judge Wynn lauds the capacity of representative bodies, and I wholeheartedly agree.
But in referencing the use of statistics in an inapposite line of cases, my good colleague is
67

 erecting a constitutional bar to those same representative bodies that have sought to utilize
a particular tool of policing as one “reasonable” option under the Fourth Amendment in
resolving the serious problems before them.
Is predictive policing the answer? Is it presumptively too injurious to individual
rights in affected areas? Richmond has not been spared the mix of peaceful protests and
violent outbreaks evidenced throughout the nation, and I hardly know what might begin to
heal this divide. I know only that communities deserve the chance to give predictive
policing a try. The Fourth Amendment—and the “reasonableness” standard at the heart of
it—is hardly inconsistent with our federalism.

States and localities, many of them

majority-minority, can choose within “reason” what law enforcement strategies work for
them. Their latitude is not infinite, but latitude there is. I do not for a moment contend that
hot spot policing is free of problems, among them the danger of overreaction, or that courts
can dispense with their customary and cautionary roles. But by stripping departments of
effective public safety programs—and then adding to the inherent dangers and difficulties
of the streets a judicial rebuke for even the most professional and minimally-intrusive
policework—courts risk inducing police officers to simply abandon inner cities as part of
their mission. See Ronald T. Hosko, Through Police Eyes—the Ferguson Effect Scare, 23
Berkeley J. Crim. L. 9, 19-24 (2018) (collecting sources and describing “depolicing”
phenomenon across departments); see also Stephen Rushin & Griffin Edwards, DePolicing, 102 Cornell L. Rev. 721, 736-38, 758-67 (2017).
The abandonment will not come with some trumpeted proclamation. No, it will be
quiet, undertaken not with a shout but with a shrug. It will be a story of silent neglect, and
68

 a tacit repositioning of resources to areas that pose fewer hazards and fewer prospects of
judicial sanction. The effects will be, in a word, tragic. See Richard A. Oppel Jr., West
Baltimore’s Police Presence Drops, and Murders Soar, N.Y. Times, June 12, 2015.
Couple an area’s rise in crime with a lack of respect shown by courts for even good
policework, and you have an America where gated communities will be safe enough and
dispossessed communities will be left to fend increasingly for themselves.
This quiet neglect and subtle abandonment by law risks hastening the exodus of
small businesses, many minority-owned, which rely upon law enforcement for protection.
It will lead to the emboldening of gangs and drug rings who will feel increasingly free to
work their less restricted will. What is more, the onus of their new free-rein will fall most
heavily on those upon whom criminal syndicates are most apt to prey: youngsters who are
led away from education and recruited as runners for the enticement of easy money, and
who in the course of it all, become claimed by an addiction that can lead to a lifelong
chemical dependency. And, tragically enough, if those at the beginning of life see their
futures compromised, those in life’s later stages who lack the strength and resources to
resist what criminal actors demand of them, will likewise be disproportionately harmed.
Crime is of course a threat to anyone at any time, but perhaps its most poignant visitations
are on the young and old.
It is true that the police have made more than their share of mistakes and that the
sad legacy of racism and mistrust hovers today over police-citizen interactions. See James
Baldwin, A Report from Occupied Territory, The Nation, July 11, 1966. It is true that
brutal incidents continue to inflict upon our fellow citizens immeasurable hurt and pain.
69

 But it is also true that stereotyping, so long an instrument of racial injustice, can be unfair
to countless individual officers as well. And it is further true that, for the future, law
enforcement for all its faults will remain one of the few protections and last lifelines a
fragile community retains against physical harm and mental despair. See, e.g., Brigham
City v. Stuart, 547 U.S. 398, 406 (2006). The question thus becomes: Is it possible to
envision the day in which police are not perceived in poorer neighborhoods as enemies but
as partners on the pathways to a better life, one where the innate gifts that lie within each
of us are free to flower most fully for the benefit of all? Crime stands as a serious obstacle
to such a life and a police force, properly reformed and trained, can help to remove it.
Professor Rod Brunson of Northeastern University has noted that “a great deal of
scholarship has demonstrated that under-policing also leaves residents feeling perpetually
underserved and unsafe.” Rod K. Brunson, If the Police Weren’t There, Wash. Post, June
14, 2020, at B1. It is not only “brutality or racism that erodes faith in law enforcement in
these [disadvantaged] neighborhoods—it’s ineffectiveness, too.”

Id. at B2.

Often,

Brunson states, residents have “seen police fail to serve them in their own neighborhoods.”
Ibid.
My friends in the majority, I fear, now deepen this perception and entrench this
reality. For top applicants will hesitate to enter a profession accorded limited judicial
understanding and little societal respect, a fact that will again diminish protection for those
who stand in sore need of it.
So much must be done. If change is to occur, part of the obligation must lie with
police themselves, and the essential efforts they must daily make to earn the trust of their
70

 communities and prove themselves responsible stewards of their power. See Brunson,
supra; Barry Friedman & Maria Ponomarenko, Democratic Policing, 90 N.Y.U. L. Rev.
1827, 1881 (2015) (noting link between effective policing and perceived legitimacy). But
another part lies with us, the courts, to wield our own authority responsibly, and not to pull
up short the sort of reasonable police conduct that took place here; namely, a timely and
measured response to homicidal dangers and lethal threats.
The majority has delivered a gut-punch to predictive policing. As the facts here so
dramatically show, the effect of its ruling is not to disarm the criminal, but to disable the
officer. The majority proceeds under the illusion that law enforcement officers will always
be there, just raring to charge in. But that is not true, and it is not what happened here. Yet
bound to this false premise, the majority fails to glimpse the reality that continued reversals
of this kind will lead to the absence of officers in those very areas where, for good and
humane reasons, their presence is needed most. In that light, while this case may now be
over for our court, the consequences have only just begun for the other America that must
suffer them.

71

 RICHARDSON, Circuit Judge, with whom Judges WILKINSON, NIEMEYER, AGEE,
QUATTLEBAUM, and RUSHING join, dissenting:
On a summer night in Richmond, Virginia, several gunshots rang out in a residential
neighborhood. That neighborhood, unfortunately, was no stranger to gun violence, with
six shootings and two homicides over the prior three months. Upon hearing the clatter of
gunfire, four uniformed, patrolling police officers froze, turned toward its source, and sped
to the scene. They arrived 35 seconds later, finding half-a-dozen men spread out near the
site of the shots and catching a glimpse of a potential victim. The officers fanned out,
instructing the men to stop, to raise their hands, and then to lift their shirts to expose their
waistbands. Flicking on flashlights, the officers looked on from a short distance, checking
for weapons. Only one man, Billy Curry, refused the officers’ instructions. That refusal
led to an attempted pat down—and the discovery of Curry’s silver revolver.
After the government indicted Curry for possessing a firearm as a felon, he moved
to suppress the silver revolver. The district court granted Curry’s motion, reasoning that
the officers’ seizure and flashlight search violated the Fourth Amendment’s prohibition on
unreasonable searches and seizures. In our initial decision, a panel of the Fourth Circuit
disagreed. So we remanded the case for the district court to consider whether the officers
had sufficient reason to pat down Curry when he refused their orders. But today, our en
banc Court affirms, concluding that the initial stop of the men at the scene of gunfire
violated the Fourth Amendment.
I respectfully dissent. The touchstone of the Fourth Amendment is reasonableness,
and the officers’ response to this evolving exigency was eminently reasonable. Within 35
72

 seconds of the gunshots, law enforcement ran into a potential active-shooter scenario. The
sight of a potential victim and multiple 911 calls reporting “shots fired” reinforced the
threat to the community. Such exigent circumstances required the officers to act, ‘now or
never,’ to protect the public. And they did so with appropriate restraint—ordering Curry
and the others to stop and show their hands.
Yet the majority holds that the officers were constitutionally prohibited from
instructing the men at the suspected scene of residential gunfire to show that they were not
holding a weapon. I fear the majority’s sweeping decision—outlawing a modest response
to a serious threat—guts the exigent-circumstances doctrine and handcuffs law
enforcement’s response to possible active-shooter situations.

As today’s result

resoundingly conflicts with the Supreme Court’s Fourth Amendment jurisprudence, I
dissent.
I.

Background
A.

The night of September 8, 2017

The residents of Creighton Court, a public-housing complex in Richmond, have
been beset by repeated violence. The latest killing took place only eleven days before the
events leading to this case. And over the previous three months, the community suffered
six shootings and yet another life taken. As a result, the Richmond Police Department
assigned its “Focus Mission Team”—a division dedicated to suppressing violent crime—
to Creighton Court. At around 9:00 p.m. on September 8, 2017, four officers from that
division were patrolling the community in a marked car when gunshots split the night.

73

 Upon hearing the gunfire, two of the officers activated their body cameras, which
provide an objective record of what happened. The government exhibit below shows the
officers’ initial location marked with an “A” at the bottom of the image. J.A. 124.

Before hearing the shots, the officers were heading south. But the gunfire came from the
direction of Walcott Place—a short residential street culminating in the cul-de-sac in the
northeast (top-right) corner of the map. So the officers quickly flipped a U-turn and drove
northeast across the dimly lit field toward Walcott Place. The arrow marks their path on
the map. The district court found that the officers “travelled two to three blocks, taking
only thirty-five seconds to arrive behind Walcott Place” at the location marked “B.” J.A.
256–57. Before coming to a stop, the officers observed a man in the darkened field wearing
a red shirt who “appeared to be maybe favoring one of his arms,” and they were “uncertain

74

 if he had been shot.” Id. at 257. As the officers stepped from the car, their radios
announced at least two 911 calls about the gunfire on Walcott Place.
Upon arriving at location “B,” the officers spotted several individuals, including
Curry, walking away from Walcott Place where the gunshots originated. Activating their
flashlights, the officers “fanned out and began approaching different individuals.” J.A.
258. In doing so, the officers stopped each of the men that they encountered in the field
walking away from the scene, including Curry.
The officers met Curry at the location marked “C.” And the district court found that
one officer addressed Curry and another man “in an authoritative tone, [stating,] ‘Let me
see your hands. Both of you. Let me see your hands.’” J.A. 258. Both men stopped and
began to raise their hands.

At this point, according to the district court, a Fourth

Amendment seizure occurred. Although the other individuals complied with the officers’
later directives to show their waistbands, Curry did not. When officers then sought to pat
Curry down, a brief scuffle ensued, and they discovered his silver revolver.
B.

The district court proceedings

Curry was indicted for possessing a firearm as a felon in violation of 18 U.S.C.
§ 922(g)(1). Moving to suppress the revolver, Curry asserted that the officers’ actions
violated the Fourth Amendment. In response, the government made two independent
arguments. First, the government claimed that the officers conducted a permissible stop
under Terry v. Ohio, 392 U.S. 1 (1968). Second, the government argued that the stop was
reasonable because of the exigent circumstances at hand.

75

 Following an evidentiary hearing, the district court rejected both arguments. As to
the first, the district court applied Terry and determined that the officers lacked the
particularized suspicion needed to justify an investigatory stop. As to the second, the court
found the situation’s “exigencies” and “safety concerns” did not excuse the supposed
prerequisite of individualized reasonable suspicion. J.A. 281 (“Despite Officer Gaines’s
legitimate concern for his own safety and the safety of his partners, the exigencies in this
situation cannot undermine the Fourth Circuit’s clear holding that ‘the Constitution
requires a particularized and objective basis for suspecting the particular person stopped
of criminal activity.’”) (citations omitted); J.A. 276 (“Even in these circumstances, with
the attendant emergency and safety concerns, the Fourth Amendment allowed the officers
only to initiate a consensual encounter, not the Terry stop they undertook.”). The district
court thus held that the initial stop was unlawful and granted Curry’s motion to suppress.
The government timely appealed. In our original panel decision, we reversed the
district court’s conclusion that the brief stop and flashlight search violated the Fourth
Amendment. So we remanded the case for the district court to consider whether the officers
could lawfully pat down Curry after he disregarded their instructions. But today, our en
banc Court affirms on the broad rationale that the initial seizure violated the Fourth
Amendment.

76

 II.

Discussion
This case concerns the reasonableness of the officers’ initial seizure—when they

asked the men to show their hands—in light of the exigent circumstances presented. 1 The
parties do not contest the district court’s factual findings—normally reviewed for clear
error. United States v. Kehoe, 893 F.3d 232, 237 (4th Cir. 2018). So accepting those
findings, we consider the district court’s Fourth Amendment analysis de novo. Id. 2
A.

The exigent-circumstances doctrine

The Fourth Amendment provides:
The right of the people to be secure in their persons, houses, papers, and
effects, against unreasonable searches and seizures, shall not be violated, and
no warrants shall issue, but upon probable cause, supported by oath or
affirmation, and particularly describing the place to be searched, and the
persons or things to be seized.
U.S. CONST. amend. IV. Its “basic purpose,” according to the Supreme Court, “is to
safeguard the privacy and security of individuals against arbitrary invasions by government
officials,” Carpenter v. United States, 138 S. Ct. 2206, 2213 (2018) (quoting Camara v.

1

The government justifies the stop of Curry solely based on the exigentcircumstances doctrine. And while it does not rely on the special-needs doctrine, the
government argues that special-needs cases may inform our assessment of what is
“reasonable” in a given exigency. See Oral Arg. 3:20−4:13, 32:05–34:25; City of
Indianapolis v. Edmond, 531 U.S. 32, 44 (2000).
2

We review for clear error the district court’s factual findings that the officers were
responding to an emergency and that legitimate safety concerns existed when they
encountered the men leaving the shooting. United States v. Moses, 540 F.3d 263, 270 (4th
Cir. 2008). But whether those legitimate concerns constituted exigent circumstances that
permitted the stop calls for a legal conclusion, which we review de novo. United States v.
Singleton, 441 F.3d 290, 293 (4th Cir. 2006).
77

 Municipal Court, 387 U.S. 523, 528 (1967)), and “to place obstacles in the way of too
permeating police surveillance,” id. at 2214 (quoting United States v. Di Re, 332 U.S. 581,
595 (1948)).
As the first clause makes plain, the “touchstone” of any Fourth Amendment analysis
is “the reasonableness in all circumstances of the particular governmental invasion of the
citizen’s personal security.” Pennsylvania v. Mimms, 434 U.S. 106, 108–09 (1977)
(internal citations omitted); see also Kansas v. Glover, 140 S. Ct. 1183, 1191 (2020). 3 This
touchstone of reasonableness requires asking if the government’s interest in undertaking a
search or seizure “outweigh[s] the degree to which the search [or seizure] invades an
individual’s legitimate expectations of privacy.” Maryland v. King, 569 U.S. 435, 461
(2013); see also Camara, 387 U.S. at 537 (evaluating a warrantless governmental intrusion
for reasonableness requires “balancing the need to search against the invasion which the
search entails”). And this balance “depends on the context within which a search [or
seizure] takes place.” New Jersey v. T.L.O., 469 U.S. 325, 337 (1985).

3

The Fourth Amendment’s Warrant Clause follows the reasonableness requirement.
That clause provides that warrants may only be issued on probable clause. See U.S. CONST.
amend. IV. But the Warrant Clause does not make a warrant a prerequisite to a search or
seizure. See California v. Acevedo, 500 U.S. 565, 581–85 (1991) (Scalia, J., concurring).
No warrant is required for certain arrests, United States v. Watson, 423 U.S. 411, 424
(1976), vehicle searches, United States v. Ross, 456 U.S. 798, 809 (1982), or personal
searches in public spaces, United States v. Arvizu, 534 U.S. 266, 277 (2002). Even the
warrantless search of a home may be justified by governmental interests: e.g., protective
sweeps, Maryland v. Buie, 494 U.S. 325, 335 (1990), and the supervision of probationers,
Griffin v. Wisconsin, 483 U.S. 868, 873 (1987). Here, Curry does not argue that his seizure
required a warrant.
78

 Recognizing that emergencies permit what normal times forbid, courts have long
held that the Fourth Amendment allows limited searches or seizures during exigent
circumstances.

“Exigencies” exist when an officer would reasonably suspect the

conditions create a need to act “now or never” to protect an important public interest—like
a threat to public safety. Roaden v. Kentucky, 413 U.S. 496, 505 (1973); Figg v. Schroeder,
312 F.3d 625, 639 (4th Cir. 2002); see also Minnesota v. Olson, 495 U.S. 91, 100 (1990)
(noting that exigencies include “prevent[ing] a suspect’s escape” or minimizing “the risk
of danger to the police or to other persons inside or outside the dwelling”).
In evaluating now-or-never responses, we have recognized that what is reasonable
for addressing public exigencies (ranging from confronting active shooters to measuring a
driver’s blood-alcohol concentration) differs from what is reasonable during routine lawenforcement operations (such as investigating drug dealing or detecting fraud). See, e.g.,
Kentucky v. King, 563 U.S. 452, 460 (2011); cf. Jacobson v. Massachusetts, 197 U.S. 11,
29 (1905) (confirming that emergencies permit broader measures to protect the public
welfare). “Reasonableness” under the Fourth Amendment is context specific: attuned to
the imminence, probability, and gravity of a threat or exigency, as well as the intrusiveness
and effectiveness of the chosen government response. See Illinois v. McArthur, 531 U.S.
326, 330–33 (2001); see also Brigham City v. Stuart, 547 U.S. 398, 405–07 (2006);
Schmerber v. California, 384 U.S. 757, 768–72 (1966); cf. Brown v. Texas, 443 U.S. 47,
50–51 (1979). Consistent with what “reasonableness” requires, the circumstances that may
create an exigency are broad, but the intrusion on personal liberty is limited—in time and
scope—by the exigency at hand. See McArthur, 531 U.S. at 331, 336; Mincey v. Arizona,
79

 437 U.S. 385, 393 (1978).

Anything more tips this balance toward the arbitrary

government action that the Fourth Amendment abhors.
The Supreme Court applies a two-step inquiry that threads the needle between
arbitrary government action and reasonable responses to exigencies. First, the Court asks
whether officers could reasonably suspect a legitimate exigency. See Missouri v. McNeely,
569 U.S. 141, 149 (2013); Brigham City, 547 U.S. at 403–06. If so, it then inquires whether
officers responded to the exigency in a reasonable manner: a context-specific inquiry
considering the gravity of the exigency, the tailoring of the officers’ response, and the
extent to which that response interferes with individual liberty. See McNeely, 569 U.S. at
149–51; Schmerber, 384 U.S. at 768–72.
For instance, the Supreme Court employed this approach in Illinois v. McArthur,
531 U.S. 326. There, the Court assessed whether exigent circumstances permitted the
seizure of a homeowner outside his home to prevent him from destroying evidence of
marijuana possession before a search warrant could be obtained. The Court noted that the
case “involve[d] a plausible claim of specially pressing or urgent law enforcement need,
i.e., ‘exigent circumstances.’” Id. at 331. Accordingly, “rather than employing a per se
rule of unreasonableness,” the Court proceeded to “balance the privacy-related and lawenforcement concerns to determine if the intrusion was reasonable.” Id.
As part of that balancing, the Court considered the gravity of the public concern,
noting that the officers had “good reason to fear that, unless restrained, [the homeowner]
would destroy the drugs.” Id. at 332. Assessing the tailoring of the government action, the
Court found that the officers “made reasonable efforts to reconcile their law enforcement
80

 needs with the demands of personal privacy” by using the “significantly less restrictive
restraint” of preventing the homeowner from entering his home unsupervised. Id. Turning
to the severity of the intrusion, the Court emphasized both that the officers “imposed the
restraint for a limited period of time,” id., and that “[t]emporarily keeping a person from
entering his home . . . is considerably less intrusive than police entry into the home itself,”
id. at 336. Balancing these considerations, the Court held that the seizure was permissible.
Id. at 333; see also Brigham City, 547 U.S. at 405–07 (evaluating the exigency and manner
of intrusion to find the officers’ actions were reasonable). 4
And so, taking the exigent-circumstances doctrine as it has been handed to us, this
case requires us to address two questions to determine whether the brief stop of Curry was
legal. First, could the officers reasonably suspect that they faced exigent circumstances?
Second, did the officers respond to those exigent circumstances reasonably? Because the
answer to both questions is ‘yes,’ the initial stop of Curry was reasonable and thus lawful.

4

The majority’s response is to suggest that while McArthur might say that it is an
exigent-circumstances case, it did not really mean to be an exigent-circumstances case.
But McArthur explicitly framed its analysis as “involv[ing] a plausible claim of specially
pressing or urgent law enforcement need, i.e., ‘exigent circumstances.’” 531 U.S. at 331
(emphasis added); see also id. at 336 (“[T]he need to preserve evidence . . . was sufficiently
urgent or pressing to justify the restriction.”) (emphasis added).
The majority also protests that McArthur was limited to seizures of things rather
than persons. But as the Supreme Court explains, the “restriction at issue” was
“[t]emporarily keeping a person from entering his home, a consequence whenever police
stop a person on the street.” Id. at 336. The majority’s mistaken belief manifests from its
more general error in cabining the totality-of-the-circumstances test to property but
applying a categorical test for persons. The Supreme Court’s totality-of-the-circumstances
framework for exigent circumstances applies with full force to searches and seizures of
persons and things alike. See, e.g., Brigham City, 547 U.S. at 403–07 (entry into home);
McNeely, 569 U.S. at 148–51 (blood draw from individual); Schmerber, 384 U.S. at
768−72 (same).
81

 B.

Officers reasonably faced exigent circumstances

We first ask whether the officers could reasonably suspect an exigency. See
McNeely, 569 U.S. at 149; McArthur, 531 U.S. at 331. In so asking, we assess the facts
and uncertainties as they unfolded at the time—rather than in hindsight from the peace of
our carefully guarded chambers. Figg, 312 F.3d at 639 (“‘[C]ourts should not engage in
‘unreasonable second-guessing’ of the officers’ assessment of the circumstances that they
faced.”) (quoting United States v. Montoya de Hernandez, 473 U.S. 531, 542 (1985)); see
also Ryburn v. Huff, 565 U.S. 469, 477 (2012).
As the Supreme Court has explained, exigent circumstances “include the need to
pursue a fleeing suspect, protect individuals who are threatened with imminent harm, or
prevent the imminent destruction of evidence.” Carpenter, 138 S. Ct. at 2223. 5 “In
addition to these well-established exigencies, the Supreme Court and this Circuit have held
that more general ‘emergencies,’ if enveloped by a sufficient level of urgency, may also
constitute an exigency.” United States v. Yengel, 711 F.3d 392, 397 (4th Cir. 2013); see
also, e.g., United States v. Taylor, 624 F.3d 626, 632 (4th Cir. 2010) (abandoned child on
the street created an exigent circumstance). One type of particularly compelling exigency
is the need to protect against threats to life and limb—such as the dangers posed by “bomb
threats, active shootings, and child abductions.” Carpenter, 138 S. Ct. at 2223; see also

5

Although exigencies “include” these scenarios, they are not so limited. See also,
e.g., Michigan v. Tyler, 436 U.S. 499, 509 (1978) (fighting a fire and investigating its
cause); Mincey, 437 U.S. at 394 (preliminary sweeps of murder scenes). Contra Majority
Op. 14–15.
82

 Brigham City, 547 U.S. at 403–04; Hunsberger v. Wood, 570 F.3d 546, 555 (4th Cir. 2009)
(missing child). 6
Here, this threshold inquiry is not—or at least should not be—particularly difficult.
As the district court found, all four officers heard “five or six gunshots” at night in a
residential neighborhood where “hundreds of families live.” J.A. 256. The officers knew
“there had been a rash of shootings and homicides” in that community, with six shootings
and two homicides in the last three months. Id. The officers arrived at the suspected scene
of the shooting, stopping in a poorly lit field, in about 35 seconds. As they stepped out of
the car, they “receiv[ed] information from dispatch that at least two calls had come in for

6

The majority asserts that, “standing alone, even a ‘possible homicide’ does not
present an ‘emergency situation’ demanding ‘immediate [warrantless] action.’” Majority
Op. 16 (quoting Mincey, 437 U.S. at 392). This characterization tests the line between
imprecision and error.
In Mincey, a narcotics bust evolved into an exchange of gunfire—leading to the
death of an officer and the wounding of others. Id. at 387–88. Law enforcement secured
the scene and emergency assistance was rendered. Id. at 388. But, without securing a
warrant, homicide detectives then launched a four-day search of the premises. Id. at 388–
89. The Supreme Court considered whether the search was constitutionally permissible.
Id. at 388. In assessing the exigency, the Court first explained that “when the police come
upon the scene of a homicide they may make a prompt warrantless search of the area to see
if there are other victims or if a killer is still on the premises.” Id. at 392 (citing Tyler, 436
U.S. at 509–10). This initial intrusion, the Supreme Court continues, was justified by “[t]he
need to protect or preserve life or avoid serious injury.” Id. (citation omitted). The Court
then reasoned that, once the scene had been secured and aid rendered, the exigency had
dissipated. Id. at 393. As a result, the “four-day search that included opening dresser
drawers and ripping up carpets can hardly be rationalized in terms of the legitimate
concerns that justify an emergency search,” and it violated the Fourth Amendment. Id.
So Mincey teaches that a response to exigent circumstances must end roughly when
the exigency does. Id.; see also Flippo v. West Virginia, 528 U.S. 11, 14 (1999). To say
that Mincey shows that even a “possible homicide” does not generally support an
immediate search or seizure leaves out important details, to say the least. See Mincey, 437
U.S. at 392 (Police “may make a prompt warrantless search” of the scene.).
83

 random gunfire, one of which was on Walcott Place,” and a potential victim approached
the rear of their vehicle. J.A. 257. The officers observed “approximately five to eight
men” in the area, but they did not yet have a suspect description. Id.
Given the shots fired and the officers’ prompt arrival at the reasonably perceived
locus of the threat, I have little trouble agreeing with the district court that the officers had
a “legitimate concern” for their own safety, J.A. 281, were confronted with “[e]xigencies
in this situation,” id., and “rightly sought to protect the community and themselves in the
event that one of these men had been the shooter,” J.A. 276. 7 The officers rushed to
respond to shots fired just seconds earlier in a densely populated residential neighborhood.
See J.A. 256–58. They were presented with the prospect that the shooter might continue
to threaten the safety of the public (not to mention the officers themselves), as well as the
prospect of retaliation from other shooters (a distinct possibility given the recent gun
violence in the community). See Brigham City, 547 U.S. at 405–06 (finding that exigent
circumstances arose from the threat of “ongoing violence” after a juvenile hit an adult).
Thus, the objective circumstances highlighted the need to protect the public and officers
from these dangers—placing this case well within the exigent-circumstances doctrine.
C.

The officers’ response to the exigency was reasonable

After identifying a legitimate exigency, we assess the response to that exigency for
reasonableness, considering (1) the gravity of the public concerns served by the seizure,
(2) the tailoring of the seizure to the exigency, and (3) the severity of the interference with

7

Recall that we review these factual findings only for clear error. See supra n.2.
84

 the individual’s liberty. See McArthur, 531 U.S. at 330–33; McNeely, 569 U.S. at 141,
149–51, 164; Schmerber, 384 U.S. at 768–72. In my view, the officers’ actions were
eminently reasonable. First, the exigent circumstances implicated compelling public
interests.

Second, the seizure—asking the men in the field to show their hands—

appropriately advanced those interests by permitting officers to determine whether anyone
was holding a gun. And third, this intrusion on Curry’s liberty was minimal in both scope
and duration.
1.

The strength of the public interest

We first assess the gravity of the public interest. Or, as the Supreme Court has
described, where on the “exigency spectrum” does this case sit? Mitchell v. Wisconsin,
139 S. Ct. 2525, 2533 (2019) (plurality). On one end of the spectrum, we have the
minimum urgency of preventing the destruction of evidence inherent in all drunk-driving
cases as the liver metabolizes alcohol. See id. And on the other end, representing the
highest order of exigencies, we have mass shootings or terrorist attacks.
Even on the minimum urgency end of that spectrum, the Supreme Court has
repeatedly found that the public interest permits, in my view, strikingly invasive seizures
and searches. Faced with suspected drunk drivers, the Court has found that the evidentiary
value of a blood draw (even when a roadside breath test had already been administered)
provides sufficient justification for the government to penetrate a stopped driver with a
needle. See id. at 2532–34; Schmerber, 384 U.S. at 768–72; see also Brigham City, 547
U.S. at 403–05 (warrantless entry into a home was warranted where the fracas had, at that
point, caused little more than a bloody lip).
85

 Yet here, the officers faced an exigency of far more substantial weight. Accounting
for the gravity, imminence, and likelihood of the threat, the public’s safety provided a
powerful impetus for the seizure.
The primary and chief concern present that evening was public safety, which we
have recognized as “the first job of any government.” Mora v. City of Gaithersburg, 519
F.3d 216, 223 (4th Cir. 2008). 8 As discussed above, the officers heard a half-dozen
gunshots in a neighborhood. J.A. 256. 9 So a strong government interest called for the first
responders to prevent any other shootings (either as retaliation or a continuation of the
opening salvo) and potentially treat wounded citizens on the scene. See Brigham City, 547
U.S. at 403–04, 406; see also Georgia v. Randolph, 547 U.S. 103, 118 (2006) (noting that
officers with reason to believe a threat of violence existed could enter a home “to determine
whether violence (or threat of violence) has just occurred or is about to (or soon will)
occur”); Florida v. J.L., 529 U.S. 266, 272 (2000) (recognizing “the serious threat that
armed criminals pose to public safety”).
The second concern present was that of police-officer safety, which the Supreme
Court has called “both legitimate and weighty.” Mimms, 434 U.S. at 110; see also Riley v.

8

Cf. John Locke, Second Treatise of Government §§ 128–31 (1690) (describing the
social contract in which man leaves the state of nature for the peace, safety, and security of
government); Robert Nozick, Anarchy, State, and Utopia 10–146 (1974) (discussing and
justifying the minimal function of the “night-watchman” state).
9

One additional contextual factor reinforces the gravity, imminence, and likelihood
of the threats that the officers confronted: The officers were patrolling this neighborhood
in response to concerns about recent gun violence, including six recent shootings and two
homicides. That alone would not justify a suspicionless stop. But the recent string of gun
violence is part of the context to be considered.
86

 California, 573 U.S. 373, 388 (2014); Warden v. Hayden, 387 U.S. 294, 298–99 (1978).
We have found this interest to justify the suspicionless seizure and detention of several
individuals based on the reasonable belief that “police officers engaged in lawful
investigatory functions would be endangered” without the seizure. Figg, 312 F.3d at 639;
see also Ryburn, 565 U.S. at 477; Olson, 495 U.S. at 100. But where gunfire has already
erupted, as it had here, the concern that an officer might be harmed while responding to the
scene is particularly acute. See J.A. 281 (district court finding that Officer Gaines had a
“legitimate concern for his own safety and the safety of his partners”).
Given the gravity, imminence, and likelihood of the threat facing police officers that
night, compelling public concerns called for action to protect the community and officers
from the threat of an unfolding and inherently uncertain situation.
2.

The tailoring

Next we turn to how the seizure advanced the public concerns of citizen and police
safety. See McNeely, 569 U.S. at 151–52; Brigham City, 547 U.S. at 406–07; cf. Brown,
443 U.S. at 50–51. Here the officers reasonably tailored their actions to the exigent
circumstances. In rapidly responding to multiple gunshots and corresponding 911 calls,
the officers ordered the first men they saw at the perceived locus of the shooting to stop
and show their hands.

Doing so was a minimally intrusive manner of effectively

establishing that none were holding a gun. See McArthur, 531 U.S. at 332.
Even so, the majority argues that the officers’ response failed to advance the public
interest because they had no way to know whether a shooter was travelling in that direction.
See Majority Op. 19–23. According to the majority, the lack of certainty distinguishes this
87

 stop from the seizure found lawful by the Second Circuit in Palacios v. Burge, 589 F.3d
556 (2d Cir. 2009). In that case, police officers sealed the exits to a club to apprehend a
group of men who ran inside after a stabbing. The district court drew the same distinction,
observing that, in Palacios, “the officers could be nearly certain that the individual or
individuals responsible for the stabbing were inside the club,” J.A. 279, whereas in Curry’s
case, the shooter could have escaped by fleeing to the north, and “several people were
standing near the apartment buildings.” J.A. 257.
But the Fourth Amendment requires reasonableness—not perfection. See Heien v.
North Carolina, 574 U.S. 54, 60 (2014) (“To be reasonable is not to be perfect . . .”).
Certainty about who will pass along a particular physical path is not required before the
public interest permits a stop along that route. See Hill v. California, 401 U.S. 797, 804
(1971) (“sufficient probability, not certainty, is the touchstone of reasonableness under the
Fourth Amendment”). For example, in United States v. Harper, this Court relied on the
“exigency of fleeing, perhaps dangerous, suspects” to approve vehicle stops without
individualized suspicion “of all persons found on a likely access route” three miles away
from a scheduled drug-delivery at a dock raided hours earlier. 617 F.2d 35, 40–41 (4th
Cir. 1980) (emphasis added); see also id. at 41 (noting that the route was “reasonably
expected to be used for their escape”); cf. United States v. Martinez-Fuerte, 428 U.S. 543,
554 (1976) (finding that a checkpoint identifying illegal aliens in 0.12% of vehicles to be

88

 effective). 10 Given that exigency, we found the vehicle checkpoint a reasonable—but far
from perfectly effective—response.
So instead of certainty, we require only reasonable tailoring based on the specific
exigency. Indeed, more severe and widespread exigencies naturally permit a different sort
of response than comparatively mild and localized ones. Compare Edmond, 531 U.S. at
44 (noting that roadway seizures could be justified by the important governmental interests
presented by “an imminent terrorist attack” or “a dangerous criminal who is likely to flee
by way of a particular route”), with Mincey, 437 U.S. at 393 (holding that, once the crime
scene had been secured, a four-day warrantless search exceeded the scope of the exigency).
By demanding perfection, the majority asks too much.
In contrast, the concurrence says that the officers could not act without “using a
roadblock, or, better yet, a perimeter.” Diaz Concurring Op. 56. But this requirement
ignores the very nature of an exigency, which calls for urgent action. For the same reason
warrants are not required in exigent circumstances, see id. at 50, neither is a well-planned
or systematic response. The threat presented here did not afford officers time to study the
problem, await further information, and formulate a discretionless programmatic response
(perhaps after forming a committee).
Instead, this potential active-shooter situation required the officers to act now—
using what knowledge, time, and resources they had—to do what they could to protect the

10

These men were the first the officers encountered when they arrived. So it cannot
be said that this brief seizure was impermissibly selective. Of course, were such improper
targeting presented here, our analysis might be different.
89

 Creighton Court community (and themselves). Given these constraints, the officers’
chosen steps were not just a reasonable approach, but seemingly the best available. It
simply “does not meet the needs of law enforcement or the demands of public safety to
require officers to walk away from a situation like the one they encountered here.”
Michigan v. Fisher, 558 U.S. 45, 49 (2009); see also United States v. Perkins, 363 F.3d
317, 328 (4th Cir. 2004) (“We cannot afford to read the Fourth Amendment to require
officers to wait . . . , and perhaps until innocent bystanders are physically harmed, before
taking reasonable, preventive measures.”).
3.

The severity of the interference with liberty

The third and final factor concerns the severity of the interference with Curry’s
individual liberty.

See McNeely, 569 U.S. at 151; McArthur, 531 U.S. at 330–33;

Schmerber, 384 U.S. at 771–72; cf. Brown, 443 U.S. at 50–51. The intrusion here was
minimal in both time and scope.
In terms of time, the seizure lasted less than a minute for the men who complied,
which was “no longer than reasonably necessary for the police, acting with diligence,” to
determine whether the men were holding a gun. McArthur, 531 U.S. at 332. Far from an
unreasonably prolonged detention, see Montoya de Hernandez, 473 U.S. at 542−44, the
seizure here—briefly ordering the men to show their hands—was temporally modest in
both absolute and relative terms. See McArthur, 531 U.S. at 332 (upholding two-hour
seizure of homeowner outside his home based on need to prevent homeowner from
destroying evidence of marijuana while officers obtained warrant).

90

 And in terms of scope, the seizure that the majority finds unreasonable required
Curry to simply show his hands. That is all. It is hard to imagine a less intrusive
government response in these circumstances. Although a seizure, the simple act of
showing one’s hands is a common and noninvasive feature of everyday life—whether a
friendly wave or a brisk handshake. Ordering the men to show their hands here appears
particularly noninvasive when compared with the ever-more invasive searches and seizures
approved by the Court, including a needle in the arm, a search of the home, and
confinement until a bowel movement could potentially reveal illegal drugs. See, e.g.,
McNeely, 569 U.S. at 148–51; Brigham City, 547 U.S. at 403–07; Montoya de Hernandez,
473 U.S. at 541, 544. And the comparative invasiveness of the order that the men show
their hands is particularly reasonable in light of the nature and gravity of a potential activeshooter scenario. Short of asking nicely, there is perhaps no less invasive way to assure
that none of the men whom the officers approached posed an armed and dangerous threat.
Thus, the initial seizure was only minimally invasive.
4.

Balancing the circumstances

Taking these three factors together, the initial seizure was a reasonable response to
the exigent circumstances that the officers confronted that night.
The Supreme Court has approved far more invasive measures in far less exigent
circumstances: From seizing a homeowner outside his home for two hours to prevent the
destruction of marijuana, McArthur, 531 U.S. at 332, to often allowing blood draws on
suspected drunk drivers to obtain the best evidence for prosecution, Mitchell, 139 S. Ct. at
2539; McNeely, 569 U.S. at 148–52, 164–65; Schmerber, 384 U.S. at 768–72. And with
91

 even limited violence afoot, the Supreme Court has permitted entry into the sanctity of the
home to prevent a potential unarmed assault. Brigham City, 547 U.S. at 403–05. In
comparison to these precedents, the threat here was more dire and the seizure less invasive,
making it a reasonable response to an evolving threat. 11
D.

The majority’s categorical limitations on the exigent-circumstances
doctrine are deeply flawed

My colleagues in the majority do not agree that the initial stop—never mind the
flashlight search—was a reasonable response to the exigency confronting the officers. But
the majority is mistaken. First, the majority errs in claiming that the officers engaged in a
“suspicionless, investigatory seizure,” Majority Op. 18 (emphasis added), even though the
seizure here is independently justified by the need to protect. Second, the majority creates
a new constitutional rule that the exigent-circumstances doctrine is categorically
inapplicable to a suspicionless seizure of a person unless the government is responding to
a “known crime” with a search tailored to either a “controlled geographic area” or a
“discrete group of people.” Majority Op. 3, 19. In doing so, the majority adopts a rule in
deep tension with the Supreme Court’s exigent-circumstances precedents.

And the

straightjacket imposed by the majority’s categorical approach will now prevent the
government from responding to exigent circumstances like those confronting Creighton

11

The majority relies on Terry to argue that, even were it inclined to apply a
balancing test (an approach it rejects), it would find Terry to be “persuasive on the question
of balancing” because, although “the Supreme Court did not apply the exigent
circumstances exception in Terry, its ruling was premised on the same general” need to
discover weapons. Majority Op. 27. But, as I explain below, Terry does not govern how
officers may act in this context.
92

 Court with even the minimal seizures employed—with no regard for the gravity and nature
of the public interests at stake.
1.

This is not a Terry stop case

I find it telling that the majority begins and ends its analysis with Terry v. Ohio.
Majority Op. 11, 31–32. In Terry, the Supreme Court confronted a different, “narrow
question”—“whether it is always unreasonable for a policeman to seize a person and
subject him to a limited search for weapons unless there is probable cause for arrest.”
Majority Op. 11 (quoting Terry, 292 U.S. at 15). The narrow answer is “no.” But since
then, Terry has come to stand for an affirmative proposition: An officer may conduct an
investigatory stop based on a reasonable suspicion of criminal activity. Terry, 392 U.S. at
20–31; United States v. Holmes, 376 F.3d 270, 275 (4th Cir. 2004); see also Majority Op.
11. These, we call “Terry stops,” and they are a “specific[]” and “well-delineated”
exception to the warrant requirement. Majority Op. 14 (quoting Mincey, 437 U.S. at 390).
And Terry stops generally require particularized suspicion of criminal activity.
This is all true, and it is also irrelevant. The government argues that the stop was
permissible under the exigent-circumstances doctrine—to protect the public and the
officers. It does not justify the encounter as a Terry investigatory stop. On the contrary, it
disavows any Terry argument.

Even so, the majority attempts to shoehorn the

government’s exigency arguments into the Terry framework. 12 Indeed, on my count, the

12

The concurrence makes a similar error when it applies the principles governing
programmatic checkpoints to this exigency. Edmond observed that “the Fourth
Amendment would almost certainly permit an appropriately tailored roadblock set up to
93

 majority analogizes to Terry more than 30 times in just over 20 pages of analysis. But let
me be clear: This is not a Terry case. And the limitations on Terry investigative seizures
should not be grafted onto this exigency-justified public-safety seizure. 13
2.

The majority’s rule

By adopting an improper framework, the majority ultimately reaches a categorical
rule in deep tension with the Supreme Court’s case-by-case balancing. See McArthur, 531
U.S. at 331 (“[R]ather than employing a per se rule of unreasonableness, we balance the
privacy-related and law enforcement-related concerns to determine if the intrusion was

thwart an imminent terrorist attack or to catch a dangerous criminal who is likely to flee by
way of a particular route.” 531 U.S. at 44. But the concurrence errs when it takes this
permissive statement—in a case about police checkpoints—to, by negative implication,
ban nonsystematic, suspicionless seizures in emergency situations.
To establish systematized searches as a limiting principle, the concurrence instead
turns to Delaware v. Prouse, 440 U.S. 648 (1979), which concerned a standing system of
“roving-patrol stops,” id. at 656, as aptly described by the concurrence. But these stops,
of course, were conducted with no exigency. Indeed, conducting arbitrary stops—not
spurred by any pressing emergency (or suspected wrongdoing)—was the very purpose of
the program in Prouse. Id. at 657. As a result, the limits established in Prouse have little
bearing on what is reasonable in response to an evolving exigency requiring immediate
action.
So I agree with the concurrence that the government’s response is part of the totality
of the circumstances, and it must be limited to the exigency. But I cannot accept the
suggestion that, on the facts presented here, the Fourth Amendment requires a roadblock
or nothing at all.
13

The majority suggests that Terry should drive our analysis because, whatever the
public safety interests at stake, they were intertwined with the government’s investigatory
interests. See Majority Op. 30 n.10. But an overlapping investigatory justification does
nothing to eliminate the public-safety exigency. See Brigham City, 547 U.S. at 405
(finding that the public-safety exigency justified the search regardless of “whether the
officers entered the kitchen to arrest respondents and gather evidence against them or to
assist the injured and prevent further violence”).

94

 reasonable.”); see also Brigham City, 547 U.S. at 405–07; McNeely, 569 U.S. at 148–52,
164−65; Taylor, 624 F.3d at 631. 14 I have grave doubts that the majority’s one-size-fitsall approach is either permissible or wise to govern the many unpredictable and fluid
exigent circumstances that confront law enforcement. Even so, the majority’s specific
categorical requirements conflict with the Supreme Court’s exigent-circumstances
doctrine.
a.

“Known crime”

Start with the majority’s “known crime” requirement. Contrary to my colleagues’
suggestion, the exigent-circumstances doctrine does not hinge on whether the
circumstances result from criminal activity but on the nature of the threat (criminal or not)
to broader public interests. 15 Indeed, the Supreme Court and our Circuit have applied the

14

To be sure, the Supreme Court’s instruction that we apply a “totality-of-thecircumstances” approach does not always foreclose “general rules” (often subject to casespecific exceptions). See, e.g., Mitchell, 139 S. Ct. at 2535 n.3; see also id. at 2537−38
(discussing the exceptions to the ‘general rule’ adopted). The plurality in Mitchell
acknowledged that “our exigent-circumstances precedent requires a ‘totality of the
circumstances’ analysis, [but] the circumstances in drunk driving cases are often typical,
and the Court should be able to offer guidance on how police should handle cases like the
one before us.” Id. at 2535 n.3 (cleaned up). So in certain classes of cases, the relevant
factors are so well defined that balancing them will point toward a particular outcome often
enough to support a general rule. But the flip side of this discussion in Mitchell is obvious:
Where the factual inputs are varied, it is not possible (nor prudent) to identify a general
rule, even subject to case-specific exceptions. The evolving exigency that police faced
here is not suited to such a rule.
15

Nor does the Supreme Court require a “known” exigency but merely an
objectively reasonable basis for perceiving an exigency. See, e.g., Brigham City, 547 U.S.
at 406; Fisher, 558 U.S. at 49. And in our Circuit, we have explained that reasonable
suspicion of an exigency is all that is required. See, e.g., Figg, 312 F.3d at 639.
95

 exigent-circumstances doctrine where a search or seizure is not focused on crime control—
i.e., burning buildings, injured civilians, and wandering children (to name a few). See, e.g.,
Brigham City, 547 U.S. at 403–04; Fisher, 558 U.S. at 47–48; Tyler, 436 U.S. at 509–10;
Taylor, 624 F.3d at 631–33.
In Brigham, for example, the Supreme Court found that exigencies permitted
officers to enter a home when they witnessed little more than a fracas leading to a bloody
lip. This entrance was justified, as the Supreme Court explained, by the need “to render
emergency assistance to an injured occupant or to protect an occupant from imminent
injury,” not by crime control. Brigham, 547 U.S. at 403; see also Fisher, 558 U.S. at 47–
48.
Similarly, in Taylor, our Circuit upheld a warrantless entry into a home based on
exigent circumstances outside the criminal context. There, officers encountered a fouryear-old girl on an “unsupervised odyssey” on a busy street. 624 F.3d at 632. We

The majority appears to require that the officers needed to await further assurance
that they faced an ‘actual’ active-shooter threat—beyond the corroborated staccato of
gunfire in a city neighborhood. But that requirement is hard to square with the Court’s
reminder that “[t]he role of a police officer includes preventing violence and restoring
order, not simply rendering first aid to casualties; an officer is not like a boxing (or hockey)
referee, poised to stop a bout only if it becomes too one-sided.” Brigham City, 547 U.S. at
406; Fisher, 558 U.S. at 49 (instructing that officers “do not need ironclad proof of a likely
serious, life-threatening injury to invoke the emergency aid exception”) (cleaned up); see
also Figg, 312 F.3d at 639 (reminding us that we have no business in “‘unreasonable
second-guessing’ of the officers’ assessment of the circumstances they faced”) (quoting
Montoya de Hernandez, 473 U.S. at 542).
To be sure, the degree of confidence that an exigency exists may be relevant to the
balance of reasonableness. See Schmerber, 384 U.S. at 770. But actual knowledge of a
threat is not required, for the police need not “know the unknowable on pain of
suppression.” Montanez v. Carvajal, 889 F.3d 1202, 1210 (11th Cir. 2018).
96

 explained that “a child of such tender age wandering alone outside the home raised the real
possibility that her caretaker was unconscious or otherwise in need of assistance,” thus
justifying the search of the home. Id.; see also United States v. Evans, 958 F.3d 1102,
1106–07 (11th Cir. 2020) (concluding that “it was not objectively unreasonable for an
officer to mistake the sounds of a dog whimpering—through the wall of a home—for a
person in distress,” and that, despite the uncertainties presented, those sounds constituted
an exigency justifying a warrantless entry into the home).
As these cases highlight, searches and seizures under the exigent-circumstances
doctrine do not require a “known crime.”
b.

“Controlled area” or “discrete group”

Next take the majority’s “controlled geographic area” or “discrete group of people”
requirement. The exigent-circumstances doctrine does not impose a one-size-fits-all
tailoring requirement in terms of time, space, or people. Instead, the acceptable scope of a
search or seizure depends on the gravity, imminence, and likelihood of the threat. See
McNeely, 569 U.S. at 148–52, 164–65; McArthur, 531 U.S. at 330–33; Welsh v. Wisconsin,
466 U.S. 740, 753 (1984); Mora, 519 F.3d at 224. We have never before followed the
majority’s lead in making the perfect the enemy of the good. As discussed above, in
Harper we found it sufficient that police set up a checkpoint to catch a fleeing suspect “on
a likely access route.” 617 F.2d at 41 (emphasis added). The government must do what it
can to respond to an exigency, even if its actions cannot guarantee success.
The majority’s rationale for requiring a “controlled geographic area” or a “discrete
group of people” appears to rest on two faulty assumptions. First, the majority suggests
97

 that the Supreme Court’s exigency cases are limited to providing guidance for searches of
the home rather than of individuals. Majority Op. 17–18; see also id. at 22 n.8. But the
same totality-of-the-circumstances balancing applies whether the government action
involves a warrantless entry into a home, a blood draw of a person, or the seizure of a
homeowner—even though different exigences present different needs and different
searches implicate different concerns. See, e.g., Brigham City, 547 U.S. at 403–07,
McNeely, 569 U.S. at 148–51; McArthur, 531 U.S. at 330–33. 16
And second, having purportedly distinguished the Supreme Court’s exigency cases,
the majority invents these categorical requirements by turning to Terry and a variety of
circuit-court cases. Majority Op. at 18–21. I am unpersuaded. For the reasons described
above, Terry fits poorly in this context. And as for the circuit-court cases, not only do they
fail to support the majority’s conclusion, but they rely on the case-specific balancing that
the majority rejects today. See, e.g., Palacios, 589 F.3d at 563–66; United States v.
Paetsch, 782 F.3d 1162, 1170–75 (10th Cir. 2015).
What caselaw teaches, common sense affirms. Say law enforcement learns of a
shooting in one of several buildings in a complex. Under the majority’s rule, the officers
would be constitutionally prohibited from stopping and demanding raised hands from
fleeing individuals just because the police have doubts about who to search (so no “discrete
group”) and have no ability to cordon off all modes of egress (so no “controlled area”).

16

For what it’s worth, even if there were different tests for different exigencies, the
test for the less offensive, brief seizure of an individual would be more permissive than that
for the more significant intrusion into the home. See, e.g., McArthur, 531 U.S. at 331.
98

 Wouldn’t it be reasonable for officers to do what they can to respond as the situation
evolves?
What if gunshots erupt during a crowded marathon? Today’s opinion will prevent
officers from simply instructing individuals to raise their hands. Unless, of course, they
can pinpoint a discrete group or exert control over the entire area.
Or take the apparent sounds of shots being fired at a music festival with thousands
in attendance, abundant modes of egress, and only so many officers operating with only so
much time. That scenario presents neither a “controlled geographic area” nor a “discrete
group of people.” Must the officers sit on their hands until enough backup arrives to cover
all the exits and establish a secure perimeter?
*

*

*

Today’s decision is a mistake. It will now be harder than ever to safeguard our
communities from the most serious of threats. Wherever our citizens and officers find
themselves—music festivals, houses of worship, or main streets—they will now be less
safe. So too for Creighton Court. That result, though tragic, would be tolerable were it
required by the Fourth Amendment (or even permitted by the Supreme Court). But where
the Fourth Amendment requires no such thing, I cannot bring myself to follow the
majority’s lead in arrogating to ourselves such power over the safety of our fellow citizens.
Barring even the most modest of steps to protect from the most serious of threats is neither
constitutional nor compassionate. I dissent.

99